PETITIONER'S W1 ~BITNO - IURC Online Services Portal...W1 ~BITNO . .,/_~ d/1'.J-17'-~-R-Ef&-oR-~-R SOLAR BUSINESS MANAGER, BURNS & MCDONNELL ON COMPETITIVENESS OF PRICE OF SOLAR PROJECT

FILED September 18, 2018

INDIANA UTILITY REGULATORY COMMISSION

OFFICIAL EXHIBITS Petitioner's Exhibit No. 6 Vectren South

Page 1 of 12

SOUTHERN INDIANA GAS AND ELECTRIC COMPANY

d/b/a VECTREN ENERGY DELIVERY OF INDIANA, INC.

(VECTREN SOUTH)

IURC CAUSE NO. 45086

REBUTTAL TESTIMONY

OF

MATTHEW R. BRINKMAN, P.E.

IURC PETITIONER'S

~BITNO . .,/_~ W1 d/1'.J-17'-~-R-Ef&-oR-~-R

SOLAR BUSINESS MANAGER, BURNS & MCDONNELL

ON

COMPETITIVENESS OF PRICE OF SOLAR PROJECT

IN COMPARISON TO CURRENT AND POTENTIAL FUTURE PRICES

SPONSORING PETITIONER'S EXHIBIT NO. 6

ATTACHMENTS MRB-R1 THROUGH MRB-R3

1 I.

2 Q.

3 A.

4

5

6 Q.

7 A.

Petitioner's Exhibit No. 6 Vectren South

Page 2of12

REBUTTAL TESTIMONY OF MATTHEW R. BRINKMAN, P.E.

INTRODUCTION

Please state your name and business address.

My name is Matthew R. Brinkman, P.E. My business address is 1850 North Central

Avenue, Phoenix, Arizona 85004.

By whom are you employed and in what capacity?

I am employed as the Solar Business Unit Manager for Burns & McDonnell's Energy

8 Group.

9

10 Q.

11

12 A.

13

14 Q.

15 A.

Are you the same Matthew R. Brinkman that previously testified in this

proceeding?

Yes, I am.

What is the purpose of your rebuttal testimony?

The purpose of my rebuttal testimony is to respond to the direct testimony of Indiana

16 Office of Utility Consumer Counselor ("OUCC") witness John E. Haselden and Alliance

17 Coal LLC ("Alliance Coal") witness Charles S. Griffey. In particular, I respond to certain

18 misconceptions created in their respective testimony that the price of constructing the

19 solar energy project {the "Solar Projecf') totaling approximately 50 megawatts of

20 alternating current ("MWac") within Vectren South's service territory is too high. I also

21 describe why it is not effective strategy to wait and hope for a decline in prices in the

22 face of the current opportunity. Finally, I address Mr. Griffey's misperception that

23 Indiana is not a good location for a solar facility.

24

25

26 II. THE COST OF THE SOLAR PROJECT IS COMPARABLE TO OR LOWER THAN

27 OTHER UTILITY-SCALE SOLAR PROJECTS

28 29 Q. Do you agree with testimony offered by the OUCC and Alliance Coal that the Solar

30 Project is too costly?

31 A. No. As I indicated in my direct testimony, the capital cost of the Solar Project proposed

32 by Vectren South is comparable in cost to similarly sized, utility-grade projects installed

1

2

3

4

5

6

7

8

9

10

11

12 Q.

13

14

15 A.

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32 Q.

33


Page 3of12

in the general geographic region. The capital cost of the Solar Project was

benchmarked against the Solar Benchmark Study for the first quarter of 2017 prepared

by the National Renewable Energy Laboratory ("NREL"). The capital costs of the

proposed Vectren South Solar Project were lower than the benchmark for utility-scale,

single-axis tracking projects using union labor in Indiana.

Furthermore, as I will explain, the Northern Indiana Public Service Company ("NIPSCO")

bid event cited by OUCC witness Haselden provides additional corroborating evidence

that the capital cost of the proposed Vectren South Solar Project is comparable to

competitively bid asset purchase projects.

In your opinion, is the potential Hoosier Energy 20-year purchase power

agreement ("PPA") with EDP Renewables relating to the Riverstart Solar Farm 200

MW project comparable to the Vectren South Solar Project?

No. Mr. Haselden is comparing a PPA for a project three to four times the size of the

proposed Vectren South Solar Project. A PPA is not comparable to an asset purchase

under which the utility will own the generation assets for thirty to fifty years. In essence,

Mr. Haselden is making an apples-to-oranges comparison. Moreover, unlike the certain

costs of the Vectren South Solar Project, PPAs typically have price escalation

provisions. Applying an assumed 3% escalation factor on the Hoosier Energy PPA, for

example, the average price of the PPA over a 20-year term would be approximately

$0.054/kWh.

Furthermore, Mr. Haselden provided no details regarding the costs and scope of the

potential Hoosier Energy/EDP Renewables PPA and as Mr. Games notes, he apparently

has not seen the PPA. The cost may not account for: cost of substation and

interconnection, congestion charges, etc. Therefore, a direct comparison of the first year

Hoosier Energy PPA cost, without considering all these other potential cost additions, to

the Vectren South levelized cost of energy from the Solar Project is inaccurate and

misleading.

Are the 16 Indiana solar projects NIPSCO summarized in its IRP Public Advisory

Slide comparable to the Solar Project?

1 A.

2

3

4

5

6

7

8

9

10

11

12

13 Q.

14

15

16

17 A.

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33


Page 4 of 12

No. Mr. Haselden again is comparing a PPA (or in this case, multiple PPAs) to an asset

purchase by Vectren South, which is not a fair or accurate comparison. The same

NIPSCO IRP Public Advisory slide Mr. Haselden references indicates an average cost

for nine bids on five solar asset sale or options as having an average bid price of

$1, 151.01/kW (DC assumed). It is not clear whether the substation and interconnection

are included in the NIPSCO capital cost. Vectren South's asset purchase price is

$1,015.00/kW DC, which includes costs for a new substation and transmission

interconnection. Accordingly, using the data included in the NIPSCO IRP Public

Advisory Slide on which Mr. Haselden relies, the proposed Vectren South Solar Project

actually has a cost that is 12% less than the average NIPSCO bids for an asset sale or

option to purchase solar assets.

Mr. Haselden suggests Vectren South could acquire the land leases from Orion

and conduct an RFP for the Engineering, Procurement and Construction ("EPC")

contract on the site. In your opinion, would that have resulted in a significant

decrease in the cost of the project?

No. As a concession to being sole-sourced as the EPC contractor for the proposed

Vectren South project, First Solar conceded to an open book process in which

competitive bids for equipment and subcontracted labor were provided to Vectren South

and Burns & McDonnell for review. An open book process can be used to induce

competitive bidding and provide transparency to the counterparty when a vendor or

contractor possesses an advantageous position in the market place, as First Solar

possessed by having photovoltaic ("PV") modules available which were not subject to

the PV module tariff.

The pricing First Solar provided for PV modules, and competitive bids received for

inverters and single-axis trackers, as well as other materials, were compared to other

equipment pricing available to Burns & McDonnell. Each of the prices was consistent

with costs in the industry.

First Solar's competitive bids for subcontracted labor for civil construction, medium

voltage installation, and substation and generation tie-in construction also were made

available to Vectren South and Burns & McDonnell. This allowed Vectren South and


Page 5 of 12

1 Burns & McDonnell to confirm competitive solicitations were obtained and verify that the

2 most cost competitive bids were used in the pricing summary.

3

4 This open book process facilitated cost comparison and/or competitive bids for

5 approximately 65% of First Solar's EPC cost. This process resulted in an EPC cost

6 which was competitive with the market based on the aforementioned NREL Solar

7 Benchmark Study, and the evidence provided by Mr. Haselden pertaining to solar asset

8 sales.

9

10 Q. How were the remaining 35% of First Solar's EPC cost evaluated?

11 A. The total project cost was compared to other cost data in the industry. The total cost

12 was comparable, if not slightly lower, than the total project costs observed in the industry

13 when factored for size, union labor, and location. Therefore, Burns & McDonnell

14 concluded the remaining 35% of the costs, which included construction indirect costs,

15 self-performed labor, margin and balance of plant materials costs, were reasonable.

16

17 Q. Are there benefits to utility project ownership that Mr. Haselden ignores?

18 A. Yes. Mr. Haselden fails to recognize the benefit of utility-ownership of the Solar Project

19 after the financial evaluation life. It is not uncommon for utilities to operate generation

20 assets for years or decades beyond the assumed life in the initial financial evaluation. In

21 this case, Mr. Games notes that Vectren South expects to operate the Solar Project for

22 fifty years, while the levelized cost of energy provided by Vectren South is based on a

23

24

25

26

27 Ill.

28

29

30 Q.

31

32

30-year design life, as is custom in the utility industry. The cost/benefit analysis

improves over time compared to a 20-year PPA.

CONCERNS ABOUT THE OPERATION AND SIZING OF THE SOLAR PROJECT ARE

MISPLACED

Mr. Haselden states that he does not have concerns about the designs of the

proposed system, but claims inverters will need to be replaced every 12 to 15

years. Do you agree with that assessment?

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

A.


