Energy storage power supply control method, system and energy storage power supply

Abstract

The present application relates to the field of energy storage power supply control technology, and specifically to an energy storage power supply control method, system and energy storage power supply, the method comprising: determining the response capability of the energy storage power supply based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data; determining the charging-related factors of the energy storage power supply by analyzing the correlation of the change trend between all grid load data and all charging power data in each valley load interval, and the extreme distribution of all grid load data, and combining the response capability to construct the control hysteresis of the energy storage power supply in each time period, and control the energy storage power supply after the current moment. The present application aims to improve the energy storage power supply's ability to absorb new energy power resources and the timeliness of the response of the energy storage power supply.

Description

Technical Field

[0001] The present application relates to the technical field of energy storage power supply control, and in particular to an energy storage power supply control method, system and energy storage power supply. Background Art

[0002] With the development and progress of society, energy storage technology has become an important part of the power system. The advantages of energy storage power sources such as high energy density, high power density, rapid response and regulation capabilities play a significant role in modern industrial production and daily life. In the process of grid dispatching and coordination, energy storage power sources can improve the grid operation efficiency and the level of new energy absorption, optimize the allocation of power resources, and enhance the grid's power supply capacity and improve the safe operation and stability of the grid.

[0003] In the process of grid dispatch coordination, energy storage power supply can quickly respond to grid load fluctuations, provide frequency regulation, peak regulation and valley filling services, optimize grid operation efficiency, and can also be used as emergency backup power supply when grid failure occurs and sudden demand increases, quickly restore power and ensure grid power supply reliability. However, the existing technology still has the problem of control strategy lag of energy storage power supply in the process of grid dispatch coordination, and may even produce the disadvantage of reverse regulation, causing energy storage power supply to be in an overcharged or discharged state for a long time; and the lack of judgment on the new energy consumption capacity caused by the control lag of energy storage power supply leads to the limitation of new energy power generation and consumption, which reduces the consumption capacity of energy storage power supply for new energy power resources and the timeliness of response of energy storage power supply. Summary of the invention

[0004] In a first aspect, an embodiment of the present application provides a method for controlling an energy storage power supply, the method comprising the following steps:

[0005] Obtain the grid frequency, grid load data, and charging power data of the energy storage power supply during the grid dispatching and coordination process at all acquisition times within a preset time period before the current time, and obtain the benchmark load of the power system;

[0006] The preset time length is divided into multiple time periods, and the base frequency deviation value of each time period is determined by analyzing the difference between the power grid frequency at all acquisition moments in each time period and the preset base frequency. The power grid load data at all acquisition moments are fitted to obtain a fitting curve, and the area of ​​the figure enclosed by the fitting curve and the preset base frequency in each time period, as well as the change trend of the fitting curve, are analyzed, and the power grid frequency imbalance factor of each time period is constructed in combination with the difference in the average level of the base frequency deviation value of each time period compared with all other time periods.

[0007] Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capacity of the energy storage power supply in each peak load interval in each time period is determined;

[0008] By analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data, the charging-related factors of the energy storage power supply in each valley-load interval in each time period are determined. Combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

[0009] Preferably, the base frequency deviation value of each time period is the average value of the difference between the power grid frequency and the preset base frequency at all acquisition moments in each time period.

[0010] Preferably, the expression of the power grid frequency imbalance factor in each time period is: ; In the formula, Indicates the grid frequency imbalance factor in each period; It represents the area of ​​the figure formed by the fitting curve and the straight line where the fundamental frequency is located in the i-th time period; It represents the cumulative result of taking the absolute value of the slope of the fitting curve at all acquisition moments in the i-th period; represents the mean of the difference in fundamental frequency deviation between the ith period and all other periods; norm() represents the normalization function.

[0011] Preferably, the determining of all suspected imbalance moments includes:

[0012] The grid frequency imbalance factor of the time period at each collection moment is used as the grid imbalance factor at each collection moment, the grid imbalance factors at all collection moments are used as the input of the threshold segmentation algorithm, the segmentation threshold is output, and the collection moment corresponding to the grid imbalance factor greater than the segmentation threshold is used as the suspected imbalance moment.

[0013] Preferably, the method for determining the response capability of the energy storage power supply in each peak load interval in each time period is:

[0014] Calculate the fractal dimension of all power grid load data in each peak load interval in each time period;

[0015] Calculate the difference between any power grid load data in each peak load interval in each time period and the mean value of all other power grid load data, record it as the load difference of any power grid load data, and multiply the cumulative sum of all the load differences in each peak load interval by the fractal dimension as the peak load trend intensity of each peak load interval in each time period;

[0016] The ratio of the number of all grid load data in each peak load interval in each time period to the peak load trend intensity is used as the response capacity of the energy storage power source in each peak load interval in each time period.

