Energy storage plan optimization method for high-proportion new energy access power grid

Abstract

The invention discloses an energy storage plan optimization method for a high-proportion new energy access power grid, and the method comprises the following steps: constructing a simulation model of a target regional power grid, and carrying out the regional division of the target regional power grid according to the simulation result of the simulation model; a network loss cost objective function is constructed, a new energy consumption risk objective function is constructed according to a region division result, constraint conditions of new energy output and grid power are set, and a multi-objective optimization model giving consideration to the network loss cost and the new energy consumption risk is obtained; and solving an optimization problem for the multi-objective optimization model to obtain an optimal new energy output as an energy storage output pre-arrangement plan. The method comprehensively and scientifically arranges the energy storage plan, gives full play to the adjustment effect of energy storage on the power grid, and maximizes the absorption environment of new energy while reducing the loss of the power grid under the condition of guaranteeing the safety of power grid equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and particularly relates to a method for optimizing an energy storage plan for high-proportion new energy access to the power grid. Background Art

[0002] The access of high-proportion and multi-type new energy will change the original topology and operation characteristics of the power grid, making the power grid change from the original conventional power network to a multi-power network with complex operation characteristics, and the power flow direction changes. Due to the uncertainty of wind power output, superimposed with renewable energy such as photovoltaic and hydropower, for the areas where the power is sent out, if the grid structure is not adaptable to the access of new energy and cannot transport and consume enough energy, it will lead to a further increase in power grid losses.

[0003] As a high-quality flexibility regulation resource, the energy storage power station has the dual attributes of power source and load, and can balance the problems of randomness, volatility and intermittency of new energy, and improve the controllability and flexibility of the new power system. Therefore, when the new energy penetration rate increases year by year, it is necessary to study a new energy storage charging and discharging method to balance the power grid losses caused by the fluctuations of renewable energy while ensuring the new energy consumption capacity.

[0004] So far, domestic and foreign scholars have conducted pioneering research on energy storage. Some researchers have proposed a calling method for energy storage power stations to participate in peak shaving considering the access of high-proportion wind power to reduce the peak shaving cost. Some researchers have proposed a coordinated operation strategy for energy storage to track the new energy power generation plan, which can relieve the system peak shaving pressure while improving the energy storage utilization rate. The above literatures construct an optimal dispatching model for energy storage power stations to participate in system peak shaving from different angles. Some researchers have proposed a pricing method for shared energy storage packages based on the master-slave game, and established a two-layer optimal allocation model for shared energy storage master-slave games with the optimization of the strategies and the maximization of the interests of wind farms and shared energy storage operators as the goals. Some people output a frequency modulation output plan based on historical frequency statistical data, and establish a charging and discharging strategy model for industrial and commercial energy storage power stations that takes into account participating in frequency modulation ancillary services. Some researchers have proposed a dynamic shared energy storage capacity use right allocation method based on the time-varying capacity requirements of each MG, and improve the utilization rate of shared energy storage through the method of "time division multiplexing" of the shared energy storage capacity use right. Some researchers have constructed a full-life cycle cost model to comprehensively analyze the operating costs when controlling multi-type energy storage, and determine the energy storage composite control scheme under the scenario of suppressing wind power fluctuations with the goal of optimal economy. Some researchers have proposed an operation optimization strategy for industrial and commercial energy storage based on rolling optimization control of the planned curve, and effectively improve the energy storage utilization rate by adjusting the planned curve. However, the above studies have not considered the problem of using the regulation function of energy storage to suppress the power grid losses caused by new energy fluctuations from the perspective of grid operators based on the limited consumption capacity and power flow sending capacity of the actual new energy centralized sending areas.

[0005] The patent application with the publication number CN119448368A proposes a method for optimizing the configuration of large-capacity energy storage in zones for new energy consumption. The power grid model is divided into multiple zones according to the geographical and electrical characteristics of the selected power grid model. After determining the zoning of the power grid model, the demand for regional energy storage is determined, and an optimal scheduling model with the goal of maximizing new energy consumption and minimizing the total cost of the energy storage system is further constructed. This solution starts from the perspective of energy storage power station operators, fails to model based on the accurate parameters of the actual operating power grid, only takes energy storage operation and investment as the cost objective function, and does not consider the grid framework loss. Summary of the Invention

[0006] The technical problem to be solved by the present invention: Aiming at the above problems of the prior art, a method for optimizing the energy storage plan for high-proportion new energy access to the power grid is provided, with the grid loss cost and new energy consumption risk as multi-objective functions, and an optimized strategy for the energy storage plan is output, providing a reference plan for power grid new energy power generation planners from the perspective of power grid operators.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0008] A method for optimizing the energy storage plan for high-proportion new energy access to the power grid, including the following steps:

