A method, system and medium for dispatching electric power resources based on virtual power plant

Abstract

The present application relates to the technical field of power resource scheduling in smart grids, and specifically to a method, system, and medium for power resource scheduling based on a virtual power plant, the method comprising: collecting three-phase current data from the power generation end of the smart grid and power data from each load end; determining the degree of abnormality of the current data from the power generation end within the time period based on the differential distribution of adjacent half-cycle curves between different phase current data within the same time period, and the degree of discreteness of the phase current data in harmonics; and optimizing and adjusting the droop coefficient in the droop control algorithm based on the mutation characteristics and trend significance of the total power data of all load ends at the same moment within each time period; and utilizing the optimized droop coefficient as the droop coefficient used when performing power resource scheduling operations with the next collected data. The present application aims to achieve reasonable scheduling of power resources and reduce the deviation in power resource scheduling based on virtual power plants.

Description

Technical Field

[0001] The present application relates to the technical field of power resource scheduling in smart grids, and in particular to a power resource scheduling method, system and medium based on a virtual power plant. Background Art

[0002] In today's power system, a large number of distributed energy resources, energy storage devices, and controllable loads are being rapidly deployed. Virtual power plant technology, as a key means of addressing the challenges of renewable energy consumption, plays a crucial role within the framework of the new smart grid system. The efficiency of power resource dispatch determines the stable and efficient integration of various renewable energy sources into the grid, and thus becomes the key to addressing renewable energy consumption.

[0003] Through its control center, a virtual power plant can monitor, predict, optimize, and trade its aggregated power resources, balance fluctuations in renewable energy, and dispatch power resources. Droop control technology is a commonly used control method that can achieve power allocation between distributed generation resources, reduce power fluctuations in the power grid, and optimize the dispatch of power resources. When using traditional droop control technology to optimize the dispatch of power resources in virtual power plants, existing methods do not fully consider the impact of the power supply stability of distributed power sources in smart grids and sudden load changes. This makes it difficult for power generation units in the droop control process to respond well to changes in the power grid, resulting in large deviations in the dispatch of power resources based on virtual power plants. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method, system and medium for power resource scheduling based on a virtual power plant. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for scheduling power resources based on a virtual power plant, the method comprising the following steps:

[0006] Collect three-phase current data of the smart grid power generation end and power data of each load end;

[0007] Based on the data characteristics of the three-phase current data and the power data of each load end, the synchronous stability value of the power supply and the load is determined; specifically:

[0008] A1. Determine the phase difference anomaly index within each time period based on the difference distribution of all adjacent half-cycle curves between different phase current data within the same time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; perform forward fusion of the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period;

[0009] A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period;

[0010] A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, to determine the synchronous stability value of power supply and load;

[0011] Based on the synchronous stability value of power supply and load, the droop coefficient in the droop control algorithm is optimized and adjusted; the optimized droop coefficient is used as the droop coefficient for the next power resource scheduling operation using the collected data.

[0012] Preferably, the two adjacent half-cycle curves are two half-cycle curves with the shortest distance between the different phase current data within the time period.

[0013] Preferably, the half-cycle curve is obtained by performing curve fitting on each phase current data in each time period; and taking the curves between each adjacent maximum point and minimum point on the fitted curve as each half-cycle curve.

[0014] Preferably, the method for determining the abnormality index is:

[0015] In the same time period, calculate the discrete degree of the time difference between all the same amplitudes in any two adjacent half-cycle curves of the current data of different phases;

[0016] The average level of the discreteness of the two half-cycle curves of all neighboring phase current data in the same time period is taken as the abnormal index of the phase difference in the time period.

[0017] Preferably, the mutation characteristics are further determined as:

[0018] The accumulated power data of all load terminals at the same moment is taken as the total load power data at the same moment;

[0019] Use the sliding t-test algorithm for all the total load power data in each time period, and output all the mutation points and the t statistics corresponding to each total load power data;

[0020] The number of mutation points and the average value of all t statistics are reversely fused, and the reverse fusion result is used as the mutation feature.

[0021] Preferably, the trend significance is determined by the probability that the output data of the stationary test performed by the addfuller function in the statsmode ls library on all the total power data in the time period is not stationary.

