Disclosure of Invention
The invention aims to provide a method, a system, a storage medium and electronic equipment for predicting electric load, which can accurately predict the load in the future of an electric load unit, and greatly improve the evaluation index of a prediction result and the robustness of load prediction.
To achieve the above object, in a first aspect, the present invention provides a method for predicting an electrical load, including: and constructing a power network map according to the topological relation of the power network. And preprocessing each power load unit according to the historical power load data, and calculating to obtain a first characteristic, a second characteristic and a third characteristic of the power load. And for each power load unit, comprehensively using the first characteristic, the second characteristic and the third characteristic of the power load as model training and prediction input, and training by using a time series model to obtain a power load prediction model corresponding to each power load unit. And loading the electric load prediction model corresponding to each electric load unit, and predicting the future electric load.
In an embodiment of the present invention, the first characteristic of the electric load is an electric load characteristic corresponding to each electric load unit, the second characteristic of the electric load is an electric load characteristic mean value of the same type of electric load units in the same distribution substation, and the third characteristic of the electric load is an electric load characteristic mean value of the other same type of electric load units in the same-region substation and in the different distribution substations.
In one embodiment of the present invention, the using the first, second, and third characteristics of the electrical load as the model training and prediction input in combination for each electrical load unit includes: and taking vectors concat corresponding first, second and third characteristics of the electric load of each electric load unit as model training and prediction input.
In one embodiment of the present invention, the training using the time-series model to obtain the electrical load prediction model corresponding to each electrical load unit includes: and selecting the average absolute error MAE by using a neural network LSTM and a loss function, thereby obtaining an electric load prediction model corresponding to each electric load unit.
In a second aspect, an embodiment of the present invention further provides a system for predicting an electrical load, including: the device comprises a network map building module, a calculating module, a prediction model building module and a prediction module. The network map building module is used for building a power network map according to the power network topological relation. The calculation module is used for preprocessing each power load unit according to historical power load data of the power load unit, and calculating to obtain a first characteristic, a second characteristic and a third characteristic of the power load unit. And constructing a prediction model module for comprehensively using the first characteristic, the second characteristic and the third characteristic of the power consumption load as model training and prediction input for each power consumption load unit, and training by using a time sequence model to obtain the power consumption load prediction model corresponding to each power consumption load unit. And the prediction module is used for loading the electric load prediction model corresponding to each electric load unit and predicting the future electric load.
In a third aspect, an embodiment of the present invention further provides a storage medium, where the storage medium stores computer-executable instructions, and the computer-executable instructions are configured to perform the above-mentioned method for predicting a power load.
In a fourth aspect, an embodiment of the present invention further provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the above-described method of power load prediction.
Compared with the prior art, the method, the system, the storage medium and the electronic equipment for predicting the electric load fully utilize the graph incidence relation of the power network graph, and except for using the electric load characteristic corresponding to the electric load unit as a first characteristic when the electric load is predicted; finding other same type of electric load units under the same power distribution station through diagram query of specific conditions, and integrating the electric load characteristics of the other same type of electric load units as second characteristics; and searching other same-type electric load units under the transformer substation in the same region and under the non-same distribution substation through a diagram under specific conditions, and integrating the electric load characteristics of the other same-type electric load units as a third characteristic. When the prediction model of the power load unit is trained, the power load rule of the unit can be learned from the first characteristic, the power load rules of other units of the same type under the same power distribution station can be learned from the second characteristic, and the power load rules of other units of the same type under the same area transformer substation can be learned from the third characteristic. The three types of electric load laws comprehensively learned by the electric load prediction model are used for predicting the load in the future of an electric load unit, so that the prediction result evaluation index and the robustness of load prediction are greatly improved.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
Fig. 1 is a flow chart illustrating a method for predicting a power load according to an embodiment of the present invention. As shown in fig. 1, in a first aspect, a method for predicting a power load according to a preferred embodiment of the present invention includes: and S1, constructing a power network map according to the power network topological relation. And S2, preprocessing each power load unit according to the historical power load data, and calculating to obtain the first characteristic, the second characteristic and the third characteristic of the power load. And S3, comprehensively using the first characteristic, the second characteristic and the third characteristic of the power load as model training and prediction input for each power load unit, and training by using a time series model to obtain a power load prediction model corresponding to each power load unit. And S4, loading the electric load prediction model corresponding to each electric load unit, and predicting the future electric load.
