Power prediction method and device based on CnnLstm algorithm
Through the power prediction method based on the CnnLstm algorithm, the feature extraction and prediction is used using meteorological data and power data, the problem of insufficient power prediction accuracy in the prior art is solved, and higher prediction accuracy is achieved.
Patent Information
- Application Number
- CN202411468684.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-21
- Publication Date
- 2025-02-07
AI Technical Summary
In the prior art, the power grid dispatching department has insufficient accuracy in the power station reporting power stations, resulting in a lack of accuracy in system power generation planning and real-time scheduling of power systems.
The power prediction method based on the CnnLstm algorithm is adopted, and the data structure and feature extraction are carried out by obtaining meteorological data and power data. In the CnnLstm algorithm model, the success rate prediction result is generated using the convolutional layer, the Lstm layer and the fully connected layer.
Through rich training data and advanced algorithm models, the accuracy of power prediction is improved, and the accuracy of system power generation planning and real-time scheduling of power systems is enhanced.
Smart Images
Figure CN119397274A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of this specification relate to the field of big data technology, and in particular to a power prediction method based on the CnnLstm algorithm. Background Art
[0002] At present, the power grid dispatching department requires power stations (including wind power and photovoltaic power) to report power forecasts, including short-term and ultra-short-term power forecasts, for system power generation planning or power system real-time dispatching. However, the current prediction accuracy is insufficient, resulting in a lack of accuracy in subsequent system power generation planning or power system real-time dispatching. Therefore, a better solution is urgently needed. Summary of the invention
[0003] In view of this, the embodiments of this specification provide a power prediction method based on the CnnLstm algorithm. One or more embodiments of this specification also involve a power prediction device based on the CnnLstm algorithm, a computing device, a computer-readable storage medium and a computer program to solve the technical defects existing in the prior art.
[0004] According to a first aspect of an embodiment of this specification, a power prediction method based on a CnnLstm algorithm is provided, comprising:
[0005] Obtaining raw data, constructing data based on the raw data, and determining training data; wherein the raw data includes meteorological data and power data;
[0006] Input the training data into the convolution layer in the CnnLstm algorithm model to determine the target features;
[0007] The target features are input into the Lstm layer and the fully connected layer to obtain the power prediction results.
[0008] In a possible implementation, data construction is performed based on the original data to determine the training data, including:
[0009] Deployment is performed based on meteorological data and power data to determine the deployment sequence;
[0010] The training data is determined based on the expanded sequence; wherein the training data is data of a target quantity dimension, and the target quantity is the sum of the number of meteorological data and the number of power data.
[0011] In a possible implementation, data construction is performed based on the original data to determine the training data, including:
[0012] Align meteorological data and power data to determine the target number of data sets;
[0013] The training data is determined based on a target number of data sets; wherein the training data is 2-dimensional data.
[0014] In a possible implementation, the method further includes:
[0015] Construct the time dimension based on sliding windows.
[0016] In a possible implementation, the meteorological data is predicted meteorological data after the current time node;
[0017] Meteorological data and power forecast results correspond one to one;
[0018] The power data is the historical power data before the current time node.
[0019] In a possible implementation, the convolution layer includes a first one-dimensional convolution layer and a second one-dimensional convolution layer;
[0020] Accordingly, the training data is input into the convolutional layer in the CnnLstm algorithm model to determine the target features, including:
[0021] Input the training data into the first one-dimensional convolutional layer to determine the first-dimensional feature data;
[0022] The first dimension feature data is input into the second one-dimensional convolutional layer to determine the target features.
[0023] In one possible implementation, the number of neurons in the fully connected layer is the same as the number of meteorological data.
[0024] According to a second aspect of an embodiment of this specification, a power prediction device based on a CnnLstm algorithm is provided, comprising:
[0025] A data acquisition module is configured to acquire raw data, perform data construction based on the raw data, and determine training data; wherein the raw data includes meteorological data and power data;
[0026] A feature determination module is configured to input the training data into the convolution layer in the CnnLstm algorithm model to determine the target features;
[0027] The power prediction module is configured to input the target features into the Lstm layer and the fully connected layer to obtain the power prediction results.
[0028] According to a third aspect of an embodiment of this specification, a computing device is provided, including:
[0029] Memory and processor;
[0030] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the power prediction method based on the CnnLstm algorithm are implemented.
[0031] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the power prediction method based on the CnnLstm algorithm are implemented.
[0032] According to a fifth aspect of the embodiments of this specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned power prediction method based on the CnnLstm algorithm.