[0017] Preferably, the method for determining the charging-related factor of the energy storage power source in each valley-load interval in each time period is:

[0018] In each time period, all grid load data and all charging power data in each valley-load interval are used as inputs of the time series decomposition algorithm, and the grid load trend item sequence and charging power trend item sequence of each valley-load interval are output;

[0019] Calculate the correlation coefficient between the grid load trend item sequence and the charging power trend item sequence in each valley load interval, and use the ratio between the correlation coefficient and the minimum value of all grid load data in each valley load interval as the energy storage related value of each valley load interval;

[0020] The time interval between the start time of each valley load interval and the collection time corresponding to the minimum value on the fitting curve in the valley load interval is calculated, and the ratio of the energy storage related value of each valley load interval to the time interval is used as the charging related factor of the energy storage power supply in each valley load interval in each time period.

[0021] Preferably, the expression of the control hysteresis of the energy storage power supply in each time period is: ; In the formula, Indicates the control hysteresis of the energy storage power supply in time period i; Represents the cumulative sum of the response capabilities of energy storage power sources in all peak load intervals in time period i; represents the cumulative sum of charging-related factors of energy storage power sources in all valley-load intervals in time period i; norm( ) represents the normalization function.

[0022] Preferably, the energy storage power source after the current moment is controlled:

[0023] The preset time before the current moment is taken as an observation period, the control hysteresis of the energy storage power supply in all time periods before the current moment is taken as the input of the threshold segmentation algorithm, and the segmentation threshold is output, which is recorded as the energy storage power supply hysteresis threshold; the time period corresponding to the control hysteresis greater than or equal to the energy storage power supply hysteresis threshold is recorded as the time period to be controlled;

[0024] The control hysteresis of all time periods to be controlled, all grid load data within each time period to be controlled, and all charging power data are used as inputs of the neural network, and the control model of the energy storage power supply is output. The control model of the energy storage power supply is used to control the time when the dispatching instructions of the energy storage power supply are issued within the same observation period after the current moment.

[0025] In a second aspect, an embodiment of the present application provides an energy storage power supply control system, the system comprising:

[0026] The energy storage power supply data acquisition module is used to obtain the grid frequency, grid load data and energy storage power supply charging data in the grid dispatching and coordination process at all acquisition moments within a preset time period before the current moment, and obtain the benchmark load of the power system;

[0027] The energy storage power source weight analysis module is used to divide the preset time into multiple time periods, and determine the base frequency deviation value of each time period by analyzing the difference between the grid frequency at all acquisition moments in each time period and the preset base frequency; fit the grid load data at all acquisition moments before the current moment to obtain a fitting curve, and construct the grid frequency imbalance factor of each time period by analyzing the area of ​​the figure enclosed by the fitting curve and the preset base frequency in each time period, as well as the change trend of the fitting curve, and combining the difference in the average level of the base frequency deviation value of each time period compared with all other time periods;

[0028] Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capacity of the energy storage power supply in each peak load interval in each time period is determined;

[0029] The energy storage power supply control module is used to determine the charging-related factors of the energy storage power supply in each valley-load interval in each time period by analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data. Combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

[0030] In a third aspect, an embodiment of the present application further provides an energy storage power supply, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned energy storage power supply control methods when executing the computer program.

[0031] It can be seen from the above embodiments that the energy storage power supply control method provided in the embodiments of the present application has at least the following beneficial effects:

[0032] The present application obtains the grid frequency factor through the fluctuation amplitude characteristics and fluctuation trend characteristics of the grid frequency, accurately evaluates the grid frequency imbalance in the grid dispatch coordination process, and prevents the grid frequency from being misjudged as a grid imbalance phenomenon when it fluctuates normally with the electricity demand; further, the control hysteresis is obtained according to the adjustment response capability of the energy storage power supply in the grid dispatch coordination process and the new energy absorption characteristics, and comprehensively analyzes the response capability of the energy storage power supply in the process of the grid load increasing with the increase of electricity demand and the ability of the energy storage power supply to absorb the excess new energy power in the grid in a timely manner when the grid load decreases with the decrease of electricity demand, which more accurately reflects the lag degree of the energy storage power supply control strategy for the smooth operation of the grid; the present application controls the time when the energy storage power supply control dispatch instruction is issued through the control hysteresis, prevents the energy storage power supply from responding in a timely manner or even generating reverse regulation due to the lag of the control strategy in the grid dispatch coordination process, and improves the energy storage power supply's ability to absorb new energy power resources and the timeliness of the response of the energy storage power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0034] Figure 1 A flowchart of a method for controlling an energy storage power supply provided in one embodiment of the present application;

[0035] Figure 2 A schematic diagram of a control hysteresis extraction process of an energy storage power supply provided in one embodiment of the present application;

[0036] Figure 3 A block diagram of an energy storage power supply control system provided for one embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the energy storage power supply control method, system and energy storage power supply proposed in the present application, its specific implementation method, structure, features and effects in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0039] The following is a detailed description of a method and system for controlling an energy storage power supply and a specific solution of an energy storage power supply provided by the present application in conjunction with the accompanying drawings.

