Province-wide Electric Load Forecasting Method Combining Meteorological and Economic Data

Abstract

The present invention belongs to the technical field of power prediction, and provides a method for predicting the provincial power load by combining meteorological and economic data. The method includes: collecting historical meteorological data and historical economic data related to the target province, and determining the first meteorological prediction data and the first meteorological response big data for the future prediction period, as well as the second meteorological prediction data and the second meteorological response big data of the associated provinces; using a first deep prediction model to predict the historical meteorological data and historical economic data to obtain a first provincial power load prediction value corresponding to the future prediction period; using a second deep prediction model to respectively perform deep prediction on the first meteorological prediction data and the first meteorological response big data, and the second meteorological prediction data and the second meteorological response big data to obtain a first influence coefficient and a second influence coefficient, and correcting the first provincial power load prediction value to obtain a second provincial power load prediction value. The present invention can significantly improve the accuracy of the result of predicting the provincial power load.

Description

Technical Field

[0001] The present invention relates to the technical field of power prediction, and more particularly, to a method for predicting the provincial power load by combining meteorological and economic data. Background Art

[0002] In today's power system management, accurate power load prediction is crucial for ensuring the stability and reliability of power supply. The change of power load is affected by various factors, among which meteorological conditions and economic activities are two key factors. Meteorological conditions, such as temperature, humidity, wind speed, etc., will directly affect the power demand of residential and industrial users. For example, in the hot summer, the large use of air conditioners will cause a sharp increase in power load; while in the cold winter, the operation of heating equipment will also have a significant impact on power load. The level and structure of economic activities will also have an important impact on power load. Economic growth is usually accompanied by an increase in power demand, and the power consumption characteristics of different industries are also different.

[0003] Currently, when predicting the power load on a provincial scale, the existing methods mainly establish prediction models based on historical meteorological data and economic data. These methods usually collect meteorological data (such as temperature, humidity, wind speed, etc.) and economic data (such as GDP, industrial output value, residential consumption, etc.) over a past period of time, and find the relationship between these data and power load through statistical analysis or machine learning algorithms, so as to predict the future power load. However, these existing methods have some obvious limitations. For example, they lack the consideration of the ability to respond to abnormal meteorological conditions in the future time period targeted by the power load prediction in this province, and the consideration of the impact of economic activities in neighboring provinces, etc.

[0004] Therefore, in order to improve the accuracy of provincial power load prediction and better meet the operation and management needs of the power system, there is an urgent need for a new method for predicting power load by combining meteorological and economic data. Summary of the Invention

[0005] In order to solve the technical problems existing in the above background art, the present invention provides a method for predicting the provincial power load by combining meteorological and economic data, an electronic device, a computer storage medium, and a computer program product.

[0006] The present invention provides a method for predicting the provincial power load by combining meteorological and economic data, including the following steps:

[0007] Collect historical meteorological data and historical economic data related to the target province, and determine the first meteorological prediction data and the first meteorological response big data for the future prediction time period corresponding to the target province, and determine the second meteorological prediction data and the second meteorological response big data for the associated province corresponding to the target province;

[0008] Use the first depth prediction model to predict the historical meteorological data and the historical economic data, and obtain the first provincial power load prediction value corresponding to the future prediction period;

[0009] Use the second depth prediction model to perform in-depth predictions on the first meteorological prediction data and the first meteorological response big data, and the second meteorological prediction data and the second meteorological response big data respectively, and obtain the first influence coefficient and the second influence coefficient;

[0010] Based on the first influence coefficient and the second influence coefficient, correct the first provincial power load prediction value to obtain the second provincial power load prediction value.

[0011] In some embodiments, using the second depth prediction model to perform in-depth predictions on the first meteorological prediction data and the first meteorological response big data, and the second meteorological prediction data and the second meteorological response big data respectively, and obtaining the first influence coefficient and the second influence coefficient includes:

[0012] Use the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological prediction data according to the first clustering index to obtain the first meteorological key semantic information; use the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological response big data according to the first clustering index to obtain the first meteorological response key semantic information;

[0013] Use the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological prediction data according to the second clustering index to obtain the second meteorological key semantic information; use the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological response big data according to the second clustering index to obtain the second meteorological response key semantic information;

[0014] Based on the first meteorological key semantic information and the first meteorological response key semantic information, and the second meteorological key semantic information and the second meteorological response key semantic information, respectively ask the third semantic analysis model about the response ability to obtain the first response ability and the second response ability; wherein, the third semantic analysis model is constructed based on a general large model;

[0015] Compare the first response ability and the second response ability with the response ability - influence coefficient comparison table to obtain the first influence coefficient and the second influence coefficient respectively.

