A deep learning-based regional power grid load prediction method and system

Abstract

The application provides a kind of based on deep learning regional power grid load prediction method and system, regional power grid load prediction method includes: integration real-time power consumption data and historical power consumption database constructs dynamic dataset;Using time series analysis model extracts current cycle power consumption characteristics and carries out similar judgment with historical same period data;If it is judged that it is not similar, obtain the first position information, judge whether the current region satisfies the first electricity condition;Combining weather forecast matching historical similar weather mode, extract the power consumption characteristic data and power generation characteristic data in future second cycle and associated cycle;Judge whether the current region satisfies the load requirement in future second cycle;If it is judged that it is not satisfied, the power of adjacent region needs to be adjusted;If it is judged that it is satisfied, continue routine monitoring, and send power borrowing early warning to adjacent region;If it is judged that it is similar, it indicates that the load condition of the current region is stable.The application improves the intelligent degree of power grid load prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid load prediction, in particular to a regional power grid load prediction method and system based on deep learning. BACKGROUND

[0002] With the development of science and technology, the daily life level of residents is also improving. With the improvement of living standards, most of the scenes need electricity, and various different factories also need large electricity, so this tests the load capacity of the regional power grid. Regional power grid load prediction refers to collecting and analyzing various data to predict the power grid load situation of a certain region in the future. Such prediction is crucial for the planning and operation of the power system, which can help power companies better manage power supply and demand and ensure the stable operation of the power grid. However, the current power grid load prediction method is not intelligent enough. SUMMARY

[0003] Therefore, the present application provides a regional power grid load prediction method and system based on deep learning, which improves the intelligence of power grid load prediction.

[0004] To solve the above problems, the present application provides a regional power grid load prediction method based on deep learning, which includes: integrating real-time power consumption data and historical power consumption database to construct a dynamic dataset; according to the dynamic dataset, using a time series analysis model to extract current period power consumption characteristics, and making a similarity judgment with historical data of the same period; if the judgment is not similar, obtaining the geographic location of the current region to obtain the first location information, and judging whether the current region meets the first electricity fee condition according to the first location information to obtain the first judgment result; combining the weather forecast to match the historical similar weather mode, extracting the power consumption feature data and power generation feature data in the future second period and the associated period; judging whether the current region meets the load requirement in the future second period according to the power consumption feature data and power generation feature data and the first judgment result; if the judgment is not satisfied, the power of the neighboring region needs to be adjusted; if the judgment is satisfied, continue to monitor daily and send a power borrowing warning to the neighboring region in advance; if the judgment is similar, it means that the load situation of the current region is stable.

[0005] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by integrating real-time power consumption data and a historical power consumption database and constructing a dynamic data set, subsequent power consumption query is more convenient, the time series analysis model extracts current period power consumption characteristics, and makes a similarity judgment with historical same period data, which can facilitate subsequent more reasonable judgment on whether it meets the requirements, and by judging whether the first electricity charge condition is met through the geographic location, and then obtaining the weather condition according to the weather forecast, the combination of the weather condition and the electricity charge condition enables more comprehensive judgment on the power consumption condition of the region, the electricity charge condition reflects the overall power consumption of the region, and the weather condition affects the power generation condition and the power generation capacity and power consumption condition of the region in response to sudden weather, therefore, the comprehensive judgment on the power consumption condition of the region according to the two can better judge the subsequent regional load, and the comprehensive judgment on whether the load requirement is met according to the comprehensive electricity charge condition, weather condition and historical power consumption condition enables more reasonable and intelligent prediction of the regional load, the comprehensive multiple factors can improve the accuracy of the prediction result, thus the power consumption load of the region can be more accurately judged, and the current regional power consumption guarantee can be better met, thereby providing more stable guarantee for the power consumption of the residents in the region.

[0006] In an example of the present application, according to the dynamic data set, the time series analysis model extracts current period power consumption characteristics, and makes a similarity judgment with historical same period data, further comprising: obtaining the power consumption in the first period of the current region from the historical power consumption database, denoted as first power consumption data; obtaining the historical power consumption in the historical first period corresponding to the first period of the current region from the historical power consumption database, denoted as historical first power consumption data; comparing the first power consumption data with the historical first power consumption data to judge whether the first power consumption data and the historical first power consumption data are similar.