[0040] See also Figure 1 , which shows a flow chart of a method for controlling an energy storage power supply provided by an embodiment of the present application, the method comprising the following steps:

[0041] Step S1: Obtain the grid frequency, grid load data and charging power data of the energy storage power supply in the grid dispatching and coordination process at all acquisition moments within a preset time period before the current moment, and obtain the reference load of the power system.

[0042] This embodiment uses the AGC (Automatic Generation Control) automatic power generation control system to obtain the grid frequency, grid load data and charging power data of the energy storage power supply in the grid dispatch coordination process at all collection times within a preset time period before the current time, and obtains the benchmark load of the power system, wherein the charging power data of the energy storage power supply is the sum of the charging power data of the photovoltaic new energy storage equipment, and the data sets the data collection frequency to f.

[0043] It should be noted that the values ​​of the preset duration and the data collection frequency f are manually set. In this embodiment, the preset duration is 24h, and the data collection frequency f is 1Hz. The implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0044] Furthermore, in order to prevent the subsequent analysis from being affected by different dimensions of the acquired power grid dispatching coordination data or data missing due to external interference, all data collected during the power grid dispatching coordination process are normalized and missing value filled.

[0045] It should be noted that there are many commonly used normalization methods and missing value filling algorithms. In this embodiment, the z-score normalization method is used to normalize the relevant data, and the median filling method is used to fill the missing values. The implementer can also use other methods such as maximum and minimum value normalization method or mean filling method to normalize the data and fill the missing values. This embodiment does not impose any special restrictions.

[0046] Among them, the z-score standardization method and the median filling method are both well-known technologies, and their specific principles are not repeated here.

[0047] Step S2: Divide the preset time length into multiple time periods, and determine the fundamental frequency deviation value of each time period by analyzing the difference between the power grid frequency at all acquisition moments in each time period and the preset fundamental frequency; fit the power grid load data at all acquisition moments to obtain a fitting curve, and construct the power grid frequency imbalance factor of each time period by analyzing the area of ​​the figure enclosed by the fitting curve and the preset fundamental frequency straight line in each time period, as well as the changing trend of the fitting curve, combined with the difference in the average level of the fundamental frequency deviation value of each time period compared with all other time periods.

[0048] In the process of grid dispatch coordination in remote areas, the responsiveness of energy storage power supplies is crucial, and the responsiveness of energy storage power supplies is closely related to the fluctuation of grid frequency. The fluctuation of grid frequency can reflect the degree of imbalance between supply and demand of the grid, and thus provide accurate timestamp data for the lag of energy storage power supply control strategy.

[0049] When the fluctuation of the grid frequency in remote areas is greater within a certain time range, it means that the grid load fluctuation caused by the change of electricity demand on the power consumption side is greater; during the grid frequency fluctuation process, the greater the trend of grid frequency peaks and troughs, the more the energy storage power supply needs to show higher response timeliness in the grid dispatch coordination process; when the control strategy of the energy storage power supply is more lagging, the imbalance between the grid load and power generation is more obvious, and the possibility of causing grid frequency imbalance is higher.

[0050] Therefore, in order to improve the timeliness of the response of energy storage power supply in the process of grid dispatching and coordination, by analyzing the change trend of grid load data points and the change of grid frequency, a grid frequency imbalance factor is constructed to determine the response characteristics of energy storage power supply, specifically:

[0051] Fitting the power grid frequency data at the current moment and all previous acquisition moments to obtain a fitting curve;

[0052] It should be noted that there are many commonly used fitting algorithms. In this embodiment, a least squares fitting algorithm with partial space constraints is used to fit the grid frequency data, wherein the constraint condition is set to 50±0.5Hz, and y=50Hz is used as the value of the preset fundamental frequency in the fitting curve. The implementer may also use other fitting methods such as polynomial fitting method. This embodiment does not impose any special restrictions on the selection of the fitting algorithm.

[0053] Among them, the least square fitting algorithm with partial space constraints is a well-known technology, and the specific process of fitting the power grid frequency is not described in detail.

[0054] Further, the preset duration is evenly divided into multiple time periods, and the length of each time period is L, wherein the length of each time period is 3 hours. The implementer may also set it according to the specific situation, and this embodiment does not impose any special restrictions.

