View
6
Download
0
Category
Preview:
Citation preview
Might the large-scale PV power generation be reached in
China? A grid parity combined with techno-economic analysis
Hongyang Zou, Huibin Du*, Marilyn A. Brown
College of Management and Economics, Tianjin University, Tianjin
300072, China
School of Public Policy, Georgia Institute of Technology, Atlanta, GA
30332, USA
I. Introduction
II. Construction and parameter description
Outline
III. Techno-economic evaluation
IV. Grid parity estimation based on learning curve
V. Conclusions and policy implications
I. Introduction
• Pressure of carbon reduction: The electricity generated from renewable
energy will be reached at 9% in 2020 in the energy planning of National
Energy Administration of China; But in 2013, this percentage was only 2.8%.
• The new-added installed capacity of PV solar in China: up to 10.6 GW in
2014, which accounted for about 20% of total new-added installed capacity
in the world.
• The PV power generation: only 9, 000 GWh in 2013 accounting for 0.166%
of the electricity generation of China. plans to be reached at 150,000GWh
in 2020.
• Might this target be reached in 2020? When and What type of PV power
generation (grid-connected or off-grid) can be applied in large scale?
Where is more suitable to use PV power generation in China?
1. Grid-connected and off-grid PV power generation system
AC DC
PV Module
Battery
Primary Load
Converter
Grid
AC DC
PV Module
Battery
Primary Load
Converter
Diesel Generator
The structure of grid-connected PV system The structure of off-grid PV system
>2200
1800-2200
1400-1800
1000-1400<1000
kWh/m2/year
Tianjin
Xigaze
Xining
ChongqingHangzhou
2. PV systems in five cities
Solar radiation (five regions):
• More solar radiation than other countries
with a similar latitude;
• distributed unevenly;
The distribution of China’s solar radiation under the optimal slope
II. System construction and parameter description
● HOMER (Hybrid Optimization of Multiple Energy Resources), a micro-grid
software developed by the HOMER Energy LLC;
● Three main functions: simulation, optimization & sensitivity analysis.
II. System construction and parameter description (cont'd)
3. Data and Parameter description
Components Capital Replacement O&M Lifetime
PV module (1kW) 3640 3640 36.4 20 years
Converter (1kW) 800 800 0 15 years
Battery 600 600 0 10 years
Diesel generator(1kW)
1300 1300 0 15000 h
2.5
2.0
1.5
1.0
0.5
0Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Max
Daily high
Daily low
Mean
Min
3.0 (kW)
The synthesized data of load profile of residential houses
The cost of each component in PV system2.39 kW
Collected from: Annual Review and Outlook for China Solar PV industry, 2013; www.solarzoom.com; www.1688.com The picture resource comes from the HOMER software
III. Techno-economic evaluation
CitiesPV capacity
(kW)
Battery
(quantity)
Converter
capacity (kW)
Generator capacity
(kW)
Grid-
connectedAll five cities 3 1 1 /
Off-grid
Chongqing 10 30 3 5
Hangzhou 7 35 3 5
Tianjin 5 35 3 5
Xining 5 30 3 5
Xigaze 5 25 3 5
1. Technical feasibility analysis
Objective function: 𝑚𝑖𝑛 NPC (Net Present Cost)
s.t.
the electricity generated from each component in a PV system ≥ the electricity demand
The optimal system configurations of grid-connected and off-grid PV systems in five cities
III. Techno-economic evaluation (cont'd)
1. Technical feasibility analysis
Grid-connected:
PV production: more than 50%;
Xigaze: PV 5913 kWh (72.7%);
Chongqing: PV 2771 kWh (51%).
Off-grid:
The same electricity consumption: 4113
kWh/year;
The excess electricity of Tianjin is the
lowest; (the most cost-effective)
51
.0%
58
.1%
66
.4%
68
.4%
72
.7%
49
.0%
41
.9%
33
.6%
31
.6%
27
.3%
0
1
2
3
4
5
6
7
8
9
Chongqing Hangzhou Tianjin Xining Xigaze
PV production Grid purchases
42
.9%
48
.4%
51
.6%
48
.7%
41
.9%
46
.0%
38
.8%
34
.8%
38
.7%
47
.6%
0123456789
1011
Chongqing Hangzhou Tianjin Xining Xigaze
AC consumption Excess electricity Losses
The electricity generation of grid-connected PV systems
The electricity generation of off-grid PV systems
10000
12500
15000
17500
20000
22500
25000
27500
30000
32500
35000
37500
0
0.2
0.4
0.6
0.8
Chongqing Hangzhou Tianjin Xining Xigaze
NP
C (
RM
B Y
uan
)
LCO
E a
nd
reta
il ele
ctri
city
pri
ce(R
MB
Yu
an
/kW
h)
LCOE Retail price NPC
III. Techno-economic evaluation (cont'd)
0
20000
40000
60000
80000
100000
0.00
0.40
0.80
1.20
1.60
2.00
2.40
Chongqing Hangzhou Tianjin Xining Xigaze
NP
C (
RM
B Y
uan
)
LCO
E a
nd
reta
il ele
ctri
city
pri
ce(R
MB
Yu
an
/kW
h)
2. Cost-effectiveness analysis
Grid-connected:
● LCOE: 0.469-0.622 RMB Yuan/kWh;
● Xigaze: the only region, LCOE<retail price;
Off-grid:
● LCOE: 1.34-2.03 RMB Yuan/kWh;
● NPC: much more than grid-connected systems
(more energy storage batteries).