[0022] Preferably, the method for determining the synchronization stability value is:

[0023] Calculate the sum of the average value of the abnormal degree of the current data at the power generation end in all time periods and the average value of the significant value of the sharp change in the power load power of the power grid in all time periods;

[0024] The correlation and the sum value are reversely fused to obtain a synchronous stability value of power supply and load.

[0025] Preferably, the optimization and adjustment method is: the optimized droop coefficient is recorded as S, and the calculation formula is: S = Z × (R + 0.5), where Z is the initial active droop coefficient in the droop control algorithm, and R is the normalized result of the synchronous stability value of power supply and load.

[0026] In a second aspect, an embodiment of the present application provides a power resource scheduling system based on a virtual power plant, the system comprising:

[0027] The power data acquisition module is used to collect the three-phase current data of the smart grid power generation end and the power data of each load end;

[0028] The power data feature extraction module is used to determine the synchronous stability value of power supply and load based on the data features of three-phase current data and power data of each load end; specifically:

[0029] A1. Determine the phase difference anomaly index within each time period based on the difference distribution of all adjacent half-cycle curves between different phase current data within the same time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; perform forward fusion of the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period;

[0030] A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period;

[0031] A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, to determine the synchronous stability value of power supply and load;

[0032] The power resource scheduling operation module is used to optimize and adjust the droop coefficient in the droop control algorithm based on the synchronous stability value of power supply and load; and use the optimized droop coefficient as the droop coefficient used in the power resource scheduling operation when the next data is collected.

[0033] In a third aspect, an embodiment of the present application also provides a power resource scheduling medium based on a virtual power plant, wherein the medium includes 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, it implements the steps of any one of the above-mentioned power resource scheduling methods based on a virtual power plant.

[0034] As can be seen from the above embodiments, the power resource scheduling method, system, and medium based on a virtual power plant provided by the embodiments of the present application have at least the following beneficial effects:

[0035] This application analyzes the difference characteristics of the current phase output by the smart grid and the harmonic information contained therein, calculates the abnormal index of the phase deviation and harmonic pollution contained in the current data at the power generation end, and then analyzes the mutation degree and trend significance of the load power, and calculates the significant value of the sharp change in the load power of the power grid. Its beneficial effect is that it can better highlight the abnormal or sharp change characteristics of the power generation end and the load end in the power grid, so that the subsequent judgment on the impact on the stability of power resource scheduling is more accurate. Combined with the matching characteristics of the fluctuation of the distributed power generation end and the fluctuation of the load end in the power grid, the synchronous stability value of the power supply and load is calculated. Its beneficial effect is that it can further optimize the droop coefficient by judging the stability of the operation of the current power grid system, thereby realizing the reasonable scheduling of power resources and reducing the deviation of power resource scheduling based on virtual power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0037] Figure 1 A flowchart of a method for scheduling power resources based on a virtual power plant according to an embodiment of the present application;

[0038] Figure 2 A flow chart of a method for determining a synchronous stability value of power supply and load provided in one embodiment of the present application;

[0039] Figure 3 A flow chart of a method for determining the degree of abnormality of current data at a power generation end provided in one embodiment of the present application;

[0040] Figure 4 A structural diagram of a power resource scheduling system based on a virtual power plant is provided as an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to further illustrate the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail the specific implementation methods, structures, features and effects of a power resource scheduling method, system and medium based on a virtual power plant proposed in this application. 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.

[0042] Unless otherwise specified and limited, terms such as "comprises", "includes" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the article or device comprising the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs.

[0043] The following describes in detail a specific scheme of a power resource scheduling method, system and medium based on a virtual power plant provided by the present application in conjunction with the accompanying drawings.

[0044] See also Figure 1 , which shows a flowchart of a method for scheduling power resources based on a virtual power plant according to an embodiment of the present application, the method comprising the following steps:

[0045] The first step is to collect the three-phase current data of the smart grid power generation end and the power data of each load end.

[0046] Smart grids can effectively manage distributed energy based on real-time data monitoring and analysis, improve dispatching efficiency through information communication, and quickly locate and take appropriate recovery measures to ensure the stable operation of the power system.