[0009] Construct a simulation model of the target regional power grid and conduct simulations, and divide the target regional power grid according to the simulation results;

[0010] Construct a grid loss cost objective function, and construct a new energy consumption risk objective function according to the regional division results. At the same time, set the constraints of new energy output and grid framework power to obtain a multi-objective optimization model that takes into account both grid loss cost and new energy consumption risk;

[0011] Solve the optimization problem of the multi-objective optimization model to obtain the optimal new energy output and grid framework power as the pre-arrangement plan for energy storage output.

[0012] Furthermore, when constructing a simulation model of the target regional power grid and conducting simulations, specifically, use the PSASP system to build the power grid model of the target region, then conduct a power flow calculation on the power grid model to obtain the power grid component parameters modeled by MATLAB. Finally, call the power grid component parameters modeled by MATLAB and obtain the new energy grid connection prediction curve to conduct a power flow calculation again.

[0013] Furthermore, when dividing the target regional power grid according to the simulation results, it includes the following steps:

[0014] Take the energy storage station in the target area power grid as the cluster center, determine the number of clusters, and calculate the electrical distance from each node in the target area power grid to the energy storage station according to the simulation results. Then, sort the electrical distances from each node to the energy storage station node from small to large;

[0015] Calculate the cluster power balance index, and dynamically divide the cluster area according to the cluster power balance index and the electrical distance. The expression is as follows:

[0016]

[0017] Among them, adjust is the energy storage area division optimization function, which dynamically reconstructs the cluster area division of new energy sites with main and backup power supply channels according to the real-time output of wind, light, and energy storage, so as to achieve the minimum value of the f min (·) function. k1 and k2 are assignment coefficients, and f min (·) is the minimum function for dynamically reconstructing the cluster area with respect to time t, the cluster power balance index x(t), and the electrical distance D(t). The expression of the cluster power balance index is:

[0018]

[0019] Among them, n k is the number of clusters k; P k，t is the net power of cluster k at time t; t is the system simulation time period. The expression of the electrical distance is as follows:

[0020]

[0021] Among them, n j is the number of nodes j; D((i,j), t) is the electrical distance between nodes i and j within the system simulation time period t.

[0022] Furthermore, the expression of the electrical distance between nodes i and j is as follows:

[0023] D ij =lg(S VQ (j,j) / S VQ (i,j))

[0024] Among them, D ij is the electrical distance between nodes i and j, and S VQ(i，j) is the element in the i-th row and j-th column of the reactive power-voltage sensitivity matrix S VQ . The expression of the reactive power-voltage sensitivity matrix is as follows:

[0025]

[0026] Among them, J Pθis the active power-voltage phase angle coefficient matrix in the simulation result; J PV is the active power-voltage amplitude coefficient matrix in the simulation result; J Qθ is the reactive power-voltage phase angle coefficient matrix in the simulation result; J QV is the reactive power-voltage amplitude coefficient matrix in the simulation result.

[0027] Furthermore, the expression of the network loss cost objective function is as follows:

[0028]

[0029] where n represents the number of system nodes; n k is the number of clusters k; P loss，ij is the loss of line ij between nodes i and j, and the expression is as follows:

[0030] P loss，ij = 2ViVjcosθ ij -G ij (V i 2 +V j 2 )

[0031] where V i and V j are the voltage amplitudes of node i and node j respectively, θ ij represents the phase angle difference generated between nodes i and j, and G ij is the conductance of the line between nodes i and j.

[0032] Furthermore, the expression of the new energy consumption risk objective function is as follows:

[0033]

[0034] where P c renew is the actual new energy generation power at node c where the new energy unit is located, k is the number of new energy units, P a w is the actual wind turbine generation power at node a where the wind turbine is located, i is the number of wind turbines; P b v is the actual photovoltaic array generation power at node b where the photovoltaic array is located, j is the number of photovoltaic arrays; P d B is the energy storage operation power at node d where the energy storage station is located; P e L is the actual operation power of the synchronous generator at node e where the synchronous generator is located, l is the number of synchronous generators;fload is the maximum load at node f where the load is located, and f is the number of load nodes; adjust is the regional division optimization function for regional division of the power grid in the target area.

[0035] Further, the constraint conditions of the new energy output and the grid power include the wind power and photovoltaic output constraints, and the expressions are as follows:

[0036]

[0037] Among them, P max w,t , P max v,t are the maximum wind power and photovoltaic installed capacities, and P w,t , P v,t are the wind power and photovoltaic installed capacities.