[0033] The embodiments of this specification provide a power prediction method and device based on the CnnLstm algorithm, wherein the power prediction method based on the CnnLstm algorithm includes: obtaining original data, constructing data based on the original data, and determining training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result. By constructing data based on the original data to determine the training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result, the information of the training data can be enriched, thereby improving the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 is a flowchart of a power prediction method based on the CnnLstm algorithm provided by an embodiment of this specification;
[0035] Figure 2 It is a first data structure schematic diagram of a power prediction method based on CnnLstm algorithm provided by an embodiment of this specification;
[0036] Figure 3 is a second data structure schematic diagram of a power prediction method based on the CnnLstm algorithm provided by an embodiment of this specification;
[0037] Figure 4 It is a schematic diagram of an algorithm structure of a power prediction method based on the CnnLstm algorithm provided by an embodiment of this specification;
[0038] Figure 5 It is a structural schematic diagram of a power prediction device based on the CnnLstm algorithm provided by an embodiment of this specification;
[0039] Figure 6 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION
[0040] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.
[0041] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.
[0042] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".
[0043] First, the terms involved in one or more embodiments of this specification are explained.
[0044] CNN: Convolutional Neural Networks, convolutional neural network.
[0045] LSTM: Long Short Term Memory, is a special recurrent neural network.
[0046] In this specification, a power prediction method based on the CnnLstm algorithm is provided. This specification also relates to a power prediction device based on the CnnLstm algorithm, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.
[0047] See also Figure 1 , Figure 1 A flowchart of a power prediction method based on the CnnLstm algorithm provided according to an embodiment of the present specification is shown, which specifically includes the following steps.
[0048] Step 101: Acquire original data, construct data based on the original data, and determine training data; wherein the original data includes meteorological data and power data.
[0049] In one possible implementation, data construction is performed based on the original data to determine the training data, including: expanding the data based on the meteorological data and the power data to determine the expansion sequence; determining the training data based on the expansion sequence; wherein the training data is data of a target quantity dimension, and the target quantity is the sum of the number of meteorological data and the number of power data.
[0050] In practical applications, the power needs to be predicted at 16 or 24 o'clock in the future according to the different ultra-short-term requirements of each province. The predicted power at the eighth o'clock is assessed for the northwest region. The input of the ultra-short-term power prediction algorithm generally includes historical actual power and future meteorological source forecasts. Depending on the algorithm used, there are generally two ways to construct ultra-short-term training data. The following will explain in detail the construction methods of these two training data.
[0051] Specifically, the first type of ultra-short-term training data construction is to directly expand and splice the historical actual power and the future 16-hour weather source forecast together as feature input, and directly predict the power of the next 16 hours. The schematic diagram of the training data construction process of this method is as follows: Figure 2 As shown. Among them, P t Refers to the power at time t, WS t Refers to the wind speed predicted by the meteorological source at time t, such as the predicted P t To P t+15 The power of the next 16 points, the ultra-short-term algorithm input includes the actual power P of the historical h points t-h To P t-1 And the meteorological source predicted wind speed WS from t to t+15 t To WS t+15 The training data input features constructed in this way are h+16 dimensions, which are suitable for the input of traditional machine learning algorithms, such as support vector regression, random forest, gradient boosting tree, and multilayer perceptron algorithms.
[0052] In a possible implementation manner, the method further includes: constructing a time dimension based on a sliding window.
[0053] Furthermore, for the Lstm algorithm, it is also necessary to construct the time dimension by sliding the time window.
[0054] In one possible implementation, data construction is performed based on the original data to determine training data, including: aligning meteorological data and power data to determine a target number of data groups; determining training data based on the target number of data groups; wherein the training data is 2-dimensional data.
[0055] Specifically, the second training data construction method is to align the historical actual power and the wind speed predicted by the meteorological source at the next 16 points in the future and splice them together as input data, and then predict the power at the next 16 points. The training data construction diagram of this method is as follows: Figure 3 shown.
[0056] The above training data construction method is suitable for the input of the CnnLstm algorithm. The time step of the input data is the same as the length of the future power point to be predicted, and the feature dimension is 2-dimensional, which is the historical actual power and the future meteorological source forecast. This method retains the time series characteristics of the original data, contains richer information, and the algorithm prediction effect will be better.
[0057] In a possible implementation, the meteorological data is the predicted meteorological data after the current time node; the meteorological data corresponds one-to-one to the power prediction result; and the power data is the historical power data before the current time node.
[0058] In practical applications, as described above, the meteorological data is the future meteorological source forecast, and the power data is the historical actual power. Moreover, the number of data of the future meteorological source forecast is the same as the number of predicted future power.
[0059] Step 102: Input the training data into the convolution layer in the CnnLstm algorithm model to determine the target features.
