Multi-objective optimization method for energy storage configuration of power distribution network with high-proportion renewable energy access

Abstract

The invention provides a high-proportion renewable energy access power distribution network energy storage configuration multi-objective optimization method, and belongs to the field of multi-objective optimization configuration.According to the method, a multi-objective optimization model is constructed, and four objectives of network loss, abandoned electricity quantity, renewable energy consumption rate and power grid income are comprehensively considered; an optimal energy storage system configuration scheme is solved by using an improved grey wolf algorithm, and specifically, non-dominated sorting, reference point selection and a simulated binary crossover mechanism (SBX) are introduced into multi-objective optimization, so that the global search capability and the local search precision of optimization are enhanced; an optimization result shows that compared with a traditional configuration scheme, the method has the advantages that the renewable energy consumption rate is remarkably increased, the abandoned electricity quantity is reduced, the network loss is reduced, the power grid income is increased, and the practicability and the economical efficiency are high.

Description

Technical Field

[0001] The invention relates to a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, and belongs to the field of multi-objective optimization configuration. Background Art

[0002] With the rapid development of renewable energy, especially wind and solar energy, the power grid is facing the challenges of unstable energy supply and large load fluctuations. The high proportion of renewable energy access has brought complexity to the operation of the power grid. Energy storage technology, as one of the key means to solve this problem, can smooth out the load fluctuations of the power grid and improve the dispatch flexibility of the power grid. However, most of the existing energy storage configuration methods only optimize a single goal, such as minimizing network losses or maximizing the renewable energy consumption rate, and fail to effectively consider multiple optimization goals and power grid constraints.

[0003] Therefore, how to comprehensively optimize the network losses, power abandonment, renewable energy consumption rate and grid revenue of the power grid under the background of a high proportion of renewable energy access is an important issue facing the current optimization of distribution network energy storage systems. Summary of the invention

[0004] According to the problem described in the background, the problem to be solved by the present invention is: to provide a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, and to improve the operating efficiency of the distribution network with a high proportion of renewable energy access by optimizing and improving the multi-objective grey wolf algorithm.

[0005] To achieve the above object, the present invention provides the following technical solution: a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access is provided, comprising the following steps:

[0006] (1) Data collection;

[0007] (2) Construct a multi-objective optimization model;

[0008] (3) Improved multi-objective grey wolf algorithm;

[0009] (4) The results are verified by improving the multi-objective grey wolf algorithm.

[0010] Preferably, the data collected in step (1) are load data of the distribution network, renewable energy power generation data and parameters of the energy storage device.

[0011] Preferably, the optimization objectives in step (2) include minimizing network losses, maximizing grid revenue, maximizing renewable energy consumption rate and minimizing power abandonment; in order to ensure that the optimization result meets the actual grid operation requirements, step (2) sets three constraints: power balance constraint, voltage constraint and capacity and power constraint of energy storage system.

[0012] Preferably, the improved multi-objective grey wolf algorithm in step (3) includes: a non-dominated sorting strategy based on a sequential search strategy; a selection strategy based on a reference point; a simulated binary crossover mechanism (SBX); and a coefficient vector reduction strategy.

[0013] Preferably, the step (4) compares the optimization result with the traditional configuration scheme to verify its effectiveness in improving the renewable energy absorption rate, reducing the amount of abandoned electricity, reducing network losses and increasing grid revenue.

[0014] Preferably, the optimization objective function for minimizing network loss is expressed by the following formula:

[0015]

[0016] Among them, P loss is the network loss, I i is the current of the ith line, R i is the resistance of the i-th circuit;

[0017] The optimization objective function for maximizing the power grid revenue is expressed by the following formula:

[0018]

[0019] Among them, E profit is the grid revenue, P charge (t) and P discharge (t) are the energy storage charging and discharging power at time t, Price peak and Price valley are the peak and valley electricity prices of the power grid respectively;

[0020] The optimization objective function for maximizing the renewable energy consumption rate is expressed by the following formula:

[0021]

[0022] Among them, η RE is the renewable energy consumption rate, P RE (t) is the power generated by renewable energy, P dump (t) is the amount of abandoned electricity. The optimization objective function for minimizing the amount of abandoned electricity is expressed by the following formula:

[0023] P dump =max(0,P RE -P TR )

[0024] Among them, P TR Transmit power to the distribution network.

