Power grid load prediction method, device, equipment, storage medium and product

Abstract

The invention discloses a power grid load prediction method, device and equipment, a storage medium and a product. The method comprises the steps of obtaining historical operation data of a to-be-predicted power grid cluster; inputting historical operation data of the to-be-predicted power grid cluster into the target prediction model, and obtaining an output load prediction result; wherein the target prediction model is a fusion model obtained by training according to the first prediction model and the second prediction model. According to the power grid load prediction method disclosed by the invention, the first prediction model and the second prediction model are combined to obtain the target prediction model, so that the accuracy of the obtained target prediction model is higher, and a more accurate load prediction result can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of power forecasting, and in particular to a power grid load forecasting method, device, equipment, storage medium and product. Background Art

[0002] The ideal scenario of smart grid is to ensure that the power supply matches the power demand of the target area. This not only helps to effectively allocate power resources and bring significant economic benefits, but also has important social value. Smart grid can cope with a variety of uncertain factors that lead to large-scale power outages, such as natural disasters, seasonal peak power consumption, terrorist attacks, urban unrest and war.

[0003] Among the existing power grid load forecasting methods, the most commonly used is the classic time series method. Among them, the autoregressive integrated moving average method is a representative model of the time series method. Its advantages are that the model is relatively simple, robust, and efficient, and can handle seasonal time series well. However, its prediction accuracy needs to be further improved. Summary of the invention

[0004] The present invention provides a power grid load prediction method, device, equipment, storage medium and product to achieve accurate prediction of power grid load.

[0005] According to one aspect of the present invention, a method for predicting power grid load is provided, comprising:

[0006] Obtain historical operation data of the power grid cluster to be predicted;

[0007] Inputting the historical operation data of the power grid cluster to be predicted into the target prediction model to obtain the output load prediction result;

[0008] The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

[0009] Furthermore, the first prediction model is a Transformer model, the second prediction model is an improved Transformer model, and the residual connection-normalization layer of the second prediction model is a membership calculation method.

[0010] Furthermore, the method for establishing the second prediction model includes:

[0011] The power grid cluster characteristic data of the set period is used as the input data of the forward generator and the reverse generator of the cloud model to obtain the target membership function;

[0012] The residual connection-normalization layer of the Transformer model is replaced by a membership calculation method according to the target membership function.

[0013] Furthermore, the training method of the target prediction model includes:

[0014] Using the power grid cluster characteristic data of a set period as a training data set, and training the first prediction model and the second prediction model respectively;

[0015] Determine the weight coefficients corresponding to the first prediction model and the second prediction model respectively, and perform weighted summation of the first prediction model and the second prediction model to obtain the target prediction model.

[0016] Further, determining the weight coefficients corresponding to the first prediction model and the second prediction model respectively includes:

[0017] Using a target weight vector as a position vector of the particle swarm algorithm, the target weight vector includes weight coefficients corresponding to the first prediction model and the second prediction model respectively;

[0018] According to the preset particle swarm algorithm parameters, combined with the particle swarm algorithm, the optimal solution of the target weight vector is obtained.

[0019] According to another aspect of the present invention, there is provided a power grid load prediction device, comprising:

[0020] A module for acquiring historical operation data of the power grid cluster to be predicted, used for acquiring historical operation data of the power grid cluster to be predicted;

[0021] A load forecasting module, used to input the historical operation data of the power grid cluster to be predicted into a target prediction model to obtain an output load forecasting result;

[0022] The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

[0023] Optionally, the first prediction model is a Transformer model, the second prediction model is an improved Transformer model, and the residual connection-normalization layer of the second prediction model is a membership calculation method.

[0024] Optionally, the method for establishing the second prediction model includes:

[0025] The power grid cluster characteristic data of the set period is used as the input data of the forward generator and the reverse generator of the cloud model to obtain the target membership function;

[0026] The residual connection-normalization layer of the Transformer model is replaced by a membership calculation method according to the target membership function.

[0027] Optionally, the training method of the target prediction model includes:

[0028] Using the power grid cluster characteristic data of a set period as a training data set, and training the first prediction model and the second prediction model respectively;

[0029] Determine the weight coefficients corresponding to the first prediction model and the second prediction model respectively, and perform weighted summation of the first prediction model and the second prediction model to obtain the target prediction model.

[0030] Optionally, determining weight coefficients corresponding to the first prediction model and the second prediction model respectively includes:

[0031] Using a target weight vector as a position vector of the particle swarm algorithm, the target weight vector includes weight coefficients corresponding to the first prediction model and the second prediction model respectively;

[0032] According to the preset particle swarm algorithm parameters, combined with the particle swarm algorithm, the optimal solution of the target weight vector is obtained.

[0033] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0034] at least one processor; and

[0035] a memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the power grid load forecasting method described in any embodiment of the present invention.