Page 6of12

No. In my experience, utility-scale central inverters do not need to be replaced every 12-

15 years. Mr. Haselden cites an EPRI publication which includes a table of ranges of

operations and maintenance costs, including replacement of inverters. However, this

same EPRI publication states:

"Historically, plant owners and managers have anticipated that central

inverter equipment will need to be replaced sometime during year 10-12

of a system's lifetime. However, over the past several years [this article is

from 2015], many have found that, with steady maintenance, central

inverters can remain operable for longer than expected and thus result in

over-budgeting. One large O&M provider claims that, for many of its

projects, as little as 25% of the funding budgeted for inverter replacement

was used by year 11."

While inverters do require periodic maintenance and replacement of components, these

costs typically are included in the operations and maintenance costs of a project. The

EPRI article later describes how owners often opt for straight-line operations and

maintenance allowances, which include inverter maintenance, throughout the project life

which allows cash reserves to build and improves the project's cash flow. Vectren South

has adopted a straight-line approach in which scheduled maintenance costs of the

inverter are included in the operations and maintenance budget. Thus, Mr. Haselden

would not see specific costs for specific inverter maintenance in the operations and

maintenance budget. A copy of the referenced portion of the EPRI publication is

attached as Petitioner's Exhibit No. 6, Attachment MRB-R1.

Mr. Haselden also references an IEEE Access article titled, "The Effect of Inverter

Failures on the Return on Investment of Solar Photovoltaic Systems" to form his opinion

regarding inverter replacements. One data set used in this article included 350 150kW

commercial systems. The second data set was derived from a 4.6 MW plant comprised

of 26 inverters. A third data set was derived from 202 systems of undisclosed size in

Taiwan. The inverters used in the first and second data sets are clearly string inverters

based on size. Based on the quantity of systems, one can reasonably deduce the third

data set also was comprised of string inverters - not utility-scale central inverters. A

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18 Q.

19

20 A.

21

22

23

24

25

26

27 Q.

28

29 A.

30

31

32

33


Page 7of12

copy of the referenced IEEE publication is attached as Petitioner's Exhibit No. 6,

Attachment MRB-R2.

Mr. Haselden also references an online article from Solar Mango, which is a resource for

rooftop installations in India. The Solar Mango Linkedln page provides:

"The Solar Mango website is dedicated to providing energy consumers

with relevant, up-to-date, and practical information and insights into using

rooftop solar power in India. The Solar Mango division within EAi works to

understand each consumer's requirements and provide them with a

solution in association with our technical partners that maximizes their

benefits from the rooftop solar system."

Since rooftop systems typically utilize string inverters, we can conclude the Solar Mango

article was written in regard to string inverters - as opposed to utility-scale central

inverters -- as well.

Is it appropriate to use string inverter data to judge the useful life of utility-scale

inverters?

No. String inverters are very different than the utility-scale inverters like those proposed

for the Vectren South Solar Project in terms of the ability to maintain and capital costs.

As a result, I do not necessarily disagree that string inverters would likely need to be

replaced every 12-15 years. However, I very much disagree that it is prudent to plan to

replace utility-scale central inverters like those proposed for the Vectren South Solar

Project every 12-15 years.

Do you agree with Mr. Griffey that "[t]he solar project cannot reliably provide

capacity near its nameplate amount?"

No. I disagree with the inference that solar PV is not reliable. Reliability is defined as

consistent performance. First Solar's solar PV Fleet has averaged 99.5% Effective

Availability over the past few years according to a New Energy Update article titled

"Common PV Performance Benchmarks Set to Boost O&M Transparency," which is

attached as Petitioner's Exhibit No. 6, Attachment MRB-R3

1


Page 8of12

2 While a solar installation will produce its nameplate capacity only under Standard Test

3 Conditions (1 ,000 W/m2 , temperature of 25°C, and an air mass index of 1.5), the annual

4 output of a solar installation can be forecast on an hourly basis using computer modeling

5 such as PVSyst. This modelling software was used to develop the energy model for the

6 Vectren South Solar Project and calculate the levelized cost of energy. The levelized

7 cost of energy calculation takes into account the variable output of the solar installation

8 due to seasonal weather variations, the time at which the energy is generated, and the

9 cost of energy at that time. Since these factors are included in the energy model used to

10 develop the cost model, which is the basis for comparison to other technologies, the

11 issue of whether the plant produces at its nameplate capacity is irrelevant.

12

13

14 Q.

15

16

17 A.

18

19

20

21

22

23

24 IV.

25

26

27 Q.

28

29

30 A.

31

32

33

Mr. Haselden questions why Vectren South did not choose a 60 MW or 100 MW

project because "larger projects are better due to economies of scale." In your

opinion, would a larger project be feasible in this instance?

No. In this case, the capacity of the Solar Project is a function of the available land. The

development purchased from Orion has a defined quantity of land, which will only

support 50 MWac of solar capacity when the layout is optimized to maximize the annual

energy production. A map of the property showing the layout of solar facilities is

attached to Mr. Games' testimony.

ANY DECREASE IN SOLAR PRICES IS SPECULATIVE COMPARED TO THE

KNOWN BENEFIT OF THE INVESTMENT TAX CREDIT

How do you respond to Mr. Haselden's criticism that the price of solar projects is

decreasing and is likely to further decrease if Vectren South were to pursue the

Solar Project at a later time?

I do not disagree that the capital cost of solar projects historically has decreased over

time. I believe the costs will continue to decrease in the future, but at a less aggressive

pace. However, I also believe this to be irrelevant as the Vectren South Solar Project

represents a unique opportunity for Vectren South to provide renewable energy within its


Page 9of12

1 service territory, which has the benefits of lower transmission costs and essentially

2 eliminates congestion as described in Mr. Joiner's testimony. Per my analysis and

3 market comparisons, the capital cost of the project is comparable with market prices for

4 utility-scale projects of similar size for the location.

5

6 Q.

7

8

9 A.

Mr. Haselden states it has not turned out to be the case that the "tariff on

polycrystalline panels made in China would cause an increase in price in due to

increased demand in the thin film alternative product." Is that accurate?

Not entirely. In September of 2017 when the Module Sale Agreement was executed

10 with First Solar, the anticipation of proposed tariffs created cost uncertainty in the solar

11 market, which resulted in many project owners and developers procuring PV modules

12 earlier than normal to facilitate construction completion prior to December 31, 2019 to

13 qualify for the Investment Tax Credit ("ITC"), which was the conservative approach to

14 securing the ITC as IRS guidance had not been issued at that time. 1 This demand

15

16

resulted in many PV module manufacturers selling out of product to be delivered in

2019. This increased demand resulted in increased PV module prices.

17 ~

18 Vectren South chose to procure PV Modules from First Solar because their modules are

19 not crystalline, are not produced in China, and thus were not subject to the proposed

20 tariffs, thereby attempting to isolate this project from cost increases associated with the

21 proposed tariffs.

22

23 When the 30% tariffs were enacted in 2018, the cost of certain PV modules did increase.

24 Therefore, the tariffs did indeed impact the cost of PV modules in the United States.

25 However, shortly after the tariffs were enacted, China revised their renewable energy

26 policy, reducing global demand for solar modules. This reduction in global demand

27 resulted in a cost reduction in PV modules in the United States. As a result, the cost of a

28 PV module today is similar to the cost in the market in September of 2017 when Vectren

29 South executed the Module Sale Agreement with First Solar.

30

31 Therefore, while I agree PV module pricing has returned to pre-tariff levels, I disagree

32 with Mr. Haselden's inference as to how the pricing returned to current levels. Moreover,

1 IRS guidance was subsequently issued on June 22, 2018.

1

2

3

4 Q.

5

6

7

8 A.

Petitioner's Exhibit No. 6 Vectren South Page 10of12

I believe Vectren South's actions were prudent given the information available to the

Company at the time.

On page 16 of this testimony, Mr. Griffey disagreed with the statement "that the

imposition of tariffs imposed at the federal level on polycrystalline modules

portend significant increases in the cost of solar through time." How do you

respond?

Similar to how I responded to Mr. Haselden, Mr. Griffey has formed an opinion based on

9 what has transpired, whereas Vectren South was faced with difficult decisions in a

1 O volatile market and made decisions to protect themselves and their customers.

11 Moreover, the solar tariffs have a five-year schedule. The tariffs have the potential to

12 increase the cost of PV modules during that five-year time frame. As discussed above,

13 other market factors have minimized the cost impact on PV modules such that the

14 overall cost of PV modules is currently similar to the costs before the tariff was enacted.

15

16 Q.

17

18

19

20

21 A.

With respect to the impact of tariffs on the cost of solar projects, Mr. Griffey

discusses a bid process for solar, wind and natural gas projects undertaken by

Public Service of Colorado. Does Mr. Griffey's implication that bid prices

decrease after the imposition of tariffs on solar modules mean there is no longer a

risk of prices increasing?

No. There is a global market for solar. As discussed above, decreased demand in

22 China has resulted in costs decreases for solar modules in the United States, which

23 have served to offset the solar tariff. A change in global demand could cause the cost of

24 PV modules to increase.

25

26 Q. Mr. Griffey quoted an excerpt containing opinions made by the CEO of XCEL

27 Energy during an earnings call in July 2018. Do you agree with the predictions

28 quoted by Mr. Griffey?

