Data feature mining method, device and equipment of energy storage system and storage medium
The typical power curve is extracted through clustering processing and double-layer bird flock algorithm, and the problem of inaccurate data mining of energy storage systems in the existing technology is solved, and accurate feature mining of photovoltaic power stations and reasonable application of energy storage systems is realized.
Patent Information
- Application Number
- CN202410975773.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-19
AI Technical Summary
The prior art fails to accurately represent the actual operating conditions of the photovoltaic power station during the mining of data characteristics of the energy storage system, and does not consider the differences in different seasons and weather scenarios, resulting in inaccurate excavation results, affecting the reasonable application of the energy storage system.
By obtaining the power data of photovoltaic power stations in various weather scenarios, clustering processing is performed to divide them into multiple power families, determining the power data of the energy storage system based on the preset regulation cloud, and using the double-layer bird flock algorithm to extract typical power curves. Represents the charge and discharge characteristics of the energy storage system.
Accurate data characteristics mining of photovoltaic power station energy storage system is achieved, and the differences in different seasons and weather scenarios are taken into account. The obtained typical power curve can represent the actual operating conditions, improving the application accuracy of the energy storage system.
Smart Images

Figure CN118981622A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of energy storage technology, and in particular relates to a data feature mining method, device, equipment and storage medium for an energy storage system. Background Art
[0002] At present, promoting the evolution of the power system to adapt to large-scale and high-proportion renewable energy has become a new direction for the development of the current power system. However, the volatility, randomness, and intermittent nature of renewable energy have brought adverse effects on many aspects of the power grid, from regulation and operation to safety control, greatly limiting the effective use of renewable energy. Therefore, the current power system usually combines energy storage systems to improve the flexibility of the entire energy system to avoid the problems caused by the grid connection of renewable energy.
[0003] At the current stage, when performing data feature mining on energy storage systems, related technologies mostly construct new charging and discharging control strategies to calculate the power data of the energy storage system, and then use the calculated power data of the energy storage system to mine the characteristics of the energy storage system. The charging and discharging control strategies of the actual photovoltaic power station energy storage system itself are not considered, resulting in the mining results being unable to represent the actual operating conditions of the photovoltaic power station energy storage system. In addition, the existing technology generally mixes the energy storage data of all weather conditions together for feature mining, without considering the differences in the operating condition characteristics of the photovoltaic power station energy storage system under different weather scenarios in different seasons. The mining results may overwhelm one or more types of energy storage system operating condition characteristics. Based on this, the results of related technologies when performing data feature mining on energy storage systems are not accurate enough, which in turn affects the reasonable application of energy storage systems. Summary of the invention
[0004] The purpose of the present invention is to provide a data feature mining method, device, equipment and storage medium for an energy storage system to solve the technical problem that the results of related technologies when performing data feature mining on energy storage systems are not accurate enough, thereby affecting the reasonable application of energy storage systems.
[0005] In order to achieve the above object, the present invention adopts the following technical solution:
[0006] In a first aspect, an embodiment of the present application provides a data feature mining method for an energy storage system, the method comprising: acquiring first data of a photovoltaic power station, the first data comprising power data of the photovoltaic power station under various weather scenarios; clustering the first data according to the weather scenario to divide the first data into a plurality of photovoltaic power station power families, each of the photovoltaic power station power families corresponding to a weather scenario; based on a preset control cloud, determining the power data of the energy storage system corresponding to each of the photovoltaic power station power families, and generating energy storage system power families corresponding to each photovoltaic power station power family; and extracting data features of each energy storage system power family through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, the typical power curve being used to indicate the charging and discharging characteristics of the energy storage system.
[0007] In combination with the first aspect, in a possible implementation, the first data of the photovoltaic power station is obtained according to a typical weather scene identification table, and the typical weather scene identification table is used to indicate the identification of various weather scenes. The method also includes: obtaining meteorological data within a target time period; classifying the meteorological data according to months to determine multiple seasonal scenes, each of the seasonal scenes including meteorological data of multiple months; dividing the meteorological data in each of the seasonal scenes according to dates, determining the meteorological data corresponding to each date in each seasonal scene, and forming the typical weather scene identification table.
[0008] In combination with the first aspect, in a possible implementation method, the first data is clustered and divided according to the weather scene to divide the first data into multiple photovoltaic power station power families, including: according to a preset typical weather scene identification table, the first data is divided into multiple categories of target data according to seasonal scenes, and each category of target data corresponds to the same seasonal scene; each category of target data is clustered and divided separately to divide each category of target data into multiple photovoltaic power station power families according to the weather scene, each of the photovoltaic power station power families corresponds to the same type of weather scene, and the data of each photovoltaic power station power family is recorded in the preset typical weather scene photovoltaic power station data table.
