An optimal configuration method for distributed energy storage in distribution networks

Abstract

The present invention relates to the technical field of power systems. An embodiment discloses a method for optimizing the configuration of distributed energy storage in a distribution network, comprising: dividing historical actual power data of distributed photovoltaics and loads into time windows to determine a probability density function of each time window, performing power sampling on the probability density function of each time window to determine a time-sampled power curve, determining investment cost, peak-shaving and valley-filling benefits, and network loss cost based on the time-sampled power curve, as well as an installation location and an installation capacity, and taking a target value of an objective function among reducing the investment cost, peak-shaving and valley-filling benefits, and network loss cost as an optimization target. While satisfying preset constraints, the installation location and the installation capacity are adjusted multiple times to obtain an optimal target value, so as to use the optimal installation location and installation capacity as the optimized configuration. This method comprehensively considers the uncertainty of renewable energy power generation and load, and can effectively improve the support capacity and economy of distributed energy storage.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for optimizing the configuration of distributed energy storage in a distribution network. Background Art

[0002] The uncertainty of distributed generation systems, such as distributed photovoltaics and loads, places significant pressure on the stable operation of the power grid, leading to a series of problems such as reverse flow, low voltage, and increased peak-to-valley differences. To more effectively manage these distributed resources, distributed energy storage systems are attracting widespread attention as a flexible and controllable supplementary energy solution. Distributed energy storage can absorb excess electricity, balance power supply and demand, and thus improve the stability and reliability of power system operations.

[0003] However, the current distributed energy storage optimization configuration technology still has some imperfections, mainly because the uncertainty between distributed renewable energy generation and load is not fully considered, resulting in low economy of energy storage configuration and low support capacity of energy storage system. Summary of the Invention

[0004] Based on this, it is necessary to propose an optimal configuration method for distributed energy storage in distribution networks to address the above problems. By comprehensively considering the uncertainty of renewable energy power generation and load, the distributed energy storage is optimized, which can effectively improve the support capacity and economy of distributed energy storage.

[0005] To achieve the above objectives, the present invention provides, in a first aspect, a method for optimizing the configuration of distributed energy storage in a distribution network, the method comprising:

[0006] Obtaining the installation location and installed capacity of distributed energy storage, and historical actual power data of targets, where the targets include distributed photovoltaics and loads;

[0007] Dividing the historical actual power data into time windows to obtain a plurality of time windows, and determining a probability density function of each time window based on a plurality of actual power data points in each time window;

[0008] Power sampling is performed on the probability density function of each time window to obtain a plurality of sampling power data points of each time window, and a time sampling power curve is determined based on the plurality of sampling power data points of all time windows;

[0009] Determining the investment cost, peak shaving and valley filling benefits, and network loss cost of the distributed energy storage according to the time-sampled power curve, the installation location, and the installation capacity;

[0010] Taking the target value of the objective function among reducing the investment cost, the peak shaving and valley filling benefit, and the network loss cost as the optimization target, the installation location and the installation capacity are adjusted multiple times while satisfying preset constraints to obtain the optimal target value of the objective function;

[0011] The installation location and installation capacity corresponding to the optimal target value are used as the optimized configuration of the distributed energy storage.

[0012] Optionally, performing power sampling on the probability density function of each time window to obtain multiple sampled power data points for each time window, and determining a time sampling power curve based on the multiple sampled power data points for all time windows includes:

[0013] Determine the covariance between the power variables of the pairwise time windows according to the window lengths of the pairwise time windows and a preset value;

[0014] Determine the covariance matrix based on the covariance between the power variables of all pairwise time windows;

[0015] Performing power sampling on the power density function of each time window according to the covariance matrix to obtain a plurality of sampled power data points of each time window;

[0016] determining an actual average power in each time window based on a plurality of actual power data points in each time window;

[0017] Determine the window index based on the actual average power of all time windows and multiple sampled power data points;

[0018] Within a first preset range, randomly adjusting the window length of each time window divided by the time window multiple times to obtain multiple window indicators corresponding to the multiple adjustments;

[0019] Among multiple window indices, the smallest window indices is used as the optimal window indices, and the time window corresponding to the optimal window indices is divided as the optimized time window;

[0020] A time-sampled power curve is determined based on the plurality of sampled power data points of all optimized time windows.

[0021] Optionally, determining the covariance between power variables in the pairwise time windows according to the window lengths of the pairwise time windows and a preset value includes:

[0022] Using the formula Determine the covariance between power variables in pairwise time windows;

[0023] The expression of the covariance matrix is:

[0024] ;

[0025] in, is the covariance between the power variable of the a-th time window and the power variable of the b-th time window, is the window length of the a-th time window, is the window length of the b-th time window, is the preset value, is the covariance matrix, is the d-th time window, The total number of time windows divided.

[0026] Optionally, determining the window index according to the actual average power of all time windows and a plurality of sampled power data points includes:

[0027] Using the formula determining the window indicator;

[0028] in, is the window index, is the total number of time windows divided, is the total number of sampled power data points in the dth time window, is the sth sampling power data point among the multiple sampling power data points in the dth time window, is the actual average power in the dth time window.

[0029] Optionally, determining a time sampling power curve according to a plurality of sampling power data points of all optimized time windows includes:

[0030] Determine multiple sampling cumulative difference values ​​according to multiple sampling power data points of all pairwise optimization time windows;

[0031] Determine the actual cumulative difference value based on the actual average power of all pairwise optimization time windows;

[0032] Determining the degree of compliance with the scene rule according to the actual cumulative difference value and multiple sampled cumulative difference values;

[0033] Within the second preset range, adjusting the preset value according to a preset step size to obtain a plurality of scene regularity compliances corresponding to different preset values;

[0034] Among multiple scene rule compliances, the smallest scene rule compliance is taken as the optimal scene rule compliance;

[0035] Taking the multiple sampling power data points of each optimization time window corresponding to the optimal scene rule compliance as the optimal sampling average power of each optimization time window;

[0036] The time sampling power curve is determined according to the optimal sampling average power of all optimized time windows.

[0037] Optionally, determining a plurality of sampled cumulative difference values ​​based on a plurality of sampled power data points in all pairwise optimized time windows includes:

[0038] Using the formula Determine the cumulative difference of multiple samples;

[0039] Determining the actual cumulative difference value based on the actual average power of all pairwise optimization time windows includes:

[0040] Using the formula determining the actual cumulative difference;

[0041] The determining of the scene regularity compliance according to the actual cumulative difference value and the plurality of sampled cumulative difference values ​​includes:

[0042] Using the formula Determining compliance with the regularity of the scenario;

[0043] in, is the vth sample cumulative difference value among multiple sample cumulative difference values, is the total number of optimization time windows divided, is the vth sampling power data point among the multiple sampling power data points of the d+1th optimization time window, is the vth sampling power data point among the multiple sampling power data points of the dth optimization time window, is the actual cumulative difference, is the actual average power of the d+1th optimization time window, is the actual average power of the dth optimization time window, is the compliance degree of the scene rules, is the total number of sample cumulative differences.