29 A.

30

31

I agree the cost of solar likely will continue to decline in the near term. However, the

30% ITC is expected to decline by 4% to 26% in 2020, another 4% to 22% in 2021, and

then continue indefinitely at 10%. While technological advances and construction

32 efficiencies could reduce the cost of solar by 4% per year, those decreases are likely to

33 be offset by the decrease in the ITC. Moreover, as Petitioner's witness Justin M. Joiner

....

1

2

3

4

5

6

7 v. 8

9

10 Q.

11

12

13

14

15

16

17 A.

18

19

20

21

22

23

24

25

26

27

28 VI.

29

30 Q.

31

32 A.

33

Petitioner's Exhibit No. 6 Vectren South Page 11 of 12

notes in his rebuttal testimony, XCEL Energy chose to add a substantial amount of solar

resources to its portfolio after that earnings call. Likewise, Vectren South also has a

good reason to capitalize on the unique opportunity to build a solar facility within its

service territory that will not be burdened by congestion or transmission issues.

VECTREN SOUTH CANNOT DELAY THE PROJECT AND OBTAIN THE ITC

BENEFIT

Mr. Haselden states: "To attain the full ITC, a small amount of work must be done

in 2019 but the proposed Solar Project does not need to be completed until

December 31, 2023. Given there is a four year window to complete a project that

should take about a year, Vectren could start construction at the interconnecting

substation by the end of 2019 and still have plenty of time to complete the

proposed Solar Project if the planned spring 2019 construction start were

delayed." Do you agree with this statement?

Not entirely. Mr. Haselden is referring to the Internal Revenue Service Notice 2018-59,

which was issued on June 22, 2018. Mr. Haselden accurately reflects the guidance

requirements related to the start of construction date but fails to recognize this guidance

also notes the IRS will apply strict scrutiny of the facts and circumstances related to

continuous construction of the project from the start of construction to final completion.

Mr. Haselden's suggestion to start the substation then delay construction of the solar

field would likely not pass IRS strict scrutiny and would not allow the project to qualify for

the ITC. In addition, slowing the construction of a project generally increases the capital

cost of the project.

CONCLUSION

What is your overall conclusion regarding the cost of the Solar Project as

compared to other projects?

I would reiterate the conclusion set forth in my direct testimony that the cost of the Solar

Project is consistent with - if not lower than - market conditions for a utility-scale PV

1

2

3

4

5

6

7 Q.

8 A.

Petitioner's Exhibit No. 6 Vectren South Page 12of12

project using union labor in Indiana. The capital cost of the project was benchmarked

against an NREL Solar Benchmark Study and the capital costs of the proposed Vectren

South project were lower than the benchmark for utility-scale, single-axis tracking

projects using union labor in Indiana. Moreover, the data Mr. Haselden provided with

respect to NIPSCO asset purchase and option bids supports the foregoing conclusions.

Does this conclude your rebuttal testimony?

Yes, at the present time.

I\ 13499001.2

VERIFICATION

I, Matthew R. Brinkman, P.E., Solar Business Manager for Burns & McDonnell, under

the penalties for perjury, affirm that the answers in the foregoing Rebuttal Testimony are true to

the best of my knowledge, information and belief.

Matthew R. Brinkman, P.E. Solar Business Manager, Burns & McDonnell

1\13507409.1

To better characterize pricing for panel washing, one O&M pro-

vider provides estimates in dollai·-per-panel ($/panel) rather than

dollar per capacity ($/kW) price metrics. This approach accounts

for differences in technology form factor (e.g., thin film require

more panels pet watt) and handling. Accordingly, per panel

washing costs can vary from ~$0.35 per panel for a simple water

spray to over $0.50 per panel for more intense washing (e.g. physical agitation). Meanwhile, vegetation management can, ac-

cording to one estimate, run between $15,000 and $30,000 per system per year based on site characteristics and acreage.

Inverter Maintenance & Replucement Historically, plant owners and managers have anticipated that

central inverter equipmem will need rn be replaced sometime

during year 10-12 of a system's lifetime. However, over the past

several years, many have found that, with steady maintenance,

central inverters can remain operable for longer thai1 expected,

and thus result in over-budgeting. One large O&M provider

claims that, for many of its·projects, as little as 25% of the fund-

ing budgeted for inverter replacement was used by year 11.

Unsurprisingly, in the budgeted replacement yeai·, system own-

ers often struggle with the decision to either pay the inverter

manufacturer for an extended 10-year warranty or to simply

replace the inverter (thereby obtaining a new warranty), regard-

less of its working condition. Rather than set aside a lump sum

of cash for inverter replacement, some enrities are now opting to

instead spread reserves across a lixed-fee maintenance schedule

Cause No. 45086 OUCC Response to Vectren's DR 1-4,

Attachment I

that builds up a cash reserve over rime, and, in turn, improves a

project's overall cash Bow. Another strategy being employed is to

group several maintenance reserves together into a major mainte-

nai1ce (i.e. contingency) reserve, thereby offering more spending

flexibility.

The cost-benefit behind string and micro inverters is not well

established. Many budget string inverter replacement aiid main-

tenai1ce similar to central inverters. Howeve1~ some have found

that string inverters do not come with the same level of warranty

and support provided for central inverters. String inverters will,

on average, require less service per inverter during the initial 10

year warranty period but by year 10-12 they will likely need to

be replaced. At the same time, while the response time for fixing

a string inverter failure may not be as critical, as only a small

portion of power is lost, more frequent visits may be required on

the whole, incurring higher O&M labor costs in the long run.

A range of experiences with inverter manufacturers colors in-

verter budgeting o'ut!ooks in terms of perceived inverter failure

rates and inverter manufacturer solvency. One O&M provider

performed a financial analysis of a manufacturer's SEC public fil-

ings that revealed a decrease in the company's reserve funds over

time. Considerarion of larger inverter replacement budget can, as

a result, be warranted.

Racking & Tracker Maintenance Costs associated with the upkeep of racking equipment are

negligible, as few long term defects are anticipated18. However,

Sources: Florida Solar Energy Center (left), Alabama Power (right)

18 EPRI is pursuing a research effort to explore racking corrosion from field exposure.

Phase 3: Evaluation of a Prototype Power Supply Utilizing " December2015

Attachment MRB-R1

IEEE Access·


Attachment I Page 33 of 40

Received August 18, 2017, accepted September 8, 201.7, date of publication September 18, 2017, date of currentversJon October 25, 2017.

Digital Object l.Umifier 10.1109/ACCESS.2017.2753246

The Effect of Inverter Failures on the Return on Investment of Solar Photovoltaic Systems TYLER J. FORMICA 1, HASSAN ABBAS KHAN2, (Member, IEEE), AND MICHAEL G. PECHT1, (Life Fellow, IEEE) 1 Center for Advanced Life Cycle En,,oineering, Univeratty of Maryland al College Park, College Park, MD 20742, USA 2Department of Electrical Engineering. School of Science and Engineering, Lahore University of Management Sciences. Lahore 54792, Pakistan

Cmresponding authm~ 'fyler J, Fonnica ([email protected])

: ABSTRACT Retum on investment (ROI) analyses of solar photovoltaic (PV) systems used for residential usage have typically. shown that at least 10 to 12 years is needed to break even, with this amount varying based on tax credits and reliability. This paper discusses the challenges with the reliability of cmTent solar photovoltaic systems and the key reliability bottlenecks, with a focus on the ROI. The problem stems primarily from reliability issues of currently available power electronics hardware. This paper's analysis of failure data shows that the short warranties and reliability concerns associated with solar PV inverters reduce the long-term ROI of residential solar PV systems by up to 10%. This paper, therefore, provides key insights for accurate ROI calculations for solar PV investments. Furthermore, methods to improve the reliability of PV inverters, such as selection of capacitors, inverter topology, and incorporating wide-bandgap semiconductor devices, are presented.

: INDEX TERMS Solar energy, solar photovoltaic systems, reliability, inverters, warranty, microelectronics, return on investment.

I. INTRODUCTION The solar photovoltaic (PV) industry has been one of the fastest growing renewable energy industries, contributing both to the secudty of the electdcity supply and the reduction of greenhouse gas emissions [l]. By the end of2016, the total installed global capacity of solar PV power has approximately been 295 GW (gigawatts) [2], a 31 % increase from 2015. An even higher trend in solar PV deployments is seen in the United States with installations of approximately 15,000 MW (megawatts) in 2016, a 100% increase from 2015 [3].

Solar PV systems work by converting sunlight into direct cun-ent (DC) electricity through solar cells, which are inte-grated into PV panels/modules [4]. Since most homes today use alternating cun-ent (AC), the DC is converted into AC through an inverter. Solar PV systems are composed of solar panels; solar inverters; mounting equipment to attacb the panels to surfaces or hold the panels in the air; a DC subsys-tem, which contains a DC combiner box (to connect multiple strings) with a DC disconnect switch for safety purposes; an AC subsystem, which in domestic deployment;s is just a switch; an electricity meter to measure the output of the system; and wiring to connect the components (see Fig. 1). A solar PV system is generally tied to the utility grid to

FIGURE 1. Configuration of a domestk grid-tied solar PV system.

21336 2169-3536 © 20171EEE. Tr

Personal use is also ~ See http://www.ieee.org/pt

deliver excess electricity to the grid during peak hours and receive electricity from the grid when the PV system is not producing enough solar power. PV systems can also act as stand-alone systems and use a solar battery to store electricity; this scenario is more common in off-grid com-munities and in developing regions where the grid power is intermittent [5], [6].