[0007] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by comparing whether the power consumption in the existing first period and the historical first power consumption data in the corresponding historical first period are similar, the power consumption in the first period is clearly known, which is convenient for subsequent judgment.

[0008] In an example of the present application, comparing the first power consumption data with the historical first power consumption data to judge whether the first power consumption data and the historical first power consumption data are similar comprises: obtaining the power consumption of a plurality of historical first periods before the first period, and adding and averaging to obtain a historical average power consumption; subtracting the power consumption of each historical first period from the historical average power consumption to obtain a plurality of first differences, and comparing the plurality of first differences; subtracting the power consumption of the historical first period corresponding to the largest first difference from the historical average power consumption to obtain a second difference; judging whether the first power consumption data and the historical first power consumption data are similar according to the second difference.

[0009] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by setting the average power consumption of the plurality of historical first periods, and making difference with the power consumption of each period, and taking the power consumption of the historical first period corresponding to the maximum difference value to make difference with the average power consumption, the similarity is judged by the difference value, so that the judgment of the historical similarity value of the current first power consumption data is more accurate, and the setting of the average power consumption enables better comprehensive feedback on the power consumption in the plurality of historical periods, so that the effectiveness and scientificity of the data are stronger, and more reasonable data support can be provided for subsequent similarity judgment, thereby facilitating improvement of the judgment accuracy.

[0010] In an example of the present application, according to the second difference value, judging whether the first power consumption data and the historical first power consumption data are similar further includes: setting a similarity range as a center point of the similarity threshold value, a range of positive and negative second difference values; if the first power consumption data falls within the similarity range, it is judged to be similar; if the first power consumption data falls outside the similarity range, it is judged to be dissimilar.

[0011] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by setting the similarity range of the similarity threshold value as the center point, the positive and negative second difference values, whether the first power consumption data is similar is judged, so that the similarity judgment of the first power consumption data is more accurate, and by the combination of the similarity threshold value and the second difference value, the similarity range is more in line with the actual situation, so that the historical power consumption of the current region can be better reflected, facilitating subsequent judgment and prediction of power consumption load, and improving the accuracy and intelligent degree of prediction.

[0012] In an example of the present application, according to the first position information, judging whether the current region satisfies the first electricity fee condition to obtain the first judgment result further includes: obtaining the total electricity fee of each same season in the history of the current region according to the first position information; calculating an electricity fee coefficient according to the total electricity fee of each same season in the history; and judging whether the first electricity fee condition is satisfied according to the electricity fee coefficient.

[0013] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by obtaining the total electricity fee of the same season in the history to calculate the electricity fee coefficient, whether the first electricity fee condition is satisfied is judged according to the electricity fee coefficient, the total electricity fee can intuitively reflect the power consumption situation of the current region, so as to serve as data support for supporting subsequent load prediction, and thus the accuracy of load prediction can be better improved.

[0014] In an example of the present application, the determining whether the first electricity cost condition is met according to the electricity cost coefficient further comprises: performing a difference between the total electricity cost of each same quarter in history in a step-by-step manner, obtaining a plurality of electricity cost difference values; performing a mean value calculation on the plurality of electricity cost difference values, obtaining the electricity cost coefficient; if the electricity cost coefficient is greater than or equal to a set threshold value, it is determined that the first electricity cost condition is met; if the electricity cost coefficient is less than the set threshold value, it is determined that the first electricity cost condition is not met.

[0015] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by performing a difference between the total electricity cost of each same quarter in history in a step-by-step manner, and performing a mean value calculation, the obtained electricity cost coefficient can be more in line with the actual situation, and can also reflect the growth of the current regional power consumption, so that the subsequent load can be better predicted, and the prediction accuracy is improved.