[0009] In combination with the first aspect, in a possible implementation, the power data of the photovoltaic power station power family corresponds to a timestamp, and based on the preset control cloud, the power data of the energy storage system corresponding to each photovoltaic power station power family is determined, and the energy storage system power family corresponding to each photovoltaic power station power family is generated, including: obtaining the energy storage unit information table and the energy storage unit operation data table of the control cloud; querying the energy storage unit code corresponding to the photovoltaic power station according to the energy storage unit information table of the control cloud; determining the power data corresponding to the timestamp of the power data of the photovoltaic power station power family in the energy storage unit operation data table of the control cloud according to the energy storage unit code; based on the power data, determine the energy storage system power family corresponding to each photovoltaic power station power family according to the timestamp.
[0010] In combination with the first aspect, in a possible implementation, data features are extracted for each energy storage system power family through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, including: determining a target energy storage system power family, wherein the target energy storage system power family is one of multiple energy storage system power families; determining a target power matrix according to the target energy storage system power family, wherein the target power matrix is used to indicate power values of energy storage systems on different dates and at different times in the target energy storage system power family; determining a target power vector in sequence according to the target power matrix, wherein the target power vector is used to indicate power values of energy storage systems at the same time on different dates in the target power matrix; based on the target power vector, data features are extracted for the target energy storage system power family through the double-layer bird flock algorithm to obtain the typical power curve.
[0011] In combination with the first aspect, in a possible implementation, based on the target power vector, the double-layer bird flock algorithm is used to extract data features of the target energy storage system power family to determine the typical power curve of the energy storage system, including: obtaining the lower limit power of the concentrated distribution interval of the target power vector through the double-layer bird flock algorithm to form a concentrated distribution interval; determining the power value of the energy storage system including the target power vector in the concentrated distribution interval to determine the interval vector of the concentrated distribution interval, and determining the distribution probability of the concentrated distribution interval; according to the interval vector, determining through the double-layer bird flock algorithm The characteristic power value of the target power vector is stored in a preset characteristic power value set; if the distribution probability of the concentrated distribution interval is not less than a preset threshold, the power value in the interval vector of the concentrated distribution interval is removed from the target power vector to determine a new target power vector, and based on the new target power vector, the characteristic power value is re-determined through the double-layer bird flock algorithm; if the distribution probability of the concentrated distribution interval is less than the preset threshold, a characteristic power value vector is determined based on all characteristic power values stored in the characteristic power value set; and the typical power curve is determined based on the characteristic power value vector.
[0012] In combination with the first aspect, in a possible implementation, determining the typical power curve according to the characteristic power value vector includes: based on a roulette algorithm, determining a target characteristic power value in the characteristic power value vector, the target characteristic power value being used to indicate a typical power value of the energy storage system at the moment indicated by the target power vector; if the moment indicated by the target power vector is not the last moment indicated by the target power matrix, storing the target characteristic power value in a preset typical power value set, and updating the target power vector in sequence, the moment indicated by the updated target power vector being the next moment of the moment indicated by the target power vector before the update; if the moment indicated by the target power vector is the last moment indicated by the target power matrix, after storing the target characteristic power value in the preset typical power value set, generating the typical power curve according to all target characteristic power values stored in the typical power value set.
[0013] In a second aspect, the embodiment of the present application further proposes a data feature mining device for an energy storage system, which is applied to an energy storage system of a photovoltaic power station, and the device includes:
[0014] A data acquisition module, used to acquire first data of the photovoltaic power station, wherein the first data includes power data of the photovoltaic power station under various weather scenarios;
[0015] A data processing module, configured to perform clustering processing on the first data according to weather scenarios, so as to divide the first data into a plurality of photovoltaic power plant power clusters, each photovoltaic power plant power cluster corresponding to a weather scenario;
[0016] A data analysis module, used to determine the power data of the energy storage system corresponding to each photovoltaic power station power group based on a preset control cloud, and generate the energy storage system power group corresponding to each photovoltaic power station power group;
[0017] The data extraction module is used to extract data features of each energy storage system power family through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, wherein the typical power curve is used to indicate the charge and discharge characteristics of the energy storage system.
[0018] In a third aspect, an embodiment of the present application further proposes a computer device, comprising a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method described in the first aspect.
[0019] In a fourth aspect, an embodiment of the present application further proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the method described in the first aspect.