[0038] Further, the constraint conditions of the new energy output and the grid power include the energy storage constraints, and the expressions are as follows:

[0039]

[0040] S t,min ≤S t ≤S t,max

[0041] 0≤P sc,t ≤P sc,max

[0042] 0≤P sd,t ≤P sd,max

[0043] Among them, S t is the stored electricity, θ i is the self-loss rate, S t,min , S t,max are the upper and lower limits of the capacity, P sc,t , P sd,t are the charging power and the discharging power respectively, P sc,max , P sd,max are the maximum charging power and the maximum discharging power respectively, δ sc,t、 δ sd,t are the charging efficiency and the discharging efficiency respectively, and Δt is the specified time interval.

[0044] Further, the constraint conditions of the new energy output and the grid power include the transmission line power constraints, and the expressions are as follows:

[0045]

[0046] Among them, P is the transmission line power, P ijmax is the maximum transmission power of the line.

[0047] Furthermore, when solving the optimization problem for the multi-objective optimization model, specifically, an optimized multi-objective particle swarm algorithm is used to solve the multi-objective optimization model. In each iteration of the solution process, the result of the previous iteration and the new energy grid connection prediction curve are input into the simulation model for simulation, and the variables that violate the constraints are corrected according to the constraints of the new energy output. Then, the target regional power grid is re-divided based on the simulation results, and the new energy consumption risk objective function is updated according to the new regional division results. Finally, the network loss cost and the new energy consumption risk are calculated as the fitness of the particle according to the network loss cost objective function and the new energy consumption risk objective function, and sorted. After completing the maximum number of iterations, the Pareto optimal solution is obtained as the frontier solution of the network loss cost objective function and the new energy consumption risk objective function.

[0048] Compared with the prior art, the advantages of the present invention are as follows:

[0049] The present invention takes the network loss cost of the power grid and the new energy consumption risk as multi-objective functions, gives full play to the regulating role of energy storage on the power grid, and while ensuring the safety of power grid equipment, maximally broadens the new energy consumption environment and reduces the power grid loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is the flowchart of the embodiment of the present invention.

[0051] Figure 2 is the model structure diagram of the embodiment of the present invention.

[0052] Figure 3 is the schematic diagram of the solution result of the multi-objective optimization model in the embodiment of the present invention.

[0053] Figure 4 is the optimized energy storage operation plan diagram in the embodiment of the present invention.

[0054] Figure 5 is the comparison diagram of the system load before and after optimization in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The present invention will be further described below in conjunction with the accompanying drawings of the specification and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.

[0056] With the development of energy storage technology and in response to national policies, the number and capacity of energy storage power stations connected to the grid are gradually increasing. Affected by issues such as grid planning and connection locations, it is necessary to give full play to the spatio-temporal complementarity of new energy power plants, balance the adverse impacts of the randomness, volatility, and intermittency of new energy on the grid, and improve the new energy consumption capacity of the regional grid and the economic efficiency of the overall grid operation.

[0057] However, existing studies have not, from the perspective of grid operators, fully utilized the regulation function of energy storage to suppress the grid framework loss caused by new energy fluctuations under the condition that the new energy centralized sending area has limited consumption capacity and power flow sending capacity in the actual grid framework. As a grid operator, the focus is on the economic operation of the grid. How can the grid minimize the power transmission loss (grid framework loss) when accepting so much new energy? Secondly, how can new energy be maximally connected to the grid?

[0058] With the gradual increase in the new energy penetration rate, restricted by the grid structure, if the energy storage plan is not properly arranged, when the wind and light coincidence rate occurs and the energy storage discharges are superimposed, exceeding the safety and stability limits of grid equipment, it will cause incalculable losses to the grid. Therefore, for the planners of the new energy output plan of the grid, how to comprehensively and scientifically arrange the energy storage plan, give full play to the regulation role of energy storage on the grid, and while ensuring the safety of grid equipment, maximize the new energy consumption environment and reduce grid losses at the same time is an urgent problem to be solved.

[0059] To solve the above problems, this embodiment proposes an energy storage plan optimization method for a high-proportion new energy connected to the grid from the perspective of grid operators. It models the actual regional grid, sets up energy storage operation constraints after dividing the grid with energy storage as the center, takes the grid loss cost and new energy consumption risk as multi-objective functions, and outputs the optimization strategy of the energy storage plan to provide a reference plan for the formulators of the new energy power generation plan of the grid.