𝐿𝐶𝑂𝐸 =𝐶ann𝐸served
the total annualized cost of the system
the total electrical load served
𝐶ann = 𝐶𝑅𝐹 𝑖, 𝑁 ∗ 𝐶NPCthe total net presented cost of the PV system
capital recovery factor
The electricity generation (MWh) of grid-connected PV systems
The electricity consumption (MWh) of off-grid PV system
Levelized cost of energy
III. Techno-economic evaluation (cont'd)
3. Environmental performance
438
659737
1057
1297
29
203 202
325384
0
300
600
900
1200
1500
Xigaze Xining Tianjin Hangzhou Chongqing
CO2 emission of grid-connected PV systems (kg/year)
CO2 emission of off-grid PV systems (kg/year)
The CO2 emissions produced by grid-connected and off-grid systems in five cities
𝑀𝑖 = 𝐸𝑝 ∗ 𝐸𝐹𝑖
𝑀𝑖 : the emissions of pollutant i (kg)
𝐸𝑝: the net grid purchases (kWh) or the consumption of fuel p (kg)
𝐸𝐹𝑖 : the emissions factor of pollutant i
◆ CO2, SO2, NOX , CO and particulate matter
◆ Grid-connected > off-grid
IV. Grid parity estimation based on learning curve
Cities Learning curve Learning rate (LR)
Xigze UCt = 8.5791∗Cumt−0.193 0.1252
Xining UCt = 9.3612∗Cumt−0.191 0.1240
Tianjin UCt = 9.7109∗Cumt−0.191 0.1240
Hangzhou UCt =11.342∗Cumt−0.190 0.1234
Chongqing UCt =12.239∗Cumt−0.184 0.1197
1. The learning curve in five cities
2. Grid parity in five cities
Grid parity in future 30 years;
Xigaze: the first region (2027-2031).
The learning curve of PV power generation in five cities
𝑈𝐶𝑡 = 𝑈𝐶0 × 𝐶𝑢𝑚𝛼
𝐿𝑅 = 1 − 2𝛼
Model Formula
One-factor 𝑆𝐶 = 𝑎 × (𝐶𝐶−𝑏)
Two-factor 𝑆𝐶 = 𝑎 × (𝐶𝐶−𝑏) × (𝐾𝑆−𝑐)
Three-factor 𝑆𝐶𝑖𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 = 𝑎 × (𝐶𝐶−𝑏) × (𝐾𝑆−𝑐)𝑄𝑑
Four-factor
𝑆𝐶
= 𝑎 × (𝐶𝐶−𝑏) × (𝐾𝑆−𝑐)𝑄𝑑(
𝑖=1
𝑛
𝑝𝑖𝑒𝑖)
Learning rate
the cost of PV power generation in time period t
the initial unit cost of PV power generation
Learning Index
Conclusions and policy implications
Xigaze has the best solar radiation and the
earliest grid parity (2027-2031).
Tianjin has the high retail electricity price and
the average solar radiation.
2. Market deployment of PV power
generation will vary across regions
(1) Remote areas in Tibei —— off-grid PV system;
(2) North China —— grid-connected PV system.
1. The diversified development of the user
market
(1) Courtyard house and single houses;
(2) High-tech industrial parks.
The LCOE of grid-connected PV systems is close
to the retail electricity price (RMB Yuan/kWh).
3. The subsidy-change strategy of PV
power generation
China’s benchmark price is reducing to 0.80, 0.88
& 0.98 RMB Yuan/kWh in 2017.
The LCOE of off-grid PV systems will drop to be
0.536 -0.870 RMB Yuan/kWh in 2050.
Grid parity will be reached in the future 30 years.
Chongqing Hangzhou Tianjin Xining Xigaze
LCOE 0.622 0.588 0.511 0.469 0.472
Retail
price0.520 0.540 0.489 0.420 0.500
Huibin Du: duhuibin@tju.edu.cn
Recommended