[0020] The embodiments of the present application can achieve the following technical effects:
[0021] Based on the method proposed in the embodiment of the present application, when applied to the energy storage system of a photovoltaic power station, first data of the photovoltaic power station is obtained, the first data including the power data of the photovoltaic power station under various weather scenarios, and then the first data is clustered according to the weather scenario to divide the first data into multiple photovoltaic power station power families, each photovoltaic power station power family corresponds to a weather scenario, and based on the preset control cloud, the power data of the energy storage system corresponding to each photovoltaic power station power family is determined, and the energy storage system power families corresponding to each photovoltaic power station power family are generated, and finally, through the double-layer bird flock algorithm, data feature extraction is performed on each energy storage system power family to determine a typical power curve of the energy storage system, and the typical power curve is used to indicate the charging and discharging characteristics of the energy storage system.
[0022] Compared with the existing technology, the data obtained by this method are all data of the actual photovoltaic power station and its energy storage system in the control cloud, and the charging and discharging control strategy of the actual photovoltaic power station energy storage system itself is fully considered, so that the typical power curve finally obtained can represent the actual photovoltaic power station energy storage system operating conditions. In addition, this method also adopts a feature mining method based on a double-layer bird flock algorithm to perform feature mining on the data of the photovoltaic power station energy storage system in different seasons and different weather scenarios, thereby forming a typical power curve of the photovoltaic power station energy storage system for each type of typical weather scenario. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:
[0024] Figure 1 A schematic diagram of a flow chart of a data feature mining method for an energy storage system provided in an embodiment of the present application;
[0025] Figure 2 A schematic diagram of a big data feature mining method based on a double-layer bird flock algorithm provided in an embodiment of the present application;
[0026] Figure 3 A schematic diagram of the structure of a data feature mining device for an energy storage system provided in an embodiment of the present application;
[0027] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0028] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.
[0029] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.
[0030] Related technologies involved in the embodiments of this application:
[0031] Control cloud: Based on cloud technology concepts such as virtualization, distribution and service, it adopts a physically distributed and logically unified architecture model, has the characteristics of resource virtualization, data standardization, and application service, and is an electric power control information system that provides cloud services for electric power control business. It is an enterprise private cloud for power grid control production and operation.
[0032] Photovoltaic power station: refers to a photovoltaic power generation system that utilizes solar energy and adopts special materials such as crystalline silicon panels, inverters and other electronic components. It is connected to the power grid and transmits electricity to the power grid.
[0033] Energy storage system: When analyzing the energy storage process, the part of the object or space range that is demarcated to determine the research object is called the energy storage system. It includes the input and output of energy and matter, and the energy conversion and storage equipment. The energy storage system often involves multiple energies, multiple devices, multiple materials, and multiple processes. It is a complex energy system that changes over time.
[0034] Energy storage unit: The smallest energy storage system consisting of a battery pack, a battery management system and a power conversion system connected to it. It is mainly used to store and convert energy for use in power systems or other equipment. Several energy storage units can form an energy storage system.
[0035] Clustering method: Clustering method is a statistical analysis method used to divide objects or samples in a data set into several different groups or categories, so that objects in the same category have high similarity, while objects in different categories have low similarity. These methods classify objects based on similarity or distance, and usually do not rely on predefined categories. Cluster analysis has a wide range of applications in data mining, pattern recognition, market segmentation and other fields. Common clustering methods include K-means, hierarchical clustering, DBSCAN, spectral clustering, fuzzy C-means, etc.
[0036] Intelligent optimization algorithm: Intelligent optimization algorithm is a general term for algorithms that use computer software programming to simulate the long-term evolution, reproduction, competition, adaptation, and natural selection of nature, the biological world, and even human beings, so as to solve complex optimization problems. These algorithms are inspired by human intelligence, the sociality of biological groups, or the laws of natural phenomena, and are designed to solve complex optimization problems such as task allocation and production scheduling. Intelligent optimization algorithms include but are not limited to fuzzy logic reasoning algorithms, neural network learning algorithms, genetic algorithms, simulated annealing algorithms, taboo algorithms, artificial immune algorithms, artificial ant colony algorithms, particle swarm algorithms, chaos optimization algorithms, quantum optimization algorithms, etc.
[0037] Charge and discharge control strategy: refers to the methods and techniques for managing and controlling energy storage devices during the charging and discharging process of the energy storage system, so as to ensure the safety of the battery, extend the battery life and achieve efficient operation of the energy storage system.
[0038] For further information, please refer to Figure 1 and Figure 2 As shown, the data feature mining method of the energy storage system proposed in the embodiment of the present application includes:
[0039] Step S10, acquiring first data of the photovoltaic power station, wherein the first data includes power data of the photovoltaic power station under various weather scenarios;
[0040] In this embodiment, the first data of the photovoltaic power station can be obtained by classification according to different weather scenes, so as to mine data features according to different weather scenes in subsequent steps.