[0055] Furthermore, the average of the differences between the grid frequency and the preset base frequency at all acquisition moments in each time period is taken as the base frequency deviation value of each time period;

[0056] It should be noted that there are many methods for measuring the difference between data. In this embodiment, the absolute value of the difference between the grid frequency and the preset baseband frequency is calculated to measure the difference between the grid frequency and the preset baseband frequency at each collection moment. In this embodiment, all methods involving measuring the difference between data use the absolute value of the difference to measure the difference between two data. The implementer may also use other methods for measuring the difference between data, such as the square or ratio of the difference. Regarding the selection of the method for measuring the difference between data, this embodiment does not impose any special restrictions.

[0057] Furthermore, by analyzing the local fluctuation trend of the grid frequency in each period, the grid frequency imbalance factor of each period is determined, which is:

[0058] Grid frequency imbalance factor in period i The expression is: ; In the formula, It represents the area of ​​the figure formed by the fitting curve and the straight line where the fundamental frequency is located in the i-th time period; It represents the cumulative result of taking the absolute value of the slope of the fitting curve at all acquisition moments in the i-th period; represents the mean of the difference in fundamental frequency deviation between the ith period and all other periods; norm() represents the normalization function.

[0059] According to the grid frequency imbalance factor of each time interval before the current moment, it can be understood that in the process of grid dispatch coordination, the greater the fluctuation of the grid frequency in a short period of time, the greater the change trend of the grid frequency, and the larger the integral area of ​​the image enclosed by the fitting curve and the fundamental frequency straight line, that is, The larger the value is, the more significant the change in the slope of the fitting curve is. The larger the difference in fundamental frequency deviation between different time periods, the greater the difference in power grid frequency fluctuation between the ith time period and all other time periods, that is, The larger the integral area, the more significant the change in the slope of the fitting curve, and the greater the difference between the fundamental frequency deviation values, the greater the grid frequency imbalance factor, indicating that the greater the fluctuation in power demand on the power consumption side or the stronger the control hysteresis of the energy storage power supply, the greater the possibility of power grid supply and demand imbalance;

[0060] On the contrary, the smaller the fluctuation of the power grid frequency in a short period of time, the smaller the change trend of the power grid frequency, and the smaller the integral area of ​​the image enclosed by the fitting curve and the fundamental frequency straight line, that is, The smaller it is, the smoother the slope change in the fitting curve is. The smaller the difference in fundamental frequency deviation between different time periods, the smaller the difference in power grid frequency fluctuation between the ith time period and all other time periods, that is, The smaller the integral area, the smoother the slope of the fitting curve changes, and the smaller the difference between the fundamental frequency deviation values, the smaller the grid frequency imbalance factor, and the smaller the possibility of grid supply and demand imbalance.

[0061] Step S3: Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capability of the energy storage power supply in each peak load interval in each time period is determined.

[0062] During the grid dispatch coordination process in remote areas, when the grid load changes with the electricity demand on the power consumption side, the energy storage power supply will accept the grid dispatch instructions to execute charging and discharging behaviors to ensure the stability of the grid load. When the grid load becomes larger, it means that the electricity demand on the power consumption side is higher. The energy storage power supply will use the stored new energy power resources to discharge and provide additional power. Conversely, when the grid load becomes smaller, the energy storage power supply will use the new energy power resources to charge to absorb excess power and prevent the grid from overloading.

[0063] The regulatory response capability of energy storage power sources in the process of grid dispatch coordination is crucial. The worse the regulatory response capability of energy storage power sources in the process of grid dispatch coordination is, the more obvious the phenomenon that the grid load increase caused by the substantial increase in electricity demand on the electricity consumption side cannot be smoothed out, the stronger the long-term trend of grid load increase, and the more obvious the phenomenon of grid load increase frequency.

[0064] Therefore, by analyzing the changing trend of the grid load data, the response capability of the energy storage power supply is determined, thereby further regulating the response capability of the energy storage power supply. The specific process is as follows:

[0065] (1) Based on step S2, the grid frequency imbalance factor in each time period is obtained. Further, the grid frequency imbalance factor in the time period of each collection moment is used as the grid imbalance factor at each collection moment. The grid imbalance factors at all collection moments are used as the input of the threshold segmentation algorithm, and the segmentation threshold is output. The collection moment corresponding to the grid imbalance factor greater than the segmentation threshold is used as the suspected imbalance moment.

[0066] It should be noted that there are many commonly used threshold segmentation algorithms. In this embodiment, the Otsu threshold segmentation algorithm is used to demarcate the power grid frequency imbalance factor. The implementer may also use other threshold segmentation algorithms. This embodiment does not impose any special restrictions on the selection of the threshold segmentation algorithm.

[0067] Among them, the principle process of Otsu threshold segmentation algorithm is a well-known technology, and the specific principle of obtaining the segmentation threshold is not repeated here.