[0037] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the power grid load forecasting method described in any embodiment of the present invention when executed.

[0038] According to another aspect of the present invention, a computer program product is provided, which includes a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the power grid load forecasting method described in any embodiment of the present invention are implemented.

[0039] The power grid load forecasting method disclosed in the present invention first obtains the historical operation data of the power grid cluster to be forecasted; then inputs the historical operation data of the power grid cluster to be forecasted into the target forecasting model to obtain the output load forecasting result; wherein the target forecasting model is a fusion model obtained by training the first forecasting model and the second forecasting model. The power grid load forecasting method disclosed in the present invention combines the first forecasting model with the second forecasting model to obtain the target forecasting model, which can make the obtained target forecasting model more accurate, thereby obtaining a more accurate load forecasting result.

[0040] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 is a flow chart of a power grid load prediction method provided according to Embodiment 1 of the present invention;

[0043] Figure 2 is a schematic diagram of a training process of a target prediction model provided according to Embodiment 1 of the present invention;

[0044] Figure 3 is a structural schematic diagram of a power grid load prediction device provided according to Embodiment 2 of the present invention;

[0045] Figure 4 It is a structural schematic diagram of an electronic device for implementing the power grid load prediction method of the third embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] Embodiment 1

[0049] Figure 1 This is a flowchart of a power grid load prediction method provided in the first embodiment of the present invention. This embodiment is applicable to the case of predicting power grid load. The method can be executed by a power grid load prediction device. The power grid load prediction device can be implemented in the form of hardware and / or software. The power grid load prediction device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0050] S110, obtaining historical operation data of the power grid cluster to be predicted.

[0051] The power grid cluster to be predicted is the power grid cluster that currently needs to be predicted, and the historical operation data of the power grid cluster to be predicted is the historical power, temperature and other data of the power grid cluster to be predicted.

[0052] S120, inputting the historical operation data of the power grid cluster to be predicted into the target prediction model to obtain the output load prediction result.

[0053] The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

[0054] In this embodiment, after obtaining the historical operation data of the power grid cluster to be predicted, it can be input into the trained target prediction model, and the output of the trained target prediction model is the load prediction result obtained by performing load prediction on the current power grid cluster to be predicted.

[0055] Furthermore, the first prediction model is a Transformer model, the second prediction model is an improved Transformer model, and the residual connection-normalization layer of the second prediction model is a membership calculation method.

[0056] Among them, the Transformer model is a deep learning model, which consists of two parts: Encoder and Decoder. Encoder and Decoder include Add&Norm (residual connection-normalization) layer, FeedForward (full connection) layer and Multi-Head Attention (multi-head attention) layer respectively. The Add&Norm layer consists of Add (residual connection) and Norm (normalization). Add can increase the depth without losing the initial features, and Norm can convert the features to a comparable scale.

[0057] In this embodiment, the first prediction model is a Transformer model, and the calculation method of its Add&Norm layer is: X=(X-Ex) / X.std, where X is the input of the Add&Norm layer, Ex is the expected value, and X.std is the standard deviation. The second prediction model is an improved Transformer model, and the calculation method of its Add&Norm layer is: X=u(x), where X is the input of the Add&Norm layer, and u(x) is the degree of membership.

[0058] Optionally, the method for establishing the second prediction model may be: using the grid cluster characteristic data of a set time period as input data for the forward generator and reverse generator of the cloud model to obtain a target membership function; and replacing the residual connection-normalization layer of the Transformer model with a membership calculation method according to the target membership function.

[0059] Specifically, the second prediction model can be obtained by optimizing the Transformer model through the cloud model. The cloud model is an uncertain transformation model that processes qualitative concepts and quantitative descriptions. The characteristic data of the power grid cluster in a set period is input as sample data into the forward generator and reverse generator of the cloud model. The reverse generator can calculate the sample mean Y and calculate the expectation Ex = Y. The main function of the forward cloud generator is to generate a normal random number En' and also generate a number with Ex as the expected value, En' 2 The normal random number x with variance is used to generate the target membership function u(x):

[0060]

[0061] After obtaining the target membership function u(x), the Add&Norm layer of the original Transformer model can be changed to cloud normalization, that is, the calculation method of all "Add&Norm" layers of the original Transformer can be changed, and the original "X=(X-Ex) / X.std" calculation method can be changed to the cloud membership calculation method "X=u(x)".

[0062] Furthermore, the training method of the target prediction model can be: using the power grid cluster characteristic data of a set time period as a training data set, and training the first prediction model and the second prediction model respectively; determining the weight coefficients corresponding to the first prediction model and the second prediction model respectively, and performing weighted summation of the first prediction model and the second prediction model to obtain the target prediction model.