[0041] Specifically, the categories of various weather scenes can be uniquely indicated by a typical weather scene identification table, that is, the first data of the above-mentioned photovoltaic power station is obtained according to the typical weather scene identification table, and the typical weather scene identification table is used to indicate the identification of various weather scenes. The method also includes: obtaining meteorological data within the target time period; classifying the meteorological data according to month, determining multiple seasonal scenes, and each seasonal scene includes meteorological data of multiple months; dividing the meteorological data in each seasonal scene according to date, determining the meteorological data corresponding to each date in each seasonal scene, and forming a typical weather scene identification table.
[0042] The target time period refers to the time range of data collected when mining data features of the energy storage system. Different seasons and weather conditions will cause different fluctuations in the power of the photovoltaic power station, thereby having different impacts on the power curve characteristics of the energy storage system. Therefore, in order to more effectively reflect the operating conditions of the energy storage system, in this embodiment, seasonal factors and weather factors within the target time period are taken into consideration, and typical weather scenes are identified by classification to form a typical weather scene identification table.
[0043] The first level is for seasonal scenes, which are classified and identified according to the overall meteorological conditions of each month. For example, the attributes of seasonal scene identification include code, name, description, etc. The code is a unique digital identifier to distinguish seasonal scenes, and the number of code digits can be selected according to the requirements of the control cloud. The name is a text identifier to help users understand. The description is used to simply record the months included in the seasonal time.
[0044] The second level classifies and identifies the weather scenes in a certain season based on the seasonal scene identification and the daily weather conditions. For example, the attributes of the weather scene identification include code, name, etc. The code is a unique digital identifier to distinguish each weather scene, and the number of code bits can be selected according to the requirements of the control cloud. The name is a text identifier to help users understand.
[0045] The above-mentioned season scene code and weather scene code are combined in levels to form a weather scene identification under a certain season scene, that is, a typical weather scene identification, which is recorded in the typical weather scene identification table.
[0046] Step S20, clustering the first data according to the weather scenario to divide the first data into a plurality of photovoltaic power plant power clusters, each photovoltaic power plant power cluster corresponding to a weather scenario;
[0047] In this embodiment, the first data is clustered and divided according to the weather scene to divide the first data into multiple photovoltaic power station power families, including: according to a preset typical weather scene identification table, the first data is divided into multiple categories of target data according to the seasonal scene, and each category of target data corresponds to the same seasonal scene; each category of target data is clustered and divided separately to divide each category of target data into multiple photovoltaic power station power families according to the weather scene, each photovoltaic power station power family corresponds to the same type of weather scene, and the data of each photovoltaic power station power family is recorded in the preset typical weather scene photovoltaic power station data table.
[0048] Specifically, the power data of the photovoltaic power station is divided into several seasonal scenes according to the meteorological conditions, for example, February, March, and April are the first seasonal scenes, May, June, and July are the second seasonal scenes, August, September, and October are the third seasonal scenes, and November, December, and January are the fourth seasonal scenes. The photovoltaic power station power data of each seasonal scene is divided into weather scenes such as sunny, cloudy to sunny, sunny to cloudy, cloudy, overcast, rain and snow using a clustering method to form photovoltaic power station power families for different typical weather scenes. At the same time, a typical weather scene photovoltaic power station data table can be created, which includes fields such as seasonal scene code, weather scene code, photovoltaic power station code, power data of the photovoltaic power station power family corresponding to the typical weather scene, and the business timestamp of each power data. The data in the photovoltaic power station power families for different typical weather scenes are recorded in the typical weather scene photovoltaic power station data table.
[0049] Step S30, based on the preset control cloud, determining the power data of the energy storage system corresponding to each photovoltaic power station power group, and generating the energy storage system power group corresponding to each photovoltaic power station power group;
[0050] Among them, the power data of the photovoltaic power station power family corresponds to the timestamp. Based on the preset control cloud, the power data of the energy storage system corresponding to each photovoltaic power station power family is determined, and the energy storage system power family corresponding to each photovoltaic power station power family is generated, including: obtaining the energy storage unit information table and the energy storage unit operation data table of the control cloud; querying the energy storage unit code corresponding to the photovoltaic power station according to the energy storage unit information table of the control cloud; according to the energy storage unit code, determining the power data corresponding to the timestamp of the power data of the photovoltaic power station power family in the energy storage unit operation data table of the control cloud; based on the power data of the energy storage unit, determine the energy storage system power family corresponding to each photovoltaic power station power family according to the timestamp.