• - - _L __ .. -~-=-- _ ...... ,...,."'""if+""" f/'\I' academic research only. I IEEE permission.

Attachment MRB-R2 nl for more information, VOLUME 5, 2017

T. J. Formica et al.: Effect of Inverter Failures on the ROI of Solar PV Systems

The payback period, which is the time it takes to recoup funds from the initial start-up costs in an investment, is estimated to be within 10-12 years for contractqr-installed residential PV systems [7]. However, cmrnnt cost-analysis literature associated with PV systems does not account for reliability issues. If there are reliability problems with a solar PV system, it can take weeks to assess the cause of failure, obtain the needed replacement parts, and make the repairs. The costs can be substantial, and the inconvenience can be significant

This article presents the different wan·anty structures offered by companies, the return on investment (ROI) chal-lenges, the reliability concerns, and candidate solutions to these concerns associated with solar energy systems. We ana-lyze w~ranty data from solar PV manufacturers and failure data from various studies available from literature.

II. SOLAR PHOTOVOLTAIC SYSTEM WARRANTIES A warranty is a guarantee from the manufacturer that defines the responsibility for the product or service provided [8]. Warranties typically provide financial security for customers purchasing a product. Under a wairanty, the costs associated with the repair or replacement of the product in the time period specified in the warranty is shifted from the customer to the manufacturer or the financer of the installation.

PV panel companies offer both performance wairanties and product warranties. A performance warranty provides the customer with the assurance that the solar PV system will operate at a power output efficiency specified by the manufacturer for a set period of time. A product warranty provides the customer assurance that the solar panels will not fail due to a manufacturiog error for a set period of time (e.g.,' physical damage due to hurricanes are typically not covered). There can also be a specific product warranty for the inverter.

Performance wairanties focus on the efficiency of the solar PV system as a somce of power (degradation based mi an original efficiency guaranteed by the manufacturer). The amount of degradation in the performance of solar panels depends on environmental and operational conditions and is generally considered to occur at a rate of 1-2% per year [9]. Performance warranties typically guarantee about 90% power output compared to the efficiency when the PV system was purchased (the original efficiency) during the 10th yeai· and 80% power output compared to the original efficiency during the 25th year of operation. For example, if the warranty guai·antees 80% output dm1ng the 25th yeai· of operation, then a 100 Wp (watt-peak) rated panel should produce at least 80 W under standard testing conditions (STCs). The STCs for solar panels include irradiance of 1000 Wfill2 cell temperature of 25 °C, and wind speed of 1 mis. For exam-ple, SunPower® guarantees at least 95% of power output compared to the origioal efficiency during the fast 5 years, and a constant degradation in efficiency of 0.4% for years 5 through 25. Thus, the efficiency by year 25 is guaranteed to be atleast 87% [10].

VOLUME 5, 2017


Attachment I

1 EE'!~c~e~s·

Product warranties for solar panels provide coverage against manufactm1ng defects and premature wear and tear. For the warranty to be honored, the customer must provide evidence that the malfunction in the product came from faulty parts when the product was bought [11]. For exam-ple, Canadian Solar's 10-year product warranty states, "Any damages caused by abrasion, improper installation or animals are exempt from this warranty," and then goes on to state that there must be proof that the malfunction can be traced back to a manufacturing error. SunPower provides a 25-year product warranty for theh' panels [12].

Warranties are also given for the solar PV ioverters to cover defects in the workmanship ·and materials associated with the inverter. Residential systems generally use a central inverter but may alternatively use string inverters or micro-inverters. String inverters can be used for a group of panels and are smaller than central inverters but larger than micro-inverters. Micro-inverters are placed on individual panels, arid each micro-inverter ties the available power to the grid. Therefore, micro-ioverters have the inherent capability of measuring an individual panel's performance.

Central inverter wai'l'anties vary from 5 to 15 yeai'S (e.g., SolarEdge offers a 12-year warranty [13]). The tech-nology associated with central inverters is improviog, but they are still the most likely components to experience failures io solai· PV systems. Micro-inve1ters gene1·ally have longer warranties than central inverters, ranging from 15 to 25 years. This is due to higher reliability associated with micro-inverters because theil' switches and energy storage parts generally have lower power processing requil'ements (e.g., whereas central ioverters are typically rated to han-dle 5 kW or higher, each micro-inverter is generally rated to handle 200-250 W [14]). ABB Group offers a 10-yeai· product warranty for their micro-inverter systems and only a 2-year product warranty for their PVS800 central invert-ers [15], [16]. Enphase Energy offers a 25-year warranty for their micro-inverters [17].

Ill. FAILURE STUDIES Failures of solar PV system components can significantly decrease the ROI for PV systems. In this section, field failure data was compiled and analyzed from three sources and categorized to determine the leading causes of failures in solar PVsystems.

The first som·ce was a SunEdison database that consisted of over 3,500 failure tickets from 350 commercial systems, approximately 150 kW, SunEdison® PV systems, operating between January 2010 and March 2012 [18], [19]. Failure tickets are issued when a system is underperforrning, and these tickets are compiled to generate data regarding the cause of failure ai1d the amount of kilowatt hours (kWh) lost due to system downtime during a failure. The kilowatt hours lost represents the energy production lost due to failures.

Table 1 is a compilation derived from the database that links the cause of failure (failure area) to the related ticket count as well as energy loss due to system downtime. In the

21337

IEEEAccess·

TABLE !. Frequency of failure tickets and associated energy Joss for each general failure area [18).

Failure Area % of Tickets % ofkWhlost

Inverter 43% 36% AC Subsystem 14% 20%

External 12% 20% Other 9% 7%

Support Structure 6% 3% DC Subsystem 6% 4% Planned Outage 5% 8%

Module 2% 1% Weather Station 2% 0%

Meter 1% 0%

table, "DC Subsystem" refers to parts that connect the solar panels to the inverter, including DC combiner boxes, wiring, and disconnects from the modules to the inverters. ''AC Sub-system" includes everything between the inverter and the generation meter [19] (e.g., wiring, switch gears, and trans-formers). The external causes of failure stem from sources that are unrelated to the reliability of the PV system (e.g., grid outages and utility-mandated shutdowns) [19]. Support struc-tures are the mounting equipment, which includes all the parts that hold the panels in place (e.g., clamps). Planned outage refers to outages that were already scheduled for preventive maintenance. Weather stations employ sensors to measure. irradiance, temperature, and wind conditions to gauge the overall system performance.

The data in Table 2 indicates that PV inverter failures constitute the highest percentage of failures in SunEdison commercial PV systems. The inverter failures (see Table 2) were further categorized by the components that appeared to have induced the inverter failure, except in the first category, where no specific component could be assigned as the cause of the failure.

No-fault-found (NFF) failures are defined as instances in which a failure was observed but the failure cause could not be identified [20]. In this study, the NFF failures were considered inte1mittent, meaning the inverte1· failed but then recovered and functioned properly again after a manual restart. The failures were assumed to be due to control software because the maintenance personnel restarted the software and then observed no failure. However, the failures could also be attributed to hardware components since there was no inves-tigation beyond restarting the inverter [18], and hardware failures are also capable of inducing software shutdowns.

Cards/boards are the printed circuit boards (PCBs) used in the inverter. All switching elements, power buffers, and heat sinks are mounted on PCBs, which are optimized for thermal management, parasitic minimization, and electdcal noise perspectives. These PCBs fail due to internal routing

21338


Attachment 1 Page 35 of 40

T. J. Formica et a/.: Effect of Inverter Failures on the ROI of Solar PV Systems

TABLE 2. Frequency of failure tickets and associated energy loss for each general failure area [!Bl.

Inverter Failure Area % of Tickets % ofkWhlost

No-Fault-Found Failures 28% 15%

Card/Board 13% 22%

AC Contactor 12% 13% Fan(s) 6% 5%

Matrix/IGBT 6% 6% Power Suouly 5% 5%

AC Fuses 4% 12% DC Contactor 4% 1%

Surge Protection 3% 1% GFI Components 3% 2%

Capacitors 3% 7% Internal Fuses 3% 4%

Internal Relay/Switch 3% 2%

DC Input Fuses 2% 1%

Other 5% 2%

issues, which result in failures in which the entire power module must be replaced. AC contactors are the primary disconnection source to switch AC power from the inverter to the grid on/off, and DC contactors operate similarly with DC power. Cooling fans are used to regulate the temperature. · IGBTs are three-terminal solid-state semiconductor switches that allow efficient power flow from the panels (DC) to the grid side (AC). Capacitors are used to temporarily store energy and provide a stable DC rail voltage to the inverter input. Fuses consist of low-resistance metallic wire inside noncombustible material used to protect current from over-loading. The impact of lightning strikes is minimized with surge protection components. Ground fault interrupter (GFI) components are used to compare the current in the neutral conductor with the ungrounded conductor.

The second data source came from Collins et al. [21], who conducted a 5-year study of failures associated with a 4.6-MW solar PV plant consisting of 26 arrays, with each array comprised of 450 PV modules and 1 inverter. Of the 237 failures observed over 5 years, 125 of the failures were attributed to the inverters (see Table 3).