[0060] In a possible implementation, the convolution layer includes a first one-dimensional convolution layer and a second one-dimensional convolution layer; accordingly, the training data is input into the convolution layer in the CnnLstm algorithm model to determine the target features, including: inputting the training data into the first one-dimensional convolution layer to determine the first dimension feature data; inputting the first dimension feature data into the second one-dimensional convolution layer to determine the target features.
[0061] In practical applications, the CnnLstm algorithm for ultra-short-term power prediction proposed in the embodiments of this specification is mainly composed of two one-dimensional convolution layers and an Lstm layer. One-dimensional convolution algorithms are commonly used in the field of signal processing and are also suitable for processing time series data. One-dimensional convolution is different from two-dimensional convolution. Two-dimensional convolution is to perform a sliding window operation on a feature map in both width and height directions, and finally multiply and sum the corresponding positions; while one-dimensional convolution is just to slide the window in the width or height direction and multiply and sum.
[0062] Step 103: Input the target features into the Lstm layer and the fully connected layer to obtain the power prediction result.
[0063] In one possible implementation, the number of neurons in the fully connected layer is the same as the number of meteorological data.
[0064] Specifically, the algorithm parameter settings of the CnnLstm algorithm are as follows:
[0065] Table 1 CnnLstm algorithm parameters
[0066] Parameter name Parameter Value The number of convolution kernels in the first layer 16 The number of convolution kernels in the second layer 32 Convolution kernel size 3 Stride size 1 Filling method valid Number of neurons in LSTM unit 20 Normalization method MaxMin Optimizer Adam Loss Function mse Iterations 10 Training batches 32
[0067] For further information, see Figure 4 The input layer data dimension of the algorithm is (n, 16, 2), where n is the number of samples, 16 is the time step, and 2 is the feature dimension. After two layers of one-dimensional convolution, the input data has a data dimension of (n, 12, 32). Then, after passing through the Lstm layer and taking only the output of the last time step, the data dimension is (n, 20). Since the predicted power of the next 16 points is to be output in the end, a fully connected layer with 16 neurons is added to obtain the final output.
[0068] The embodiments of this specification provide a power prediction method and device based on the CnnLstm algorithm, wherein the power prediction method based on the CnnLstm algorithm includes: obtaining original data, constructing data based on the original data, and determining training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result. By constructing data based on the original data to determine the training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result, the information of the training data can be enriched, thereby improving the accuracy of the prediction.
[0069] Corresponding to the above method embodiment, this specification also provides a power prediction device embodiment based on the CnnLstm algorithm. Figure 5 FIG. 1 shows a schematic diagram of a power prediction device based on a CnnLstm algorithm provided by an embodiment of the present specification. Figure 5 As shown, the device comprises:
[0070] The data acquisition module 501 is configured to acquire raw data, perform data construction based on the raw data, and determine training data; wherein the raw data includes meteorological data and power data;
[0071] The feature determination module 502 is configured to input the training data into the convolution layer in the CnnLstm algorithm model to determine the target feature;
[0072] The power prediction module 503 is configured to input the target feature into the Lstm layer and the fully connected layer to obtain the power prediction result.
[0073] In a possible implementation, the data acquisition module 501 is further configured to:
[0074] Deployment is performed based on meteorological data and power data to determine the deployment sequence;
[0075] The training data is determined based on the expanded sequence; wherein the training data is data of a target quantity dimension, and the target quantity is the sum of the number of meteorological data and the number of power data.
[0076] In a possible implementation, the data acquisition module 501 is further configured to:
[0077] Align meteorological data and power data to determine the target number of data sets;
[0078] The training data is determined based on a target number of data sets; wherein the training data is 2-dimensional data.
[0079] In a possible implementation, the data acquisition module 501 is further configured to:
[0080] Construct the time dimension based on sliding windows.
[0081] In a possible implementation, the data acquisition module 501 is further configured to:
[0082] The meteorological data is the predicted meteorological data after the current time node;
[0083] Meteorological data and power forecast results correspond one to one;
[0084] The power data is the historical power data before the current time node.
[0085] In a possible implementation, the feature determination module 502 is further configured to:
[0086] The convolution layer includes a first one-dimensional convolution layer and a second one-dimensional convolution layer;
[0087] Accordingly, the training data is input into the convolutional layer in the CnnLstm algorithm model to determine the target features, including:
[0088] Input the training data into the first one-dimensional convolutional layer to determine the first-dimensional feature data;
[0089] The first dimension feature data is input into the second one-dimensional convolutional layer to determine the target features.
[0090] In a possible implementation, the power prediction module 503 is further configured to:
[0091] The number of neurons in the fully connected layer is the same as the number of meteorological data.