[0016] In some embodiments, the clustering distances of the first clustering index and the second clustering index are different, and the clustering distance of the first clustering index is greater than that of the second clustering index; wherein, the clustering distance represents the lowest value of similar data belonging to data points or two clustering clusters that can be clustered.

[0017] In some embodiments, the first clustering index is determined by the following method:

[0018] Retrieve the abnormal meteorological loss record data of a target province for several years, and vectorize each of the abnormal meteorological loss record data;

[0019] Integrate the abnormal meteorological loss record data of non - recent years after vectorization into first loss feature data, calculate the first Euclidean distance between the second loss feature data of the most recent year after vectorization and the first loss feature data, and determine the clustering distance of the first clustering index according to the first Euclidean distance and the standard clustering distance.

[0020] In some embodiments, the second clustering index is determined by the following method:

[0021] Retrieve the abnormal meteorological loss record data of an associated province for several years, and vectorize each of the abnormal meteorological loss record data;

[0022] Integrate the abnormal meteorological loss record data of non - recent years after vectorization into third loss feature data, calculate the second Euclidean distance between the fourth loss feature data of the most recent year after vectorization and the third loss feature data, and determine the clustering distance of the second clustering index according to the second Euclidean distance and the standard clustering distance.

[0023] In some embodiments, if the clustering distance of the first clustering index is less than that of the second clustering index, the method further includes:

[0024] Use a preset adjustment coefficient to increase the clustering distance of the first clustering index at least once so that it is greater than that of the second clustering index.

[0025] In some embodiments, the step of correcting the first provincial - wide power load prediction value based on the first influence coefficient and the second influence coefficient to obtain a second provincial - wide power load prediction value includes:

[0026] Perform a multiplication operation on the first influence coefficient and the second influence coefficient with the first provincial - wide power load prediction value at the same time to obtain the corrected second provincial - wide power load prediction value.

[0027] The present invention also provides an electronic device, including: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory and executes the method described in any one of the preceding items.

[0028] The present invention also provides a computer storage medium, on which a computer program is stored, and when the computer program is run by a processor, it executes the method described in any one of the preceding items.

[0029] The present invention also provides a computer program product, which includes a computer program stored in a computer storage medium, and when the computer program is executed by a processor of an electronic device, it implements the method described in any one of the preceding items.

[0030] The beneficial effects of the present invention are as follows:

[0031] The power load forecasting method of the present invention takes into account the meteorological response capabilities of the target province during the future forecasting period, and also takes into account the meteorological response capabilities of related provinces. By accurately evaluating the impact of various factors on the load, it can effectively improve the accuracy of power load forecasting, provide a reliable basis for power system planning and dispatching, help allocate resources reasonably, and reduce supply risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0033] Figure 1 is a schematic flowchart of a method for forecasting the provincial power load by combining meteorological and economic data disclosed in an embodiment of the present invention;

[0034] Figure 2 is a schematic structural diagram of a second deep forecasting model disclosed in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] The following describes exemplary embodiments of the present disclosure, including various details of the embodiments of the present disclosure to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following.

[0036] As Figure 1 shown, an embodiment of the present invention discloses a method for forecasting the provincial power load by combining meteorological and economic data, including the following steps:

[0037] S10. Collect historical meteorological data and historical economic data related to the target province, determine the first meteorological prediction data and the first meteorological response big data for the future prediction period corresponding to the target province, and determine the second meteorological prediction data and the second meteorological response big data for the associated province corresponding to the target province.