[0016] In an example of the present application, the extracting the power consumption feature data and the power generation feature data in the future second period and the associated period in combination with the weather forecast matching the historical similar weather mode comprises: obtaining the weather forecast in the future second period in the current region, obtaining the first weather condition; querying the historical first weather condition similar to the first weather condition in the weather system, and obtaining the historical second period power consumption corresponding to the historical first weather condition, obtaining the historical second power consumption data and obtaining the historical first power generation data corresponding to the historical first weather condition; the matching mode of the historical first power generation data and the historical second power consumption data is the same as the historical first weather condition; obtaining the number of power generation weather in the first weather condition, denoted as the first weather number, obtaining the number of adverse weather in the first weather condition, denoted as the second weather number; according to the first weather number and the second weather number, the corresponding historical first weather condition is found; wherein, the power generation weather is sunny and cloudy, and the adverse weather is all weather conditions except the power generation weather.

[0017] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by obtaining the number of power generation weather and adverse weather in the first weather condition to determine the corresponding historical first weather condition, by this way, the finding of the historical first weather condition is more rapid, and the found historical first weather condition is more in line with the actual situation, and it is also convenient to understand the weather condition in the period, so that the power consumption and power generation are considered more comprehensively, and then the subsequent load prediction is more accurate.

[0018] In an example of the present application, the searching for the corresponding historical first weather condition according to the first weather number and the second weather number further comprises: if there is no historical first weather condition matching the first weather number and the second weather number, reducing the first weather number by a set number to match; if there is a matching historical first weather condition, determining whether there are multiple historical first weather conditions; if there are multiple historical first weather conditions, taking the average of the power consumption in the multiple historical first weather conditions as the historical second power consumption data, and taking the average of the power generation in the multiple historical first weather conditions as the historical first power generation data.

[0019] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by setting to reduce by a set number to match when there is no matching number of historical first weather conditions, the matching range is larger, and similar weather conditions can be obtained as reference data, thereby facilitating subsequent power load prediction under the weather conditions, and making the power prediction more accurate. At the same time, by setting to determine whether there are multiple historical first weather conditions when there is a matching condition, the average is taken as the historical first power generation data and power consumption data when there are multiple conditions, which makes the obtained historical first power generation data and historical first power consumption data more consistent with the current situation, thereby making the subsequent prediction more accurate.

[0020] In an example of the present application, the determining whether the current region meets the load requirement in the future second period according to the power consumption feature data, the power generation feature data, and the first determination result further comprises: if the historical second power consumption data is greater than or equal to the historical first power generation data, determining that the load requirement is not met regardless of whether the first electricity fee condition is met; if the historical second power consumption data is less than the historical first power generation data and the first electricity fee condition is met, determining that the load requirement is not met; and if the historical second power consumption data is less than the historical first power generation data and the first electricity fee condition is not met, determining that the load requirement is met.

[0021] Compared with the prior art, the technical effects achieved by adopting the technical scheme are: by setting to determine the load requirement under different conditions, the determination method can cover multiple different conditions, thereby making the prediction of the load more comprehensive and comprehensive in actual use, and making the obtained prediction result more accurate and more in line with the actual situation. When the historical second power consumption data is greater than or equal to the historical first power generation data, it indicates that the current power consumption is too large, and the region is difficult to bear the current power consumption, so it is determined that the load requirement is not met. When the first electricity fee condition is met, it indicates that the power consumption of the current region is growing rapidly, so attention needs to be paid to the possibility of load exceeding.

[0022] The application further provides a regional power grid load prediction system based on deep learning, which is used to implement the regional power grid load prediction method of any one of the above, and comprises: an acquisition module, which is used to acquire regional electricity consumption, power generation, first position information and first weather conditions; a judgment and calculation module, which is used to judge and calculate whether the current region meets the load requirement; and a control module, which is used to control the electricity.

[0023] Compared with the prior art, the technical effects achieved by adopting the technical scheme are as follows: the acquisition module is arranged to acquire the electricity consumption, power generation and other information of the region, so that the acquisition is faster, and subsequent calculation is facilitated, and then the judgment and calculation module is arranged to quickly judge the load requirement, thereby improving the convenience and accuracy of the judgment, so that the intelligent degree of the system is higher, and finally the control module is arranged to control the electricity, so that the use is more convenient.