[0047] Key technologies for virtual power plants include coordinated control, smart metering, and information and communication technologies. Coordinated control can be used to aggregate distributed energy resources, energy storage, and controllable loads within a virtual power plant, enabling it to achieve the high-demand power output of smart grids.

[0048] However, the uncertainty of distributed power generation in the power grid and sudden changes in load will affect the accuracy of coordinated control, and thus the power supply quality during the power resource scheduling process. Therefore, this embodiment collects and analyzes the data characteristics of current data from the smart grid generation end and power data from each load end to determine the stability of the current power grid system operation. Furthermore, when using droop control technology to optimize the power resources of the virtual power plant, the droop coefficient is optimized to achieve reasonable scheduling of power resources and reduce the deviation of power resource scheduling based on the virtual power plant.

[0049] This application uses an electric energy meter to collect three-phase current data from the power generation end of the smart grid, and uses the electric energy meter to collect power data from each load end in the smart grid, and takes the cumulative sum of the power data collected from all load ends at the same time as the total load power of the grid.

[0050] In the embodiment of the present application, the time interval for collecting current and power data is set to 2 milliseconds, and the single data collection time is set to 10 seconds. In other embodiments of the present application, the implementer can set it by himself.

[0051] At this point, the three-phase current data of the smart grid generation end and the power data of each load end can be obtained through the above method.

[0052] The second step is to determine the synchronization stability value of power supply and load based on the data characteristics of the three-phase current data and the power data of each load end.

[0053] When applying droop control technology for optimized scheduling in the power resource scheduling process based on virtual power plants, it is easy to ignore the impact of the mismatch between power supply fluctuations and load fluctuations in the power grid on the grid frequency or voltage change response, resulting in the set droop coefficient being difficult to effectively realize power resource scheduling.

[0054] Specifically, if the droop coefficient is too large, the power generation unit may respond too strongly to grid changes, thereby causing system oscillations; if the droop coefficient is too small, the power generation unit may not respond sufficiently and may be unable to adjust the output power in time to cope with grid changes.

[0055] Nowadays, more and more distributed power sources are connected to smart grids, and the load changes in the grid may fluctuate dramatically at any time. The mismatch between the fluctuations of the two increases the difficulty of maintaining stable grid operation. The uncertainty of distributed power generation in the grid leads to phase differences and harmonic pollution in the three-phase AC power provided by the power generation end, which in turn affects the efficiency of power resource scheduling. Accordingly, this application determines the synchronous stability value of power supply and load based on the data characteristics of three-phase current data and power data.

[0056] In the embodiment of the present application, the flow chart of the method for determining the synchronous stability value of power supply and load is shown in the attached figure. Figure 2 As shown, specifically:

[0057] A1. Within the same time period, based on the difference distribution of all two adjacent half-cycle curves between different phase current data, determine the phase difference anomaly index within each time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; forward fuse the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period.

[0058] Extract relevant features of the three-phase alternating current at the power grid's generating end. Because current and power data may experience large random fluctuations in the short term, this embodiment of the application divides each acquisition time into 15 equal time periods. The specific number of time periods can be set by the implementer.

[0059] Distributed power sources in smart grids may be connected to the grid at different locations and times, which can cause abnormal phase differences in the three-phase AC power, affecting the balance of the grid. Abnormal phase differences manifest as inconsistent time delays between the currents of different phases.

[0060] To analyze this feature, the embodiment of the present application uses the i-th time period as an example to fit each phase current data in the three-phase current data using a polynomial fitting technique to obtain a corresponding fitting curve, and the fitting curve is a sine waveform. The polynomial fitting technique is a well-known technique and will not be described in detail.

[0061] The maximum and minimum points of the fitting curve are calculated, and the fitting curve between adjacent maximum and minimum points is used as the half-cycle curve of the corresponding single-phase current, thereby dividing each single-phase current data into multiple half-cycles.

[0062] In three-phase current data, the current data for each phase is arranged alternately. Under normal circumstances, there is a constant phase difference between the three phases of three-phase AC power. This means that the time difference between the current of each phase and the current of the next phase at the same amplitude is constant.