[0051] In the basic information table of the energy storage unit of the control cloud, according to the photovoltaic power station code, query the energy storage unit code included in its equipped energy storage system, so as to form the equipped energy storage system of the photovoltaic power station. According to the energy storage unit code and the business timestamp corresponding to each power data in the photovoltaic power station power family of different typical weather scenarios, find the actual power data of the energy storage unit at the corresponding time in the energy storage unit operation data table of the control cloud. After adding the power of all energy storage units under the same business timestamp, the power family of the energy storage system in different typical weather scenarios is formed. Add a data table of the energy storage system of the photovoltaic power station in typical weather scenarios, which includes fields such as seasonal scene code, weather scene code, photovoltaic power station code, power data of the energy storage system power family under the corresponding typical weather scenario, and business timestamp of each power data. Record the data in the energy storage system power family of different typical weather scenarios into the data table of the energy storage system of the photovoltaic power station in typical weather scenarios.
[0052] Step S40 , extracting data features from each energy storage system power group through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, wherein the typical power curve is used to indicate charging and discharging characteristics of the energy storage system.
[0053] The characteristic form of the power curve of the energy storage system is directly related to the power fluctuation of the photovoltaic power station. In order to mine the typical power curve that can represent the operating characteristics of the energy storage system in the context of smoothing the power fluctuation of the photovoltaic power station in reality, this embodiment adopts a big data feature mining method based on a double-layer bird flock algorithm to mine the typical power curve of the energy storage system for each type of typical weather scenario.
[0054] In this embodiment, the step of extracting data features for each energy storage system power family through a double-layer bird flock algorithm includes: determining a target energy storage system power family, wherein the target energy storage system power family is one of a plurality of energy storage system power families; determining a target power matrix according to the target energy storage system power family, wherein the target power matrix is used to indicate power values of energy storage systems at different dates and times in the target energy storage system power family; determining a target power vector in sequence according to the target power matrix, wherein the target power vector is used to indicate power values of energy storage systems at the same time and on different dates in the target power matrix; and extracting data features for the target energy storage system power family based on the target power vector through the double-layer bird flock algorithm to obtain the typical power curve.
[0055] Specifically, first read the energy storage system power family data of a typical weather scenario to form a target power matrix C. The element p in the matrix C is ij is the power value of the energy storage system at the jth moment on the i-th day, where i∈[1,2,…,m], j∈[1,2,…,n];
[0056]
[0057] Afterwards, the energy storage system power values at the same time on different days are sequentially extracted from the energy storage system power matrix C to form the above-mentioned target power vector P j , P j =[p 1j p 2j …p mj ] T .
[0058] Furthermore, based on the target power vector, the step of extracting data features of the target energy storage system power family through a double-layer bird flock algorithm includes: obtaining the lower limit power of the concentrated distribution interval of the target power vector through the double-layer bird flock algorithm to form a concentrated distribution interval; determining the power value of the energy storage system including the target power vector in the concentrated distribution interval to determine the interval vector of the concentrated distribution interval, and determining the distribution probability of the concentrated distribution interval; according to the interval vector, determining the characteristic power value of the target power vector through the double-layer bird flock algorithm, and storing the characteristic power value in a preset characteristic power value set; if the distribution probability of the concentrated distribution interval is not less than a preset threshold, then removing the power value in the interval vector of the concentrated distribution interval from the target power vector to determine a new target power vector, and re-determining the characteristic power value through the double-layer bird flock algorithm based on the new target power vector; if the distribution probability of the concentrated distribution interval is less than the preset threshold, then determining the characteristic power value vector based on all the characteristic power values stored in the characteristic power value set; and determining the typical power curve based on the characteristic power value vector.
[0059] Specifically, we search for the interval where the energy storage system power value is most concentrated at the jth moment on different days. Since the vector P j In the subsequent mining process, the initial vector will be used as a benchmark to transform the vector to obtain the corresponding concentrated distribution interval, thereby obtaining the vector P j Multiple concentrated distribution intervals, for this purpose, the vector P j The initial vector is named P j1 , that is, P j1 =P j , in the vector P j1 Search for the energy storage system power value p ij The concentrated distribution interval L j1 , L j1 =[x,x+l].
[0060] Specifically, l is the length of the concentrated distribution interval, and the length l is a positive constant, which can be dynamically determined according to the weather scene and the maximum power family of the actual photovoltaic power station energy storage system, that is, l = k × P max . Where k is the weather scene weight, P max It is the maximum value of the power family of the actual photovoltaic power station energy storage system.
[0061] Specifically, x is the lower limit power of the concentrated distribution interval, and the lower limit power x is a continuous variable, x∈[p jmin ,p jmax ], where p jmin is the energy storage system power vector P j1 Minimum power of energy storage system, p jmax is the energy storage system power vector P j1 The present invention adopts the bird flock algorithm to find the lower limit power x, and the vector P j1 The power value of the energy storage system falls into the concentrated distribution interval L j1 The maximum number of points is taken as the objective function, and the objective function is shown in the following formula.