[0044] Optionally, the objective function is expressed as:

[0045] ;

[0046] in, is the target value of the objective function, is the investment cost, For the peak shaving and valley filling benefits, is the network loss cost, is the energy storage quantity of the distributed energy storage, is the unit cost of the distributed energy storage configuration power, is the configured power of the kth energy storage in the distributed energy storage, is the unit cost of the configuration capacity of the distributed energy storage, is the configuration cost of the kth energy storage in the distributed energy storage, is the discount rate of the distributed energy storage, is the service life of the kth energy storage in the distributed energy storage, is the life span of the kth energy storage in the distributed energy storage, is the total number of days in a year, The typical value of is 365. is the total peak shaving duration of the distributed energy storage, is the peak shaving power unit benefit of the distributed energy storage at time t, is the peak shaving power of the kth energy storage in the distributed energy storage at the tth moment, is the total valley-filling time of the distributed energy storage, is the unit valley-filling power benefit of the distributed energy storage at time t, is the valley-filling power of the kth energy storage in the distributed energy storage at the tth moment, is the total number of nodes of the mth bus, is the unit cost of network loss power of the distributed energy storage, is the network loss power of the distributed energy storage between the i-th node and the j-th node of the m-th bus.

[0047] Optionally, the preset constraints include power flow constraints and energy storage charging and discharging constraints;

[0048] The expression of the power flow constraint condition is:

[0049] ;

[0050] The expression of the energy storage charge and discharge constraint condition is:

[0051] ;

[0052] in, is the active power injected into the mth bus at the tth moment, is the active power injected into the i-th node of the m-th bus at the t-th moment in the time sampling power curve corresponding to the distributed photovoltaic system, is the active power injected into the i-th node of the m-th bus at the t-th moment in the time-sampled power curve corresponding to the load, is the total number of nodes of the mth bus, It is 1 or 0, 1 represents that the distributed energy storage is installed at the i-th node of the m-th bus, and 0 represents that the distributed energy storage is not installed at the i-th node of the m-th bus. is the active power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the voltage amplitude of the ith node of the mth bus at the tth moment in the time-sampled power curve corresponding to the load, is the voltage amplitude of the jth node of the mth bus at the tth moment in the time-sampled power curve corresponding to the load, is the branch admittance between the i-th and j-th nodes of the m-th busbar, is the cosine function, is the phase difference between the i-th and j-th nodes of the m-th bus, is a sine function, is the reactive power injected by the mth bus at time t, is the reactive power injected by the i-th node of the m-th bus at time t in the time sampling power curve corresponding to the distributed photovoltaic system, is the reactive power injected into the ith node of the mth bus at time t in the time-sampled power curve corresponding to the load, is the reactive power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the minimum voltage amplitude of the load at the i-th node of the m-th bus, is the maximum voltage amplitude of the load at the i-th node of the m-th bus, is the minimum active power of the line between the i-th node and the j-th node of the m-th bus, is the active power of the line between the ith and jth nodes of the mth bus at the tth moment, is the maximum active power of the line between the ith and jth nodes of the mth bus, is the state of charge of the kth energy storage in the distributed energy storage at the tth moment, is the state of charge of the kth energy storage in the distributed energy storage at time t-1, is the charge and discharge power of the kth energy storage in the distributed energy storage at the tth moment, is the charging time of the energy stored in the distributed energy storage, is the charging and discharging efficiency of the energy storage in the distributed energy storage, is the rated capacity of the kth energy storage in the distributed energy storage, is the current predicted capacity of the kth energy storage in the distributed energy storage, is the rated power of the kth energy storage in the distributed energy storage at the tth moment.

[0053] Optionally, the method further includes:

[0054] Obtain the remaining usable capacity, usage days, all-day charge and discharge power, and all-day charge and discharge average power of the kth energy storage in the distributed energy storage, as well as the average value and standard deviation of the all-day ambient temperature;

[0055] The remaining usable capacity, usage days, all-day charge and discharge power, and all-day charge and discharge average power of the k-th energy storage in the distributed energy storage, as well as the average value and standard deviation of the all-day ambient temperature are input into a preset energy storage capacity prediction model to obtain the remaining usable capacity of the k-th energy storage in the distributed energy storage for multiple days in the future;

[0056] The life usage limit of the kth energy storage in the distributed energy storage is determined based on the future multi-day remaining usage capacity and the life threshold of the kth energy storage in the distributed energy storage, or the current predicted capacity of the kth energy storage in the distributed energy storage is determined based on the future multi-day remaining usage capacity of the kth energy storage in the distributed energy storage.

[0057] Optionally, the probability density function of the dth time window is expressed as:

[0058] ;

[0059] in, is the probability density function of the d-th time window, is the power variable, is the total number of actual power data points in the dth time window, is the bandwidth, is the kernel function, is the wth actual power data point among the multiple actual power data points in the dth time window.

[0060] To achieve the above-mentioned object, the present invention provides, in a second aspect, a device for optimizing the configuration of distributed energy storage in a distribution network, the device comprising:

[0061] an acquisition module, configured to acquire the installation location and capacity of distributed energy storage, and historical actual power data of targets, wherein the targets include distributed photovoltaics and loads;

[0062] a division determination module, configured to divide the historical actual power data into time windows to obtain a plurality of time windows, and determine a probability density function of each time window based on a plurality of actual power data points in each time window;

[0063] A sampling determination module is used to perform power sampling on the probability density function of each time window to obtain multiple sampling power data points of each time window, and determine a time sampling power curve based on the multiple sampling power data points of each time window;

[0064] a determination module, configured to determine the investment cost, peak shaving and valley filling benefit, and network loss cost of the distributed energy storage according to the time-sampled power curve, the installation location, and the installation capacity;

[0065] an adjustment and optimization module, configured to take as an optimization target a target value of an objective function among reducing the investment cost, the peak shaving and valley filling benefit, and the network loss cost, and adjust the installation location and the installation capacity multiple times while satisfying preset constraints to obtain an optimal target value of the objective function;

[0066] The optimization configuration module is used to use the installation location and installation capacity corresponding to the optimal target value as the optimized configuration of the distributed energy storage.

[0067] To achieve the above-mentioned object, the present invention provides, in a third aspect, a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the method as described in any one of the first aspects.

[0068] To achieve the above-mentioned objectives, the present invention provides a computer device in a fourth aspect, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the method as described in any one of the first aspects.

