Optimal Placement and Sizing of Wind Farm in Vietnamese Power System Based on Particle Swarm Optimization

Vietnam is a long coast country with high potential for wind energy development. However, the cost for wind farm project is a big challenge for the power system planning, while the location and capacity of the wind turbines have the large impact on the operation efficiency of the wind farm. Therefore, problem of optimal placement and size of wind farm attracts great interest of the practitioner and researchers. In this paper, the particle swarm optimization algorithm has been applied to optimize the placement and sizing of wind farm. The model has been tested on the simplified Vietnamese power system to determine the optimal location as well as installation capacity of the wind farm in order to minimize system operation cost while ensuring the capacity limit of transmission lines, power balance, and other constraints. The results in the paper can be applied for power system planning, design, and wind farm projects

Trang 1

Optimal Placement and Sizing of Wind Farm in Vietnamese Power System Based on Particle Swarm

Optimization

Dinh Thanh Viet

The University of Danang

Da Nang, Vietnam

dtviet@ac.udn.vn

Tran Quoc Tuan

Alternative Energies and Atomic Energy Commission (CEA), National Institute for Solar Energy (INES)

Le Bourget-du-Lac, France quoctuan.tran@cea.fr

Vo Van Phuong

Da Nang Power Company, Ltd

Da Nang, Vietnam phuongvv@cpc.vn

Abstract—Vietnam is a long coast country with high

potential for wind energy development However, the cost for

wind farm project is a big challenge for the power system

planning, while the location and capacity of the wind turbines

have the large impact on the operation efficiency of the wind

farm Therefore, problem of optimal placement and size of wind

farm attracts great interest of the practitioner and researchers

In this paper, the particle swarm optimization algorithm has

been applied to optimize the placement and sizing of wind farm

The model has been tested on the simplified Vietnamese power

system to determine the optimal location as well as installation

capacity of the wind farm in order to minimize system operation

cost while ensuring the capacity limit of transmission lines,

power balance, and other constraints The results in the paper

can be applied for power system planning, design, and wind

farm projects

Keywords— particle swarm optimization; power system

planning; Vietnamese power system; renewable energy; wind

farm

I INTRODUCTION With the rapid pace of urbanization and industrialization

in the last 10 years, many experts predict that in the next 10

years, Vietnam is in high risk of power shortage The demand

for energy in Vietnam, especially electricity in the period

2020-2030, is huge Demand for energy is increasing, while

energy supply is facing many challenges, especially domestic

traditional energy sources such as hydropower, coal, oil and

gas are gradually depleted and difficult to develop

In this context, the exploitation of renewable energy

sources is very important for Vietnam in terms of economy,

society, energy security and sustainable development [1], [8]

Under Revised National Power Development Master Plan for

the 2011-2020 period with the vision to 2030, the Government

of Vietnam set a target of producing 10.7% of electricity from

renewable sources by 2030 [1], [10] Specifically, by 2020,

the total wind power capacity will reach about 800 MW and

by 2030 is 6,000 MW [1] as showed in Table I

TABLE I DATA OF WIND ENERGY DEVELOPMENT

PLANNING IN VIETNAM

Energy

source

Power (MW) Percentage of energy

production (%)

2020 2025 2030 2020 2025 2030

In order to encourage investors to invest in wind power

projects in Vietnam, the Government of Vietnam has many

preferential mechanisms and policies Typically, the

Government has issued feed-in-tariff (FIT) prices according

to Decision No 39/2018/QD-TTg on the mechanism to support the development of wind power projects in Vietnam

In addition to the FIT price, the government encourages the Electricity of Vietnam to prioritize the purchase of all electricity generation by wind power plants Besides, these plants are able to receive tax incentives at the highest level, while some other taxes, such as equipment import tax, environmental tax, land tax, etc are reduced [2]

With great potential to exploit such incentive mechanisms,

in the coming years, wind energy sources in Vietnam are expected to be strongly invested and developed Currently, the total installed capacity of wind power in Vietnam is 197 MW;

263 MW of wind power capacity is under construction; 412

MW is in the process of approval of basic appraisal Approximately 4,236 MW has been approved and the total registered wind power capacity is 10,729 MW The total wind potential in Vietnam is estimated at 27,200 MW, as shown in Table II [12]:

TABLE II WIND POTENTIAL BY WIND GRADE AND REGION

Low Medium High

There were some studies on integrating renewable energy sources into Vietnamese power system and Vietnamese electricity market, such as [7], [8], [10], [12], [13], [14], [15] However, investigation of these publications leads to an essential need of building a computational model to determine the optimal location and the installation capacity of wind power plants into the Vietnamese power system in the current period, especially for the calculation and approval of investment projects for wind power plants

A Objective function

It is assumed that the investment rate for 1 MW of wind power at all buses is equal, so this component can be ignored

in the objective function Hence, the objective function of the problem is to optimize the operating costs of power plants in the presence of wind power plants, as shown in the formula (1) [3], [11]:

min

Trang 2

Where:

 ∂: voltage angle of generator,

 U m: voltage magnitude of generator,

 : active power injection of generator in bus k,

 : reactive power injection of generator in bus k,

 : cost of active power injection of generator in bus

k,

 : cost of reactive power injection of generator in bus

k,

 : cost of active power injection of wind generator in

bus k

B Constraints in power system

The optimization process must ensure that the system

parameters meet the conditions of power balance, stability and

transmission constraints, including:

Power balance constraints:

where s k is the power of the power plant at bus k; d k is the load

at bus k; M k,b is the incident matrix of the power grid and f b is

the power transmitted through branch b

Limitation on power generation capacity of power plants:

, ≤ ≤ , (3)

Constraints on transmission capacity of lines

Limit on voltage at the buses:

The limit on total installation capacity of wind power

plants in the whole network:

Constraints on maximum capacity of wind power plant at

each bus:

C Particle Swarm Optimization

Particle swarm optimization (PSO) is one of the optimal

algorithms proposed by J Kennedy and R Eberhart in 1995

PSO is an optimal tool that provides a swarm-based search

process in which each individual changes its position over

time A potential solution for each problem can be represented

as an individual in the swarm, flying in the search space with

d dimensions The speed and position adjustment of each

individual can be calculated using the current velocity and the

distance from pbest to gbest by formulas (9) and (10) [4]:

Where:

i = 1,2 n, in which n is the population size, : the inertia weight,

c1, c2: the acceleration coefficients,

rand1, rand2: random numbers within 0 and 1, : position of particle i in the iteration k, : velocity of particle i in the iteration k, : the best position of particle i to the iteration k, : the best position of the swarm to the iteration k

In this paper, the binary particle swarm optimization has been used in combination with the selective method to create the optimal placement and sizing of wind farm model [5], [6] The flow chart diagram of the proposed algorithm is shown in Fig 1

Fig 1 Flow chart diagram of the proposed algorithm

Trang 3

III CASESTUDY:OPTIMALPLACEMENTAND

A Vietnamese power system in the year of 2030

In order to study and evaluate the possibility of optimally

integrating wind energy into Vietnamese power system in the

future, stemming from the research and evaluation of the

actual power grid [7], the authors have built the simplified

Vietnamese 500kV power system model consists of 31 buses,

41 branches as shown in Fig 2 Buses number and name of

each bus are listed in Table VII in Appendix

Fig 2 Simplified-Vietnamese power system in the year of 2030

In particular, the power source and load parameters in each

area are referred to the nearest bus In Fig 2, the buses in the

power system are represented by a blue circle, the size of the

circle is drawn proportionally to the load referred to that bus

Buses are linked together through transmission lines

B System parameters

In the paper, the data set of the installed capacity of power

plants, parameters of transmission lines and actual demand

have been collected Based on this data set and information on

the 500kV power development plan by 2030, data from the

Revised National Power Development Master Plan for the

2011-2020 period with the vision to 2030 [1], the authors built

the data set for the simplified power system in Fig 2

C Simulation and results

The methodology, described in part II has been applied to

solve the problem of optimal placement and sizing of wind

power plants model The calculation scenarios are given by

the authors based on the Revised National Power

Development Master Plan for the 2011-2020 period with the

vision to 2030 and the development strategy of renewable

energy of Vietnam by 2030 with a vision to 2050 [9], with the

different penetration levels of wind power plants into the

Vietnamese power system The three specific scenarios have been investigated as follows:

 Scenario of the year 2020: maximum penetration of wind power plants into Vietnamese power system is limited to 800 MW

Each scenario has been simulated with 100 iterations using proposed algorithm (Fig 1)

First, we run the simulation for the first scenario (scenario

of the year 2020) Fig 3 and Table III show the performance and results of the model From the first iteration to 100th iteration, the objective function is reduced from

$2,694,233.907 to $2,618,864.524 (-2.80%) As we can see from the Fig 3, the model almost reaches its optimum after 44 iterations

Fig 3 Optimization performance of the model in scenario of the year

2020

TABLE III SCENARIO OF THE YEAR 2020–SIMULATION

RESULTS

Scenario

Total optimized wind capacity installation

Operation cost (USD)

Percent-age of cost saving

Before optimization

After optimization

Scenario of the year

2020

780 2,694,233 2,618,864.5 2.80%

Iteration 2.61

2.62 2.63 2.64 2.65 2.66 2.67 2.68 2.69

Trang 4

In the scenario of the year 2020, after running the

simulation with 100 iterations, the model proposed to install

the total of 780MW wind power plants at the buses 3, 5, 6, 18,

29, 30 The proposed installation capacity at each bus is shown

in Table IV

TABLE IV PROPOSED LOCATION AND CAPACITY OF

WIND POWER PLANTS IN SCENARIO OF THE YEAR 2020

Bus Wind power plants capacity (MW)

We run the same simulation for scenario of the year 2025

and scenario of the year 2030 Fig 4 and Fig 5 show the

performance of the model for the two scenarios In the

scenario of the year 2025, the model almost reaches the

optimum after 72 iterations, while it takes 84 iterations to

reach optimum in the scenario of the year 2030

Summary results for all scenarios are shown in Table V

Thus, it can be seen that when the penetration of wind power

plants into the power system increases, the percentage of cost

saving in optimization performance increases

2025

2030 TABLE V SIMULATION RESULTS FOR ALL SCENARIOS

Scenario

Total optimized wind capacity installation

Operation cost (USD)

Percent-age of cost saving

Before optimization

After optimization

Scenario of the year

2020

780 2,694,233 2,618,864.5 2.80% Scenario of

the year

2025

1,980 2,694,233 2,506,265.5 6.98% Scenario of

the year

2030

6,000 2,694,233 2,159,906.8 19.83%

In the scenario of the year 2030, when the installed capacity of wind power plants reaches 6,000 MW, the optimal economic operation of wind power plants help reducing the total operation cost nearly 20%, showing the importance of the integration of wind power plants and other renewable energy sources into the Vietnamese power system

Table VI summarized the result of the optimal location and installation capacity of wind power plants at buses in the Vietnamese power system in the three mentioned scenarios

TABLE VI OPTIMAL PLACEMENT AND INSTALLATION

CAPACITY OF WIND FARMS

Bus

Wind power installation capacity (MW)

Scenario of the year 2020

Iteration

2.52

2.54

2.56

2.58

2.6

Iteration

2.16 2.165 2.17 2.175 2.18 2.185 2.19 2.195 2.2

Trang 5

Wind power installation capacity (MW)

Scenario of the

year 2020

Tot

Fig 6, Fig 7 and Fig 8 show the voltage diagrams at buses

in the mentioned power system corresponding to the three

scenarios It can be seen that the voltage at buses in the power

system is within the allowable limits with the participation of

wind power plants Moreover, the voltage difference between

the buses tends to decrease thanks to the participation of these

wind farms

Fig 6 Buses voltage in scenario of the year 2020

IV CONCLUSION

In the paper, a model to optimize the location and installation capacity of the wind farms has been built in order

to optimally integrate wind power plants into the power system The optimization model was developed based on the binary particle swarm optimization algorithm, combined with selective algorithms to optimize the operating costs of power plants in the system The proposed model has been used to calculate and test on simplified 500kV power system in