[0062]
[0063]
[0064] After that, the records are concentrated in the distribution interval L j1 Contains vector P j1 The energy storage system power value is constructed to construct the concentrated distribution interval vector A j1 , A j1 =[p' 1j p' 2j …p' rj ] Τ . Calculate the concentrated distribution interval L j1 The distribution probability F j1 .
[0065]
[0066] Among them, r is the concentrated distribution interval vector A j1 Power value points of energy storage system.
[0067] Then, according to the concentrated distribution interval vector A j1 , calculate the vector P j1 The characteristic power value E j1 , E j1 is a continuous variable, E j1 ∈[p' jmin ,p' jmax ], where p' jmin is the concentrated distribution interval vector A j1 Minimum power of energy storage system, p' jmax is the concentrated distribution interval vector A j1 The maximum power of the energy storage system in the middle. In this embodiment, the bird flock algorithm is used to find the optimal vector P j1 The characteristic power value E j1, the characteristic power value E j1 To the concentrated distribution interval vector A j1 The shortest Euclidean distance of all elements in is taken as the objective function, and the objective function is shown in the following formula.
[0068]
[0069] For comparison, if the distribution probability F j1 If the distribution probability F is less than the preset threshold ε, then j = j + 1, that is, in the subsequent steps, we start to find the characteristic power value of the photovoltaic power station energy storage system at time j + 1. j1 is greater than the set threshold ε, then from vector P j1 Remove the concentrated distribution interval vector A j1 The power value in forms a new vector P j2 , refer to the above steps to obtain the concentrated distribution interval L j2 , concentrated distribution interval L j2 The distribution probability F j2 And the characteristic power value E j2 .
[0070] Among them, when setting the threshold ε, it can be adjusted according to the distribution degree of the power value of the actual photovoltaic power station energy storage system. For example, in sunny weather scenarios, the power value distribution of the photovoltaic power station energy storage system is relatively concentrated, so the threshold ε can be appropriately increased; in cloudy weather scenarios, the power value distribution of the photovoltaic power station energy storage system is relatively dispersed, so the threshold ε can be appropriately reduced.
[0071] Repeat the above steps until the distribution probability F js Less than the set threshold ε, all characteristic power values E js It is stored in the preset characteristic power value set to form the characteristic power value vector E of the photovoltaic power station energy storage system at time j j , E j =[E j1 E j2 …E jg ] Τ . Where s is the vector P j The transformation vector number is the vector P j The concentrated distribution interval number obtained by optimization, s∈[1,2,…,g]. g is the vector P j The total number of transformation vectors, that is, vector P j The total number of concentrated distribution intervals obtained by optimization.
[0072] Furthermore, according to the characteristic power value vector, a typical power curve is determined, including: based on a roulette algorithm, a target characteristic power value is determined in the characteristic power value vector, the target characteristic power value is used to indicate the typical power value of the energy storage system at the moment indicated by the target power vector; if the moment indicated by the target power vector is not the last moment indicated by the target power matrix, the target characteristic power value is stored in a preset typical power value set, and the target power vector is updated in sequence, the moment indicated by the updated target power vector is the next moment of the moment indicated by the target power vector before the update; if the moment indicated by the target power vector is the last moment indicated by the target power matrix, after storing the target characteristic power value in the preset typical power value set, a typical power curve is generated according to all target characteristic power values stored in the typical power value set.
[0073] In this embodiment, the characteristic power value vector E of the photovoltaic power station energy storage system at time j is j In the roulette algorithm, the characteristic power value vector E is selected j A certain characteristic power value E in js As the typical power value p of the photovoltaic power station energy storage system at time j j The characteristic power value E js As the typical power value p of the photovoltaic power station energy storage system at time j j The individual selection probability of is as follows:
[0074]
[0075] Finally, the typical power value of the photovoltaic power station energy storage system at each moment is calculated in turn until the moment indicated by the target power vector is the last moment indicated by the target power matrix. That is, after the calculation of the typical power value of the photovoltaic power station energy storage system at all moments is completed, a typical power curve P that can represent the operating characteristics of the energy storage system in the context of smoothing the power fluctuation of the photovoltaic power station in practice can be obtained. optimal =[p1 p2…p n ].
[0076] As a feasible implementation method, a typical power table of a photovoltaic power station energy storage system is added for typical power data. The table includes fields such as seasonal scene code, weather scene code, photovoltaic power station code, typical power data of the photovoltaic power station energy storage system under typical weather scenes, and the corresponding time of each power data. The data in the typical power curve of the photovoltaic power station energy storage system obtained by mining is recorded in the typical power table of the photovoltaic power station energy storage system.