[0038] Among them, the historical meteorological data includes information such as temperature, humidity, wind speed, etc., and these data will affect the electricity consumption behaviors of residents and industrial users. For example, in a certain target province, during past summers with high temperatures, the electricity consumption of residents' air conditioners increased significantly. By collecting the temperature data and the corresponding electricity load data in summers over the past years, it can provide a basis for subsequent analysis. Historical economic data such as GDP, industrial output value, and residents' consumption reflect the electricity demand of the economic activities in this province. For example, when the industrial output value increases, the production electricity consumption of factories will increase accordingly.

[0039] Next, determine the first meteorological prediction data and the first meteorological response big data (both including multiple data) for the future prediction period (such as the next month, quarter, half year, etc.) corresponding to the target province. The first meteorological prediction data is an estimate of the meteorological conditions in the target province during the future prediction period. For example, predict the average temperature, number of rainy days, probability and degree of flood disasters, etc. in the target province for the next month. The first meteorological response big data involves data related to the province's ability to respond to meteorological changes. For example, the number of power generation equipment reserved in the province to cope with extreme weather (such as floods caused by heavy rains), the construction and maintenance of emergency transmission lines, the repair of river reservoirs, the construction of flood control facilities (completed, planned), the procurement of flood control equipment, etc. For example, if the target province may face high temperature weather during the future prediction period, the first meteorological prediction data will reflect the number of high temperature days and the temperature range, while the first meteorological response big data may show that the province has prepared sufficient raw materials for cooling equipment production to cope with high temperatures. These data reflect the ability to respond to abnormal meteorology, which is related to the impact of abnormal meteorology on production enterprises in the region, and the impact size is related to the electricity load. For example, when the target province's ability to respond to abnormal meteorology is poor, more enterprises in the region will suspend production due to abnormal meteorology, and the electricity load of this target province will be lower than expected.

[0040] Meanwhile, the second meteorological prediction data and the second meteorological response big data of the associated province corresponding to the target province are also determined. The associated province refers to a province that has a relatively high economic association or dependence with the target province, especially neighboring provinces. The meteorological and response data of these provinces also have an impact on the power load of the target province. For example, if it is predicted based on the second meteorological prediction data of an associated province that there will be severe rainstorm weather in the future, and the second meteorological response big data shows that the province's material reserves for dealing with rainstorms are insufficient and may need to obtain materials from the target province, for the production of these materials, it will cause factories in the target province to increase power consumption, thus affecting the power load of the target province.

[0041] S20. Use the first depth prediction model to predict the historical meteorological data and the historical economic data to obtain the first provincial-wide power load prediction value corresponding to the future prediction period.

[0042] Among them, the first depth prediction model is a model based on deep learning algorithms (such as CNN, Transformer, etc.). It can use the historical meteorological and economic data of the target province to predict the power load of the target province in the future prediction period. For example, the first depth prediction model learns that in the past few years, when the GDP increased by 10% and the average summer temperature increased by 2°C, the power load increased by 15%. Based on the expected GDP growth and temperature changes in the future prediction period, the first depth prediction model predicts the first provincial-wide power load prediction value.

[0043] However, the above first provincial-wide power load prediction value does not take into account the meteorological response ability and the influence of the associated province, so it is a preliminary prediction result.

[0044] S30. Use the second depth prediction model to respectively conduct in-depth predictions on the first meteorological prediction data and the first meteorological response big data, and the second meteorological prediction data and the second meteorological response big data to respectively obtain the first influence coefficient and the second influence coefficient.

[0045] Among them, the second depth prediction model (such as CNN, Transformer, etc.) is used to analyze the matching degree between the meteorological data and the response ability of the province. The second depth prediction model first analyzes the first meteorological prediction data and the first meteorological response big data of the target province to obtain the first influence coefficient. For example, when the first influence coefficient is 0.8, it indicates that the target province's preparation for abnormal meteorology is insufficient, resulting in a greater impact on it by flood disasters during the future prediction period, leading to the interruption of enterprise production, and thus the actual power load of the target province is lower than expected.

[0046] Similarly, the second depth prediction model also analyzes the second meteorological prediction data and the second meteorological response big data of the associated provinces to obtain a second influence coefficient. For example, if a severe meteorological disaster is predicted in an associated province and its response capacity is insufficient, and a large amount of materials need to be purchased from the target province, the second influence coefficient is, for example, 1.2. Obviously, the second influence coefficient at this time will cause the actual power load of the target province to be higher than expected.