[0069] The embodiment of the present invention has the following beneficial effects: the above method obtains the installation location and installation capacity of the distributed energy storage, and the historical actual power data of the target, wherein the target includes distributed photovoltaic and load, and then divides the historical actual power data into time windows to obtain multiple time windows, and determines the probability density function of each time window according to the multiple actual power data points of each time window, and then performs power sampling on the probability density function of each time window to obtain multiple sampled power data points of each time window, and determines the time sampling power curve according to the multiple sampled power data points of each time window, and then determines the investment cost, peak shaving and valley filling benefits and network loss cost of the distributed energy storage according to the time sampling power curve, as well as the installation location and installation capacity, and reduces the target function between the investment cost, peak shaving and valley filling benefits and network loss cost. The target value of the number is the optimization target. Under the condition of meeting the preset constraints, the installation position and installation capacity are adjusted multiple times to obtain the optimal target value of the objective function. Finally, the installation position and installation capacity corresponding to the optimal target value are used as the optimized configuration of distributed energy storage; that is, by obtaining the historical actual power data of distributed photovoltaic and load, integrating the probability density function and time sampling technology, it is possible to obtain the time sampling power curve that simulates the distributed photovoltaic power generation and load changes in different time periods, and comprehensively consider the uncertainty of new energy power generation and load, so as to accurately determine the investment cost, peak shaving and valley filling benefits and network loss cost of distributed energy storage, and on this basis, with reducing the investment cost, peak shaving and valley filling benefits and network loss cost of distributed energy storage as the optimization target, thereby achieving the optimal configuration of distributed energy storage, which can effectively improve the support capacity and economy of distributed energy storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0071] in:

[0072] Figure 1 A schematic diagram of a method for optimizing configuration of distributed energy storage in a distribution network according to an embodiment of the present application;

[0073] Figure 2 A schematic diagram of an optimized configuration illustrated in an embodiment of the present application;

[0074] Figure 3 A schematic diagram of an optimized configuration device for distributed energy storage in a distribution network according to an embodiment of the present application;

[0075] Figure 41 is a diagram of the internal structure of a computer device in some embodiments. DETAILED DESCRIPTION

[0076] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0077] The uncertainty of distributed generation systems, such as distributed photovoltaics and loads, places significant pressure on the stable operation of the power grid, leading to a series of problems such as reverse flow, low voltage, and increased peak-to-valley differences. To more effectively manage these distributed resources, distributed energy storage systems are attracting widespread attention as a flexible and controllable supplementary energy solution. Distributed energy storage can absorb excess electricity, balance power supply and demand, and thus improve the stability and reliability of power system operations.

[0078] However, the current distributed energy storage optimization configuration technology still has some imperfections, mainly because the uncertainty between distributed renewable energy generation and load is not fully considered, resulting in low economy of energy storage configuration and low support capacity of energy storage system.

[0079] In response to the above problems, this application proposes a method for optimizing the configuration of distributed energy storage in the distribution network. By comprehensively considering the uncertainty of renewable energy power generation and load, the distributed energy storage is optimized, which can effectively improve the support capacity and economy of distributed energy storage. The specific implementation principle will be described in detail in the following embodiments.

[0080] In a first aspect, the present application provides a method for optimizing the configuration of distributed energy storage in a distribution network.

[0081] See also Figure 1 , is a schematic diagram of a method for optimizing configuration of distributed energy storage in a distribution network according to an embodiment of the present application, the method comprising:

[0082] Step 110: Obtain the installation location and installation capacity of the distributed energy storage, as well as the historical actual power data of the target, where the target includes distributed photovoltaics and loads.

[0083] The installation location and installation capacity of the distributed energy storage here are initial values, which need to be adjusted and optimized during the optimization configuration of the distributed energy storage. Therefore, they can be set by the operator according to actual needs and are not limited here.

[0084] Step 120: Divide the historical actual power data into time windows to obtain multiple time windows, and determine the probability density function of each time window based on multiple actual power data points in each time window.

[0085] Regarding the method of dividing the time window, in some embodiments, assuming that the historical actual power data is the power data of one year, the power data of each day can be randomly divided into time windows within the preset division range according to multiples of the preset division step to obtain multiple time windows; wherein the window length of each time window can be different, the preset division range can be [1, 24], and the preset division step can be 1 hour.

[0086] Regarding the determination method of the probability density function, in some embodiments, a non-parametric kernel density estimation method may be used to determine the probability density function of each time window based on multiple actual power data points in each time window.

[0087] Step 130: Power sampling is performed on the probability density function of each time window to obtain a plurality of sampled power data points of each time window, and a time sampling power curve is determined based on the plurality of sampled power data points of all time windows.

[0088] Regarding the power sampling method, in some embodiments, a Monte Carlo sampling method may be used to perform power sampling on the probability density function of each time window to obtain multiple sampled power data points for each time window.

[0089] Regarding the method for determining the time sampling power curve, in some embodiments, all sampling power data points corresponding to all time windows can be fitted to obtain the time sampling power curve; in other embodiments, the sampling average power of each time window can be determined based on multiple sampling power data points of each time window, and then the sampling average powers of all time windows can be fitted to obtain the time sampling power curve.

[0090] Step 140: Determine the investment cost, peak shaving and valley filling benefits, and network loss cost of the distributed energy storage based on the time-sampled power curve, the installation location, and the installation capacity.

[0091] It should be noted that the historical actual power data of this application includes the historical actual power data of distributed photovoltaics and the historical actual power data of loads. Correspondingly, the time-sampled power curve here also includes the time-sampled power curve of distributed photovoltaics and the time-sampled power curve of loads.

[0092] Furthermore, by comprehensively considering the uncertainty of distributed photovoltaics and loads, that is, combining the time-sampled power curve of distributed photovoltaics and the time-sampled power curve of loads to determine the investment cost, peak-shaving and valley-filling benefits, and network loss costs of distributed energy storage, the investment cost, peak-shaving and valley-filling benefits, and network loss costs of distributed energy storage can be accurately determined, avoiding the situation where the determined investment cost, peak-shaving and valley-filling benefits, and network loss costs are inconsistent with the actual situation due to the uncertainty of distributed photovoltaics and loads, thereby resulting in low economy of energy storage configuration and low support capacity of the energy storage system.

[0093] Step 150: Taking the target value of the objective function between reducing investment cost, peak shaving and valley filling benefits and network loss cost as the optimization target, the installation location and installation capacity are adjusted multiple times while satisfying the preset constraints to obtain the optimal target value of the objective function.

[0094] The preset constraints refer to the necessary conditions for maintaining the normal operation of the distribution network. The preset constraints can be obtained and set in advance by the operator based on a lot of experience, experiments or statistics.

[0095] It should be noted that, under the condition that the preset constraints are met, that is, under the necessary conditions for the normal operation of the distribution network, the investment cost, peak shaving and valley filling benefits and network loss costs of the distributed energy storage can be changed by adjusting the installation location of the distributed energy storage and the installation capacity corresponding to the installation location, thereby causing the target value of the objective function between the investment cost, peak shaving and valley filling benefits and the network loss cost to change accordingly. Since the optimization goal is to reduce the target value of the objective function between the investment cost, peak shaving and valley filling benefits and the network loss cost, multiple adjustments will also result in multiple target values ​​when the preset constraints are met. Among the multiple target values, the smallest target value can be taken as the optimal target value.

[0096] Step 160: The installation location and installation capacity corresponding to the optimal target value are used as the optimized configuration of the distributed energy storage.

[0097] It should be noted that, under the condition of satisfying the preset constraints, multiple target values ​​can be obtained by adjusting the installation location of the distributed energy storage and the installation capacity corresponding to the installation location multiple times. That is, each target value has a corresponding installation location and the installation capacity corresponding to the installation location. Therefore, after determining the optimal target value, the installation location and installation capacity corresponding to the optimal target value can be used as the optimized configuration of the distributed energy storage.