[0068] (2) Further, the mean of the grid load data at all collected moments between each pair of adjacent suspected imbalance moments is calculated, and the time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval;

[0069] (3) Further, by analyzing the complexity of all grid load data in each peak load interval in each time period, and the difference between each grid load data and the average level of all other grid load data, the response capability of the energy storage power source in each peak load interval in each time period is determined, specifically:

[0070] Firstly, the fractal dimension of all power grid load data in each peak load interval in each time period is calculated to characterize the complexity of power grid load data.

[0071] It should be noted that there are many methods for calculating fractal dimension. In this embodiment, the Hurst exponent method is used to calculate the fractal dimension of the power grid load data. The implementer can also use other methods such as the box counting method to calculate the fractal dimension. The specific calculation process of the fractal dimension in this embodiment will not be repeated.

[0072] Secondly, the difference between any power grid load data in each peak load interval in each time period and the mean value of all other power grid load data is calculated, which is recorded as the load difference of any power grid load data. The cumulative sum of all the load differences in each peak load interval is multiplied by the fractal dimension, which is taken as the peak load trend intensity of each peak load interval in each time period.

[0073] Furthermore, the ratio of the number of all grid load data in each peak load interval in each time period to the peak load trend intensity is used as the response capacity of the energy storage power source in each peak load interval in each time period.

[0074] According to the response capability of the energy storage power supply in each peak load interval in each time period, it can be understood that the milder the phenomenon of the increase frequency of the grid load, the longer the duration of the peak load of the grid load, and the more grid load data contained in the peak load interval; and the energy storage power supply discharges in time to weaken the trend of the grid load to produce peak load conditions, and the smaller the amplitude of the grid load to produce peak load, the smaller the fractal dimension of the peak load data sequence corresponding to the peak load interval; the smaller the difference between the grid load data in the peak load interval and the mean of all its grid load data, the stronger the response capability of the energy storage power supply;

[0075] Therefore, if the peak load interval contains more grid load data, the smaller the amplitude of the peak load generated by the grid load, and the smaller the difference between the grid load data in the peak load interval and the average of all grid load data, the stronger the response capability of the energy storage power supply; conversely, if the peak load interval contains less grid load data, the larger the amplitude of the peak load generated by the grid load, and the larger the difference between the grid load data in the peak load interval and the average of all grid load data, the weaker the response capability of the energy storage power supply.

[0076] Step S4: by analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data, the charging-related factors of the energy storage power supply in each valley-load interval in each time period are determined, and combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

[0077] As an important component of renewable energy power generation, the regulatory response capability of energy storage power supply in the process of grid dispatch coordination is not only reflected in the regulatory recovery capability under peak load conditions of the grid, but also in the charging level of energy storage power supply when the grid load produces valley load conditions due to a sudden decrease in electricity demand on the electricity consumption side. Energy storage power supply absorbs excess renewable energy power from the grid through charging, avoiding grid overload while reducing renewable energy power abandonment.

[0078] The stronger the absorption capacity of the energy storage power source in the grid dispatch coordination process, the stronger the match between the valley load condition caused by the reduction in electricity demand on the power consumption side and the charging behavior of the energy storage power source, that is, the stronger the correlation between the intensity of the grid load reduction trend caused by the reduction in electricity demand on the power consumption side and the intensity of the increasing trend of the energy storage power supply charging power, the smaller the valley load peak value generated by the grid load caused by the excess electricity in the grid being absorbed by the energy storage power source through charging behavior, and the shorter the duration of the valley load condition generated in the grid load reaching the minimum peak of the grid load.

[0079] Therefore, by analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data, the charging-related factors of the energy storage power source in each valley-load interval in each period are determined, specifically:

[0080] In each time period, all grid load data and all charging power data in each valley-load interval are used as inputs of the time series decomposition algorithm, and the grid load trend item sequence and charging power trend item sequence of each valley-load interval are output;

[0081] There are many commonly used time series decomposition algorithms. The STL decomposition algorithm is used in this embodiment. The implementer may also use other time series decomposition algorithms such as the SEATS decomposition algorithm. Regarding the selection of the time series decomposition algorithm, this embodiment does not impose any special restrictions.

[0082] Among them, the process of using the STL decomposition algorithm to obtain the trend item sequence is a well-known technology. Therefore, the specific process of using the STL decomposition algorithm to obtain the power grid load trend item sequence and the charging power trend item sequence will not be repeated.

[0083] Further, the correlation coefficient between the grid load trend item sequence and the charging power trend item sequence in each valley load interval is calculated, and the ratio between the correlation coefficient and the minimum value of all grid load data in each valley load interval is used as the energy storage related value of each valley load interval;

[0084] It should be noted that there are many methods for calculating the correlation coefficient between sequences. In this embodiment, the absolute value of the Pearson correlation coefficient between the grid load trend item sequence and the charging power trend item sequence is calculated to measure the correlation between the changing trends of the grid load data and the charging power data. The implementer may also use the Spearman correlation coefficient or the Kendall rank correlation coefficient to measure the degree of correlation between the sequences. This embodiment does not impose any special restrictions on the selection of the method for calculating the correlation coefficient between the sequences.