[0047] S40. Based on the first influence coefficient and the second influence coefficient, correct the first provincial power load prediction value to obtain a second provincial power load prediction value.

[0048] Since the first provincial power load prediction value obtained in the foregoing steps does not take into account the meteorological response capacity of the target province itself and the influence of the associated provinces, it is corrected by the first influence coefficient and the second influence coefficient.

[0049] For example, if the first provincial power load prediction value is 1 million kWh, the first influence coefficient indicates that the power load is reduced by 10% due to the insufficient meteorological response capacity of the target province itself, and the second influence coefficient indicates that the demand power load of the associated province needs to be increased by another 5%. Then the finally corrected second provincial power load prediction value is 100×(1 - 10%)×(1 + 5%) = 945,000 kWh. The prediction value obtained in this way more accurately takes into account the influence of various factors on the power load, making up for the limitation of the existing method that lacks consideration of the meteorological response capacity and the influence of the economic activities of the associated provinces.

[0050] The power load prediction method of the present invention takes into account the meteorological response capacity of the target province in the future prediction period, and also takes into account the meteorological response capacity of the associated provinces at the same time. By accurately evaluating the influence of various factors on the load, it can effectively improve the accuracy of power load prediction, provide a reliable basis for power system planning and dispatching, help allocate resources reasonably, and reduce supply risks.

[0051] In some embodiments, using the second depth prediction model to perform deep prediction on the first meteorological prediction data and the first meteorological response big data, and the second meteorological prediction data and the second meteorological response big data respectively to obtain a first influence coefficient and a second influence coefficient, including:

[0052] Using the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological prediction data according to the first clustering index to obtain first meteorological key semantic information; using the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological response big data according to the first clustering index to obtain first meteorological response key semantic information;

[0053] Use the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological prediction data according to the second clustering index to obtain the second key meteorological semantic information; use the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological response big data according to the second clustering index to obtain the second key meteorological response semantic information;

[0054] Based on the first key meteorological semantic information and the first key meteorological response semantic information, and the second key meteorological semantic information and the second key meteorological response semantic information, respectively, ask the third semantic analysis model about the response ability to obtain the first response ability and the second response ability; wherein, the third semantic analysis model is constructed based on a general large model;

[0055] Compare the first response ability and the second response ability with the response ability - influence coefficient comparison table to obtain the first influence coefficient and the second influence coefficient respectively.

[0056] In this embodiment, as Figure 2 shown, the second depth prediction model of the present invention includes three semantic analysis models, namely the first semantic analysis model, the second semantic analysis model, and the third semantic analysis model. Among them, the first semantic analysis model and the second semantic analysis model are constructed by hierarchical clustering algorithms, convolutional neural networks (i.e., CNN), etc.

[0057] Among them, the hierarchical clustering algorithm can perform hierarchical clustering analysis on text data. In meteorological semantic analysis, agglomerative hierarchical clustering can start from individual words or phrases and gradually merge according to semantic similarity to form different levels of semantic categories; divisive hierarchical clustering is the opposite, starting from the entire text set and gradually splitting into smaller semantic subsets. Through the hierarchical clustering algorithm, the semantic information in meteorological prediction data and meteorological response big data can be classified according to the degree of similarity, so as to screen out the key semantic information. For example, abnormal meteorological data of the same type and degree can be grouped into one category, and semantic information related to meteorological disaster response can be grouped into one category. Then, data with a meteorological abnormality degree higher than the preset abnormality degree and a meteorological response ability higher than the preset response ability are screened out, so as to obtain the above-mentioned first / second key meteorological semantic information and first / second key meteorological response semantic information.

[0058] The third semantic analysis model is built based on general large models, such as DeepSeek, GPT, Wenxin Yiyan, etc. The two types of key semantic information obtained previously can be respectively input into the third semantic analysis model and questions can be asked, such as "The additional semantic information is the meteorological prediction data and meteorological response ability data of a certain province in a certain future period. Please evaluate the matching degree between the response ability of the target province and the above meteorological prediction data, that is, the response ability level." After in-depth thinking, the third semantic analysis model outputs the above first response ability and second response ability. A response ability - influence coefficient comparison table is pre-constructed, and the above first influence coefficient and second influence coefficient can be quickly determined by querying this comparison table, providing a quantitative basis for correcting the power load prediction value.