Bus

1 1.01 1.02 1.03 1.04 1.05 1.06

Before wind power installation After wind power installation

Bus

1 1.01 1.02 1.03 1.04 1.05 1.06 1.07

Buses voltage

Before wind power installation After wind power installation

Trang 6

Vietnam The collected data has been used to build a compact

grid consisting of 31 nodes and 41 branches This grid then

has been tested with the proposed model with 03 scenarios

corresponding to the different penetration level of wind power

plants according to the Revised National Power Development

Master Plan for the 2011-2020 period with the vision to 2030

The program has calculated and provided the results of the

optimal location as well as the optimal installation capacity of

wind power plants at each bus in the power system in

accordance to the objective function, while ensuring the

constraints on power balance in the power system, ensuring

compliance with the transmission limits of 500kV lines and

ensuring the voltage at the buses within the allowed limits

The simulation results show that the greater the degree of

penetration of wind energy sources into the power system, the

higher the economic efficiency of the power plants operation,

while the voltage stability of the buses is maintained

The research results as well as the model proposed in the

paper can be applied to the optimal placement and sizing of

wind farms and can be useful for the power system

development planning and approval

REFERENCES

[1] The Vietnamese Prime Minister, “Approval of the Revised National

Power Development Master Plan for the 2011-2020 Period with the

Vision to 2030”, Decision No 428/QD-TTg dated March 18, 2016

[2] The Vietnamese Prime Minister, “Amending several articles of

Decision no.37/2011/QD-TTg dated June 29, 2011 of the Prime

Minister on provision of assistance in development of wind power

projects in Vietnam”, Decision No 39/2018/QD-TTg dated September

10, 2018

[3] S Frank, S Rebennack, "An introduction to optimal power flow:

Theory, formulation, and examples", IIE Transactions, vol 48, no 12,

pp 1172-1197, 2016

[4] J Kennedy, R Eberhart, “Particle Swarm Optimization”, Proceeding

IEEE International Conference on Neural Networks, vol 4, pp 1942–

1948, 1995

[5] J Kennedy, R Eberhart, “A decrete binary version of Particle Swarm

Algorithm”, Proceeding IEEE International Conference on System, pp

4104–4109, 1997

[6] T M Khalil, A V Gorpinich, “Selective Particle Swarm

Optimization”, International journal of multidisciplinary sciences and

engineering, vol 3, no 4, April 2012

[7] D T Viet, V V Phuong, M Q Duong, M P Khanh, A Kies, B

Schyska, "A Cost-Optimal Pathway to Integrate Renewable Energy

into the Future Vietnamese Power System", 4th International

Conference on Green Technology and Sustainable Development

(GTSD), pp 144-149, 2018

[8] A Kies, B Schyska, D T Viet, L Bremen, D Heinemann, S

Schramm, “Large-Scale Integration of Renewable Power Sources into

the Vietnamese Power System”, Energy Procedia, vol 125, pp 207–

213, 2017

[9] The Vietnamese Prime Minister, “Approving the development strategy

of renewable energy of Vietnam by 2030 with a vision to 2050”,

Decision No 2068/QD-TTg dated November 25, 2015

[10] D T Viet, V V Phuong, M Q Duong, A Kies, B Schyska, Y.K Wu,

“A Short-Term Wind Power Forecasting Tool for Vietnamese Wind

Farms and Electricity Market”, 4th International Conference on Green

Technology and Sustainable Development (GTSD), pp 130-135, 2018

[11] R D Zimmerman, C E Murillo-Sanchez, Matpower 6.0 User’s

Manual, December 2016

[12] N Q Khanh, N T Hang, N T Khanh, N T Hoan, “Analysis of future

generation capacity scenarios for Vietnam”, Green Innovation and

Development Centre (GreenID), 2018

[13] T P Nam, D T Viet, Development of distributed generation in power

market - Proc Conf UK-VN CECE, 2012

[14] M Q Duong, T V Dinh, H V P Nguyen, V T Nguyen, N T N Tran, T T M Le, Effects of FSIG and DFIG Wind Power Plants on

Ninh Thuan Power Grid, Vietnam, GMSARN International Journal,

2018

[15] D T Viet, P C Tien, Impact of Ancillary Services and Its Prices on Large-Scale Solar and Wind Power Penetration in Electricity Market,

4th International Conference on Green Technology and Sustainable Development (GTSD), pp 114-121, 2018

APPENDIX

TABLE VII NAMEOF31BUSES

Định dạng
Số trang	6
Dung lượng	336,7 KB