[0098] In the embodiment of the present application, by obtaining the historical actual power data of distributed photovoltaic and load, integrating the probability density function and time sampling technology, it is possible to obtain a time-sampled power curve that simulates the changes in distributed photovoltaic power generation and load in different time periods, which fully considers the uncertainty of new energy power generation and load, so as to accurately determine the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage. On this basis, the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage are optimized with the reduction of the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage as the optimization goal, thereby achieving the optimal configuration of distributed energy storage, which can effectively improve the support capacity and economy of distributed energy storage.

[0099] In addition, this method of the present application also has the following advantages: through the optimized configuration of the present application, it is possible to better cope with the volatility between distributed renewable energy power generation and load, thereby reducing the problems of low voltage, reverse flow and other problems caused by imbalance between supply and demand in the distribution network, and improving the stability and reliability of power supply; based on the probabilistic analysis of distributed photovoltaics and load fluctuations, a more intelligent and flexible grid dispatching plan can be formulated to improve the grid regulation efficiency, reduce the impact of dispatching difficulty and uncertainty on grid security; with the support of distributed energy storage, the urban distribution network can better adapt to the large-scale access of new energy such as distributed photovoltaics, enhance the flexibility and adaptability of the grid, and provide strong support for energy transformation; through accurate cost-benefit analysis, the present application can optimize the investment plan of energy storage, reduce unnecessary construction investment, and improve the operational economy of the entire distribution network. At the same time, it can effectively utilize the peak shaving and valley filling function to reduce the operating cost of the grid during peak hours and improve the overall economic benefits.

[0100] In a feasible implementation, step 130 in the above embodiment performs power sampling on the probability density function of each time window to obtain multiple sampled power data points for each time window, and determines a time sampling power curve based on the multiple sampled power data points of all time windows, including: determining the covariance between the power variables of each time window based on the window length of each time window and a preset value; determining a covariance matrix based on the covariance between the power variables of all time windows; performing power sampling on the power density function of each time window based on the covariance matrix to obtain multiple sampled power data points for each time window; determining the actual average power of each time window based on the multiple actual power data points of each time window; determining a window index based on the actual average power of all time windows and the multiple sampled power data points; within a first preset range, randomly adjusting the window length of each time window divided by the time window multiple times to obtain multiple window indexes corresponding to the multiple adjustments; among the multiple window indexes, taking the smallest window index as the optimal window index, and dividing the time window corresponding to the optimal window index as the optimized time window; and determining the time sampling power curve based on the multiple sampled power data points of all optimized time windows.

[0101] The preset value and the first preset range can be obtained by the operator through a large amount of actual experience, experiments or statistics and can be set in advance.

[0102] Regarding the setting method of the preset value, in some embodiments, the preset value can be set to any integer from 1 to 5 by default.

[0103] The setting method of the first preset range is similar to the preset division range in the above embodiment. In some embodiments, the first preset range can also be set to [1, 24].

[0104] It should be noted that the purpose of this application is to optimize the time window by randomly adjusting the window length of each time window divided within the first preset range multiple times to obtain the most reasonable time window division, that is, the optimal time window division.

[0105] In an embodiment of the present application, by randomly adjusting the window lengths of each time window of the time window division multiple times within a first preset range, and selecting the division method that minimizes the window index as the optimal time window division, the actual power data can be segmented more reasonably, and the characteristics of distributed photovoltaics and loads in different time periods can be better reflected. In this way, the load and power generation conditions in different time periods can be more accurately reflected, and the accuracy of the optimized configuration can be improved.

[0106] Furthermore, by adjusting the window length of the time window and calculating the covariance between the power variables in each time window, the correlation in different time periods can be comprehensively considered, which helps to more accurately simulate the changes in distributed photovoltaics and loads. Then, by performing power sampling based on a time window with an appropriate sample size, the time-sampled power curve can be determined more accurately, thereby more accurately predicting the investment cost, peak shaving and valley filling benefits, and network loss costs of distributed energy storage.

[0107] In a feasible implementation, the method of determining the covariance between power variables in the two time windows according to the window lengths of the two time windows and a preset value in the above embodiment includes:

[0108] Using the formula Determine the covariance between power variables in pairwise time windows;

[0109] The expression of the covariance matrix in the above embodiment is:

[0110] ;

[0111] in, is the covariance between the power variable of the a-th time window and the power variable of the b-th time window, is the window length of the a-th time window, is the window length of the b-th time window, is the default value, is the covariance matrix, is the d-th time window, The total number of time windows divided.

[0112] In the embodiments of the present application, a rigorous formula for calculating the covariance between power variables in two time windows is provided from a mathematical perspective. The accuracy of the calculated covariance can be ensured from the rigor of mathematical logic, and the above-mentioned calculation formula is preferably shown to provide reference, understanding and calculation for technical personnel.

[0113] In addition, by adopting the covariance calculation formula provided in the above-mentioned preferred embodiment of the present application to calculate the covariance, a more accurate covariance matrix can be obtained, thereby improving the optimization process of time window division, so that the optimized distributed energy storage configuration more accurately reflects the actual situation of distributed photovoltaics and loads, thereby improving the overall performance of distributed energy storage in the distribution network.

[0114] In a feasible implementation, the above embodiment determines the window index based on the actual average power of all time windows and multiple sampled power data points, including:

[0115] Using the formula Determine window indicators;

[0116] in, is the window indicator, is the total number of time windows divided, is the total number of sampled power data points in the dth time window, is the sth sampling power data point among the multiple sampling power data points in the dth time window, is the actual average power in the dth time window.

[0117] In the embodiments of the present application, a rigorous calculation formula for the window index is provided from a mathematical perspective. The accuracy of the calculated window index can be ensured from the rigor of mathematical logic, and the above calculation formula is preferably shown to provide reference, understanding and calculation for technical personnel.

[0118] In addition, by adopting the calculation formula of the window index preferably provided in the above-mentioned application to calculate the window index, the deviation between the sampled power data and the actual average power of each time window can be comprehensively considered, that is, the smaller the window index, the higher the degree of fit between the sampled power data and the actual power data, and the more accurately the actual changes in distributed photovoltaics and loads can be reflected, thereby making the distributed energy storage optimization configuration based on these data more accurate, and can more effectively improve the investment cost-effectiveness and system support capacity of energy storage, reduce network loss costs, and improve the overall stability and reliability of the distribution network.

[0119] Furthermore, the window indicator-based optimization strategy of the present application not only takes into account the randomness of short-term power changes, but also takes into account the smoothness of long-term trends, making the optimization results more practical and flexible, thereby providing more optimized and intelligent support for the operation of the distribution network.

[0120] In a feasible implementation method, the time sampling power curve is determined according to multiple sampling power data points of all optimized time windows in the above embodiment, including: determining multiple sampling cumulative difference values ​​according to multiple sampling power data points of all pairwise optimized time windows; determining the actual cumulative difference value according to the actual average power of all pairwise optimized time windows; determining the scene regularity compliance according to the actual cumulative difference value and the multiple sampling cumulative difference values; within a second preset range, adjusting the preset value according to a preset step size to obtain multiple scene regularity compliances corresponding to different preset values; among the multiple scene regularity compliances, taking the minimum scene regularity compliance as the optimal scene regularity compliance; taking the multiple sampling power data points of each optimized time window corresponding to the optimal scene regularity compliance as the optimal sampling average power of each optimized time window; and determining the time sampling power curve according to the optimal sampling average power of all optimized time windows.