[0063] In this embodiment, in order to obtain the target prediction model, the first prediction model and the second prediction model can be trained respectively, and the training data set is the characteristic data of the power grid cluster in the set time period, that is, the characteristic data such as the temperature characteristic value, power characteristic value, load label value, etc. of the power grid cluster collected in a set time period. The training data set is imported into the first prediction model and the second prediction model respectively, and the model parameters are adjusted through training so that the performance of the trained first prediction model and the second prediction model meets the preset requirements. The first prediction model and the second prediction model are represented as F1 and F2 respectively, and the target prediction model is represented as F3, and F3 = (F1 + F2) W, where W is the weight vector, W = (W1, W2), and W1 and W2 are the weight coefficients corresponding to the first prediction model and the second prediction model respectively. After training F1 and F2 to the optimal, the weight vector W is determined to obtain the target prediction model F3.

[0064] Optionally, the method for determining the weight coefficients corresponding to the first prediction model and the second prediction model respectively may be: using the target weight vector as the position vector of the particle swarm algorithm, the target weight vector includes the weight coefficients corresponding to the first prediction model and the second prediction model respectively; according to preset particle swarm algorithm parameters, combined with the particle swarm algorithm, the optimal solution of the target weight vector is obtained.

[0065] Specifically, the particle swarm algorithm (PSO) is an evolutionary computing technology that originated from the study of bird flock predation behavior. Its basic idea is to find the optimal solution through collaboration and information sharing between individuals in the group. In the particle swarm algorithm, q particles can be randomly generated, each particle contains two vectors, position and velocity. The position vector of the i-th particle is the weight coefficient vector W of the n+1 particle layer. i =(w i,1 ,...,w i,n+1 ), where w i,j ∈[0,1], 1≤j≤n+1, w i,1 +w i,2 +...+w i,n+1 =1, velocity vector V i =(v i,1 ,...,v i,n+1 )(v i,j ∈[-1,1], 1≤j≤n+1) is used to control W i The update speed of the vector Vi and W i The update formula at the tth iteration is:

[0066] V i，t =ω×V i，t-1 +C1×r1×(P best，i -W i，t-1 )+C2×r2×(P gbest -W i，t-1 )

[0067] W i，t =W i，t-1 +V i，t

[0068] Among them, ω is the inertia weight coefficient, C1 and C2 are the cognitive coefficient and social coefficient (also known as the acceleration constant), r1 and r2 are two standard normal distribution random numbers, P best,i is the optimal position vector of the ith particle, P gbest is the optimal position vector of all particles, and P best,i and P gbest All by W i Initialization. In each iteration, the optimization objective function of the i-th particle is calculated as follows:

[0069] f = avg_error t =||Y G -X G ·W i ||

[0070] Among them, X G and Y G are the predicted values ​​and observed values ​​of all particle layers at the current time. In each iteration of the particle swarm algorithm, P is first updated according to the optimization objective function value. best,i and P gbest , if the optimization objective function value of the i-th particle in this iteration is less than P best,i The optimization objective function value is, let P best,i =W i ; if P best,i The optimization objective function value is less than P gbest The optimization objective function value is, let P gbest =P best,i Then update V according to the formula i and W i When the predefined maximum number of iterations is reached, the algorithm ends and P gbest It is the optimal solution for the final target weight vector W.

[0071] According to the training method of the particle swarm algorithm, after obtaining the optimal solution of the target weight vector W, the target prediction model F3 = (F1 + F2) W can be determined. The target prediction model F3 can take the historical operation data of the power grid cluster to be predicted as input and output the corresponding load prediction result.

[0072] Figure 2 This is a schematic diagram of the training process of a target prediction model provided by an embodiment of the present invention. As shown in the figure, the target prediction model is obtained by fusing the Transformer model with the improved Transformer model. In the improved Transformer model, all residual connection-normalization layers are replaced by the membership calculation method. The Transformer model and the improved Transformer model are represented as F1 and F2, respectively. F1 and F2 are trained respectively, and then the target weight vector W is determined by the particle swarm algorithm to determine the target prediction model, that is, F3=(F1+F2)W.

[0073] In a specific application scenario, the total power supply data of 31 power grid clusters released by a power company from January 2021 to January 2023 is used for load forecasting. First, the characteristic data of the 31 power grid clusters of the power company collected in real time in the server are obtained to obtain 31 data sets D1, D2, ... D31. After data preprocessing, the collected data is input into the trained Transformer model, the improved Transformer model and the target prediction model respectively. The load forecast result is obtained according to the model output, and the prediction error is obtained according to the calculation formula of the mean absolute percentage error (MAPE). The results are as follows:

[0074] Table 1 Prediction results of different models for 31 data sets

[0075]

[0076]

[0077] According to the results shown in Table 1, it can be seen that the load forecasting results obtained by the target forecasting model have the lowest average absolute percentage error and the most accurate forecasting results.