[0025] Preferably, the power balance constraint function is expressed by the following formula:

[0026] P grid (t)+P RE (t)+P discharge (t) = P load (t)+P charge (t)

[0027] Among them, P grid (t) is the power generation at time t, P load (t) is the load power at time t;

[0028] The voltage constraint function is expressed by the following formula:

[0029]

[0030] Among them, U min and U max is the upper and lower limits of the grid voltage, U i is the voltage of the i-th node;

[0031] The capacity and power constraint functions of the energy storage system are expressed by the following formula:

[0032]

[0033] 0≤E(i)≤E max

[0034] Among them, P charge (i) and P discharge (i) are the charging and discharging powers of the i-th energy storage device, and The maximum powers for charging and discharging respectively; the charging and discharging power and energy storage capacity of the energy storage equipment should be within the specified maximum range to avoid exceeding the physical and technical limitations of the equipment.

[0035] Preferably, in the non-dominated sorting strategy based on the sequential search strategy, the number of individuals in the non-dominated sorting method is determined by the number of the dominated individuals:

[0036]

[0037] Where: front(p) is the number of the frontier where individual p is located; q is the set composed of dominant individuals p;

[0038] In the reference point selection strategy, the additional points corresponding to each objective function are calculated:

[0039]

[0040] ASF(s,w),w=(τ,…,τ),τ=10 -6

[0041] The i-th target will have an additional target vector The m additional vectors can form an m-dimensional hyperplane, based on which the intercept a can be calculated i (i=1,2,…,m), the objective function can be expressed as:

[0042]

[0043] The simulated binary crossover mechanism (SBX) is more likely to generate new individuals by simulating binary crossover in the early stage of iterative operation, thereby enhancing the search ability of the algorithm. As the iterative operation progresses to the later stage, the method of generating new individuals is more likely to be the hunting mechanism of the gray wolf algorithm, thereby enhancing the convergence performance of the algorithm;

[0044] In the coefficient vector decreasing strategy, the coefficient vector a changes from linear decrease to nonlinear decrease, and the calculation formula is as follows:

[0045] a=2(1-(t / t max ) 2 )

[0046] Where: t is the current number of iterations; t max The maximum number of iterations set.

[0047] Preferably, the calculation steps of the improved multi-objective grey wolf algorithm include the following steps:

[0048] (3.1) Set the number of individuals in the population N and the maximum number of iterations t max , initialize a, A, C and population P, and generate reference point Z;

[0049] (3.2) Calculate the target values ​​of each dimension of the individual, and select the top three levels of α, β, and δ wolves using a non-dominated sorting strategy based on a sequential search strategy;

[0050] (3.3) Update the population Pt and generate the offspring population Qt. If the random number is greater than the upper limit of the set value, perform binary crossover mutation, otherwise update the population according to the original mechanism of the gray wolf algorithm;

[0051] (3.4) The updated population is R t =(P t ∪Q t ), generate a new population St in Rt through a selection strategy based on reference points;

[0052] (3.5) If the number of iterations is less than t max , then return to step 2 to continue iterating, otherwise output the current population as the optimization result.

[0053] The beneficial effects of the present invention are as follows: the energy storage system configuration scheme under the optimization algorithm significantly improves the renewable energy consumption rate of the regional power grid, reduces the amount of abandoned electricity, reduces network losses, increases power grid revenue, realizes the optimal configuration of the regional power grid energy storage system, improves the economy and stability of system operation, and verifies the practicality and effectiveness of the optimization algorithm for the optimal configuration of energy storage for the access of a high proportion of renewable energy to the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of the method of the present invention;

[0055] Figure 2 This is the flow chart for solving the improved multi-objective grey wolf algorithm;

[0056] Figure 3 The optimization results of the energy storage configuration scheme and the simulation results of the original configuration scheme;

[0057] Figure 4 The transmission power after energy storage configuration and the transmission power under the original scheme configuration;

[0058] Figure 5 The wind and solar power consumption curve after energy storage configuration and the wind and solar power output curve under the original scheme configuration;

[0059] Figure 6 The distributed energy storage output diagrams are for PV 8 and PV 9 and for wind power 2 and PV 12;

[0060] Figure 7 The network loss after energy storage configuration and the network loss under the original solution configuration;

[0061] Figure 8 It is a time-of-use electricity price list;

[0062] Fig. 9 Parameters related to the energy storage device; DETAILED DESCRIPTION

[0063] The embodiments of the present invention are further described below in conjunction with the accompanying drawings:

[0064] Example 1

[0065] like Figure 1 As shown, the present invention comprises the following steps:

[0066] (1) Data collection;

[0067] The collected data include load data of distribution network, renewable energy generation data and parameters of energy storage equipment.