[0121] The second preset range and the preset step size can be obtained by the operator through a large amount of actual experience, experiments or statistics and can be set in advance.

[0122] Regarding the setting method of the second preset range and the preset step size, in some embodiments, the second preset range can be set to [1, 5], and the preset step size can be set to 1.

[0123] It should be noted that the purpose of this application is to optimize the compliance between the sampling generation scenario and the actual scenario by adjusting the preset value according to the preset step size within the second preset range, so as to obtain the most regular sampling generation scenario, that is, the optimal sampling generation scenario; among which, the optimal sampling generation scenario is the time sampling power curve.

[0124] Regarding the method for determining the optimal sampling power curve, in some embodiments, the optimal sampling average power of all optimized time windows may be fitted to obtain the time sampling power curve.

[0125] In the embodiment of the present application, by considering time correlation and optimizing the conformity of the scene rules, it is ensured that the sampled generation scene can be closer to the actual scene, and the prediction accuracy of the sampled power data for the actual power data is improved. This not only makes the energy storage configuration more accurate, but also enhances the stability and reliability of the entire distribution network.

[0126] Furthermore, by adjusting the preset values ​​and optimizing the sampling process using the covariance matrix, the sampling accuracy is improved, making the actual power data and the sampled power data more consistent, thereby more accurately predicting the investment cost of distributed energy storage, peak shaving and valley filling benefits, and network loss costs.

[0127] In addition, by comprehensively considering the historical actual power data of distributed photovoltaics and loads, combined with probability density functions and time sampling technology, the power generation and consumption conditions in each time period can be simulated more accurately. This enables the configuration of distributed energy storage to adapt to this uncertainty while maximizing economic benefits.

[0128] In a feasible implementation, the method of determining multiple sampled cumulative difference values ​​based on multiple sampled power data points in all pairwise optimized time windows in the above embodiment includes:

[0129] Using the formula Determine the cumulative difference of multiple samples;

[0130] In the above embodiment, determining the actual cumulative difference value based on the actual average power of all pairwise optimization time windows includes:

[0131] Using the formula Determine the actual cumulative difference;

[0132] In the above embodiment, determining the scene regularity compliance based on the actual cumulative difference value and multiple sampled cumulative difference values ​​includes:

[0133] Using the formula Determine the compliance with the scene rules;

[0134] in, is the vth sample cumulative difference value among multiple sample cumulative difference values, is the total number of optimization time windows divided, is the vth sampling power data point among the multiple sampling power data points of the d+1th optimization time window, is the vth sampling power data point among the multiple sampling power data points of the dth optimization time window, is the actual cumulative difference, is the actual average power of the d+1th optimization time window, is the actual average power of the dth optimization time window, is the conformity of the scene rules, is the total number of sample cumulative differences.

[0135] In the embodiments of the present application, rigorous calculation formulas for the sampling cumulative difference value, the actual cumulative difference value and the compliance with the scene rules are provided from a mathematical perspective. The accuracy of the calculated sampling cumulative difference value, the actual cumulative difference value and the compliance with the scene rules can be ensured from the rigor of mathematical logic, and the above-mentioned calculation formulas are preferably shown to provide reference, understanding and calculation for technical personnel.

[0136] In addition, by adopting the calculation formulas for the sampled cumulative difference, actual cumulative difference and scene regularity compliance provided in the above-mentioned preferred embodiment of this application to calculate the sampled cumulative difference, actual cumulative difference and scene regularity compliance, it can be ensured that the sampled data is highly consistent with the actual power data, thereby improving the accuracy and practicality of the distributed energy storage optimization configuration and providing a solid foundation for the stable operation of the distribution network.

[0137] In a feasible implementation, the objective function in the above embodiment is expressed as:

[0138] ;

[0139] in, is the target value of the objective function, is the investment cost, To reduce peaks and fill valleys, is the network loss cost, is the amount of energy stored in distributed energy storage, is the unit cost of distributed energy storage configuration power, is the configured power of the kth energy storage in the distributed energy storage, is the unit cost of distributed energy storage configuration capacity, is the configuration cost of the kth energy storage in the distributed energy storage, is the discount rate of distributed energy storage, is the service life of the kth energy storage in the distributed energy storage, is the life span of the kth energy storage in the distributed energy storage, is the total number of days in a year, The typical value of is 365. is the total peak shaving duration of distributed energy storage, is the unit profit of peak shaving power of distributed energy storage at time t, is the peak shaving power of the kth energy storage in the distributed energy storage at the tth moment, is the total valley-filling time of distributed energy storage, is the unit valley-filling power benefit of distributed energy storage at time t, is the valley-filling power of the kth energy storage in the distributed energy storage at the tth moment, is the total number of nodes of the mth bus, is the unit cost of network loss power of distributed energy storage, is the network loss power of distributed energy storage between the i-th node and the j-th node on the m-th bus.

[0140] In the embodiments of the present application, a rigorous expression of the objective function is provided from a mathematical perspective. The accuracy of the expression of the objective function can be ensured from the rigor of mathematical logic, and the above expression is preferably shown to provide reference, understanding and calculation for technical personnel.

[0141] In addition, by using the expression of the objective function preferably provided in the above-mentioned application to calculate the target value and determine its optimal target value, it is possible to effectively guide the optimal configuration of distributed energy storage, thereby achieving a dual improvement in the economic benefits and technical performance of the distributed energy storage system.

[0142] It is understandable that the design of the objective function of this application takes into account the core aspects of energy storage cost-benefit analysis, namely investment cost, peak shaving and valley filling benefits, and network loss costs. By accurately mathematically modeling and calculating these factors, the objective function can effectively guide the optimal configuration of distributed energy storage, thereby achieving a dual improvement in the economic benefits and technical performance of the distributed energy storage system.

[0143] In a feasible implementation, the preset constraints in the above embodiment include power flow constraints and energy storage charging and discharging constraints.

[0144] The expression of the power flow constraint condition is:

[0145] ;

[0146] The expression of energy storage charging and discharging constraint condition is:

[0147] ;

[0148] in, is the active power injected into the mth bus at the tth moment, is the active power injected into the i-th node of the m-th bus at time t in the time sampling power curve corresponding to the distributed photovoltaic system, is the active power injected into the ith node of the mth bus at time t in the time sampling power curve corresponding to the load, is the total number of nodes of the mth bus, It is 1 or 0, 1 means that the distributed energy storage is installed at the i-th node of the m-th bus, and 0 means that the distributed energy storage is not installed at the i-th node of the m-th bus. is the active power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the voltage amplitude of the ith node of the mth bus at the tth moment in the time-sampled power curve corresponding to the load, is the voltage amplitude of the jth node of the mth bus at the tth moment in the time sampling power curve corresponding to the load, is the branch admittance between the i-th and j-th nodes of the m-th busbar, is the cosine function, is the phase difference between the i-th and j-th nodes of the m-th bus, is a sine function, is the reactive power injected by the mth bus at time t, is the reactive power injected by the i-th node of the m-th bus at time t in the time sampling power curve corresponding to the distributed photovoltaic system, is the reactive power injected into the ith node of the mth bus at time t in the time sampling power curve corresponding to the load, is the reactive power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the minimum voltage amplitude of the load at the i-th node of the m-th bus, is the maximum voltage amplitude of the load at the i-th node of the m-th bus, is the minimum active power of the line between the i-th node and the j-th node of the m-th bus, is the active power of the line between the ith and jth nodes of the mth bus at the tth moment, is the maximum active power of the line between the ith and jth nodes of the mth bus, is the state of charge of the kth energy storage in the distributed energy storage at the tth moment, is the state of charge of the kth energy storage in the distributed energy storage at the time t-1, is the charging and discharging power of the kth energy storage in the distributed energy storage at the tth moment, is the charging time of distributed energy storage, is the charging and discharging efficiency of energy storage in distributed energy storage, is the rated capacity of the kth energy storage in the distributed energy storage, is the current predicted capacity of the kth energy storage in the distributed energy storage, is the rated power of the kth energy storage in the distributed energy storage at the tth moment.