[0059] In some embodiments, the clustering distances of the first clustering index and the second clustering index are different, and the clustering distance of the first clustering index is greater than that of the second clustering index; wherein, the clustering distance represents the lowest value of similar data that can be clustered between two data points or two clustering clusters.

[0060] In this embodiment, when the first semantic analysis model performs semantic clustering and screening on meteorological data, the clustering distance is an important index, which plays a key role in judging the similarity between data points (such as meteorological prediction data like heavy rain, and meteorological response data like flood control measure preparation) and determining the clustering result. It should be noted that the clustering in the present invention is for the same type of data, such as clustering meteorological data of different temperatures, clustering meteorological data of different rainfall amounts, or clustering the maintenance of emergency power transmission lines and the repair data of river reservoirs (maintenance category).

[0061] The clustering distance represents the lowest value of similar data that can be clustered between two data points or two clustering clusters. For example, if the approximation degree between the two meteorological data of "high temperature weather of 30 degrees Celsius" and "hot climate of 35 degrees Celsius" is higher than the above clustering distance, then the two are clustered into one category, such as both being clustered as "high temperature"; while the approximation degree between "high temperature weather" and "heavy rainfall weather" is lower than the above clustering distance, then the two are not clustered into one category. By adopting different clustering distances, different degrees of clustering of meteorological prediction data and meteorological response big data can be obtained, so as to obtain different meteorological key semantic information and meteorological response key semantic information.

[0062] Since the solution of the present invention is to predict the power load of the target province, the meteorological response ability of the province itself should be the main factor, supplemented by the meteorological response ability of the associated provinces. Therefore, the present invention sets the clustering distance of the first clustering index to be greater than the clustering distance of the second clustering index. By setting it in this way, more detailed features are retained in the first meteorological key semantic information and the first meteorological response key semantic information compared to the second meteorological key semantic information and the second meteorological response key semantic information, which is beneficial for the third semantic analysis model to analyze and obtain a more accurate response ability. In addition, the influence degree of the associated provinces is usually not too large. Therefore, the accuracy requirement for the second influence coefficient is set to be relatively low. Correspondingly, the clustering distance of the second clustering index is set to be smaller, that is, fewer detailed features are retained in the second meteorological key semantic information and the second meteorological response key semantic information after clustering, which can significantly improve the analysis rate of the third semantic analysis model for the second response ability.

[0063] In some embodiments, the first clustering index is determined by the following method:

[0064] Retrieve the abnormal meteorological loss record data of the target province in several years, and vectorize each of the abnormal meteorological loss record data;

[0065] Integrate the abnormal meteorological loss record data of non-nearest years after vectorization into the first loss feature data, calculate the first Euclidean distance between the second loss feature data of the nearest year after vectorization and the first loss feature data, and determine the clustering distance of the first clustering index according to the first Euclidean distance and the standard clustering distance.

[0066] In some embodiments, the second clustering index is determined by the following method:

[0067] Retrieve the abnormal meteorological loss record data of the associated provinces in several years, and vectorize each of the abnormal meteorological loss record data;

[0068] Integrate the abnormal meteorological loss record data of non-nearest years after vectorization into the third loss feature data, calculate the second Euclidean distance between the fourth loss feature data of the nearest year after vectorization and the third loss feature data, and determine the clustering distance of the second clustering index according to the second Euclidean distance and the standard clustering distance.

[0069] In this embodiment, first, an example is given for the determination method of the clustering distance of the first clustering index:

[0070] Retrieve the abnormal meteorological loss record data of the target province in the past several years. These data record various loss situations in the target province in different past years due to abnormal meteorology (such as extreme weather like heavy rain, typhoon, high temperature, etc.), which may include economic losses (such as losses of affected crops, repair costs of damaged infrastructure, etc.), casualty situations and other relevant information.

[0071] Then, vectorize each abnormal meteorological loss record data, that is, convert these non-numerical data (such as the loss situations described in text) into the form of numerical vectors. For example, use techniques such as Word Embedding to convert the text description into a vector, and each dimension in the vector represents a certain feature of the data.