[0068] (2) Construct a multi-objective optimization model;

[0069] The optimization objectives include minimizing network losses, maximizing grid revenue, maximizing renewable energy consumption rate, and minimizing power curtailment;

[0070] Minimizing network losses reduces the losses during power transmission by optimizing the grid line configuration and energy storage system capacity. The optimization objective function is expressed by the following formula:

[0071]

[0072] Among them, P loss is the network loss, I i is the current of the ith line, R i is the resistance of the i-th circuit;

[0073] Maximizing the benefits of the power grid Through the charging and discharging operations of energy storage equipment, the peak-valley electricity price difference is optimized to maximize the benefits of the power grid. The optimization objective function is expressed by the following formula:

[0074]

[0075] Among them, E profit is the grid revenue, P charge (t) and P discharge (t) are the energy storage charging and discharging power at time t, Price peak and Price valley are the peak and valley electricity prices of the power grid respectively;

[0076] Maximizing the renewable energy consumption rate can absorb more renewable energy through the energy storage system and reduce the amount of power wasted. The optimization objective function is expressed by the following formula:

[0077]

[0078] Among them, η RE is the renewable energy consumption rate, P RE (t) is the power generated by renewable energy, P dump (t) is the amount of power wasted.

[0079] Minimizing the amount of power curtailment reduces the amount of power curtailment caused by line transmission limitations by optimizing the energy storage capacity and configuration location. The optimization objective function is expressed by the following formula:

[0080] P dump =max(0,P RE -P TR )

[0081] Among them, P TRTransmit power to the distribution network;

[0082] In order to ensure that the optimization results meet the actual grid operation requirements, three constraints are set: power balance constraint, voltage constraint, and capacity and power constraint of energy storage system;

[0083] The power balance constraint ensures that the power input and output of the power grid are balanced in each period. The function is expressed by the following formula:

[0084] P grid (t)+P RE (t)+P discharge (t) = P load (t)+P charge (t)

[0085] Among them, P grid (t) is the power generation at time t, P load (t) is the load power at time t;

[0086] The voltage constraint ensures that the grid voltage is maintained within a safe range. The function is expressed by the following formula:

[0087]

[0088] Among them, U min and U max is the upper and lower limits of the grid voltage, U i is the voltage of the i-th node;

[0089] The capacity and power constraints of the energy storage system limit the charging and discharging power and energy storage capacity range of the energy storage system. The function is expressed by the following formula:

[0090]

[0091] 0≤E(i)≤E max

[0092] Among them, P charge (i) and P discharge (i) are the charging and discharging powers of the i-th energy storage device, and The maximum powers for charging and discharging respectively; the charging and discharging power and energy storage capacity of the energy storage equipment should be within the specified maximum range to avoid exceeding the physical and technical limitations of the equipment.

[0093] (3) Improved multi-objective grey wolf algorithm;

[0094] The improved multi-objective grey wolf algorithm includes:

[0095] A non-dominated sorting strategy based on a sequential search strategy is used to sort the solution set and select the optimal solution;

[0096] A reference point selection strategy is used to optimize the diversity of solutions and select the solution with the best fitness;

[0097] Simulated binary crossover mechanism (SBX) for generating new solutions in the population and enhancing global search capabilities;

[0098] The coefficient vector decreasing strategy gradually reduces the coefficient vector a to improve the convergence and local search accuracy of the algorithm in the later stage;

[0099] In the non-dominated sorting strategy based on the sequential search strategy, the number of individuals in the non-dominated sorting method is determined by the number of the dominated individuals:

[0100]

[0101] Where: front(p) is the number of the frontier where individual p is located; q is the set composed of dominant individuals p;

[0102] In the reference point selection strategy, the additional points corresponding to each objective function are calculated:

[0103]

[0104] ASF(s,w),w=(τ,…,τ),τ=10 -6

[0105] The i-th target will have an additional target vector The m additional vectors can form an m-dimensional hyperplane, based on which the intercept a can be calculated i (i=1,2,…,m), the objective function can be expressed as:

[0106]

[0107] If the intercept does not exist, the intercept is directly set to the maximum value under the target. Select the reference line closest to the normalized individual from the reference lines composed of reference points. The reference points with fewer corresponding individuals are more likely to be retained;

[0108] Simulated binary crossover mechanism (SBX): In the early stage of iterative operation, it is more likely to generate new individuals by simulating binary crossover, thereby enhancing the algorithm's search ability. As the iterative operation progresses to the later stage, the method of generating new individuals is most likely the hunting mechanism of the gray wolf algorithm, thereby enhancing the convergence performance of the algorithm;

[0109] In the coefficient vector decreasing strategy, the coefficient vector a changes from linear decrease to nonlinear decrease, and the calculation formula is as follows:

[0110] a=2(1-(t / t max ) 2 )

[0111] Where: t is the current number of iterations; t max The maximum number of iterations set.