[0063] In the embodiment of the present application, the flow chart of the method for determining the abnormality degree of the current data of the power generation end is shown in the attached figure. Figure 3 As shown, specifically:

[0064] (1) In the same time period, based on the difference distribution of all two adjacent half-cycle curves between different phase current data, the abnormal index of phase difference in each time period is determined.

[0065] Preferably, in an embodiment of the present application, within the same time period, the degree of discreteness of the time difference between all the same amplitudes in any two adjacent half-cycle curves between different phase current data is calculated; and the average level of the discreteness of all two adjacent half-cycle curves between all phase current data within the same time period is used as the abnormality index of the phase difference within the time period.

[0066] The two adjacent half-cycle curves are two half-cycle curves with the shortest distance between the current data of different phases within the time period.

[0067] In one embodiment of the present application, in a certain half-cycle curve, each moment corresponds to an amplitude, and the absolute value of the difference between the half-cycle curve and the adjacent half-cycle curve in the next phase current data at the same amplitude at the corresponding moment is calculated respectively, and the standard deviation of all the absolute values ​​is calculated. The average value of the standard deviation calculated for all two adjacent half-cycle curves between all adjacent phase current data in the i-th time period is used as the abnormal index of the phase difference in the time period, which is recorded as D i .

[0068] It should be understood that the obtained D i The larger the value is, the more likely it is that there is a phase anomaly in the current data at the power generation end during this time period.

[0069] Since the first and last parts of the single-phase current data within the time period may contain data portions less than half a cycle length, these data portions less than half a cycle length are not included in the calculation.

[0070] (2) Obtain the average level of dispersion of all phase current data in harmonics in each time period.

[0071] When distributed power sources are connected to the grid, they may introduce harmonics due to some power electronic equipment, thus affecting the quality of the grid.

[0072] To analyze the harmonic pollution of the power generation end current, this application adopts a harmonic detection method based on the Mallat algorithm to obtain the harmonic data of each phase current data. Among them, the harmonic detection method based on the Mallat algorithm is a well-known technology and will not be described in detail.

[0073] Preferably, in the embodiment of the present application, the average level of the discreteness of the harmonic data of all phase current data in each time period is used as the harmonic distortion value of the time period.

[0074] In one embodiment of the present application, taking the i-th time period as an example, the average value of the coefficient of variation of the harmonic data of all phase current data in the i-th time period is taken as the harmonic distortion value of the time period, which is recorded as E i .

[0075] It should be understood that the obtained E i The larger the value is, the greater the coefficient of variation of the harmonic data of all phase current data in this time period, that is, the greater the dispersion of the harmonic data, indicating that the harmonic pollution of the power generation end current in this time period is greater.

[0076] (3) The abnormal index of the phase difference in the same time period is forwardly fused with the average level of the discrete degree to determine the abnormal degree of the current data at the power generation end in the time period.

[0077] It can be understood that forward fusion is a fusion method such as addition and multiplication between data. The specific forward fusion method is determined by the implementer according to the actual situation, and this application does not impose any special restrictions.

[0078] Preferably, in the embodiment of the present application, taking the i-th time period as an example, the abnormality degree F of the current data of the power generation end in the time period is calculated. i The expression is: F i =D i ×E i Where, F i Indicates the abnormality of the current data at the power generation end in the i-th time period, D i Indicates the abnormal index of phase difference in the i-th time period, E i Indicates the harmonic distortion value of the i-th time period.

[0079] In other embodiments of the present application, the expression for the abnormality degree of the current data at the power generation end during the time period can also be set as: i =D i +E i .

[0080] It should be understood that the obtained F i The larger it is, the more serious the phase difference and harmonic pollution problems of the power grid current in this period of time are.

[0081] A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period.

[0082] In addition to the impact of power generation on power resource scheduling, rapid changes in load power in smart grids can also cause fluctuations in grid frequency, affecting grid stability and the efficiency of power resource scheduling. Therefore, it is necessary to further extract relevant features of load power.

[0083] For all total load power data in the i-th time period, this application uses a sliding t-test algorithm to obtain its mutation characteristics. The algorithm inputs all total load power data in the time period, sets the significance level to 0.05, and outputs the t-statistic corresponding to each total load power data point and all mutation points. The sliding t-test algorithm is a well-known technique and will not be described in detail.