[0149] It should be noted that It is 1 or 0, 1 represents that the distributed energy storage has installed energy storage at the i-th node of the m-th bus, and 0 represents that the distributed energy storage has not installed energy storage at the i-th node of the m-th bus. If the distributed energy storage has installed energy storage at the i-th node of the m-th bus, the i-th node of the m-th bus is the installation position, and the installation capacity of the installation position here can also be adjusted. Therefore, the present application can change the installation position by adjusting 0 or 1, and in the case of 1, the installation capacity corresponding to the installation position can also be adjusted. Then, during the adjustment process, as long as the flow constraints and energy storage charging and discharging constraints are met, it can be ok.

[0150] In the embodiments of the present application, rigorous expressions of power flow constraints and energy storage charge and discharge constraints are provided from a mathematical perspective. The accuracy of the expressions of power flow constraints and energy storage charge and discharge constraints can be ensured from the rigor of mathematical logic, and the above expressions are preferably shown to provide reference, understanding and calculation for technical personnel.

[0151] In addition, by adopting the expressions of the flow constraints and energy storage charge and discharge constraints provided in the above-mentioned preferred embodiment of the present application, it is possible to combine the flow constraints and the energy storage charge and discharge constraints to optimize the installation location and capacity of the distributed energy storage, thereby ensuring both the economy of the energy storage system and the stability of the system. This not only improves the flexibility and responsiveness of the distribution network, but also ensures the smooth operation of the distributed power generation system and improves the overall performance of the distribution network.

[0152] It is understandable that by setting power flow constraints, it is possible to ensure that the installation location of distributed energy storage ensures grid stability while reasonably distributing electricity, avoiding line overload problems caused by improper energy storage installation, thereby avoiding grid safety risks caused by instability during power transmission; by setting energy storage charge and discharge constraints, it is ensured that each energy storage can be charged and discharged according to its maximum design capacity while meeting grid demand, which not only improves the economic benefits of energy storage, but also ensures the normal operation of its charge and discharge cycle.

[0153] In a feasible implementation, the method in the above embodiment further includes: obtaining the remaining usable capacity, number of usage days, daily charge and discharge power, and daily average charge and discharge power of the kth energy storage in the distributed energy storage, as well as the average value and standard deviation of the daily ambient temperature; inputting the remaining usable capacity, number of usage days, daily charge and discharge power, and daily average charge and discharge power of the kth energy storage in the distributed energy storage, as well as the average value and standard deviation of the daily ambient temperature into a preset energy storage capacity prediction model to obtain the future multi-day remaining usable capacity of the kth energy storage in the distributed energy storage; determining the life usage limit of the kth energy storage in the distributed energy storage based on the future multi-day remaining usable capacity of the kth energy storage in the distributed energy storage and a life threshold, or determining the current predicted capacity of the kth energy storage in the distributed energy storage based on the future multi-day remaining usable capacity of the kth energy storage in the distributed energy storage.

[0154] The preset energy storage capacity prediction model here refers to a trained deep learning model that can be directly used to predict the remaining usage capacity for multiple days in the future based on the input remaining usage capacity, number of usage days, all-day charging and discharging power and all-day charging and discharging average power, as well as the average and standard deviation of the all-day ambient temperature.

[0155] In some embodiments, the remaining usage capacity, number of days of use, daily charge and discharge power, and average power of charge and discharge, as well as the average value and standard deviation of the ambient temperature of each energy storage that has been used, and the future multi-day remaining usage capacity of each energy storage that has been used can be obtained, and then the remaining usage capacity, number of days of use, daily charge and discharge power, and average power of charge and discharge, as well as the average value and standard deviation of the ambient temperature of each energy storage that has been used, and the future multi-day remaining usage capacity of each energy storage that has been used can be input into the initial energy storage capacity prediction model for training. After training to a certain extent, a trained preset energy storage capacity prediction model can be obtained; wherein, the future multi-day remaining usage capacity of each energy storage that has been used can be used as the true value of the initial energy storage capacity prediction model, that is, by comparing the true value with the future multi-day remaining usage capacity output during the training process one by one, it can be determined whether the initial energy storage capacity prediction model is well trained and has met the expected requirements.

[0156] It should be noted that the remaining usage capacity for multiple days in the future refers to the remaining usage capacity for each day in the future. For example, there are R days in the future, so the remaining usage capacity for multiple days in the future includes the remaining usage capacity on the first day in the future, the remaining usage capacity on the second day in the future,..., the remaining usage capacity on the rth day in the future,..., the remaining usage capacity on the Rth day in the future.

[0157] Furthermore, if the remaining usage capacity on the next r day is less than or equal to the life threshold, it means that the energy storage has only r days left in its life. The life threshold can be set to 80% of the standard capacity set by the energy storage at the factory. Of course, it can also be set by the operator according to actual needs.

[0158] In the embodiments of the present application, the service life limit of energy storage is taken into consideration. That is, by combining the service life limit to optimize the configuration of distributed energy storage, not only can the flexibility and adaptability of the power grid be enhanced to better cope with the uncertainty of photovoltaic power generation and load demand, but also the operating efficiency and economic benefits can be further improved.

[0159] In a feasible implementation, the probability density function of the d-th time window in the above embodiment is expressed as:

[0160] ;

[0161] in, is the probability density function of the d-th time window, is the power variable, is the total number of actual power data points in the dth time window, is the bandwidth, is the kernel function, is the wth actual power data point among the multiple actual power data points in the dth time window.

[0162] In the embodiments of the present application, a rigorous expression of the probability density function is provided from a mathematical perspective. The accuracy of the expression of the probability density function can be ensured from the rigor of mathematical logic, and the above expression is preferably shown to provide reference, understanding and calculation for technical personnel.

[0163] In addition, by adopting the expression of the probability density function preferably provided in the above-mentioned application, a good balance can be achieved between flexibility and accuracy to cope with the uncertainty of distributed photovoltaic power generation and load changes, and it can also provide more accurate basic data support for the subsequent distributed energy storage optimization configuration based on time sampling, so that the optimization results are closer to actual operational needs.