The third data source was deiived from Huang et aL [22], who analyzed failure data gathered by the Industrial Tech-nology Research Institute consisting of 202 PV systems in Taiwan over a 3-year span. Among the 202 PV systems, 62 experienced failures within the 3-year span. 60% of the failures in the 62 systems were attributed to the inverter (see Table 4).

A survey was conducted by Zaman et al. [23] that also indicates inverters contribute to the most failures. For the study, solar PV users and stakeholders in Australia reported

VOLUME 5, 2017

i I I

I ' I I I I

I

T. I. Formica et al.: Effect of Inverter Failures on the ROI of Solar PV Systems

TABLE 3. Distribution of failures observed at a 4.6-MW PV plant over 5 years [2 l].

. Failure Area % of Tickets Inverter 53%

AC Subsystem 14% DC Subsystem 14%

Module 12% Other

(lightning) 7%

TABLE 4. Distribution of failures observed in 202 PV systems over 3 years [22].

Failure Area % of Tickets Inverter 60%

Balance of System Components 28%

PVModules 12%

the failures they had observed in their solar PV systems. Of the 29 respondents, 26 of the problems repo1ted were related to the inverter, including 10 instances of a complete functional failure of the inverter.

IV. COMPONENTS CONTRIBUTING TO SOLAR PHOTOVOLTAIC SYSTEM FAILURES The data provided in the above-mentioned studies indicates that PV inverters comprise the most unreliable component in PV systems. In industry, solar PV system manufacturers openly admit the high likelihood of solar PV inveiters failing. For example, SolarCity New Zealand [24] states on their website, "The inverter, which has a 10-year warranty, is likely to be the only piece of equipment you will need to replace."

Inverter topology significantly affects the reliability of solar PV systems. Table 5 summarizes the benefits and weak-nesses of the three general inve1ter categories.

The failure of an inverter is usually precipitated by the capacitors, insulated-gate bipolar transistors (IGBTs), or metal-oxide-semiconductor-field-effect transistors (MOSFETs) that comprise the inverter [25]. With regards to specific switching requirements and operation, IGBTs pe1form well in high-voltage, high-temperature conditions where high-power processing is required [26]. MOSFETs, on the other hand, provide an efficient alternative to IGBTs in inverter topologies where higher switching speeds are required at relatively low power processing require-ments [26].

A. CAPACITORS

The most common capacitors used in inverters are electrolytic and film; film capacitors are far more reliable but more expen-sive than electrolytic capacitors, with the pdce difference varying based on size [27]. Aluminum electrolytic capacitors

VOLUME 5, 2017


Attachment 1

IE~54er~s·

TABLE 5. The Effect of inverter topology on reliability.

Inverter Topolol!V Effect on Reliability Typically handles 200-300 W. Low power processing requirements lead to longer

Micro-inverter lifetimes and warranty periods of 25 years. Each micro-inverter is attached

to an individual panel Typically handles 1-10

kW. Medium power processing requirements lead to shorter lifetimes

Residential Size Central Inverter than micro-inverters and wananty periods of 5-15 years. These are generally

used in residential applications

Typically handles greater than 100 kW. High power

Utility-Scale Central Inverter processing requirements lead to frequent

maintenance and upkeep in utility applications.

have been estimated to be approximately one-third the price of film capacitors per amount of energy storage needed [28]. Although film capacitors offer improved reliability compared to electrolytic capacitors, replacing electrolytic capacitors with film capacitors in PV inverters is not cost-effective in all applications due to the higher price and smaller capaci-tance per volume ratio associated with film capacitors [29]. Schimpf and Norum estimate the capacitance per volume

· ratio of electrolytic capacitors to be 20 times greater than film capacitors [29]. .

Solar PV inverters with a single standard electrolytic capacitor (DC-link) are estimated to have a lifetime of about five years before a failure [30]. Electrolytic capacitors in solar PV inverters fail due to temperature cycling, power cycling, and high internal capacitor temperature [31]-[33]. Tempera-ture cycling is particularly prominent in micro-inverter appli-cations when the inverters are placed outdoors on individual panels. Electrolytic capacitors are significantly more prone to catastrophic failures than film capacitors [35]. In a catas-trophic failure, a capacitor is completely non-functional and must be replaced. Sometimes the electrolytic capacitor will explode, which subsequently damages other components. Catastrophic failures usually occur in poorly sealed capac-itors when ripple currents cause high internal temperatures leading to the vaporization of the electrolyte [32]. Film capacitors rarely fail catastrophically, rather they tend to fail due to degradation, which decreases perf01mance [33}.

B. IGBTs AND MOSFETS

Semiconductor devices used in solar PV inve1ters, such as IGBTs and MOSFETs, fail due to electrical degradation in

21339

IEEEAccess·

the components or mechanical degradation associated with the electronic packaging [33]. Transistor failures frequently occur in PV inverters operating in high-voltage, high-cunent, or extreme temperature· conditions exceeding the manufac-turer's specifications [33]-[35] in the form of bond wire lift-off or deterioration of the die attach.

To test the reliability of IGBTs in solar PV inverter appli-cations, Sandia National Laboratories [38] studied the effects of high-temperature and high-voltage conditions on IGBTs. The IGBTs were stressed at various conditions, such as their maximum rated cmTent of 61 A at 25 °C for 45 min, and at temperatures above their rated current, such as 90 °C. The study did not specify how many IGBTs were used but stated that most IGBTs performed at a satisfactory level. However, in a few cases the IGBTs degraded significantly and in one case the IGBT degraded so drastically to the point it would have caused a complete failure in a solar PV inverter. MOSFETs tend to fail due to high junction temperatures [33].

In addition to electronic devices and capacitors, which perform core inverter operations, there are other inverter-related components which may result in system downtime. For instance, AC fuses may cause the inverter to stop func-tioning [36] if there are short circuits, which can occur when the insulation surrounding the wiring of the PV system is exposed. Pecan Street, a company that compiles data regard-ing energy needs and water supply, conducted a study of 255 residential solar PV systems over a period of 4 years [37]. Fifty-four of these solar PV systems reported minor main-tenance issues within the time period. Of the 54 reports, 13 experienced PV inverter failures due to blown AC fuses in the inverters.

V. ANALYSIS OF THE EFFECT OF FAILURES AND DEGRADATION ON RETURN ON INVESTMENT ROI analyses often assume components in solar PV systems will last 25 years without experiencing failures that constitute a replacement and only assume a constant maintenance cost to account for repairs [7], [38]. Central inverter wananties are most often between 5 to 15 years, and, as discussed in Section ill, these inverters are likely to suffer multiple failures in 25 years. This section takes into account the effect of out-of-wananty PV inverter failures and module performance degradation on a 25-year ROI of a typical PV system setup in Florida taking Yang et al. [7] work as baseline. This residen-tial 6.7-kW PV system in Gainesville, Florida, was installed by a contractor and qualified for the following benefits:

a) Feed-in tariffs (FITs): FITs offer lucrative rates for supplying electdcity back to the grid to encourage users to invest in renewable energy. In this case, the FIT was $0.21/kWh.

b) State tax-rebate: Fl01ida offered arebate for Gainesville that was calculated to be $82.50/year.

c) U.S. Federal Solar Investment Tax Credit: An upfront cost subsidy was offered in the form of a 30% tax credit for start-up c0sts (purchasing the system and installation costs).

21340


Attachment I Page 37 of 40


For simplicity, Yang et al. [7] did not account for inflation and future changes to tax credits offered by federal, state, and local governments, which are difficult to predict and quantify accUl'ately. We took these assumption as the baseline for fut'-ther analysis and incorporated the effect of panel degradation (not included in [7]) as well as the effect of multiple inverter failures in the ROI calculation.

Panel efficiency degradation must be accounted for in the calculations for more accurate predictions. Our analysis uses the minimum efficiency guaranteed each year by Suntech in their perf01mance wan·anty [39] for their STP-280 panels as these panels were used in [7]. However, we do not account for any module replacements, which are not a significant failure area, as indicated in Section ill. The energy produced over the lifetime is also averaged for yearly calculations. Fmther, an SMA inverter [ 40] with a 5-year wananty is taken for analysis (same as [7]) with decreasing inverter costs at 10% (case 1) and 15% per year (case 2). It is expected that the cost of inverters will decrease between 10% to 15% per year [41] in the future, and it is important to incorporate this effect for realistic projections. In order to calculate ROI, the equation shown atthe top of the next page has been used and is adopted from [7].

Note that the n01malized inverter replacement costs depend on the number of replacements included in the ROI anal-ysis. For instance, in the case of two replacements in the 25-year lifetime of the system, we include inverter replace-ments at 8.33 and 16.66 years and incorporate the projected decrease in the cost along with the overall cost n01malized over 25 years. The starting cost of the inverter was taken as $2647.00 (at the time of installation) and the annual mainte-nance cost was taken as $168.00 per year (same as [7]).

A plot showing the 25-year ROI for Yang et al. and our analysis is given in Fig. 2. Yang et al. predicts an ROI of 2.45 which, we believe, is unlikely to be the case in practice as it does not include the effect of panel degradation and potential inverter replacements. After incmporating panel degradation, the best-case scenario (albeit highly unlikely

2.5 ~- I I *Yang et al. 2 4 _ _ ___ 1 -+-case 1 - 10% annual · I reduction In Inverter Cost

I -X-case2-15% annual 2.3 -+···-----·---t reduction in Inverter Cost

kkl·T--1·-a 2.2 ~ . f'~/i"----- i I

2.1 +-------·1· ·-·----t~?.(::r--2.0 -~-------. ------i-l-----+---

1 I I I I 1.9-'--'r-~-+-~--+--~-+-~-r-~ 0 2 3 4

No. of Inverter Replacements In 25 Years

FIGURE 2. ROI of Gainesville, Florida, case study with the additions of inverter replacement costs and panel degradation [7].