[0085] The calculation method of the Pearson correlation coefficient is a well-known technique, and its specific calculation principle will not be described in detail.

[0086] Furthermore, the time interval between the start time of each valley load interval and the collection time corresponding to the minimum value on the fitting curve in the valley load interval is calculated, and the ratio of the energy storage related value of each valley load interval to the time interval is used as the charging related factor of the energy storage power supply in each valley load interval in each time period.

[0087] According to the charging correlation factor of the energy storage power supply in each valley-load interval in each time period, it can be understood that in the process of grid dispatch coordination, the stronger the correlation between the intensity of the grid load reduction trend and the intensity of the energy storage power supply charging power increase trend, the greater the correlation coefficient between the grid load trend item sequence and the charging power trend item sequence, and the smaller the minimum value in the grid load data in the valley-load interval due to the charging behavior of the energy storage power supply, the shorter the duration of the valley-load phenomenon generated in the grid load by the energy storage power supply absorbing the excess new energy power in the grid load to reach the lowest peak, that is, the smaller the time interval between the start time of the valley-load interval and the collection time corresponding to the minimum value on the fitting curve in the valley-load interval, the greater the charging correlation factor of the energy storage power supply, indicating that the energy storage power supply has a stronger ability to absorb new energy power resources;

[0088] On the contrary, in the process of grid dispatch coordination, the smaller the correlation coefficient between the grid load trend item sequence and the charging power trend item sequence, and the larger the minimum value in the grid load data in the valley load interval due to the charging behavior of the energy storage power supply, the longer the duration of the valley load phenomenon generated in the grid load when the energy storage power supply absorbs the excess new energy power in the grid load reaches the lowest peak, that is, the longer the time interval between the start time of the valley load interval and the collection time corresponding to the minimum value on the fitting curve in the valley load interval, the smaller the charging correlation factor of the energy storage power supply, indicating that the energy storage power supply has a weaker ability to absorb new energy power resources.

[0089] In the process of coordinated grid dispatching, the stronger the energy storage power source's ability to respond to and recover from the increase in grid load caused by changes in electricity demand on the power consumption side and its ability to absorb new energy power, the less lag in the control strategy of the energy storage power source in the coordinated grid dispatching, and the more it can promptly carry out peak shaving and valley filling according to changes in grid load, thereby ensuring grid operation efficiency and stability.

[0090] Therefore, the control hysteresis of the energy storage power supply in any time interval can be determined by the response capability of the energy storage power supply and the charging-related factors, specifically:

[0091] Control hysteresis of energy storage power supply in time period i The expression is: ; In the formula, Represents the cumulative sum of the response capabilities of energy storage power sources in all peak load intervals in time period i; represents the cumulative sum of charging-related factors of energy storage power sources in all valley-load intervals in time period i; norm( ) represents the normalization function.

[0092] According to the control hysteresis of the energy storage power supply in each time period, it can be understood that when the control strategy hysteresis of the energy storage power supply within a certain time range is more serious, the control hysteresis of the energy storage power supply is greater, and the energy storage power supply is less able to discharge in time to maintain the balance of grid operation when the grid load increases with the increase of electricity demand, and is less able to charge in time to absorb new energy power resources when the grid load decreases with the decrease of electricity demand. That is, the cumulative result of the response capacity of the energy storage power supply in all peak load intervals is smaller, and the cumulative result of the charging-related factors in all valley load intervals is smaller;

[0093] On the contrary, when the control strategy lag of the energy storage power supply is more serious within a certain time range, the control lag of the energy storage power supply is smaller, the more balanced the grid operation is maintained when the energy storage power supply is discharged, and the stronger the ability to absorb new energy power resources is, that is, the greater the cumulative result of the response capability of the energy storage power supply in all peak load intervals, and the greater the cumulative result of the charging-related factors in all valley load intervals.

[0094] Preferably, the schematic diagram of the control hysteresis extraction process of the energy storage power supply provided in this embodiment is as follows: Figure 2 shown.

[0095] The preset time before the current moment is taken as an observation period, the control hysteresis of the energy storage power supply in all time periods before the current moment is taken as the input of the threshold segmentation algorithm, and the segmentation threshold is output, which is recorded as the energy storage power supply hysteresis threshold; the time period corresponding to the control hysteresis greater than or equal to the energy storage power supply hysteresis threshold is recorded as the time period to be controlled;

[0096] It should be noted that there are many commonly used threshold segmentation algorithms. In this embodiment, the Otsu threshold segmentation algorithm is used. The implementer may also use other threshold segmentation methods. This embodiment does not impose any special restrictions on the selection of the threshold segmentation algorithm.