[0084] Preferably, in the embodiment of the present application, the number of mutation points and the average value of all t statistics are reversely fused, and the reverse fusion result is used as the mutation feature.

[0085] In one implementation of the embodiment of the present application, the expression for calculating the abnormal mutation value of the total load power in the time period is: Where G i represents the abnormal mutation value of the total load power in the i-th time period, X represents the number of mutation points of the total load power data in this time period, and Y represents the average t-statistic of all total load power data in this time period.

[0086] It should be understood that the obtained G i That is, it indicates the mutation degree of the total load power data within the time period. The more mutation points there are and the smaller the average value of the t statistic is, the more likely there is a mutation in a set of sequences, that is, G i The larger the value is, the greater the degree of mutation of the load power is. The numerator is added with 1 to avoid G i When the value is zero, the data has certain fluctuations. At this time, the size of Y can reflect the degree of fluctuation.

[0087] In addition, a sharp change in load power may be manifested as a rapid change with a clear trend, in which there is no mutation. Therefore, the embodiment of the present application uses the adfull er function in the statsmode ls library to test the stability of load power. The input is the total load power data in the i-th time period, and the output is the probability of the data being unstable, which is recorded as the trend significance degree P of the total load power in the i-th time period. i The process of load power stability test by the addfuller function in the statsmode ls library is a well-known technique and will not be described in detail. It should be understood that the obtained P i The larger it is, the more unstable the total load power in the power grid is and the more obvious the corresponding trend characteristics are.

[0088] In other implementations of the embodiments of the present application, a moving average algorithm, an exponential smoothing method, etc. may also be used to calculate the trend significance of the total load power data in the i-th time period.

[0089] Preferably, in an embodiment of the present application, the method for calculating the significant value of the grid charge power with a sharp change characteristic is: forwardly fuse the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, and use the result of the forward fusion as the significant value of the sharp change in the grid load power in the time period.

[0090] In one embodiment of the present application, the expression for calculating the significant value of the rapid change in power grid load power is: i =G i ×P i Where, H i Indicates the significant value of the sharp change in power load in the i-th time period, G i Indicates the abnormal mutation value of the total load power in the i-th time period, P i Indicates the trend significance of the total load power in the i-th time period.

[0091] In other implementations of the embodiment of the present application, the calculation expression for the significant value of the sharp change in the power load of the power grid during the time period can also be set as: i =G i +P i .

[0092] It should be understood that the obtained H i The larger it is, the faster the load of the power grid changes during this period, and the more likely it is to affect the stability of power resource scheduling.

[0093] A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, determines the synchronous stability value of power supply and load.

[0094] Furthermore, the fluctuation of distributed power generation in the smart grid does not match the fluctuation of load, which may lead to the activation of power generation energy, affect its absorption rate, and thus reduce the efficiency of power resource scheduling.

[0095] Through the above analysis, the fluctuation of distributed power generation and load can be respectively expressed by the abnormal degree F of the current data at the generating end in the i-th time period. i , the significant value H of the rapid change of power load in the i-th time period i The size is reflected.

[0096] Therefore, the sequence obtained by arranging the abnormality degree of the current data at the power generation end in all time periods of a single acquisition time in ascending time order is recorded as the abnormality degree sequence M, and the sequence obtained by arranging the significant values ​​of the sharp changes in the power grid load power in all time periods in ascending time order is recorded as the significant value sequence N.

[0097] The correlation between sequences M and N is calculated, denoted as T. The larger the obtained T, the more closely the fluctuation state at the power generation end matches the fluctuation state at the load end. In this embodiment, the absolute value of the correlation coefficient between sequences M and N is calculated using the Pearson correlation coefficient as T. The Pearson correlation coefficient is well known in the art and will not be further described.

[0098] Then calculate the mean of all elements in sequence M and the mean of all elements in sequence N, and record the sum of the two means as U. The smaller the obtained U, the smaller the current anomaly at the power generation end and the sharp change at the load end.