VOLUME 5, 2017

T. J. Formica et al; Effect of Inverter Failures on the ROI of Solar PV Systems


Attachment I

IEE'E'A~t@ss·

-(Normalized Inverler replacement Cost +A11nual Maintenance)] PV System ROI = 25 x l [(av. A11nual Energy Yield .x FIT+ Tax Rebate) }

Staitup Costs - . Upfornt Cost Subsidy

with zero inverter replacements) shows the ROI closer to 2.2. We further modele!f the ROI for scenarios in which the inverter fails multiple times without being covered by the 5-year inverter warranty during the 25-year guaranteed per-formance lifetime of the panels. It can be concluded that each inverter replacement will cause the overall ROI of the system to decrease. Further, two cases for inverter replacements are taken where the annual decrease in the inverter cost is 10% (case 1) and 15% (case 2). A worst-case scenaiio where the inverter fails every 5 years (just after the warranty expires) and requires 4 replacements shows a reduction in ROI to 1.95 for case 1 and 2.07 for case 2, which amounts to a decrease of up to 10% due to inverter failures.

It should be noted that this analysis does not account for potential shipping costs when the inverter needs to be replaced and the energy lost due to downtime when the PV system is not producing energy in the event of an inverter replacement. There is variation in the amount of these costs and whether these costs are covered by the manufacturer. Consumers should be cautious with these issues when nego-tiating contractor-based installations where contractors or system installers must be required to maintain an inventory for the likelihood of an inverter failure. This will prevent long system downtime associated with inverter failures due to assessment and shipping delays.

Many failure studies indicate that PV inverters have a lifetime of 1-20 years until failure [21), [22), [38]; this time-frame varies signilicantly based on power cycling conditions, inverter size, temperature cycling, inverter components (types of capacitors used, semiconductor materials used, etc.), and other conditions. Therefore, analysis on possible inverter fail-ures will result in better estimation of ROI and suitability of the investment within a particular incentive scheme. A signif-icant variation in ROI is observed (see Fig.2) due to inverter replacement costs, and therefore this variation must be accounted for in a detailed assessment at the planning stage.

VI. EMERGING TECHNOLOGIES TO IMPROVE THE RELIABILITY OF PV INVERTERS Two major advances in electronics promise to improve the reliability of PV system inverters and, in turn, solar PV systems. The first advance pertains to wide-bandgap semiconductors (e.g., silicon carbide (SiC) and gallium nitride (GaN)), which are capable of providing high-temperature operation, long-term performance, and improved efficiency of the inverters compared to inverters employing silicon (Si)-based semiconductors. This advance improves the energy production and reliability of the PV inverter and, in turn, the overall ROI. The second advance involves

VOLUMES, 2017

imp1·oved inverter design topologies, including the develop-ment of micro-inverters.

SiC and GaN are wide-bandgap materials with supe-rior conduction and switching properties compared to Si. When used in MOSFETs and IGBTs, they can with-stand higher reverse. voltages and temperatures and achieve higher frequencies than Si-based transistors [42]-[46]. Hinata et al. [ 42] tested a solai· PV inverter using all SiC semi-conductors with overall efficiency of 99% (mass-produced inverters have not yet reached 99% efficiency). Their SiC-based inverter design also achieved 50 times as many power cycles to failure as a Si-based design used for comparison. 500 thermal cycles with parameters of -40 °C and 175 °C showed failures in the Si-based inverter and no noticeable degradation in the SiC-based inverter. Sintamarean et al. [52] designed PV inverters to compare the pe1formances of a Si IGBT-based solar inverter with a SiC MOSFET-based inverter in high-power applications (IO kW or higher). They achieved a switching frequency of 50 kHz for the SiC-based inverter compared to 16 kHz for the Si-based inverter. Sintamarean et al. [46] also concluded that, for particular set-tings, a SiC MOSFET-based inverter was more cost-effective and reliable due to higher switching frequency, which allowed 40% and 70% lower inductance and capacitance requirements, respectively. However, in general, the price of SiC- and GaN-based inverters is still considerably higherthan Si-based inverters.

Micro-inverters promise improved reliability compared to central inverters due to lower power processing require-ments for switches and energy storage elements. Each micro-inverter is typically connected to a 200-250 Wp panel, and the need for electrolytic capacitors is largely eliminated. Film capacitors, which are more reliable but have 1/20 the capacitance per volume ratio of aluminum electrolytic counterparts [32), can be used due to these lower power processing requirements [47). In addition to better reliabil-ity, there are two more significant advantages of deploying micro-inverters. First, unlike a central inverter, if a single micro-inverter fails, only the module that the micro-inverter is attached to will fail, and the rest of the PV system will remain functional. Conversely, if a central inverter fails, all panels (attached to the inverter) stop delivering any power to the loads/utility until the inverter is replaced/fixed. Second, micro-inverters have a much better performance ratio (PR) compared to central inverters in deployments where there is a high shading effect [48) because micro-inverters deploy module-level MPPTs (maxim.um power point tracking) as opposed to string-level MPPT operation for central inverters [49). MPPT is a process by which the input

21341

IEEEAccess·

impedance of the inverter is changed to match the maximum power load-line from the panels [33].

Micro-inverters also have several disadvantages. First, the upfront system cost of PV deployments with micro-inverters may be considerably higher compared to central systems due to the larger number of micro-inverters used in the system along with higher installation costs. For instance, purchas-ing several 200-250 W rated micro-inverters in a residential PV system compared to one 3-10 kWp central inverter will result in a higher start-up cost. Moreover, micro-inverters may not necessarily be easy to replace because they may not be readily accessible in some systems without completely dis-mantling portions of the array. This may limit the ROI due to paneVsystem downtime associated with inverter replacement. Second, because they are placed outside on each individual panel, they are exposed to environmental conditions, such as high temperature and moisture, which decrease the reliability of PCBs and solder joints that are typically not built to last the 25-year warranty period of solar PV systems, especially when

·exposed to volatile outdoor climates [50]. However, no fail-ure studies specific to micro-inverters are available in the literature to corroborat_e their: suitability for a 25-year period. Despite these anticipated challenges, the 25-year warranties associated with micro-inverters is a significant advantage ·over central inverters, which typically offer lower warranty periods. Therefore, micro-inverter-based PV deployments are likely to have a higher and more predictable ROI owing to the likelihood that inverter replacements will be 1ower•(if the initial costs for central and micro-inverter-based PV systems are the same).

Improvement in the reliability of solar PV inverters is a growing research area. In-depth analysis of the effects of the design, usage, and power conversion requirements of the inverter on reliability will undoubtedly lead to more reliable inverters in the years to come [51]-[56].

VII. CONCLUSIONS While solar photovoltaic (PV) systems are generally reliable, failures associated with inverters will decrease the return on investment (ROI). The data presented in this study confirms that central inverters are the least reliable component in solar PV systems. Our study shows that the 25-year ROI of solar PV systems may vary from approximately 2 to 2.2 due to central inverter replacement costs, and even higher variance may occur when factors such as system downtime due to an inverter failure are incorporated. Central inverter warranties are most often less than 15 years, whereas the PV panels and mounting equipment are likely to last 25 years. Cun·ent ROI studies of solar PV systems often overlook repair and replacement costs associated with inverters. They also do not account for downtime during which the PV system is not pe1forming while the user is in the process of filing a wairanty claim and is waiting for the manufacturer to inves-tigate and make a decision or for the inverter to be repaired or replaced. Even if an inverter is covered under wananty, inverter manufacturers are not always obligated to cover the costs of system downtime, shipping, and reinstallation of the

21342


Attachment 1 Page 39 of40


replacement inverter. These factors all decrease the long-term ROI of solar PV systems, and consumers must guard against these issues when choosing contractor-deployed PV systems.

The 25-year wairanties of some emerging micro-inverter manufacturers are likely to allow residential users to experi-ence fewer system losses due to replacements as compared to system losses with central inverters. AB the cost of micro-inverters decreases, their longer warranties for residential PV users can allow a higher and more predictable 25-year ROI as compared to current estimates.

REFERENCES [1] P. Vithayasrlchareon, G, Mills, !llld I. F. Mac.Gill, "Impact of electric

vehicles and solar PV on future generation portfolio investment," IEEE Ti·ans. Susta/Ji. E11ergy, vol. 6, no. 3, pp. 899--908, Jul. 2015.

[2] J. Pyper, "Global solar PV capacity will approach 295 GW in 2016." Renewable Energy Focus, Global Data, (Dec. 2016). [Online]. Avail-able: http://www.renewableenergyfocus.com/view/45114/global-solar-pv-capacity-will-approach-295-gw-in-2016-says-globaldata/

[3] "Solar industry data: Solar industry growing at a record pace." Safar Energy Industries Association. Accessed: Aug. 18, 2017. http:/fwww.seiaorg/research-resources/solar-industry-data

[4] S. Islam, S. Mekhilef, and M. Hasan, "Single phase transformer!ess inverter topologies for grid-tied photovoltaic system: A review,'' Renew. Sustain. Energy Rim, vol. 45, pp. 69-86, May 2015.