[0060] As Figure 1 shown, the method of this embodiment includes the following steps:

[0061] S1) Build a simulation model of the target regional grid and conduct simulations, and divide the target regional grid according to the simulation results;

[0062] S2) Build a grid loss cost objective function, and build a new energy consumption risk objective function according to the regional division results. At the same time, set the constraints of new energy output and grid framework power to obtain a multi-objective optimization model that takes into account both grid loss cost and new energy consumption risk;

[0063] S3) Solve the optimization problem of the multi-objective optimization model to obtain the optimal new energy output and grid framework power as the pre-arrangement plan for energy storage output.

[0064] The method of this embodiment is based on the joint simulation of PSASP and Matlab, and is based on the actual model of the regional power grid. The following will explain each step.

[0065] In step S1, when constructing the simulation model of the target regional power grid, the network structure diagram of the 220 kV voltage level in the target area is drawn through the geographical wiring diagram function in PSASP. Considering that the power flow calculation needs to be called multiple times in the subsequent optimization process and its calculation time has a great impact on the optimization speed, in order to ensure the calculation speed under multiple iterations, this embodiment extracts the power grid data and processes the external network in the form of equivalent machines. While ensuring that the simulation model conforms to the actual power grid as much as possible, the power flow calculation time is reduced.

[0066] In step S1, when performing the simulation, specifically perform power flow calculation on the power grid model to obtain the power grid component parameters modeled by MATLAB, and then call the power grid component parameters modeled by MATLAB and obtain the new energy grid-connected prediction curves containing wind, light, and load day-ahead prediction data from major platforms for MATLAB power flow calculation.

[0067] In order to achieve automatic data interaction between MATLAB and PSASP, this embodiment performs MATLAB programming by calling the PSASP power flow calculation data, extracts the PSASP power flow calculation result data and calculation results, so that the accurate power grid parameter database built by the PSASP system for the power grid model can be called by the MATLAB program.

[0068] After completing the call of the PSASP power flow calculation results, store them in the form of a matrix for easy reading and calculation processing of the power flow results by MATLAB. The form of the entire flowchart is as Figure 2 shown.

[0069] In step S1 of this embodiment, when dividing the target regional power grid according to the simulation results, specifically divide the region according to the power grid structure, equipment parameters, and the distribution location and installed capacity of new energy power sources, and optimize the active power output system within the region for the optimal problem of the active power flow of the power grid. At the same time, the regional division has the following several goals. (1) The new energy output and load within each region should be maintained as balanced as possible, so that the power of the tie lines connecting each region is maintained at a low level, reducing the dependence between regions. (2) There should be a strong electrical connection between the internal nodes of each region, so that the new energy power sources and reactive power sources within the region can efficiently regulate voltage and power. The specific steps are as follows:

[0070] S101) Theoretical calculation of electrical distance, including:

[0071] Taking the energy storage station in the target area power grid as the cluster center, determine the number of clusters, and calculate the electrical distance from each node in the target area power grid to the energy storage station according to the simulation results. Then, sort the electrical distances from each node to the energy storage station node from small to large;

[0072] The existing definition of electrical distance adopts the expression form of the distance between two points in analytic geometry to characterize the tightness of electrical connection between two nodes. Find the characteristic quantity of the influence degree of other nodes on the energy storage station node. Calculate the electrical distance by calculating the reactive power-voltage sensitivity matrix, and the expression is as follows:

[0073]

[0074] Among them, J Pθ is the active power-voltage phase angle coefficient matrix in the simulation result; J PV is the active power-voltage amplitude coefficient matrix in the simulation result; J Qθ is the reactive power-voltage phase angle coefficient matrix in the simulation result; J QV is the reactive power-voltage amplitude coefficient matrix in the simulation result.

[0075] The expression of the electrical distance between nodes i and j is as follows:

[0076] D ij =lg(S VQ (j,j) / S VQ (i,j))

[0077] Among them, D ij is the electrical distance between nodes i and j, S VQ(i，j) is the element in the i-th row and j-th column of the reactive power-voltage sensitivity matrix S VQ , and D ij reflects to a certain extent the influence of the voltage of node j on the voltage of node i. The greater the influence, the smaller the value of D ij . Sort the electrical distances from each node to the energy storage station node from small to large, and the expression of the electrical distance is as follows:

[0078]

[0079] Among them, n j is the number of nodes j; D((i,j), t) is the electrical distance between nodes i and j during the system simulation time period t;

[0080] S102) Calculate the cluster power balance index, including:

[0081] To avoid large-scale power transmission between clusters, the new energy output within the cluster should be made as equal as possible to the load demand, that is, the net power within the cluster should be as small as possible, so as to evaluate the acceptance capacity of new energy in each region. The expression of the cluster power balance index is as follows:

[0082]

[0083] Among them, n k is the number of clusters k; P k，t is the net power of cluster k at time t; t is the system simulation time period; the cluster power balance degree x aims to balance the net power complementarity level between nodes in the network, so as to give full play to the cluster's autonomous ability and effectively reduce the long-distance transmission of a large amount of power flow within the power grid, thereby reducing the network loss;

[0084] S104) Dynamically divide the cluster area according to the cluster power balance index and the electrical distance, and the expression is as follows:

[0085]

[0086] Among them, k1 and k2 are assignment coefficients, and f min (·) is the minimum function for dynamically reconstructing the cluster area with respect to time t, the cluster power balance index x(t), and the electrical distance D(t). The optimization function for the energy storage area division under high proportion of new energy access is represented by adjust. According to the real-time output of wind, light, and energy storage, dynamically reconstruct the cluster area division of new energy sites with main and backup supply channels to achieve the minimum value of the f min (·) function.

[0087] In this embodiment, the network loss cost and the new energy consumption risk are incorporated into the research of the energy storage operation strategy through step S2, and a multi-objective optimization model that can take into account the network loss cost and the new energy consumption risk under the coordinated operation of clean energy power stations and energy storage power stations is constructed.

[0088] For an operating power grid, under the condition of meeting the system operation constraints, considering the economy, reliability, and new energy consumption capacity of the power grid operation, a multi-objective optimization scheduling model that takes into account the network loss cost of the power grid and the new energy consumption risk is established.

[0089] As the penetration rate of new energy gradually increases, some regions have changed from the "load side" to the "power supply side" and bear a large amount of cross-network fees generated due to the inability of new energy to be locally consumed. Therefore, in this embodiment, the network loss cost is used as an important indicator of the economy of power grid operation. Under the condition of meeting the power grid operation constraints, the network loss cost is minimized, and the expression of the network loss cost objective function is as follows:

[0090]

[0091] Among them, n represents the number of system nodes; n k is the number of clusters k; P loss，ij is the loss of line ij between nodes i and j, and the expression is as follows:

[0092] P loss，ij = 2ViVjcosθ ij -G ij (V i 2 +V j 2 )

[0093] Among them, V i and V j are the voltage amplitudes of node i and node j respectively, θ ij represents the phase angle difference generated between nodes i and j, and G ij is the conductance of the line between nodes i and j.

[0094] In addition, when the new energy penetration rate is too high and the system's adjustable resource consumption capacity is insufficient, it will face a relatively large new energy consumption risk. The new energy consumption risk in this embodiment mainly considers the risks of light curtailment and wind curtailment. The expression of the new energy consumption risk objective function is as follows:

[0095]

[0096] Among them, P c renew is the actual new energy power generation of the new energy unit at node c where the new energy unit is located, k is the number of new energy units, P a w is the actual power generation of the wind turbine at node a where the wind turbine is located, i is the number of wind turbines; P b v is the actual power generation of the photovoltaic array at node b where the photovoltaic array is located, j is the number of photovoltaic arrays; P d B is the energy storage operation power at node d where the energy storage station is located; P e L is the actual operation power of the synchronous generator at node e where the synchronous generator is located, l is the number of synchronous generators; P fload is the maximum load at node f where the load is located, f is the number of load nodes; adjust is the regional division optimization function for regional division of the target regional power grid.

[0097] In step S2 of this embodiment, when setting the constraint conditions for new energy output and grid power, the regional network structure and power generation, supply, and consumption levels are specifically considered, and the constraint conditions for clean energy power stations, energy storage, and sub-region power grid operations are divided according to system requirements, including:

[0098] (1) Wind power and photovoltaic output constraints, with the expression as follows:

[0099]

[0100] Among them, P max w,t and P max v,t are the maximum installed capacities of wind power and photovoltaic power, and P w,t and P v,t are the installed capacities of wind power and photovoltaic power.

[0101] (2) Energy storage constraints, with the expression as follows:

[0102]

[0103] S t,min ≤S t ≤S t,max

[0104] 0 ≤ P sc,t ≤P sc,max

[0105] 0 ≤ P sd,t ≤P sd,max

[0106] Among them, S t is the stored electricity, θ i is the self-loss rate, S t,min and S t,max are the upper and lower limits of the capacity, P sc,t and P sd,t are the charging power and discharging power respectively, P sc,max and P sd,max are the maximum charging and discharging powers respectively, δ sc,t and δ sd,t are the charging efficiency and discharging efficiency respectively, and Δt is the specified time interval.