[0024] After adopting the technical scheme of the application, the following technical effects can be achieved:

[0025] (1) By integrating real-time electricity consumption data and historical electricity consumption database and constructing a dynamic data set, subsequent electricity consumption query is more convenient, and the time sequence analysis model extracts current period electricity consumption characteristics and makes similar judgment with historical same period data, so that subsequent judgment on whether the requirements are met can be made more reasonably, and whether the first electricity fee condition is met is judged through the geographic position, and then the weather condition is obtained according to the weather forecast, so that more comprehensive judgment can be made on the electricity consumption of the region through the combination of the weather condition and the electricity fee condition, the electricity fee condition reflects the overall electricity consumption of the region, and the weather condition affects the power generation situation and the power generation capacity and electricity consumption of the region in response to sudden weather, so that the electricity consumption of the region is comprehensively judged according to the two, which can better judge the subsequent regional load, and the electricity fee condition, the weather condition and the historical electricity consumption are comprehensively judged to determine whether the load requirement is met, so that the prediction of the regional load is more reasonable and intelligent, and the comprehensive multiple factors can improve the accuracy of the prediction result, so that the electricity load of the region can be more accurately judged, and the electricity of the residents in the region can be better guaranteed, so that more stable guarantee can be provided for the electricity of the residents in the region. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical scheme of the embodiments of the application, the drawings to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0027] Figure 1A flow chart of a regional power grid load prediction method based on deep learning provided for an embodiment of the present application is shown in the figure.

[0028] Figure 2 A module diagram of a regional power grid load prediction system based on deep learning provided for an embodiment of the present application is shown in the figure.

[0029] Explanation of reference signs:

[0030] 100, regional power grid load prediction system; 110, control module; 120, judgment calculation module; 130, control module. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application are described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0032] [First embodiment]

[0033] Referring to Figure 1 , the present application provides a regional power grid load prediction method based on deep learning, comprising:

[0034] Step S100: integrate real-time power consumption data and historical power consumption database to construct a dynamic data set;

[0035] Step S200: according to the dynamic data set, adopt a time series analysis model to extract current period power consumption characteristics, and make a similarity judgment with historical data of the same period;

[0036] Step S210: if the judgment is not similar, obtain the geographical position of the current region to obtain first position information, and judge whether the current region meets the first electricity fee condition according to the first position information to obtain a first judgment result;

[0037] Step S300: combine the weather forecast to match the historical similar weather mode, extract the power consumption feature data and power generation feature data in the future second period and the associated period; and judge whether the current region meets the load requirement in the future second period according to the power consumption feature data and power generation feature data and the first judgment result;

[0038] Step S310: if the judgment is not satisfied, the power of the neighboring region needs to be adjusted;

[0039] Step S320: if the judgment is satisfied, continue daily monitoring, and send a power borrowing warning to the neighboring region in advance;

[0040] Step S220: If the similarity is judged, it indicates that the load condition of the current region is stable

[0041] Specifically, in daily use, the power generation and power consumption of each region are uploaded to the power grid system for record keeping, and the power consumption of the region is taken as the first data, and a historical power consumption database is established according to all historical data existing in the power grid system.

[0042] Further, when load prediction is needed, first, all power consumptions in the first period of the current region are obtained, the first period is a man-made setting value, which can be a week, two weeks or a month, etc., which can be adjusted according to actual needs. After obtaining, the corresponding historical power consumption of the historical first period is found in the historical power consumption database, which is recorded as the historical first power consumption period, wherein the historical first period is the date segment corresponding to the first period, which is pushed forward, for example, if the first period is from October 10th to 15th of a month, the historical first period can be from October 10th to 15th of the previous year, and so on.

[0043] Further, the average value of the power consumptions of the plurality of historical first periods is obtained, and the power consumptions of each historical first period are subtracted to obtain a plurality of first difference values, and the plurality of first difference values are compared to obtain a maximum value. The power consumption of the historical first period corresponding to the maximum value is subtracted from the historical average power consumption to obtain a second difference value, and a similarity threshold is set, which is a man-made setting value and can be changed according to actual conditions. The similarity range is set as the center point of the similarity threshold, and the range of positive and negative second difference values, i.e. when the similarity threshold is 16 billion kilowatt hours, and the second difference value is 2 billion kilowatt hours, the similarity range is [14, 18], and the first power consumption data is judged according to the similarity range. When it is judged to be outside the similarity range, the weather condition and power consumption are obtained according to the first position information.