[0077] As another aspect of an embodiment of the present application, an embodiment of the present application also provides a data feature mining device for an energy storage system. The device can be a software module. The software module includes a number of instructions stored in a memory. A processor can access the memory and call the instructions for execution to complete the data feature mining method for the energy storage system described in the above-mentioned embodiments.
[0078] In some embodiments, the data feature mining device of the energy storage system can also be constructed by hardware devices. For example, the data feature mining device of the energy storage system can be constructed by one or more chips, and each chip can work in coordination with each other to complete the data feature mining method of the energy storage system described in each of the above embodiments. For another example, the data feature mining device of the energy storage system can also be constructed by various logic devices, such as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (AcornRISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components.
[0079] For details, please refer to Figure 3 , Figure 3 The figure shows a schematic diagram of the structure of a data feature mining device for an energy storage system, as shown in the figure, comprising:
[0080] The data acquisition module 210 is used to acquire first data of the photovoltaic power station, where the first data includes power data of the photovoltaic power station under various weather scenarios;
[0081] A data processing module 220, configured to perform clustering processing on the first data according to the weather scenario, so as to divide the first data into a plurality of photovoltaic power plant power clusters, each photovoltaic power plant power cluster corresponding to a weather scenario;
[0082] The data analysis module 230 is used to determine the power data of the energy storage system corresponding to each photovoltaic power station power group based on the preset control cloud, and generate the energy storage system power group corresponding to each photovoltaic power station power group;
[0083] The data extraction module 240 is used to extract data features from each energy storage system power group through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, where the typical power curve is used to indicate the charge and discharge characteristics of the energy storage system.
[0084] Further, see Figure 4 , Figure 430 is a schematic diagram of a computer device provided in an embodiment of the present application. The computer device 30 includes one or more processors 31 and a memory 32. The memory 32 is connected to the one or more processors 31, for example, connected to the processor 31 via a bus.
[0085] The processor 31 is configured to support the computer device to perform the corresponding functions in the method in the above method embodiment. The processor 31 can be a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.
[0086] The memory 32 is used to store program codes, etc. The memory 32 may include a volatile memory (VM), such as a random access memory (RAM); the memory 32 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory 32 may also include a combination of the above-mentioned types of memories.
[0087] The memory 32 may be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the data feature mining method for the energy storage system in the embodiment of the present application. The processor 31 executes various functional applications and data processing of the data feature mining method for the energy storage system and the data feature mining device for the energy storage system by running the non-volatile software programs, instructions and modules stored in the memory 32, that is, realizes the functions of each module or unit of the data feature mining method for the energy storage system and the data feature mining device for the energy storage system provided in the above method embodiment.
[0088] The memory 32 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function. The data storage area may store data created based on the use of the data feature mining device of the energy storage system, etc. In some embodiments, the memory 32 may optionally include a memory remotely arranged relative to the processor 31, and these remote memories may be connected to the data feature mining device of the energy storage system via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0089] One or more modules are stored in the memory 32. When executed by one or more processors 31, the data feature mining method of the energy storage system in any of the above method embodiments is executed, for example, the method steps described in the above method embodiments are executed to realize the functions of the modules described in the above device embodiments.
[0090] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the method of the aforementioned embodiment.
[0091] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.
[0092] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0093] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0094] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A data feature mining method for an energy storage system, characterized in that: An energy storage system applied to a photovoltaic power station, the method comprising: Acquire first data of the photovoltaic power station, where the first data includes power data of the photovoltaic power station under various weather scenarios; performing clustering processing on the first data according to the weather scenario to divide the first data into a plurality of photovoltaic power plant power clusters, each photovoltaic power plant power cluster corresponding to a weather scenario; Based on the preset control cloud, determine the power data of the energy storage system corresponding to each photovoltaic power station power group, and generate the energy storage system power group corresponding to each photovoltaic power station power group; By using a double-layer bird flock algorithm, data features are extracted from each energy storage system power group to determine a typical power curve of the energy storage system, where the typical power curve is used to indicate the charging and discharging characteristics of the energy storage system.
2. The method according to claim 1, characterized in that: The first data of the photovoltaic power station is obtained according to a typical weather scene identification table, where the typical weather scene identification table is used to indicate identifications of various weather scenes. The method further includes: Obtain meteorological data within the target time period; Classifying the meteorological data according to months to determine a plurality of seasonal scenes, each of the seasonal scenes including meteorological data of a plurality of months; The meteorological data in each of the seasonal scenes is divided according to date, and the meteorological data corresponding to each date in each seasonal scene is determined to form the typical weather scene identification table.