[0107] (3) Transmission line power constraints, with the expression as follows:

[0108]

[0109] Among them, P is the transmission line power, and P ij max is the maximum transmission power of the line.

[0110] In step S3 of this embodiment, when solving the optimization problem for the multi-objective optimization model, specifically, the multi-objective optimization model is solved using the multi-objective particle swarm optimization algorithm (MOPSO). In each iteration of the solution process, the result of the previous iteration and the new energy grid connection prediction curve are input into the simulation model for simulation. Variables that violate the constraints are corrected according to the constraints of new energy output. Then, the target regional power grid is re-divided according to the simulation results, and the new energy consumption risk objective function is updated according to the new regional division results. Finally, the network loss cost and new energy consumption risk are calculated as the fitness of the particle according to the network loss cost objective function and the new energy consumption risk objective function, and sorted. After completing the maximum number of iterations, the Pareto optimal solution is obtained as the frontier solution of the network loss cost objective function and the new energy consumption risk objective function.

[0111] Each objective function of the multi-objective optimization model represents the objective to be optimized. Usually, improving the value of one objective function may damage other objective functions. Therefore, the goal of MOPSO is to find a set of non-dominated solutions or Pareto optimal solutions on the basis of meeting the constraints, and these solutions cannot be outperformed by other solutions under all objective functions. The specific steps are as follows:

[0112] 1) Data initialization. Read the structure of the power grid, parameter model parameters, MOPSO algorithm parameters, etc. after external equivalence. At the same time, initialize the particle population, and each particle individual in the population corresponds to an energy storage operation plan scheme including new energy output and grid power in a scheduling period.

[0113] 2) Input the particle individual as a system variable into the simulation model. The simulation model calls the power grid component parameters modeled by MATLAB and obtains the particle individual for MATLAB power flow calculation. After correcting the variables that violate the constraints, the simulation model follows the energy storage operation plan scheme corresponding to the particle individual, and performs simulation according to the power grid component parameters and the new energy grid connection prediction curve. After re-dividing the target regional power grid according to the simulation results, the new energy consumption risk objective function is updated according to the expression of the new energy consumption risk objective function with the new regional division results, and the network loss cost and new energy consumption risk corresponding to the simulation results are calculated as the fitness of the particle according to the network loss cost objective function and the new energy consumption risk objective function, and sorted. These solutions are better than other solutions in all objective functions, and are set as the individual optimal position of the particle.

[0114] 3) Select non-dominated solutions from the individual optimal positions of all particles as the optimal position of the entire population.

[0115] 4) At the same time, update the velocity and position of the particle, set the dynamic inertia weight, and obtain the updated particle extreme value pbest.

[0116] 5) The Pareto optimal solutions saved using the external archive set are used to select the population extreme value gbest from the external set according to the crowding distance of the optimal solutions by the roulette wheel method.

[0117] 6) A small probability random mutation mechanism is introduced to generate a small probability perturbation of ±30% to the position of the particle on the original position, increasing the optimization ability of the particle for the global optimal front.

[0118] 7) Return to step (2) until the maximum number of iterations is completed, and output the final energy storage optimization result.

[0119] The method of this embodiment is verified through experiments below.

[0120] Through the geographical wiring diagram function in PSASP, the network structure diagram of the 220 kV voltage level in the target area is drawn, and the output power of various clean energy sources is obtained from the hydrological monitoring platform, the OMS system plan reporting platform, the provincial and local integrated platform, etc., and the regional load is obtained from the load forecasting platform. The relevant parameter values of MOPSO are: the particle swarm size is 100, the maximum number of iterations is 100, c1 is taken as 0.1, c2 is taken as 0.2, and the mutation probability is 0.1.

[0121] Using the MOPSO algorithm to iterate 100 times to generate 100 Pareto optimal solutions, the frontier solutions of the network loss cost and the new energy consumption risk are obtained as Figure 3 shown. Three optimization schemes are extracted from Figure 3 as shown in Table 1.