[0072] Integrate the abnormal meteorological loss record data of non-nearest years (for example, 2024 is the nearest year, and those before 2024 are non-nearest years) after vectorization into the first loss feature data. Here, the integration includes operations such as statistical calculation and aggregation of the data of each non-nearest year. For example, calculate the average value, total value, etc. of the same type of losses in each year to form a comprehensive feature vector to represent the overall abnormal meteorological loss characteristics of non-nearest years. For the abnormal meteorological loss record data of the nearest year after vectorization, form the second loss feature data separately, which represents the feature vector of the abnormal meteorological loss situation in the nearest year.

[0073] Calculate the Euclidean distance between the second loss feature data and the first loss feature data. This Euclidean distance characterizes the similarity degree between the two vectors. Obviously, when the Euclidean distance between the two is larger, it means the similarity degree between the two is lower. Determine the clustering distance of the first clustering index according to the calculated Euclidean distance and the pre-set standard clustering distance (a reference distance value set according to experience, historical data or specific business requirements). Specifically, determine the corresponding coefficient value according to the Euclidean distance, and then multiply the coefficient value by the standard clustering distance. For example, the coefficient value corresponding to a certain Euclidean distance is 1.2, and the standard clustering distance is 65% (similarity threshold, that is, the above-mentioned lowest value, namely the threshold), and the calculated clustering distance of the first clustering index is 1.2 * 65% = 78%. For example, when the similarity degree between two meteorological prediction data reaches 78%, cluster them into one category, for example, both are classified as "high temperature".

[0074] It should be noted that there should be a positive correlation between the Euclidean distance and the above-mentioned clustering distance. That is, the larger the Euclidean distance, the lower the similarity between the abnormal meteorological losses in the most recent year and those in earlier years, indicating greater abnormality. At this time, a larger coefficient value (such as 1.2) is set, resulting in a larger clustering distance, so that more detailed features are retained in the meteorological key semantic information and the key semantic information for meteorological response, thereby improving the accuracy of the influence coefficient for correcting the predicted value of the provincial power load; conversely, a smaller coefficient value (such as 1.1) is set, and correspondingly, the clustering distance is smaller, so that fewer detailed features are retained in the meteorological key semantic information and the key semantic information for meteorological response, that is, not so many detailed features are required.

[0075] The method for determining the clustering distance of the second clustering index is the same and will not be elaborated here.

[0076] In some embodiments, if the clustering distance of the first clustering index is less than the clustering distance of the second clustering index, the method further includes:

[0077] Using a preset adjustment coefficient to increase the clustering distance of the first clustering index at least once so that it is greater than the clustering distance of the second clustering index.

[0078] In this embodiment, since the clustering distances of the first clustering index and the second clustering index are calculated independently based on the abnormal meteorological loss record data of the target province and the associated province respectively, the clustering distance of the first clustering index may be less than the clustering distance of the second clustering index. At this time, it is necessary to use a preset adjustment coefficient to gradually increase the clustering distance of the first clustering index until it is greater than the clustering distance of the second clustering index. The preset adjustment coefficient is, for example, 1.1, 1.2, and the present invention does not make specific limitations on this.

[0079] In some embodiments, the correcting the first provincial power load prediction value based on the first influence coefficient and the second influence coefficient to obtain a second provincial power load prediction value includes:

[0080] Performing a multiplication operation on the first influence coefficient and the second influence coefficient simultaneously with the first provincial power load prediction value to obtain the corrected second provincial power load prediction value.

[0081] An embodiment of the present invention also discloses an electronic device, including: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory and executes the method as described in any one of the preceding items.

[0082] An embodiment of the present invention also discloses a computer storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the method described in any of the preceding items.

[0083] An embodiment of the present invention also discloses a computer program product, which includes a computer program stored in a computer storage medium. When the computer program is executed by a processor of an electronic device, it implements the method described in any of the preceding items.

[0084] The various embodiments of the systems and technologies described above in this document 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 a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor. The programmable processor can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0085] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0086] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0087] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions of this disclosure can be achieved, and this is not limited herein.