[0097] Furthermore, the control hysteresis of all time periods to be controlled, all grid load data and all charging power data in each time period to be controlled are used as the input of the Long Short-term Memory Networks (LSTM), and the control model of the energy storage power supply is output. The control model of the energy storage power supply is used to control the dispatching instruction issuance time of the energy storage power supply in the same observation period after the current moment until the end of the observation period, and the end time of the observation period is used as the starting time. According to the change characteristics of the grid frequency, grid load data and charging power data in the observation period, according to the process of step S1-step S4, the dispatching instruction issuance time of the energy storage power supply in the next observation period is determined. Among them, the absolute value loss MAE is used as the loss function of the long short-term memory neural network, and Adma is used as the optimization algorithm of the long short-term memory neural network.

[0098] Among them, the long short-term memory neural network is a well-known technology, and the specific process of its training will not be described in detail.

[0099] So far, this application has determined the control lag of the energy storage power supply by analyzing the changing trend of the grid frequency and the correlation between the grid load data and the charging power data of the energy storage power supply, and combined with the neural network model to regulate the response time of the energy storage power supply, thereby improving the energy storage power supply's ability to absorb new energy power resources and the timeliness of the response of the energy storage power supply.

[0100] Based on the same inventive concept as the above method, the embodiment of the present application further provides an energy storage power supply control system, comprising:

[0101] The energy storage power supply data acquisition module is used to obtain the grid frequency, grid load data and energy storage power supply charging data in the grid dispatching and coordination process at all acquisition moments within a preset time period before the current moment, and obtain the benchmark load of the power system;

[0102] The energy storage power source weight analysis module is used to divide the preset time into multiple time periods, and determine the base frequency deviation value of each time period by analyzing the difference between the grid frequency at all acquisition moments in each time period and the preset base frequency; fit the grid load data at all acquisition moments before the current moment to obtain a fitting curve, and construct the grid frequency imbalance factor of each time period by analyzing the area of ​​the figure enclosed by the fitting curve and the preset base frequency in each time period, as well as the change trend of the fitting curve, and combining the difference in the average level of the base frequency deviation value of each time period compared with all other time periods;

[0103] Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capacity of the energy storage power supply in each peak load interval in each time period is determined;

[0104] The energy storage power supply control module is used to determine the charging-related factors of the energy storage power supply in each valley-load interval in each time period by analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data. Combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

[0105] The present application provides a block diagram of a power storage control system, such as Figure 3 shown.

[0106] Based on the same inventive concept as the above method, an embodiment of the present application also provides an energy storage power supply, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above energy storage power supply control methods are implemented.

[0107] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0109] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for controlling an energy storage power supply, characterized in that: The method comprises the following steps: Obtain the grid frequency, grid load data, and charging power data of the energy storage power supply during the grid dispatching and coordination process at all acquisition times within a preset time period before the current time, and obtain the benchmark load of the power system; The preset time length is divided into multiple time periods, and the base frequency deviation value of each time period is determined by analyzing the difference between the power grid frequency at all acquisition moments in each time period and the preset base frequency. The power grid load data at all acquisition moments are fitted to obtain a fitting curve, and the area of ​​the figure enclosed by the fitting curve and the preset base frequency in each time period, as well as the change trend of the fitting curve, are analyzed, and the power grid frequency imbalance factor of each time period is constructed in combination with the difference in the average level of the base frequency deviation value of each time period compared with all other time periods. Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capacity of the energy storage power supply in each peak load interval in each time period is determined; By analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data, the charging-related factors of the energy storage power supply in each valley-load interval in each time period are determined. Combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

2. A method for controlling an energy storage power supply according to claim 1, characterized in that: The base frequency deviation value of each time period is the average value of the difference between the power grid frequency at all acquisition moments in each time period and the preset base frequency.

3. A method for controlling an energy storage power supply according to claim 1, characterized in that: The expression of the power grid frequency imbalance factor in each period is: ; In the formula, Indicates the grid frequency imbalance factor in each period; It represents the area of ​​the figure enclosed by the fitting curve and the straight line where the fundamental frequency lies in the i-th time period; It represents the cumulative result of taking the absolute value of the slope of the fitting curve at all acquisition moments in the i-th period; represents the mean of the difference in fundamental frequency deviation between the ith period and all other periods; norm() represents the normalization function.

4. The energy storage power supply control method according to claim 1, characterized in that: The determination of all suspected imbalance moments includes: The grid frequency imbalance factor of the time period at each collection moment is used as the grid imbalance factor at each collection moment, the grid imbalance factors at all collection moments are used as the input of the threshold segmentation algorithm, the segmentation threshold is output, and the collection moment corresponding to the grid imbalance factor greater than the segmentation threshold is used as the suspected imbalance moment.