[0044] Further, the total amount of electricity charges of each quarter of the current region is obtained to calculate the electricity charge coefficient, the total amount of electricity charges of each same quarter of the history is subtracted step by step, a plurality of electricity charge difference values are obtained, and the average value of the plurality of electricity charge difference values is obtained to obtain the electricity charge coefficient, and compared with a set threshold value, wherein the set threshold value is a man-made setting value and can be adjusted according to actual conditions, and the result of whether the first electricity charge condition is met is obtained.

[0045] Further, while combining the weather forecast with the historical similar weather mode, the power consumption characteristic data and the power generation characteristic data in the future second period and the associated period are extracted, that is, the first weather condition of the future second period in the current area is obtained, the historical similar weather is searched according to the number of power generation weather and severe weather, when there is no historical first weather condition matching the first weather number and the second weather number, the first weather number is reduced by a set number for matching, wherein the set number is a man-made setting value, which can be adjusted according to the actual situation, and can be set to 1. When there is a matching historical first weather condition, the average of the power consumption in the multiple historical first weather conditions is taken as the historical second power consumption data, and the average of the power generation in the multiple historical first weather conditions is taken as the historical first power generation data.

[0046] Further, the deep neural network is used to fuse multi-dimensional data for load prediction, wherein the multi-dimensional data includes the power consumption characteristic data and the power generation characteristic data and the first judgment result, that is, the historical second power consumption data, the historical first power generation data and the first judgment result are used to judge whether the load of the area in the future second period meets the requirements.

[0047] Preferably, the method is applied to a deep learning model for training, and historical data is substituted for calculation, so that the prediction result of the model is more accurate, and the subsequent model is established successfully and is convenient for use.

[0048] Preferably, the historical power consumption database is established by obtaining the daily power consumption and the historical power consumption, so that the subsequent power consumption query is more convenient, and whether the power consumption in the existing first period is similar to the historical first power consumption data in the corresponding historical first period is compared, so that the power consumption in the first period is clearly known, and a more reasonable judgment on whether it meets the requirements can be made subsequently, and whether the first electricity charge condition is met is judged by the geographical position, and then the weather condition is obtained according to the weather forecast, so that a more comprehensive judgment on the power consumption of the area can be made by combining the weather condition and the electricity charge condition. The electricity charge condition reflects the overall power consumption of the area, and the weather condition affects the power generation condition and the power generation capacity and power consumption of the area in response to sudden weather, so that the power consumption of the area is comprehensively judged according to the two, which can better judge the subsequent area load, and the electricity charge condition, the weather condition and the historical power consumption are comprehensively judged to determine whether the load requirement is met, so that the prediction of the area load is more reasonable and intelligent, and the comprehensive multiple factors can improve the accuracy of the prediction result, so that the power load of the area can be more accurately judged, and the power supply of the residents in the area can be better guaranteed, so that more stable guarantee can be provided for the power supply of the residents in the area.

[0049] Specifically, according to the dynamic data set, the time series analysis model is used to extract the current period power consumption characteristics, and the similarity is judged with the historical data of the same period, which also includes: obtaining the power consumption in the first period of the current region from the historical power consumption database, which is recorded as the first power consumption data; obtaining the historical power consumption in the first period of the corresponding first period of the current region from the historical power consumption database, which is recorded as the historical first power consumption data; comparing the first power consumption data with the historical first power consumption data to determine whether the first power consumption data and the historical first power consumption data are similar.

[0050] Preferably, by comparing the power consumption in the existing first period with the historical first power consumption data of the corresponding historical first period, the power consumption in the first period is clearly understood, which is convenient for subsequent judgment.