3. The method according to claim 1, characterized in that The step of clustering the first data according to the weather scenario to divide the first data into a plurality of photovoltaic power plant power clusters specifically includes: According to a preset typical weather scene identification table, the first data is divided into multiple categories of target data according to seasonal scenes, and each category of target data corresponds to the same seasonal scene; Each type of target data is clustered and divided respectively, so as to divide each type of target data into multiple photovoltaic power station power families according to the weather scenario. Each photovoltaic power station power family corresponds to the same type of weather scenario, and the data of each photovoltaic power station power family is recorded in a preset typical weather scenario photovoltaic power station data table.
4. The method according to claim 1, characterized in that: The power data of the photovoltaic power station power group corresponds to a timestamp, and the step of determining the power data of the energy storage system corresponding to each photovoltaic power station power group based on the preset control cloud, and generating the energy storage system power group corresponding to each photovoltaic power station power group specifically includes: Obtaining the energy storage unit information table and energy storage unit operation data table of the control cloud; According to the energy storage unit information table of the control cloud, query the energy storage unit code corresponding to the photovoltaic power station; According to the energy storage unit code, determining, in the energy storage unit operation data table of the control cloud, power data corresponding to the timestamp of the power data of the photovoltaic power station power family; Based on the power data and according to the timestamp, the energy storage system power group corresponding to each photovoltaic power station power group is determined.
5. The method according to claim 1, characterized in that The step of extracting data features from each energy storage system power family by using a double-layer bird flock algorithm to determine a typical power curve of the energy storage system includes: Determine a target energy storage system power family, where the target energy storage system power family is one of a plurality of energy storage system power families; Determine a target power matrix according to the target energy storage system power family, wherein the target power matrix is used to indicate the power values of the energy storage system at different dates and times in the target energy storage system power family; According to the target power matrix, sequentially determine target power vectors, wherein the target power vectors are used to indicate power values of the energy storage system at the same time on different dates in the target power matrix; Based on the target power vector, data features are extracted from the target energy storage system power family through the double-layer bird flock algorithm to obtain the typical power curve.
6. The method according to claim 5, characterized in that The step of extracting data features of the target energy storage system power family based on the target power vector by using the double-layer bird flock algorithm to obtain the typical power curve specifically includes: The lower limit power of the concentrated distribution interval of the target power vector is obtained by optimizing the double-layer bird flock algorithm to form a concentrated distribution interval; Determine a power value of the energy storage system including the target power vector within the concentrated distribution interval to determine an interval vector of the concentrated distribution interval, and determine a distribution probability of the concentrated distribution interval; According to the interval vector, determining the characteristic power value of the target power vector by optimizing the double-layer bird flock algorithm, and storing the characteristic power value in a preset characteristic power value set; If the distribution probability of the concentrated distribution interval is not less than a preset threshold, the power value in the interval vector of the concentrated distribution interval is removed from the target power vector to determine a new target power vector, and the characteristic power value is re-determined according to the new target power vector by using the double-layer bird flock algorithm; If the distribution probability of the concentrated distribution interval is less than a preset threshold, determining a characteristic power value vector according to all characteristic power values stored in the characteristic power value set; The typical power curve is determined according to the characteristic power value vector.
7. The method according to claim 6, characterized in that The step of determining the typical power curve according to the characteristic power value vector specifically includes: Based on a roulette algorithm, determining a target characteristic power value in the characteristic power value vector, wherein the target characteristic power value is used to indicate a typical power value of the energy storage system at the time indicated by the target power vector; If the time indicated by the target power vector is not the last time indicated by the target power matrix, the target characteristic power value is stored in a preset typical power value set, and the target power vector is updated in sequence, and the time indicated by the updated target power vector is the next time after the time indicated by the target power vector before the update; If the moment indicated by the target power vector is the last moment indicated by the target power matrix, then after storing the target characteristic power value in a preset typical power value set, the typical power curve is generated according to all target characteristic power values stored in the typical power value set.
8. A data feature mining device for an energy storage system, characterized in that: An energy storage system applied to a photovoltaic power station, the device comprising: A data acquisition module, used to acquire first data of the photovoltaic power station, wherein the first data includes power data of the photovoltaic power station under various weather scenarios; A data processing module, configured to perform clustering processing on the first data according to weather scenarios, so as to divide the first data into a plurality of photovoltaic power plant power clusters, each photovoltaic power plant power cluster corresponding to a weather scenario; A data analysis module, used to determine the power data of the energy storage system corresponding to each photovoltaic power station power group based on a preset control cloud, and generate the energy storage system power group corresponding to each photovoltaic power station power group; The data extraction module is used to extract data features of each energy storage system power family through a double-layer bird flock algorithm to determine a typical power curve of the energy storage system, wherein the typical power curve is used to indicate the charge and discharge characteristics of the energy storage system.
9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1 to 7.
10. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the method according to any one of claims 1 to 7.