[0164] In order to better understand the embodiments of the present application, this application provides a schematic diagram of the relevant application scenario, please refer to Figure 2, is a schematic diagram of the optimized configuration illustrated in an embodiment of the present application, in which 1 to 33 are respectively the 1st to 33rd nodes, PV1 to PV4 are respectively the 1st to 4th photovoltaics in the distributed photovoltaic, and ESS1 and ESS2 are respectively the installation locations of the 1st and 2nd energy storage after the distributed energy storage is optimized in the present application; wherein, the installation capacity corresponding to the installation location is not shown.

[0165] In a second aspect, the present application provides a device for optimizing the configuration of distributed energy storage in a distribution network.

[0166] See also Figure 3 , is a schematic diagram of an optimized configuration device for distributed energy storage in a distribution network according to an embodiment of the present application, wherein the device 310 includes:

[0167] An acquisition module 311 is configured to acquire the installation location and capacity of distributed energy storage, as well as historical actual power data of targets, where the targets include distributed photovoltaics and loads;

[0168] A division determination module 312 is configured to divide the historical actual power data into time windows to obtain a plurality of time windows, and determine a probability density function of each time window based on a plurality of actual power data points in each time window;

[0169] The sampling determination module 313 is used to perform power sampling on the probability density function of each time window to obtain multiple sampling power data points of each time window, and determine a time sampling power curve based on the multiple sampling power data points of each time window;

[0170] Determination module 314, for determining the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage based on the time-sampled power curve, as well as the installation location and installation capacity;

[0171] The adjustment and optimization module 315 is used to optimize the target value of the objective function between reducing investment cost, peak shaving and valley filling benefits, and network loss cost, and adjust the installation location and installation capacity multiple times while satisfying preset constraints to obtain the optimal target value of the objective function;

[0172] The optimization configuration module 316 is used to use the installation location and installation capacity corresponding to the optimal target value as the optimized configuration of the distributed energy storage.

[0173] In the embodiment of the present application, the relevant contents of the acquisition module 311, the division determination module 312, the sampling determination module 313, the determination module 314, the adjustment optimization module 315 and the optimization configuration module 316 can be referred to. Figure 1 The contents of the illustrated embodiments are not described in detail here.

[0174] It should be noted that the device 310 of the present application also includes some other modules. It can be understood that the method of the present application and the device 310 have a one-to-one correspondence. Therefore, the other modules of the device 310 of the present application are the contents corresponding to the method of the present application in the above embodiment.

[0175] In the embodiment of the present application, by obtaining the historical actual power data of distributed photovoltaic and load, integrating the probability density function and time sampling technology, it is possible to obtain a time-sampled power curve that simulates the changes in distributed photovoltaic power generation and load in different time periods, which fully considers the uncertainty of new energy power generation and load, so as to accurately determine the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage. On this basis, the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage are optimized with the reduction of the investment cost, peak shaving and valley filling benefits, and network loss cost of distributed energy storage as the optimization goal, thereby achieving the optimal configuration of distributed energy storage, which can effectively improve the support capacity and economy of distributed energy storage.

[0176] In addition, this method of the present application also has the following advantages: through the optimized configuration of the present application, it is possible to better cope with the volatility between distributed renewable energy power generation and load, thereby reducing the problems of low voltage, reverse flow and other problems caused by imbalance between supply and demand in the distribution network, and improving the stability and reliability of power supply; based on the probabilistic analysis of distributed photovoltaics and load fluctuations, a more intelligent and flexible grid dispatching plan can be formulated to improve the grid regulation efficiency, reduce the impact of dispatching difficulty and uncertainty on grid security; with the support of distributed energy storage, the urban distribution network can better adapt to the large-scale access of new energy such as distributed photovoltaics, enhance the flexibility and adaptability of the grid, and provide strong support for energy transformation; through accurate cost-benefit analysis, the present application can optimize the investment plan of energy storage, reduce unnecessary construction investment, and improve the operational economy of the entire distribution network. At the same time, it can effectively utilize the peak shaving and valley filling function to reduce the operating cost of the grid during peak hours and improve the overall economic benefits.

[0177] In a third aspect, the present application further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes a method for optimizing configuration of distributed energy storage in a distribution network in the above-mentioned method embodiment.

[0178] In a fourth aspect, the present application further provides a computer device including a memory and a processor, wherein the memory stores a computer program. When the computer program is executed by the processor, the processor executes a method for optimizing configuration of distributed energy storage in a distribution network in the above-mentioned method embodiment.

[0179] Figure 4 The internal structure diagram of the computer device in some embodiments is shown. The computer device can be a terminal, a server, or a gateway. Figure 4As shown, the computer device includes a processor, a memory, and a network interface connected via a system bus.

[0180] The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the various steps in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can implement the various steps in the above method embodiment. It will be understood by those skilled in the art that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0181] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0182] Among them, any reference to memory, storage, database or other media used in the various embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0183] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0184] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for optimizing the configuration of distributed energy storage in a distribution network, characterized in that: The method comprises: Obtaining the installation location and installed capacity of distributed energy storage, and historical actual power data of targets, where the targets include distributed photovoltaics and loads; Dividing the historical actual power data into time windows to obtain a plurality of time windows, and determining a probability density function of each time window based on a plurality of actual power data points in each time window; Power sampling is performed on the probability density function of each time window to obtain a plurality of sampling power data points of each time window, and a time sampling power curve is determined based on the plurality of sampling power data points of all time windows; Determining the investment cost, peak shaving and valley filling benefits, and network loss cost of the distributed energy storage according to the time-sampled power curve, the installation location, and the installation capacity; Taking the target value of the objective function among reducing the investment cost, the peak shaving and valley filling benefit, and the network loss cost as the optimization target, the installation location and the installation capacity are adjusted multiple times while satisfying preset constraints to obtain the optimal target value of the objective function; The installation location and installation capacity corresponding to the optimal target value are used as the optimized configuration of the distributed energy storage.

2. The method according to claim 1, characterized in that The power sampling of the probability density function of each time window to obtain a plurality of sampled power data points of each time window, and determining a time sampling power curve according to the plurality of sampled power data points of all time windows, comprises: Determine the covariance between the power variables of the pairwise time windows according to the window lengths of the pairwise time windows and a preset value; Determine the covariance matrix based on the covariance between the power variables of all pairwise time windows; Performing power sampling on the power density function of each time window according to the covariance matrix to obtain a plurality of sampled power data points of each time window; determining an actual average power in each time window based on a plurality of actual power data points in each time window; Determine the window index based on the actual average power of all time windows and multiple sampled power data points; Within a first preset range, randomly adjusting the window length of each time window divided by the time window multiple times to obtain multiple window indicators corresponding to the multiple adjustments; Among multiple window indices, the smallest window indices is used as the optimal window indices, and the time window corresponding to the optimal window indices is divided as the optimized time window; A time-sampled power curve is determined based on the plurality of sampled power data points of all optimized time windows.

3. The method according to claim 2, characterized in that The step of determining the covariance between power variables in the two time windows according to the window lengths of the two time windows and a preset value includes: Using the formula Determine the covariance between power variables in pairwise time windows; The expression of the covariance matrix is: ； in, is the covariance between the power variable of the a-th time window and the power variable of the b-th time window, is the a-th time window, is the window length of the a-th time window, is the b-th time window, is the window length of the b-th time window, is the preset value, is the covariance matrix, is the d-th time window, The total number of time windows divided.