In an embodiment of the present invention, the first characteristic of the electric load is an electric load characteristic corresponding to each electric load unit, the second characteristic of the electric load is an electric load characteristic mean value of the same type of electric load units in the same distribution substation, and the third characteristic of the electric load is an electric load characteristic mean value of the other same type of electric load units in the same-region substation and in the different distribution substations.
In one embodiment of the present invention, the using the first, second, and third characteristics of the electrical load as the model training and prediction input in combination for each electrical load unit includes: and taking vectors concat corresponding first, second and third characteristics of the electric load of each electric load unit as model training and prediction input.
In one embodiment of the present invention, the training using the time-series model to obtain the electrical load prediction model corresponding to each electrical load unit includes: and selecting the average absolute error MAE by using a neural network LSTM and a loss function, thereby obtaining an electric load prediction model corresponding to each electric load unit.
In a second aspect, an embodiment of the present invention further provides a system for predicting an electrical load, including: the device comprises a network map building module, a calculating module, a prediction model building module and a prediction module. The network map building module is used for building a power network map according to the power network topological relation. The calculation module is used for preprocessing each power load unit according to historical power load data of the power load unit, and calculating to obtain a first characteristic, a second characteristic and a third characteristic of the power load unit. And constructing a prediction model module for comprehensively using the first characteristic, the second characteristic and the third characteristic of the power consumption load as model training and prediction input for each power consumption load unit, and training by using a time sequence model to obtain the power consumption load prediction model corresponding to each power consumption load unit. And the prediction module is used for loading the electric load prediction model corresponding to each electric load unit and predicting the future electric load.
In an embodiment of the present invention, the first characteristic of the electric load is an electric load characteristic corresponding to each electric load unit, the second characteristic of the electric load is an electric load characteristic mean value of the same type of electric load units in the same distribution substation, and the third characteristic of the electric load is an electric load characteristic mean value of the other same type of electric load units in the same-region substation and in the different distribution substations.
In one embodiment of the present invention, the using the first, second, and third characteristics of the electrical load as the model training and prediction input in combination for each electrical load unit includes: and taking vectors concat corresponding first, second and third characteristics of the electric load of each electric load unit as model training and prediction input.
In one embodiment of the present invention, the training using the time-series model to obtain the electrical load prediction model corresponding to each electrical load unit includes: and selecting the average absolute error MAE by using a neural network LSTM and a loss function, thereby obtaining an electric load prediction model corresponding to each electric load unit.
In a third aspect, an embodiment of the present invention further provides a storage medium, where the storage medium stores computer-executable instructions, and the computer-executable instructions are configured to perform the above-mentioned method for predicting a power load.
The storage medium may be any available medium or data storage device that can be accessed by a computer, including but not limited to magnetic memory (e.g., floppy disks, hard disks, magnetic tape, magneto-optical disks (MOs), etc.), optical memory (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor memory (e.g., ROMs, EPROMs, EEPROMs, nonvolatile memory (NAND FLASH), Solid State Disks (SSDs)), etc.
In a fourth aspect, fig. 5 shows a block diagram of an electronic device according to another embodiment of the invention. The electronic device 1100 may be a host server with computing capabilities, a personal computer PC, or a portable computer or terminal that is portable, or the like. The specific embodiment of the present invention does not limit the specific implementation of the electronic device.
The electronic device 1100 includes at least one processor (processor)1110, a Communications Interface 1120, a memory 1130, and a bus 1140. The processor 1110, the communication interface 1120, and the memory 1130 communicate with each other via the bus 1140.
The communication interface 1120 is used for communicating with network elements including, for example, virtual machine management centers, shared storage, etc.
Processor 1110 is configured to execute programs. Processor 1110 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present invention.
The memory 1130 is used for executable instructions. The memory 1130 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 1130 may also be a memory array. The storage 1130 may also be partitioned and the blocks may be combined into virtual volumes according to certain rules. The instructions stored by the memory 1130 are executable by the processor 1110 to enable the processor 1110 to perform the method of power load prediction in any of the method embodiments described above.
Fig. 2 is a schematic diagram of a power plant map of a method of power load prediction according to an embodiment of the invention. As shown in fig. 2, the hub substation includes a plurality of regional substations, each of which includes a plurality of distribution stations, each of which includes a plurality of residential meters or a plurality of commercial meters. The power equipment map is merely an example, and the nodes of the power equipment may include residential meters, commercial meters, industrial meters, distribution substations, regional substations, hub substations, and the like. Because the electrical transmission is directional, the priority of the nodes in the graph may be considered hub substation nodes > regional substation nodes > distribution substation nodes > residential meters/commercial meters/industrial meters. Wherein the entities in the graph (i.e., nodes in the graph) represent power equipment nodes, and the relationships in the graph (i.e., edges in the graph) represent two power equipment nodes directly connected by wires.