[5] M. Das and V. Aganval, "Novel high-perfonnance stand-alone solar PV system with high-gain high-efficiency DC-DC converter power stages," IEEE Tran." Ind. Appl., vol. 51, no. 6, pp. 4718-4728, Nov. 2015.

[ 6] H. A. Khan and S. Pervaiz, "Technological review on solar PV in Pakistan: Scope, practices and reconunendations for optimized system design," Re11ew. Sustain. Energy Rev., vol. 23, pp. 147-154, Jul. 2013.

[7] D. Yang, H. A. Latchman, D. Tingling, and A. A. Amal'Singh, "Design and return on investment analysis of residential solar PV systems,'' IEEE Potentials, vol. 34, no. 4, pp. 11-17, Jul. 2015.

[8] M. G, Pecht, "Establishing a. relationship between warranty and reliabil-ity;" IEEE Trans. Electron. Packag. Mamif., vol. 29, no. 3, pp. 184-190, Jul. 2006.

[9] R. Strecker. "Solar panels-Lifetime productivity and maintenance costs." Boston Solar. {Sep. 2011). [Online], Available: http://www.bostonsolar. us/blogfsolar-p!lllels-lifetime-prodnctivity-and-maintenance-costs

[10] M. A. Maehlum. (Jnn, 2013), Solar Panel Warranty Comparison. Energy Infonnative, accessed: Aug. 18, 2017. [Online]. Available: http://energyinfoanative.org/solar-panel-warranly-comparison/

[11] Limited W01wnty Statement P/!otovoltalc Diamond Module Products. Canadian Solar, accessed: Aug. 18, 2017. [Online]. Available: http://www. canadiansolar.com/downloads/warrantiesfWarranty_St!llldard_PV _ Module_en.pdf .

[12] Limited Product and Power Warran/)' for PV Modules. SunPower, accessed: Aug. 18, 2017. [Online]. Available: http://ns.sunpower.com/ sites/sunpower/files/media-librruy/warranties/w1'sunpower-limited-product-and-power-warranty-pv-modules.pdf

[13] Solar Edge Product Warrant)• Program. SolarEdge, accessed: Ayg. 18, 2017. [Online]. Available: http://www.solaredge.us/groups/us/ service/warranty

[14] Grid-Connected Solar Microi11verler Reference Design Usi11g a dsPIC Digital Signal• Controller, Microchip Technol. Inc., Chandler, AZ, USA, Aug. 2011.

[15] Central· Inverters PVSBOO Wannm')' and Sen•ice Qtferlng. ABB Group, accessed: Aug. 18, 2017. [Online]. Available: https://library.e.abb.com/ public/7488c3149 l 241476cl 257d89002aca52/17267_ Warranty_and_ service_offeriug_3AUA0000133536_RevC_lowres.pdf

[16] ABB Micro Jm1erter System MICR0-0.25!0.3/0.3HV-I-OUTD 0.25kW to 0.3kW. ABB Group, accessed: Aug. 18, 2017. [Online]. Available: https://library.e.abb.com/public/3b4b2359a4986e2685257dff005e1834/ MICR0-0.25-0.3-0.3HV-Rev0.l.pdf

[17] (Jan. 2014), Enphase Energy M215 Micro-lm•erler 25-Year Limited Warra11t}~N011h America. Enphase Energy, accessed: Aug. 18, 2017. [Online]. Available: http://www2.enphase.com/entrnst/25-year-limited-warranty/

[18] A. Golnas, "PV system reliability: An operator's perspective," IEEE J. Plwtovolt., vol. 3, no. l, pp. 416-421, Sep. 2013.

VOLUME 5, 2017

I ! i

I I

I I

I 1

T. J. Formica et al: Effect of Inverter Failures on the ROI of Solar PV Systems

[19] A. Golnas, Telephone Interview. Tech. Rep., Oct. 2015. [20] R. Bakhshi, S. Kunche, and M. G. Pecht, "Intermittent failures

in hardware and software," J. Electron. Packag., vol. 136, no. 1, pp. 011014-1--011014-5, Mar. 2014.

[21] E. Collins, M. Dvorack, J. Mahn, and M. Quintana, "Reliability and availability analysis of a fielded photovoltaic system," in Proc. 24th IEEE Photovolt. Specialist Conj. (PVSC), Jun. 2009, pp. 002316--002321.

[22] H. S. Huang, J, S. Jao, K. L. Yen, and C. T. Tsai, "Performance and availability analyses of PV generation systems in Taiwan," World Acad. Sci. Eng. Tee/mo/., vol. 54,no. 6,pp. 731-735,2011.

[23] A. Zaman et aL, "PV system reliability-Preliminary findings from thePV module and system fault reporting Website," in Proc. Asia-Pacific Solar Res. Co11f., 2014.

[24] Warra11ties & Mai11tennnce. SolarC!ty, accessed: Aug. 18, 2017. [Online]. Available: http://www.solarcity.co.nz/residential/prodnct/solar-life-expectancy/

[25] R. Olay, "Solru· inve1ter components can raise efficiencies," Solar Ind. Mag., vol. 5, no. 12, Jan. 2013.

[26] C. Black and C. Bull. (Oct 1999). IGBTs or MOSFETs: Which is Better for Your Design?. Electronic Design. [Online], Available: http://electronicdesign.com/power/igbts-or-mosfets-which-better-your-design

[27] G. Terzulli and B. W. Peace. Fi/111 Tec/uwlogy to Replace Electrolytic Tech-nology. AVX Corporation, accessed: Aug. 18, 2017. [Online]. Available: https://www.avx.com/docs/techinfo/filmtech.pdf

[28] Alu111inum Electrolytic Capacitors vs Film Capacitors. Heynen, Electronic Component Solutions, accessed: Aug. 18, 2017. [Online]. Available: http://www.heynen.com/aluminnm-electrolytic-capacitors-vs-film-capacitors

[29] F. Schimpf and L. Norum, ''Effective use of film capacitors in single-phase PY-inverters by active powel'decoupling," in Proc. 36thAnnu. Conj. IEEE Ind. Electron. (!ECON), Nov. 2010, pp. 2784-2789.

[30] C. Rodriguez and G. Amaratunga, "Long-lifetime power inverter for photovoltaic AC modules," IEEE Trans. Ind. mectro11., vol. 55, no. 7, pp. 2593-2601, Jul. 2008.

[31] M. C. Russell and G. RaySolar, "The promise of reliable inverters for PV systems: The micro-inverter solution,'' GTM Res., New York, NY, USA, Tech. Rep., Jun. 2010. [Online]. Available: https://www. greenteclnnediacom/articles/readlthe-promise-of-reliable-inverters-for-pv-systems-the-micro-inve1ter-solution

[32] J. Flicker, "Capacitor reliability in photovoltaic inverters,'' Sandia Nat Lab., Albuquerque, NM, USA, Tech. Rep., Jun. 2015. [Online]. Available: https://www.researchgate.net/publication/27944867 l_Capacitor_ Reliability_in_Photovoltaic_Inverters

[33] J. Flicker and R. Kaplar, "Reliability of power conversion systems in photovoltaic applications," Ph.D. dissertation, Dept. JET Power Energy Series, Inst. Eng. Technol, Stevenage, U.K., 2016, pp. 398-403.

[34] A. Ristow, M. Begovic, A. Pregelj, and A. Rohatgi, "Development of a methodology for improving photovoltaic inverter reliability," IEEE Trans. Ind. Electron., vol. 55, no. 7, pp. 2581-2592, Jul. 2008.

[35] R. Kaplar et al., "PV inverter performance and reliability; What is the role of the IGBT7" in Proc. 37th IEEE Photovolt. Specialists Conj. (PVSC), Jnn. 2011, pp. 001842-001847.

[36] Fuse and 1jpes of Fuses. Electrical Technology, accessed: Nov. 7, 2014. [Online]. Available: http://www.electricaltechnology.org/2014/11/fuse-cypes-of-fuses.html

[37] P. Street, "Minor maintenance issues proving difficult to detect for many solar PV system owners," GTM Res., Pecan Street GTM PV Mai111e11a11ce Rep., vol. 2, no. 5, pp. 1-10, Feb. 2015. [Online]. Avail-able: http://www.pecanstreet.org/wordpress/wp-contentluploads/2015/02/ Pecan-Street-GTM-PV-Maintenance-Report-Feb-2015.pdf

[38] E. Shouman, E.T. Shenaway, and N. M. Khattab, "Mru·ketfinancial analy-sis and costperformancefor photovoltaic technology through international and national perspective with case study for Egypt," Elsevier, Renew. Sustain. Energy Rel'., Vol. 57, pp. 540-549, May 2016.