[0122] Table 1 Comparison of multi-objective optimization schemes

[0123] Scheme Network loss (10,000 kWh) Abandonment ratio of wind and light 1 19.51 0.2459 2 19.32 0.251 3 19.23 0.2568

[0124] According to Figure 3 it can be seen that there is no scheme that can simultaneously optimize the network loss and the wind and light consumption ratio. Among them, the network loss of the third scheme is the lowest, only 192,300 kW, but the corresponding new energy consumption risk is the largest; the first scheme is exactly the opposite, the new energy consumption risk is the smallest, but the corresponding network loss is the highest. Therefore, a suitable scheme should be selected in combination with the requirements of the wind and light abandonment ratio and the network loss to achieve the balance of the network loss cost and the wind and light abandonment ratio. In this embodiment, scheme 2 is selected, and the energy storage operation plan of scheme 2 is used as the pre-arrangement plan for the energy storage output. The energy storage operation plan under scheme 2 is as Figure 4 shown. The comparison of the system load curves before and after implementing scheme 2 is as Figure 5 shown. It can be seen that after implementing scheme 2, the system load at the same time point is significantly reduced, achieving the effect of energy storage optimization.

[0125] In summary, the present invention proposes an optimization method for energy storage planning in a power grid with a high proportion of new energy access. Based on the actual power grid model, energy storage is used as a flexible resource for the power grid, and the relationships among photovoltaic power, wind power, hydropower, energy storage, and load are fully considered to explore the coordinated optimal operation mode of the power grid under multiple objectives. On the basis of ensuring that the system operation constraints are met, a multi-objective optimal dispatching model of the power grid that combines network loss cost and new energy consumption risk is constructed, and a multi-objective particle swarm algorithm is used to solve this problem. The following beneficial effects are achieved:

[0126] 1. Through the combined simulation of PSASP and MATLAB, compared with the existing research schemes that use fixed models or IEEE node topology models in power grid modeling methods, it can accurately model based on the actual operating regional power grid.

[0127] 2. In terms of cluster division, the f min (·) function is adopted to dynamically reconstruct the cluster area according to time, the cluster power balance degree index, and the electrical distance. Combining the predicted wind and light output in advance, the cluster area division of new energy sites with main and backup power supply channels is dynamically reconstructed, expanding the new energy grid connection channels and optimizing the new energy grid connection path to the greatest extent.

[0128] 3. Combining the predicted wind, light, and load data in advance that can be obtained by the new energy output plan arranger of the power grid, and performing simulations based on the accurate power grid model of the region, a set of optimized energy storage pre-day plan formulation schemes is provided. The final output result is a set of pre-arranged energy storage output plans from 0 to 24 o'clock, providing a reference scheme for the formulator of the new energy power generation plan of the power grid.

[0129] The above is only the preferred embodiment of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A method for optimizing an energy storage plan for a power grid with a high proportion of new energy access, characterized in that It includes the following steps: Construct a simulation model of the target regional power grid and conduct simulations. According to the simulation results, divide the power grid in the target area; Construct a network loss cost objective function, and construct a new energy consumption risk objective function according to the regional division results. At the same time, set the constraints of new energy output and grid power, and obtain a multi-objective optimization model that takes into account network loss cost and new energy consumption risk; Solve the optimization problem for the multi-objective optimization model to obtain the optimal new energy output and grid power as the pre-arrangement plan for energy storage output.

2. The energy storage plan optimization method for high-proportion new energy access to the power grid according to claim 1, wherein When constructing a simulation model of the target regional power grid and conducting simulations, specifically, use the PSASP system to build the power grid model of the target area, then conduct a power flow calculation on the power grid model to obtain the power grid component parameters, and finally obtain the power grid component parameters and obtain the new energy grid connection prediction curve to conduct a power flow calculation again.

3. The energy storage plan optimization method for high proportion of new energy accessing the power grid according to claim 1, characterized in that, When dividing the power grid in the target area according to the simulation results, it includes the following steps: Take the energy storage stations in the target regional power grid as cluster centers, determine the number of clusters, and calculate the electrical distances from each node in the target regional power grid to the energy storage stations according to the simulation results. Then, sort the electrical distances from each node to the energy storage station nodes from small to large; Calculate the cluster power balance index, and dynamically divide the cluster area according to the cluster power balance index and the electrical distance. The expression is as follows: Among them, adjust is the optimization function for energy storage area division. According to the real-time output of wind, light, and energy storage, it dynamically reconstructs the cluster area division of new energy sites with main and backup power supply channels to achieve the minimum value of the f min (·) function, where k1 and k2 are assignment coefficients, and f min (·) is the minimum function for dynamically reconstructing the cluster area with respect to time t, the cluster power balance index x(t), and the electrical distance D(t). The expression for the cluster power balance index is as follows: where n k is the number of clusters k; P k，t is the net power of cluster k at time period t; t is the system simulation time period, and the expression of the electrical distance is as follows: where n j is the number of nodes j; D((i, j), t) is the electrical distance between nodes i and j during the system simulation time period t.