[0078] The power grid load forecasting method disclosed in the present invention combines the original Transformer model with the Transformer model improved by using the cloud model to obtain a target forecasting model, and uses a particle swarm algorithm to optimize the weight coefficients in the target forecasting model, so that the obtained target forecasting model has a higher accuracy rate, thereby obtaining a more accurate load forecasting result.

[0079] Embodiment 2

[0080] Figure 3 A schematic diagram of the structure of a power grid load prediction device provided in the second embodiment of the present invention is shown in FIG. Figure 3 As shown, the device includes: a historical operation data acquisition module 310 for the power grid cluster to be predicted and a load prediction module 320.

[0081] The module 310 for acquiring historical operation data of the power grid cluster to be predicted is used to acquire historical operation data of the power grid cluster to be predicted.

[0082] The load forecasting module 320 is used to input the historical operation data of the power grid cluster to be forecasted into the target forecasting model to obtain the output load forecasting result.

[0083] The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

[0084] Optionally, the first prediction model is a Transformer model, the second prediction model is an improved Transformer model, and the residual connection-normalization layer of the second prediction model is a membership calculation method.

[0085] Optionally, the method for establishing the second prediction model includes:

[0086] The characteristic data of the power grid cluster in a set period of time is used as the input data of the forward generator and the reverse generator of the cloud model to obtain the target membership function; according to the target membership function, the residual connection-normalization layer of the Transformer model is replaced by the membership calculation method.

[0087] Optionally, the training method of the target prediction model includes:

[0088] The power grid cluster characteristic data of the set time period is used as the training data set to train the first prediction model and the second prediction model respectively; the weight coefficients corresponding to the first prediction model and the second prediction model are determined, and the first prediction model and the second prediction model are weighted summed to obtain the target prediction model.

[0089] Optionally, the method for determining the weight coefficients corresponding to the first prediction model and the second prediction model respectively may be:

[0090] The target weight vector is used as the position vector of the particle swarm algorithm, and the target weight vector includes weight coefficients corresponding to the first prediction model and the second prediction model respectively; according to the preset particle swarm algorithm parameters, combined with the particle swarm algorithm, the optimal solution of the target weight vector is obtained.

[0091] The power grid load prediction device provided in the embodiment of the present invention can execute the power grid load prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0092] Embodiment 3

[0093] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0094] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0095] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0096] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a power grid load forecasting method.

[0097] In some embodiments, the power grid load forecasting method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the power grid load forecasting described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the power grid load forecasting method in any other appropriate manner (e.g., by means of firmware).

[0098] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0099] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0100] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0101] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0102] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0103] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0104] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0105] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting power grid load, characterized in that: include: Obtain historical operation data of the power grid cluster to be predicted; Inputting the historical operation data of the power grid cluster to be predicted into the target prediction model to obtain the output load prediction result; The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

2. The method according to claim 1, characterized in that: The first prediction model is a Transformer model, the second prediction model is an improved Transformer model, and the residual connection-normalization layer of the second prediction model is a membership calculation method.

3. The method according to claim 2, characterized in that The method for establishing the second prediction model comprises: The power grid cluster characteristic data of the set period is used as the input data of the forward generator and the reverse generator of the cloud model to obtain the target membership function; The residual connection-normalization layer of the Transformer model is replaced by a membership calculation method according to the target membership function.

4. The method according to claim 1, characterized in that: The training method of the target prediction model includes: Using the power grid cluster characteristic data of a set period as a training data set, and training the first prediction model and the second prediction model respectively; Determine the weight coefficients corresponding to the first prediction model and the second prediction model respectively, and perform weighted summation of the first prediction model and the second prediction model to obtain the target prediction model.

5. The method according to claim 4, characterized in that Determining weight coefficients corresponding to the first prediction model and the second prediction model respectively includes: Using a target weight vector as a position vector of the particle swarm algorithm, the target weight vector includes weight coefficients corresponding to the first prediction model and the second prediction model respectively; According to the preset particle swarm algorithm parameters, combined with the particle swarm algorithm, the optimal solution of the target weight vector is obtained.

6. A power grid load prediction device, characterized in that: include: A module for acquiring historical operation data of the power grid cluster to be predicted, used for acquiring historical operation data of the power grid cluster to be predicted; A load forecasting module, used to input the historical operation data of the power grid cluster to be predicted into a target prediction model to obtain an output load forecasting result; The target prediction model is a fusion model trained based on the first prediction model and the second prediction model.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the power grid load forecasting method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the power grid load forecasting method according to any one of claims 1 to 5 when executed.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the power grid load forecasting method according to any one of claims 1 to 5 are implemented.

Images