[0099] Preferably, in an embodiment of the present application, the method for calculating the synchronous stability value of power supply and load is: reversely fusing the correlation T with the sum value U to obtain the synchronous stability value of power supply and load.

[0100] It can be understood that reverse fusion is a fusion method such as subtraction and division between data. The specific reverse fusion method is determined by the implementer based on the actual situation and is not specifically limited in this application.

[0101] In one implementation of the embodiment of the present application, the expression for calculating the synchronous stability value W of the power supply and the load is: Where W represents the synchronous stability value of power supply and load, T represents the correlation between sequences M and N, and U represents the sum of the mean of all elements in sequence M and the mean of all elements in sequence N.

[0102] It should be understood that the larger T is, the more sequences M and N have the same abnormality variation trend, and the smaller U is, the less significant the drastic changes in the two sequences are. That is, the larger the synchronous stability value W is, the more stable the power supply and load of the power grid are.

[0103] The third step is to optimize and adjust the droop coefficient in the droop control algorithm based on the synchronous stability value of power supply and load; and use the optimized droop coefficient as the droop coefficient for the next power resource scheduling operation when collecting data.

[0104] The embodiment of the present application calculates the synchronous stability value W of power supply and load by analyzing the abnormal characteristics of the output current of the distributed power supply in the power grid and the rapid change characteristics of the load, and combining the matching degree of the fluctuations of the two.

[0105] It should be understood that this value reflects the degree of synchronization and stability between the power supply and load of the power grid during the acquisition period. A smaller W indicates less stable power generation and load in the power grid. During power resource scheduling, to improve system stability, it is necessary to increase the droop coefficient to enhance rapid response to grid system changes. A larger W can be used to reduce the droop coefficient during power resource scheduling to avoid excessive oscillations in the power grid.

[0106] This application optimizes and adjusts the initial active power droop coefficient in the droop control algorithm. The specific calculation formula of the optimized droop coefficient S is: S = Z × (R + 0.5), where Z is the initial active power droop coefficient in the droop control algorithm, and R is the normalized result of the synchronous stability value of the power supply and load. This application uses the tanh activation function for normalization. The tanh activation function is a well-known technology and will not be repeated here.

[0107] For the first collection period, the initial active power droop control technology is used for adjustment. Subsequently, the S value calculated in the previous collection period is used as the optimized droop coefficient value for the current collection period. In this way, power resources are dispatched according to the actual operation of the power grid, making the allocation of power resources in the smart grid more efficient and flexible.

[0108] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of a power resource scheduling system based on a virtual power plant provided by an embodiment of the present application. In this embodiment, the various units included in the terminal are used to execute the various steps in the embodiment corresponding to a power resource scheduling method based on a virtual power plant. Figure 4 , the system comprising:

[0109] The power data acquisition module is used to collect the three-phase current data of the smart grid power generation end and the power data of each load end;

[0110] The power data feature extraction module is used to determine the synchronous stability value of power supply and load based on the data features of three-phase current data and power data of each load end; specifically:

[0111] A1. Determine the phase difference anomaly index within each time period based on the difference distribution of all adjacent half-cycle curves between different phase current data within the same time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; perform forward fusion of the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period;

[0112] A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period;

[0113] A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, to determine the synchronous stability value of power supply and load;

[0114] The power resource scheduling operation module is used to optimize and adjust the droop coefficient in the droop control algorithm based on the synchronous stability value of power supply and load; and use the optimized droop coefficient as the droop coefficient used in the power resource scheduling operation when the next data is collected.

[0115] Based on the same inventive concept as the above method, an embodiment of the present application also provides a power resource scheduling medium based on a virtual power plant, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned power resource scheduling methods based on a virtual power plant.

[0116] The various embodiments in this application are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0117] It should be noted that, unless otherwise specified and limited, terms such as "include", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, the phrase "including a ..." defines an element, does not exclude the presence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items.

[0118] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not invented herein.