[0112] The multi-objective gray wolf algorithm may fall into the local optimum during the calculation process. At the same time, the operation stability is poor, the solution distribution space is poor, and the non-dominated sorting and Archive update in the calculation process take a lot of time. Therefore, the multi-objective gray wolf algorithm is improved, such as Figure 2 As shown, the calculation steps of the improved multi-objective grey wolf algorithm include the following steps:

[0113] (3.1) Set the number of individuals in the population N and the maximum number of iterations t max , initialize a, A, C and population P, and generate reference point Z;

[0114] (3.2) Calculate the target values ​​of each dimension of the individual, and select the top three levels of α, β, and δ wolves using a non-dominated sorting strategy based on a sequential search strategy;

[0115] (3.3) Update the population Pt and generate the offspring population Qt. If the random number is greater than the upper limit of the set value, perform binary crossover mutation, otherwise update the population according to the original mechanism of the gray wolf algorithm;

[0116] (3.4) The updated population is R t =(P t ∪Q t ), generate a new population St in Rt through a selection strategy based on reference points;

[0117] (3.5) If the number of iterations is less than t max , then return to step 2 to continue iterating, otherwise output the current population as the optimization result.

[0118] like Figure 3-Figure 9 As shown, step (4) verifies the result by improving the multi-objective grey wolf algorithm;

[0119] Taking the actual regional power grid of Nancun Town, Pingdu as an example, 6 substations, 12 photovoltaic power stations and their associated electricity users, and 3 wind power stations in the region are selected as research objects, and a 21-node diagram of the regional power grid is constructed. Taking into account the economic and technical characteristics of energy storage equipment, as well as the investment willingness of energy storage construction companies, this paper sets the number of accessible energy storage devices to 2 and the number of energy storage devices allowed to be connected to each node to 1. The time-of-use electricity price used in the solution process is as follows: Figure 8 As shown, the relevant parameters of the energy storage device are as follows Fig. 9 shown.

[0120] By solving the improved multi-objective grey wolf algorithm model and combining the typical design scheme of energy storage equipment, the energy storage configuration scheme of the 21-node regional power grid is obtained as follows:

[0121] The photovoltaic 8 site is set to have a power of 630kW and a battery capacity of 2260kWh, and the photovoltaic 9 site is set to have a power of 800kW and a battery capacity of 2525kWh. The specific optimization effects are as follows: Figure 3 As shown above;

[0122] After configuring the energy storage device, the transmission power to the main grid is as follows: Figure 4 As shown on the left;

[0123] After configuring the energy storage device, the wind and solar power consumption is as follows Figure 5 As shown on the left;

[0124] The output diagram of the energy storage device at PV 8 and PV 9 is as follows Figure 6 As shown above;

[0125] The system network loss after configuring energy storage is shown in the figure below: Figure 7 As shown above;

[0126] In order to verify the practicality and effectiveness of the optimization algorithm, a 21-node network diagram identical to the actual regional power grid was built in the MATLAB simulation software, and the original configuration scheme and system parameters were input for simulation. The regional power grid optimization results are shown in the figure below. Figure 3 As shown below;

[0127] The transmission power under the original configuration is as follows Figure 4 As shown on the right;

[0128] The wind and solar power consumption under the original scheme configuration is as follows Figure 5 As shown on the right;

[0129] The output diagram of the original configuration of 2 wind power stations and 12 photovoltaic stations is as follows: Figure 6 As shown below;

[0130] The system network loss diagram under the original solution configuration is as follows Figure 7 As shown below;

[0131] Comparison of configuration scheme results shows that under the energy storage configuration strategy obtained by the optimization algorithm, the daily revenue of the power grid is increased by 240 yuan, the network loss is reduced by 324kWh, the on-site consumption rate of renewable energy is increased by 5.21%, and the power abandonment rate is reduced by 1.53%.

[0132] The optimization results were compared with the traditional configuration scheme, verifying their effectiveness in improving the renewable energy consumption rate, reducing power abandonment, reducing network losses and increasing grid revenue.

Claims

1. A multi-objective optimization method for energy storage configuration in a distribution network with a high proportion of renewable energy access, characterized in that: The following steps are involved: (1) Data collection; (2) Construct a multi-objective optimization model; (3) Improved multi-objective grey wolf algorithm; (4) The results are verified by improving the multi-objective grey wolf algorithm.

2. According to the multi-objective optimization method for energy storage configuration of distribution network with high proportion of renewable energy access as described in claim 1, it is characterized in that: The data collected in step (1) are load data of the distribution network, renewable energy generation data and parameters of the energy storage device.