[0051] Specifically, comparing the first power consumption data with the historical first power consumption data to determine whether the first power consumption data and the historical first power consumption data are similar includes: obtaining the power consumption of a plurality of historical first periods before the first period, and adding and averaging to obtain the historical average power consumption; the historical average power consumption is subtracted from the power consumption of each historical first period to obtain a plurality of first difference values, and the plurality of first difference values are compared; the power consumption of the historical first period corresponding to the largest first difference value is subtracted from the historical average power consumption to obtain a second difference value; according to the second difference value, it is determined whether the first power consumption data and the historical first power consumption data are similar.

[0052] Preferably, by setting the average power consumption of a plurality of historical first periods, and subtracting the power consumption of each period, and taking the power consumption of the historical first period corresponding to the maximum difference value to subtract the average power consumption, the similarity is judged by the difference value, so that the judgment of the historical similarity value of the current first power consumption data is more accurate, and the setting of the average power consumption makes the power consumption in a plurality of historical periods have better comprehensive feedback, so that the effectiveness and scientificity of the data are stronger, and more reasonable data support can be provided for subsequent similar judgment, which is convenient for improving the judgment accuracy.

[0053] Specifically, according to the second difference value, it is determined whether the first power consumption data and the historical first power consumption data are similar, which also includes: setting the similarity range as the range of positive and negative second difference value with the similarity threshold value as the center point; if the first power consumption data falls within the similarity range, it is judged to be similar; if the first power consumption data falls outside the similarity range, it is judged to be dissimilar.

[0054] Preferably, the first power consumption data is judged whether similar by setting a similar threshold as a center point and a similar range of positive and negative second difference values, so that the similar judgment of the first power consumption data is more accurate, and the combination of the similar threshold and the second difference value makes the similar range more suitable for the actual situation, so that the current region's historical power consumption situation can be better reflected, facilitating subsequent judgment and prediction of power load, and improving the accuracy and intelligence of the prediction.

[0055] Specifically, the first judgment result obtained by judging whether the current region meets the first electricity cost condition according to the first location information further comprises: obtaining the total electricity cost of each same quarter in the current region according to the first location information; calculating an electricity cost coefficient according to the total electricity cost of each same quarter in the history; and judging whether the first electricity cost condition is met according to the electricity cost coefficient.

[0056] Preferably, the electricity cost coefficient is calculated by obtaining the total electricity cost of the same quarter in the history, so as to judge whether the first electricity cost condition is met according to the electricity cost coefficient. The total electricity cost can intuitively reflect the power consumption situation of the current region, thereby supporting the subsequent load prediction data, and further improving the accuracy of load prediction.

[0057] Specifically, the judgment of whether the first electricity cost condition is met according to the electricity cost coefficient further comprises: calculating a plurality of electricity cost differences by pairwise and step-by-step difference of the total electricity cost of each same quarter in the history; calculating the electricity cost coefficient by averaging the plurality of electricity cost differences; if the electricity cost coefficient is greater than or equal to a set threshold, it is judged that the first electricity cost condition is met; and if the electricity cost coefficient is less than the set threshold, it is judged that the first electricity cost condition is not met.

[0058] Preferably, the electricity cost coefficient is calculated by pairwise and step-by-step difference of the total electricity cost of each same quarter in the history and averaging, so that the obtained electricity cost coefficient is more suitable for the actual situation, and can also reflect the growth of the power consumption situation of the current region, thereby better predicting the subsequent load and improving the accuracy of the prediction.

[0059] Specifically, the historical first weather condition similar to the first weather condition is queried in the weather system, and the power consumption of the historical second period corresponding to the historical first weather condition is obtained to obtain the historical second power consumption data and the historical first power generation data corresponding to the historical first weather condition further comprises: the matching mode of the historical first weather condition of the historical first power generation data and the historical second power consumption data is the same; the number of power generation weather in the first weather condition is obtained, which is recorded as a first weather number, and the number of severe weather in the first weather condition is obtained, which is recorded as a second weather number; the corresponding historical first weather condition is found according to the first weather number and the second weather number; wherein, the power generation weather is sunny and cloudy, and the severe weather is all weather conditions except the power generation weather.

[0060] Preferably, the corresponding historical first weather condition is found by acquiring the number of power generation weather and bad weather in the first weather condition, so that the finding of the historical first weather condition is more rapid, and the found historical first weather condition is more consistent with the actual situation, and the weather condition in the period can be understood, so that the power consumption and power generation are considered more comprehensively, and the subsequent load prediction is more accurate.