5. The energy storage power supply control method according to claim 1, characterized in that: The method for determining the response capability of the energy storage power supply in each peak load interval in each time period is as follows: Calculate the fractal dimension of all power grid load data in each peak load interval in each time period; Calculate the difference between any power grid load data in each peak load interval in each time period and the mean value of all other power grid load data, record it as the load difference of any power grid load data, and multiply the cumulative sum of all the load differences in each peak load interval by the fractal dimension as the peak load trend intensity of each peak load interval in each time period; The ratio of the number of all grid load data in each peak load interval in each time period to the peak load trend intensity is used as the response capacity of the energy storage power source in each peak load interval in each time period.

6. The energy storage power supply control method according to claim 1, characterized in that: The method for determining the charging-related factors of the energy storage power source in each valley load interval in each time period is as follows: In each time period, all grid load data and all charging power data in each valley-load interval are used as inputs of the time series decomposition algorithm, and the grid load trend item sequence and charging power trend item sequence of each valley-load interval are output; Calculate the correlation coefficient between the grid load trend item sequence and the charging power trend item sequence in each valley load interval, and use the ratio between the correlation coefficient and the minimum value of all grid load data in each valley load interval as the energy storage related value of each valley load interval; The time interval between the start time of each valley load interval and the collection time corresponding to the minimum value on the fitting curve in the valley load interval is calculated, and the ratio of the energy storage related value of each valley load interval to the time interval is used as the charging related factor of the energy storage power supply in each valley load interval in each time period.

7. The energy storage power supply control method according to claim 1, characterized in that: The expression of the control hysteresis of the energy storage power supply in each time period is: ; In the formula, Indicates the control hysteresis of the energy storage power supply in time period i; Represents the cumulative sum of the response capabilities of energy storage power sources in all peak load intervals in time period i; represents the cumulative sum of charging-related factors of energy storage power sources in all valley-load intervals in time period i; norm( ) represents the normalization function.

8. The energy storage power supply control method according to claim 1, characterized in that: The energy storage power source after the current moment is controlled: The preset time before the current moment is taken as an observation period, the control hysteresis of the energy storage power supply in all time periods before the current moment is taken as the input of the threshold segmentation algorithm, and the segmentation threshold is output, which is recorded as the energy storage power supply hysteresis threshold; the time period corresponding to the control hysteresis greater than or equal to the energy storage power supply hysteresis threshold is recorded as the time period to be controlled; The control hysteresis of all time periods to be controlled, all grid load data within each time period to be controlled, and all charging power data are used as inputs of the neural network, and the control model of the energy storage power supply is output. The control model of the energy storage power supply is used to control the time when the dispatching instructions of the energy storage power supply are issued within the same observation period after the current moment.

9. An energy storage power supply control system, implementing an energy storage power supply control method as claimed in claim 1, characterized in that: The system comprises: The energy storage power supply data acquisition module is used to obtain the grid frequency, grid load data and energy storage power supply charging data in the grid dispatching and coordination process at all acquisition moments within a preset time period before the current moment, and obtain the benchmark load of the power system; The energy storage power source weight analysis module is used to divide the preset time into multiple time periods, and determine the base frequency deviation value of each time period by analyzing the difference between the grid frequency at all acquisition moments in each time period and the preset base frequency; fit the grid load data at all acquisition moments before the current moment to obtain a fitting curve, and construct the grid frequency imbalance factor of each time period by analyzing the area of ​​the figure enclosed by the fitting curve and the preset base frequency in each time period, as well as the change trend of the fitting curve, and combining the difference in the average level of the base frequency deviation value of each time period compared with all other time periods; Based on the grid frequency imbalance factor of the time period in which all the acquisition moments are located, all suspected imbalance moments are determined, and the mean of the grid load data of all the acquisition moments between each pair of adjacent suspected imbalance moments is calculated. The time period between the adjacent suspected imbalance moments corresponding to the mean value greater than or equal to the benchmark load is recorded as the peak load interval, and the time period between the remaining adjacent suspected imbalance moments is recorded as the valley load interval; based on the complexity of all grid load data in each peak load interval, and the difference between each grid load data and the average level of all other grid load data, the response capacity of the energy storage power supply in each peak load interval in each time period is determined; The energy storage power supply control module is used to determine the charging-related factors of the energy storage power supply in each valley-load interval in each time period by analyzing the correlation between the changing trends of all grid load data and all charging power data in each valley-load interval, as well as the extreme distribution of all grid load data. Combined with the response capability, the control hysteresis of the energy storage power supply in each time period is constructed to control the energy storage power supply after the current moment.

10. An energy storage power supply, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the energy storage power supply control method as described in any one of claims 1-8 are implemented.

Images