4. The energy storage plan optimization method for a power grid with a high proportion of new energy access according to claim 3, characterized in that, The expression of the electrical distance between nodes i and j is as follows: D ij = lg(S VQ (j, j) / S VQ (i, j)) Among them, D ij is the electrical distance between nodes i and j, and S VQ(i，j) is the element in the i-th row and j-th column of the reactive power-voltage sensitivity matrix S VQ . The expression of the reactive power-voltage sensitivity matrix is as follows: Among them, J Pθ is the active power-voltage phase angle coefficient matrix in the simulation result; J PV is the active power-voltage amplitude coefficient matrix in the simulation result; J Qθ is the reactive power-voltage phase angle coefficient matrix in the simulation result; J QV is the reactive power-voltage amplitude coefficient matrix in the simulation result.

5. The energy storage plan optimization method for a power grid with a high proportion of new energy access according to claim 1, characterized in that The expression of the network loss cost objective function is as follows: Among them, n represents the number of system nodes; n k is the number of clusters k; P loss，ij is the loss of line ij between nodes i and j, and the expression is as follows: P loss，ij = 2ViVjcosθ ij -G ij (V i 2 +V j 2 ) Among them, V i and V j are the voltage amplitudes of nodes i and j respectively, and θ ij represents the phase angle difference generated between nodes i and j, and G ij is the conductance of the line between nodes i and j.

6. The energy storage plan optimization method for high proportion of new energy accessing the power grid according to claim 1, characterized in that The expression of the new energy consumption risk objective function is as follows: Among them, P c renew is the actual power generation of new energy at node c where the new energy unit is located, k is the number of new energy units, P a w is the actual power generation of the wind turbine at node a where the wind turbine is located, i is the number of wind turbines; P b v is the actual power generation of the photovoltaic array at node b where the photovoltaic array is located, j is the number of photovoltaic arrays; P d B is the energy storage operation power at node d where the energy storage station is located; P e L is the actual operation power of the synchronous generator at node e where the synchronous generator is located, l is the number of synchronous generators; P fload is the maximum load at node f where the load is located, f is the number of load nodes; adjust is the regional division optimization function used to divide the power grid in the target area.

7. The energy storage plan optimization method for high proportion of new energy accessing the power grid according to claim 1, characterized in that The constraints of the new energy output and grid power include the wind power and photovoltaic output constraints. The expression is as follows: Among them, P max w,t , P max v,t are the maximum wind power and photovoltaic installed capacities, and P w,t , P v,t are the wind power and photovoltaic installed capacities.

8. The energy storage plan optimization method for a power grid with a high proportion of new energy access according to claim 1, characterized in that The constraints of the new energy output and grid power include the energy storage constraints. The expression is as follows: S t,min ≤ S t ≤ S t,max 0 ≤ P sc,t ≤ P sc,max 0 ≤ P sd,t ≤ P sd,max Among them, S t is the stored electricity, θ i is the self-loss rate, S t,min , S t,max are the upper and lower limits of the capacity, P sc,t , P sd,t are the charging power and the discharging power respectively, P sc,max , P sd,max are the maximum charging power and the maximum discharging power respectively, δ sc,t , δ sd,t are the charging efficiency and the discharging efficiency respectively, and Δt is the specified time interval.

9. The energy storage plan optimization method for high proportion of new energy accessing the power grid according to claim 1, characterized in that The constraints of the new energy output and grid power include the transmission line power constraints. The expression is as follows: Among them, P is the power of the transmission line, and P ij max is the maximum transmission power of the line.

10. The energy storage plan optimization method for a power grid with a high proportion of new energy access according to claim 1, characterized in that When solving the optimization problem for the multi-objective optimization model, specifically, use the optimized multi-objective particle swarm algorithm to solve the multi-objective optimization model. During the solution process, in each iteration, input the results of the previous iteration and the new energy grid connection prediction curve into the simulation model for simulation, correct the variables that violate the constraints according to the constraints of the new energy output and grid power, then re-divide the power grid in the target area according to the simulation results and update the new energy consumption risk objective function according to the new regional division results. Finally, calculate the network loss cost and new energy consumption risk as the fitness of the particles according to the network loss cost objective function and the new energy consumption risk objective function, and conduct sorting. After completing the maximum number of iterations, obtain the Pareto optimal solution as the frontier solution of the network loss cost objective function and the new energy consumption risk objective function.

Images