4. The method according to claim 2, characterized in that Determining the window index based on the actual average power of all time windows and multiple sampled power data points includes: Using the formula determining the window indicator; in, is the window index, is the total number of time windows divided, is the total number of sampled power data points in the dth time window, is the sth sampling power data point among the multiple sampling power data points in the dth time window, is the actual average power in the dth time window.

5. The method according to claim 2, characterized in that The step of determining a time sampling power curve according to a plurality of sampling power data points of all optimized time windows includes: Determine multiple sampling cumulative difference values ​​according to multiple sampling power data points of all pairwise optimization time windows; Determine the actual cumulative difference value based on the actual average power of all pairwise optimization time windows; Determining the degree of compliance with the scene rule according to the actual cumulative difference value and multiple sampled cumulative difference values; Within the second preset range, adjusting the preset value according to a preset step size to obtain a plurality of scene regularity compliances corresponding to different preset values; Among multiple scene rule compliances, the smallest scene rule compliance is taken as the optimal scene rule compliance; Taking the multiple sampling power data points of each optimization time window corresponding to the optimal scene rule compliance as the optimal sampling average power of each optimization time window; The time sampling power curve is determined according to the optimal sampling average power of all optimized time windows.

6. The method according to claim 5, characterized in that The method of determining a plurality of sampled cumulative difference values ​​based on a plurality of sampled power data points in all pairwise optimized time windows includes: Using the formula Determine the cumulative difference of multiple samples; Determining the actual cumulative difference value based on the actual average power of all pairwise optimization time windows includes: Using the formula determining the actual cumulative difference; The determining of the scene regularity compliance according to the actual cumulative difference value and the plurality of sampled cumulative difference values ​​includes: Using the formula Determining compliance with the regularity of the scenario; in, is the vth sample cumulative difference value among multiple sample cumulative difference values, is the total number of optimization time windows divided, is the vth sampling power data point among the multiple sampling power data points of the d+1th optimization time window, is the vth sampling power data point among the multiple sampling power data points of the dth optimization time window, is the actual cumulative difference, is the actual average power of the d+1th optimization time window, is the actual average power of the dth optimization time window, is the compliance degree of the scene rules, is the total number of sample cumulative differences.

7. The method according to claim 1, characterized in that The expression of the objective function is: ； in, is the target value of the objective function, is the investment cost, For the peak shaving and valley filling benefits, is the network loss cost, is the energy storage quantity of the distributed energy storage, is the unit cost of the distributed energy storage configuration power, is the configured power of the kth energy storage in the distributed energy storage, is the unit cost of the configuration capacity of the distributed energy storage, is the configuration cost of the kth energy storage in the distributed energy storage, is the discount rate of the distributed energy storage, is the service life of the kth energy storage in the distributed energy storage, is the life span of the kth energy storage in the distributed energy storage, is the total number of days in a year, The typical value of is 365. is the total peak shaving duration of the distributed energy storage, is the peak shaving power unit benefit of the distributed energy storage at time t, is the peak shaving power of the kth energy storage in the distributed energy storage at the tth moment, is the total valley-filling time of the distributed energy storage, is the unit cost of valley-filling power of the distributed energy storage at time t, is the valley-filling power of the kth energy storage in the distributed energy storage at the tth moment, is the total number of nodes of the mth bus, is the unit cost of network loss power of the distributed energy storage, is the network loss power of the distributed energy storage between the i-th node and the j-th node of the m-th bus.

8. The method according to claim 1, characterized in that The preset constraints include power flow constraints and energy storage charging and discharging constraints; The expression of the power flow constraint condition is: ； The expression of the energy storage charge and discharge constraint condition is: ； in, is the active power injected into the mth bus at the tth moment, is the active power injected into the i-th node of the m-th bus at the t-th moment in the time sampling power curve corresponding to the distributed photovoltaic system, is the active power injected into the i-th node of the m-th bus at the t-th moment in the time-sampled power curve corresponding to the load, is the total number of nodes of the mth bus, It is 1 or 0, 1 represents that the distributed energy storage is installed at the i-th node of the m-th bus, and 0 represents that the distributed energy storage is not installed at the i-th node of the m-th bus. is the active power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the voltage amplitude of the ith node of the mth bus at the tth moment in the time-sampled power curve corresponding to the load, is the voltage amplitude of the jth node of the mth bus at the tth moment in the time-sampled power curve corresponding to the load, is the branch admittance between the i-th and j-th nodes of the m-th busbar, is the cosine function, is the phase difference between the i-th and j-th nodes of the m-th bus, is a sine function, is the reactive power injected by the mth bus at time t, is the reactive power injected by the i-th node of the m-th bus at time t in the time sampling power curve corresponding to the distributed photovoltaic system, is the reactive power injected into the ith node of the mth bus at time t in the time-sampled power curve corresponding to the load, is the reactive power injected by the distributed energy storage at the i-th node of the m-th bus at time t, is the minimum voltage amplitude of the load at the i-th node of the m-th bus, is the maximum voltage amplitude of the load at the i-th node of the m-th bus, is the minimum active power of the line between the i-th node and the j-th node of the m-th bus, is the active power of the line between the ith and jth nodes of the mth bus at the tth moment, is the maximum active power of the line between the ith and jth nodes of the mth bus, is the state of charge of the kth energy storage in the distributed energy storage at the tth moment, is the state of charge of the kth energy storage in the distributed energy storage at time t-1, is the charge and discharge power of the kth energy storage in the distributed energy storage at the tth moment, is the charging time of the energy stored in the distributed energy storage, is the charging and discharging efficiency of the energy storage in the distributed energy storage, is the rated capacity of the kth energy storage in the distributed energy storage, is the current predicted capacity of the kth energy storage in the distributed energy storage, is the rated power of the kth energy storage in the distributed energy storage.

9. The method according to claim 7 or 8, characterized in that The method further comprises: Obtain the remaining usable capacity, usage days, all-day charge and discharge power, and all-day charge and discharge average power of the kth energy storage in the distributed energy storage, as well as the average value and standard deviation of the all-day ambient temperature; The remaining usable capacity, usage days, all-day charge and discharge power, and all-day charge and discharge average power of the k-th energy storage in the distributed energy storage, as well as the average value and standard deviation of the all-day ambient temperature are input into a preset energy storage capacity prediction model to obtain the remaining usable capacity of the k-th energy storage in the distributed energy storage for multiple days in the future; The life usage limit of the kth energy storage in the distributed energy storage is determined based on the future multi-day remaining usage capacity and the life threshold of the kth energy storage in the distributed energy storage, or the current predicted capacity of the kth energy storage in the distributed energy storage is determined based on the future multi-day remaining usage capacity of the kth energy storage in the distributed energy storage.

10. The method according to claim 1, characterized in that The expression of the probability density function of the dth time window is: ； in, is the probability density function of the d-th time window, is the power variable, is the total number of actual power data points in the dth time window, is the bandwidth, is the kernel function, is the wth actual power data point among the multiple actual power data points in the dth time window.

Images