[39] Suntech. 280 Watt Polycrystalline Solar Module. Accessed: Aug. 18, 2017. [Online]. Available: http://www.ecodirect.com/Suntech-STP280-24-VD-280W-35V-PV-Panel-p/suntech-stp280-24vd.htm

[40] SMA. (Mar. 21, 2016). SMA A111erica LLC Factory Wanwity. Accessed: Aug. 18, 2017. [Online]. Available: https://pdf.wholesalesolar. comlinveiter%20pdf%20folder/Sunny-Boy-Wru.Tanty-8-13. pdf

[41] V. Shah and J. Booream-Phelps, "Solar grid parity in a low oil price era," Deutsche Bank, Frankfurt, Germany, Tech. Rep., Feb. 2015. [Online]. Available: https://www.db.com/cr/en/concrete-deutsche-bank-repmt-solar-grid-parity-in-a-low-oil-price-era.htm

VOLUME 5, 2017


Attachment I

IE~/4tteiss·

[ 42] Y. Hinata, M. Horio, Y. Ikeda, R. Yamada, and Y. Takahashi, "Full SIC power module with advanced structure and its solar inverter application,'' in Proc. 28th Annu. IEEE Appl. Power Electron. Conf. Expa. (APEC), Mar. 2013, pp. 604-607.

[43] M. Oneill. Silicon Carbide Diodes Make Solar Power Systems More Efficient. EE Times, accessed: Aug. 18, 2017. [Online]. Available: http://www.eetlmes.com/document.asp?doc_id=1273 I 88&

[44] Reaching for the High Fruit: Finding Room for SiC and GaN iii the Solar Im1erler Market. Lux Research, accessed: Apr. I, 2013. [Ouline]. Avail-able: https://portal.luxresearchinc.com/reseru·ch/repmt_excerptll3342

[45] U. Schwarzer, S. Buschhom, and K. Vogel, "System benefits for solar inverter using SiC semiconductor modules," in Proc. PCIM E111:, May 2014, pp. 1-8.

[46] N. C. Sintamarean, E. P. Eni, B. Blaabjerg, R. Teodorescu, and H. Wang, "Wide-band gap devices in PV systems-Opportunities and challenges," in Proc. Int. Power Electron. Co11f., 2014, pp. 1912-1919.

[47] H. Haibing, S. Harb, N. H Kutkut, Z. J. Shen, and!. Batarseh, "A single-stage microinverter without using eletrolytic capacitors," IEEE Trans. Power Electron., vol. 28, no. 6, pp. 2677-2687, Jun. 2013.

[48] K. Sinapis et al., "A comprehensive study on partial shading response of c-Si modules and yield modeling of string inverter and module level power electronics," Solar Energy, vol. 135, pp. 731-741, Oct, 2016.

[ 49] Y. Sin and R. Winston, "Power conversion in concentrating photovoltaic systems: Central, string, and micro-inverters," Prog. Photovolt., Res.Appl., vol. 22, no. 9, pp. 984-992, 2014.

[501 C. Tulkoff. Reliability Challenges for Solar E/ectro11ics. DfR Solutions, Solar Energy Industry Association, Solar Electric Power Association, accessed: Aug. 18, 2017.

[51] G. Adinolfi, "Multiobjective optimal design of photovoltaic synchronous boost converters assessing efficiency, reliability, and cost savings," IEEE Trans. Ind. Jnfonnat., vol. 11, no. 5, pp. 103&-1048, Oct. 2015.

[52] P. Zhang, "Reliability evaluation of grid-connected photovoltaic power systems," IEEE Trans. Sustain. Energy, vol. 3, no. 3, pp. 379-389, Jul. 2012.

[53] G. Graditi, "Photovoltaic optimizer boost converters: Temperature infln-ence and electro-thermal design," Appl. Energy, vol. 115, pp. 140-150, Feb. 2014.

[54] S. Harb and R. S. Balog, "Reliability of candidate photovoltaic module-integrated-inverter (PV-MII) topologies-A usage model approach," IEEE Trans. Power Electron., vol. 28, no. 6, pp. 3019-3027, Jun. 2013.

[55] P. S. Shenoy, K. A. Kim, B. B. Johnson, and P. T. Krein, "Differ-ential power processing for increased energy production and reliability of photovoltaic systems," IEEE 1hzns. Power Electro1L, vol. 28, no. 6, pp. 2968-2979, Jun, 2013.

[56] F. Spertino, "Power conditioning units in grid-connected photovoltaic systems: A comparison with different technologies and wide range of power ratings," Solar Energy, vol. 108, pp. 219-229, Oct 2014.

TYLER J. FORMICA received the M.Sc. degree in mechanical engineering from the University of Maryland at Coliege Park. He researched solar photovoltaic system reliability with the Center for Advanced Life Cycle Engineering while at the University of Maryland at College Park. He is cun-enUy a Systems Reliability Engineer with Raytheon Company.

HASSAN ABBAS KHAN received the M.Sc. and Ph.D. degrees in elecai-cal and electronic engineering from The University of Manchester, U .K., in 2006 and 2010, respectively. His current work is on low cost solar PV deployments.

MICHAEL G. PECHT (S'78-M'83-SM'90-F'92-1F' 18) received theM.S. degree in electrical engiueedng and the Ph.D. degreein engineering mechan-ics from the University of Wisconsin. He is an ASME Fellow. He received the IEEE Reliability Society's Lifetime Achievement Award in 2008.

21343

...

Common PV performance benchmarks set to boost O&M transparency I New Energy Up ... Page 1of2

AOOutUs J lndustryReports i ContactUs J EventsandConferences

Home Wind PV

I US Biogas Nov 5, 2018 -Nov 6, 2018, San Diego, USA Monetize your Production-and-Use Biagas cycle

CSP Energy Storage Tidal

Common PV performance benchmarks set to boost O&M transparency

Jul 20,2016

A new European report has provided a basis for standardized PV performance indicators but more detailed

metrics will be required to maximize the gains from an increasingly competitive Operations and Maintenance

sector, experts told PV Insider.

While much of the utlllty-scale PV sector Is embracing performance

benchmarks In order to assess Operations and Maintenance (O&M)

performance and drive up plant avallablllty, a lack of comprehensive

standardized performance metrics Is llmltlng potential gains.

Key Performance Indicators (KP ls} help to underpin supply

guarantees, provide performance comparisons over time and

between plants, and measure the efficiency of O&M services.

'1fwe can use the same KPls, we would have additional benefits

because we see a lot of consolidation Jn the market, and all these

Image credit Rapid Eye

companies would like a common base to measure and compare the performance of the assets," VassUJs

Papaeconomou, managing director at Alectris, a solar asset management service provider, said.

Papaeconomou was among a group of European Industry leaders consulted by Solar?ower Europe to produce

the 'O&M Best Practice Guidelines for PV solar plants.'

The guldellnes, published In June. propose basic unlversal performance metrfcsfor professional O&M services.

The report collated views from O&M Service Providers, Asset owners, Asset Managers, Technical Advisors and

manufacturers.

Since there is currently no standardized ·Avallablllty' calculatlon, engineers must often review how AvallablHcyts

defined and calculated In order to establish performance guarantees, according co Heidi Marie Larson, director of

Solar Generation at Leldos' Renewable Generation Services division.

Avallablllty benchmarks currently In use typically range from around 95% to 100%, according to Industry experts.

"Avallablllty Is probably the most common metric that we see In O&M providers' performance measurements,

though the Industry has not yet standardized how Avallablllty for a PV plant should be determined, and what should and should not be Included," Larson noted.

Full coverage

The European O&M Best Practice Guidelines distinguish between Plant Performance KPls which "directly reflect

the performance of the plant and are under the duties of the O&M Contractor" and O&M Contractor KP ls which

''reflect the performance of the service ...

While the Plant Performance KPls are quantitative and measure the plant performance ratio, plant avallablllty,

uptime and energy output, the O&M Contractor KP ls are both quantitative and qualitative.

Performance should be rated against the lndlctors and translated Into bonus schemes and liquidated damages,

ensuring the assetownerwlll be compensated for poor reaction times or avallablllty performance, the guidellnes

said.

Examples of Availability-related Bonus Schemes and Liquidated Damages

Attachment MRB-R3

Sign up to the newsletter B

Waste to Energy Blockcha1n ln Energy

Get more information of thi NEW data heavy report here

I Upcoming Events • US Blogas

san Diego, USA Nov 5, 201 S to Nov 6, 201 S

• Solar PV Operations USA(• Energy Storage,

Rooftop Solar & Smart Grtds)

San Dlego, USA

Nov 7, 2018 to Nov 8, 2018

~ Wind O&M Canada

Toronto

Nov 12, 201Sto Nov 13, 2018

~ CSP Madrid (formerly CSP Seville)

Madrid

Nov 13, 2018 to Nov 14, 2018

• Africa New Energy

Cape Town, South Africa

Nov 26, 2018 to Nov27, 2018

., Offshore Wind Europe

London, UK

Nov 27, 2018 to Nov 28, 2018

httn://analvsis.newenerQVundate.com/nv-111:siuc1' '-u• 11111u1 ,-., • .,~. i'ormance-benchmarks-seL. 9/16/2018

Common PV performance benchmarks set to boost O&M transparency I New Energy Up ...

• es re O.n "i>Pl'"'"i" RJ.~·~:to:>;i ·"'" aopcilcab!e !Ulprtilll! l'IJ)rlod

P

Documents

PETITIONER'S W1 ~BITNO - IURC Online Services Portal...W1 ~BITNO . .,/_~ d/1'.J-17'-~-R-Ef&-oR-~-R SOLAR BUSINESS MANAGER, BURNS & MCDONNELL ON COMPETITIVENESS OF PRICE OF SOLAR PROJECT