3. According to the multi-objective optimization method for distribution network energy storage configuration with high proportion of renewable energy access as described in claim 1, it is characterized in that: The optimization objectives in step (2) include minimizing network losses, maximizing grid revenue, maximizing renewable energy consumption rate, and minimizing power abandonment; In order to ensure that the optimization result meets the actual grid operation requirements, step (2) sets three constraints: power balance constraint, voltage constraint, and capacity and power constraint of the energy storage system.

4. According to claim 1, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: The improved multi-objective grey wolf algorithm in step (3) includes: Non-Dominated Sorting strategy based on sequential search strategy; Selection strategy based on reference points; Simulated Binary Crossover (SBX); Coefficient vector reduction strategy.

5. According to claim 1, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: The step (4) compares the optimization results with the traditional configuration scheme to verify the effectiveness in improving the renewable energy consumption rate, reducing the amount of power wasted, reducing network losses and increasing grid revenue.

6. According to claim 3, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: The optimization objective function of minimizing network loss is expressed by the following formula: Among them, P loss is the network loss, I i is the current of the ith line, R i is the resistance of the ith circuit; The optimization objective function for maximizing the power grid revenue is expressed by the following formula: Among them, E profit is the grid revenue, P charge (t) and P discharge (t) are the energy storage charging and discharging power at time t, Price peak and Price valley are the peak and valley electricity prices of the power grid respectively; The optimization objective function for maximizing the renewable energy consumption rate is expressed by the following formula: Among them, η RE is the renewable energy consumption rate, P RE (t) is the power generated by renewable energy, P dump (t) is the amount of power wasted. The optimization objective function for minimizing the amount of power wasted is expressed by the following formula: P dump =max(0,P RE -P TR ) Among them, P TR Transmit power to the distribution network.

7. According to claim 3, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: The power balance constraint function is expressed by the following formula: P grid (t)+P RE (t)+P discharge (t)=P load (t)+P charge (t) Among them, P grid (t) is the power generation at time t, P load (t) is the load power at time t; The voltage constraint function is expressed by the following formula: Among them, U min and U max is the upper and lower limits of the grid voltage, U i is the voltage of the i-th node; The capacity and power constraint functions of the energy storage system are expressed by the following formula: 0≤E(i)≤E max Among them, P charge (i) and P discharge (i) are the charging and discharging powers of the i-th energy storage device, and The maximum powers for charging and discharging respectively; the charging and discharging power and energy storage capacity of the energy storage equipment should be within the specified maximum range to avoid exceeding the physical and technical limitations of the equipment.

8. According to claim 4, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: In the non-dominated sorting strategy based on the sequential search strategy, the number of individuals in the non-dominated sorting method is determined by the number of the dominated individuals: Where: front(p) is the number of the frontier where individual p is located; q is the set composed of dominant individuals p; In the reference point selection strategy, the additional points corresponding to each objective function are calculated: The i-th target will have an additional target vector The m additional vectors can form an m-dimensional hyperplane, based on which the intercept a can be calculated i (i=1,2,...,m), the objective function can be expressed as: The simulated binary crossover mechanism (SBX) is more likely to generate new individuals by simulating binary crossover in the early stage of iterative operation, thereby enhancing the search ability of the algorithm. As the iterative operation progresses to the later stage, the method of generating new individuals is more likely to be the hunting mechanism of the gray wolf algorithm, thereby enhancing the convergence performance of the algorithm; In the coefficient vector decreasing strategy, the coefficient vector a changes from linear decrease to nonlinear decrease, and the calculation formula is as follows: a=2(1-(t / t max ) 2 ) Where: t is the current number of iterations; t max The maximum number of iterations is set.

9. According to claim 4, a multi-objective optimization method for energy storage configuration of a distribution network with a high proportion of renewable energy access, characterized in that: The calculation steps of the improved multi-objective grey wolf algorithm include the following steps: (3.1) Set the number of individuals in the population N and the maximum number of iterations t max , initialize a, A, C and population P, and generate reference point Z; (3.2) Calculate the target values ​​of each dimension of the individual, and select the top three levels of α, β, and δ wolves using a non-dominated sorting strategy based on a sequential search strategy; (3.3) Update the population Pt and generate the offspring population Qt. If the random number is greater than the upper limit of the set value, perform binary crossover mutation, otherwise update the population according to the original mechanism of the gray wolf algorithm; (3.4) The updated population is R t =(P t ∪Q t ), generate a new population St in Rt through a selection strategy based on reference points; (3.5) If the number of iterations is less than t max , then return to step 2 to continue iterating, otherwise output the current population as the optimization result.

Images