[0092] The embodiments of this specification provide a power prediction method and device based on the CnnLstm algorithm, wherein the power prediction device based on the CnnLstm algorithm includes: obtaining original data, constructing data based on the original data, and determining training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result. By constructing data based on the original data to determine the training data; wherein the original data includes meteorological data and power data; inputting the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; inputting the target features into the Lstm layer and the fully connected layer to obtain the power prediction result, the information of the training data can be enriched, thereby improving the accuracy of the prediction.
[0093] The above is a schematic scheme of a power prediction device based on the CnnLstm algorithm of this embodiment. It should be noted that the technical scheme of the power prediction device based on the CnnLstm algorithm and the technical scheme of the power prediction method based on the CnnLstm algorithm belong to the same concept. For the details not described in detail in the technical scheme of the power prediction device based on the CnnLstm algorithm, please refer to the description of the technical scheme of the power prediction method based on the CnnLstm algorithm.
[0094] Figure 6 The block diagram of a computing device 600 according to an embodiment of the present specification is shown. The components of the computing device 600 include but are not limited to a memory 610 and a processor 620. The processor 620 is connected to the memory 610 via a bus 630, and the database 650 is used to store data.
[0095] The computing device 600 also includes an access device 640 that enables the computing device 600 to communicate via one or more networks 660. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 640 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).
[0096] In one embodiment of the present specification, the above components of the computing device 600 and Figure 6 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 6 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.
[0097] The computing device 600 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 600 may also be a mobile or stationary server.
[0098] Among them, the processor 620 is used to execute the following computer executable instructions, which, when executed by the processor, implement the steps of the power prediction method based on the CnnLstm algorithm. The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the power prediction method based on the CnnLstm algorithm belong to the same concept. The details of the technical scheme of the computing device that are not described in detail can be found in the description of the technical scheme of the power prediction method based on the CnnLstm algorithm.
[0099] An embodiment of the present specification also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the power prediction method based on the CnnLstm algorithm.
[0100] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the power prediction method based on the CnnLstm algorithm above belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be found in the description of the technical scheme of the power prediction method based on the CnnLstm algorithm above.
[0101] An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the power prediction method based on the CnnLstm algorithm.
[0102] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the power prediction method based on the CnnLstm algorithm are of the same concept, and the details not described in detail in the technical scheme of the computer program can be found in the description of the technical scheme of the power prediction method based on the CnnLstm algorithm.
[0103] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0104] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.
[0105] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.
[0106] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0107] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.
Claims
1. A power prediction method based on CnnLstm algorithm, characterized in that: include: Acquire original data, perform data construction based on the original data, and determine training data; wherein the original data includes meteorological data and power data; Input the training data into the convolution layer in the CnnLstm algorithm model to determine the target features; The target features are input into the Lstm layer and the fully connected layer to obtain the power prediction result.
2. The method according to claim 1, characterized in that The step of constructing data based on the original data and determining training data includes: Perform deployment based on the meteorological data and the power data, and determine a deployment sequence; The training data is determined based on the expansion sequence; wherein the training data is data of a target quantity dimension, and the target quantity is the sum of the number of the meteorological data and the number of the power data.
3. The method according to claim 1, characterized in that: The step of constructing data based on the original data and determining training data includes: Aligning the meteorological data and the power data to determine a target number of data groups; Training data is determined based on the target number of data sets; wherein the training data is 2-dimensional data.
4. The method according to claim 2, characterized in that: Also includes: Construct the time dimension based on sliding windows.
5. The method according to claim 1, characterized in that The meteorological data is the predicted meteorological data after the current time node; The meteorological data corresponds to the power prediction result one by one; The power data is historical power data before the current time node.
6. The method according to claim 1, characterized in that The convolution layer includes a first one-dimensional convolution layer and a second one-dimensional convolution layer; Accordingly, the training data is input into the convolutional layer in the CnnLstm algorithm model to determine the target features, including: Inputting the training data into the first one-dimensional convolutional layer to determine first-dimensional feature data; The first dimensional feature data is input into the second one-dimensional convolutional layer to determine the target feature.
7. The method according to claim 1, characterized in that The number of neurons in the fully connected layer is the same as the number of meteorological data.
8. A power prediction device based on CnnLstm algorithm, characterized in that: include: A data acquisition module is configured to acquire raw data, perform data construction based on the raw data, and determine training data; wherein the raw data includes meteorological data and power data; A feature determination module is configured to input the training data into a convolutional layer in a CnnLstm algorithm model to determine target features; The power prediction module is configured to input the target feature into the Lstm layer and the fully connected layer to obtain a power prediction result.
9. A computing device, characterized in that include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the power prediction method based on the CnnLstm algorithm described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the power prediction method based on the CnnLstm algorithm as described in any one of claims 1 to 7.