[0088] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A method for predicting the province's power load by combining meteorological and economic data, characterized in that: The steps include: Collect historical meteorological data and historical economic data related to the target province, and determine first meteorological forecast data and first meteorological response big data for a future forecast period corresponding to the target province, and determine second meteorological forecast data and second meteorological response big data for associated provinces corresponding to the target province; Use the first deep prediction model to predict the historical meteorological data and the historical economic data to obtain a first province-wide power load prediction value corresponding to the future prediction period; Use a second deep prediction model to perform deep prediction on the first meteorological prediction data and the first meteorological response big data, the second meteorological prediction data and the second meteorological response big data, respectively, to obtain a first influence coefficient and a second influence coefficient, respectively; Correcting the first predicted value of the power load for the entire province based on the first influence coefficient and the second influence coefficient to obtain a second predicted value of the power load for the entire province; Use the second deep prediction model to perform deep prediction on the first meteorological prediction data and the first meteorological response big data, the second meteorological prediction data and the second meteorological response big data, respectively, to obtain a first influence coefficient and a second influence coefficient, respectively, including: Using the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological forecast data according to the first clustering index to obtain the first meteorological key semantic information; using the first semantic analysis model to perform semantic clustering and screening processing on the first meteorological response big data according to the first clustering index to obtain the first meteorological response key semantic information; Using the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological forecast data according to the second clustering index to obtain second meteorological key semantic information; using the second semantic analysis model to perform semantic clustering and screening processing on the second meteorological response big data according to the second clustering index to obtain second meteorological response key semantic information; Based on the first meteorological key semantic information and the first meteorological response key semantic information, the second meteorological key semantic information and the second meteorological response key semantic information, respectively, a response capability question is asked to the third semantic analysis model to obtain a first response capability and a second response capability; wherein the third semantic analysis model is constructed based on a general large model; The first coping capability, the second coping capability and a coping capability-influence coefficient comparison table are compared to obtain the first influence coefficient and the second influence coefficient respectively.

2. A provincial power load forecasting method combining meteorological and economic data according to claim 1, characterized in that: The clustering distances of the first clustering indicator and the second clustering indicator are different, and the clustering distance of the first clustering indicator is greater than the clustering distance of the second clustering indicator; wherein the clustering distance is the minimum value indicating that two data points or two clusters are similar data that can be clustered.

3. A method for predicting provincial power load by combining meteorological and economic data according to claim 2, characterized in that: The first clustering index is determined by: Retrieving abnormal weather loss record data of the target province for several years, and vectorizing each abnormal weather loss record data; The abnormal meteorological loss record data of non-recent years after vectorization processing are integrated into the first loss characteristic data, and the first Euclidean distance between the second loss characteristic data of the most recent year after vectorization processing and the first loss characteristic data is calculated. The clustering distance of the first clustering indicator is determined based on the first Euclidean distance and the standard clustering distance.

4. A method for predicting provincial power load by combining meteorological and economic data according to claim 2, characterized in that: The second clustering index is determined by: Retrieving abnormal weather loss record data of related provinces for several years, and vectorizing each abnormal weather loss record data; The abnormal meteorological loss record data of non-recent years after vectorization processing are integrated into the third loss characteristic data, and the second Euclidean distance between the fourth loss characteristic data of the most recent year after vectorization processing and the third loss characteristic data is calculated. The clustering distance of the second clustering indicator is determined based on the second Euclidean distance and the standard clustering distance.

5. A method for predicting provincial power load by combining meteorological and economic data according to claim 4, characterized in that: If the clustering distance of the first clustering indicator is smaller than the clustering distance of the second clustering indicator, the method further includes: The clustering distance of the first clustering indicator is increased at least once using a preset adjustment coefficient to make it greater than the clustering distance of the second clustering indicator.

6. A method for predicting provincial power load by combining meteorological and economic data according to claim 1, characterized in that: The step of correcting the first province-wide power load forecast value based on the first influence coefficient and the second influence coefficient to obtain a second province-wide power load forecast value includes: The first influence coefficient and the second influence coefficient are multiplied by the first province-wide power load forecast value at the same time to obtain a revised second province-wide power load forecast value.

Images