Fig. 3 is a simplified flowchart of a method for predicting a power load according to an embodiment of the present invention. In practical application, as shown in fig. 3, after the method of the present invention is started, a power network map is constructed, and meanwhile, historical power load data is obtained and is subjected to data preprocessing, so as to obtain a first characteristic of the power load. And searching other same type of electric load units under the same power distribution station through a diagram of specific conditions, and synthesizing the electric load characteristics of the other same type of electric load units to obtain a second characteristic. And through the map query of specific conditions, other electric load units of the same type under the transformer substation in the same region and under the non-same distribution substation are found, and the electric load characteristics of the other electric load units are integrated to obtain a third characteristic. And comprehensively using the first characteristic, the second characteristic and the third characteristic, and obtaining a power load prediction model through an LSTM neural network and an RMSE loss function. And predicting the future electric load by loading the electric load prediction model corresponding to each electric load unit.
FIG. 4 is a schematic flow diagram of model training/prediction for a method of power load prediction according to an embodiment of the present invention. As shown in fig. 4, the first characteristic of the electrical load is calculated as follows: the manner of calculation of the electrical load characteristics is as follows. The smart meter collects data every 15 minutes, and for a 10 day power load, the time series contains 960 load data, such as [ x1, x2, x3, … …, x959, x960 ]. Then calculating the power load characteristics of the sequence, including: 1. the trend characteristics of the time series can be calculated by a moving average method; 2. the periodic characteristics of the time sequence can be calculated through the autocorrelation coefficients, namely, enough phase differences are traversed, the maximum autocorrelation coefficient is found, and the phase difference corresponding to the maximum autocorrelation coefficient is the period; 3. the environmental characteristics of the time series, namely the temperature and the humidity of the environment where the electric load acquisition equipment is located, are collected through a sensor beside the equipment. The second characteristic of the electric load is the average value of the characteristics of the electric loads of other electric load units of the same type under the same power distribution station. In the power network map, the second characteristic of the electrical load is calculated as follows: for example, for (residential meter 1), by querying (residential meter 1) - (distribution station) - (residential meter) through a graph, all (residential meter) nodes associated with (residential meter 1) in two degrees are found, and the first characteristics of their power consumption loads are averaged, i.e., the second characteristics of the power consumption loads of (residential meter 1). And similarly, calculating the second characteristic of the electric load of the commercial electric meter/industrial electric meter. And the third characteristic of the electric load is the average value of the characteristics of the electric loads of other electric load units of the same type in the transformer substation in the same region and in a non-same distribution substation. In the power network map, the third characteristic of the electrical load is calculated as follows: for example, for (commercial electric meter 1), through a graph query of (commercial electric meter 1) - (distribution station x) - (regional substation) - (distribution station y) - (commercial electric meter), all (commercial electric meter) nodes which are associated with (commercial electric meter 1) four degrees and are not under the same distribution station are found, and the average of the power load characteristics of the nodes is the third power load characteristic of (commercial electric meter 1). And similarly, calculating the third characteristic of the electricity load of the residential electric meter/industrial electric meter.
In summary, the method, system, storage medium, and electronic device for predicting electrical load of the present invention make full use of the graph association relationship of the power network map, and when predicting the electrical load, use the electrical load characteristic corresponding to the electrical load unit as the first characteristic; finding other same type of electric load units under the same power distribution station through diagram query of specific conditions, and integrating the electric load characteristics of the other same type of electric load units as second characteristics; and searching other same-type electric load units under the transformer substation in the same region and under the non-same distribution substation through a diagram under specific conditions, and integrating the electric load characteristics of the other same-type electric load units as a third characteristic. When the prediction model of the power load unit is trained, the power load rule of the unit can be learned from the first characteristic, the power load rules of other units of the same type under the same power distribution station can be learned from the second characteristic, and the power load rules of other units of the same type under the same area transformer substation can be learned from the third characteristic. The three types of electric load laws comprehensively learned by the electric load prediction model are used for predicting the load in the future of an electric load unit, so that the prediction result evaluation index and the robustness of load prediction are greatly improved.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.