[0119] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for dispatching power resources based on a virtual power plant, characterized in that: The method comprises the following steps: Collect three-phase current data of the smart grid power generation end and power data of each load end; Based on the data characteristics of the three-phase current data and the power data of each load end, the synchronous stability value of the power supply and the load is determined; specifically: A1. Determine the phase difference anomaly index within each time period based on the difference distribution of all adjacent half-cycle curves between different phase current data within the same time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; perform forward fusion of the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period; A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period; A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, to determine the synchronous stability value of power supply and load; Based on the synchronous stability value of power supply and load, the droop coefficient in the droop control algorithm is optimized and adjusted; the optimized droop coefficient is used as the droop coefficient for the next power resource scheduling operation using the collected data.

2. The method for dispatching power resources based on a virtual power plant according to claim 1, wherein: The two adjacent half-cycle curves are two half-cycle curves with the shortest distance between the current data of different phases within the time period.

3. The power resource scheduling method based on a virtual power plant according to claim 2, characterized in that: The half-cycle curve is: curve fitting is performed on each phase current data in each time period; and the curve between each adjacent maximum point and minimum point on the fitted curve is used as each half-cycle curve.

4. The method for dispatching power resources based on a virtual power plant according to claim 1, wherein: The method for determining the abnormality index is: In the same time period, calculate the discrete degree of the time difference between all the same amplitudes in any two adjacent half-cycle curves of the current data of different phases; The average level of the discreteness of the two half-cycle curves of all neighboring phase current data in the same time period is taken as the abnormal index of the phase difference in the time period.

5. The method for dispatching power resources based on a virtual power plant according to claim 1, wherein: The mutation characteristics are further determined as: The accumulated power data of all load terminals at the same moment is taken as the total load power data at the same moment; Use the sliding t-test algorithm for all the total load power data in each time period, and output all the mutation points and the t statistics corresponding to each total load power data; The number of mutation points and the average value of all t statistics are reversely fused, and the reverse fusion result is used as the mutation feature.

6. The method for dispatching electric power resources based on a virtual power plant according to claim 1, characterized in that: The trend significance is determined by the probability that the output data of the stationary test of all the total power data in the time period is not stationary using the adfuller function in the statsmodels library.

7. The method for dispatching electric power resources based on a virtual power plant according to claim 1, characterized in that: The method for determining the synchronization stability value is: Calculate the sum of the average value of the abnormal degree of the current data at the power generation end in all time periods and the average value of the significant value of the sharp change in the power load power of the power grid in all time periods; The correlation and the sum value are reversely fused to obtain a synchronous stability value of power supply and load.

8. The method for dispatching electric power resources based on a virtual power plant according to claim 1, wherein: The optimization adjustment method is: the optimized droop coefficient is recorded as S, and the calculation formula is: S=Z×(R+0.5), where Z is the initial active droop coefficient in the droop control algorithm, and R is the normalized result of the synchronous stability value of power supply and load.

9. A power resource scheduling system based on a virtual power plant, implementing a power resource scheduling method based on a virtual power plant as described in claims 1-8, characterized in that: The system comprises: The power data acquisition module is used to collect the three-phase current data of the smart grid power generation end and the power data of each load end; The power data feature extraction module is used to determine the synchronous stability value of power supply and load based on the data features of three-phase current data and power data of each load end; specifically: A1. Determine the phase difference anomaly index within each time period based on the difference distribution of all adjacent half-cycle curves between different phase current data within the same time period; obtain the average level of dispersion of all phase current data in harmonics within each time period; perform forward fusion of the phase difference anomaly index within the same time period with the average level of dispersion to determine the degree of anomaly of the power generation end current data within that time period; A2, based on the mutation characteristics and trend significance of the total power data of all load terminals at the same time in each time period, determines the significant value of the sharp change in grid load power in each time period; A3, combining the correlation and numerical value between the abnormal degree of the current data at the power generation end in all time periods and the significant value of the sharp change in the power grid load power in all time periods, to determine the synchronous stability value of power supply and load; The power resource scheduling operation module is used to optimize and adjust the droop coefficient in the droop control algorithm based on the synchronous stability value of power supply and load; and use the optimized droop coefficient as the droop coefficient used in the power resource scheduling operation when the next data is collected.

10. A power resource scheduling medium based on a virtual power plant, 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 processor implements the steps of a power resource scheduling method based on a virtual power plant as described in any one of claims 1 to 8.

Images