[0061] Specifically, according to the first weather number and the second weather number, the corresponding historical first weather condition is found, which further includes: if there is no historical first weather condition matching the first weather number and the second weather number, the first weather number is reduced by a set number for matching; if there is a matching historical first weather condition, it is judged whether there are multiple historical first weather conditions; if there are multiple historical first weather conditions, the average of the power consumption in the multiple historical first weather conditions is taken as the historical second power consumption data, and the average of the power generation in the multiple historical first weather conditions is taken as the historical first power generation data.

[0062] Preferably, by setting to reduce a set number for matching when there is no matching number of historical first weather conditions, the matching range is larger, and similar weather conditions can be obtained as reference data, so that the subsequent power consumption load prediction under the weather condition is facilitated, and the power consumption prediction is more accurate. Also by setting to judge whether there are multiple historical first weather conditions when there is a matching condition, the average is taken as the historical first power generation data and power consumption data when there are multiple conditions, so that the obtained historical first power generation data and historical first power consumption data are more consistent with the current situation, and the subsequent prediction is more accurate.

[0063] Specifically, according to the historical second power consumption data, the historical first power generation data and the first judgment result, it is judged whether the current region in the second period meets the load requirement, which further includes: if the historical second power consumption data is greater than or equal to the historical first power generation data, it is judged that the load requirement is not met regardless of whether the first electricity fee condition is met; if the historical second power consumption data is less than the historical first power generation data, and the first electricity fee condition is met, it is judged that the load requirement is not met; if the historical second power consumption data is less than the historical first power generation data, and the first electricity fee condition is not met, it is judged that the load requirement is met.

[0064] Preferably, by setting the judgment load requirements in different cases, the judgment mode can cover a variety of different cases, and the predicted load can be more comprehensive and comprehensive in actual use, so that the prediction result is more accurate and more reflects the actual situation. When the historical second power consumption data is greater than or equal to the historical first power generation data, it indicates that the current power consumption is too large, and the region is difficult to support the current power consumption, so the judgment is not satisfied. When the first electricity fee condition is met, it indicates that the power consumption of the current region is growing rapidly, so there is a possibility of load exceeding, which needs attention.

[0065] The application also provides a regional power grid load prediction system 100 based on deep learning, which is used to realize the regional power grid load prediction method of any one of the above. The regional power grid load prediction system 100 comprises: an acquisition module 110, which is used to acquire regional power consumption, power generation, first position information and first weather conditions; a judgment and calculation module 120, which is used to judge and calculate whether the current region meets the load requirement; and a control module 130, which is used to control the power.

[0066] Preferably, the acquisition module 110 is used to acquire the power consumption, power generation and other information of the region, so that the acquisition is faster and convenient for subsequent calculation, and the judgment and calculation module 120 is used to quickly judge the load requirement, thereby improving the convenience and accuracy of the judgment, so that the intelligent degree of the system is higher, and finally the control module 130 is used to control the power, so that it is more convenient to use.

[0067] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the application, but not to limit them; although the application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for predicting regional power grid load based on deep learning, characterized in that, include: Integrate real-time electricity consumption data with historical electricity consumption databases to construct dynamic datasets; Based on the dynamic dataset, a time series analysis model is used to extract the electricity consumption characteristics of the current cycle, and a similarity judgment is made with historical data of the same period. If the location is determined to be dissimilar, the geographical location of the current region is obtained to obtain the first location information. Based on the first location information, it is determined whether the current region meets the first electricity fee condition to obtain the first judgment result. Based on the first location information, obtain the total electricity bill for each historical quarter in the current region; Calculate the electricity cost factor based on the total electricity cost for each of the same historical quarters; Determine whether the first electricity fee condition is met based on the electricity fee coefficient; The total electricity bill for each of the same historical quarters is subtracted pairwise to obtain multiple electricity bill differences. The average of the multiple electricity cost differences is used to calculate the electricity cost coefficient. If the electricity fee coefficient is greater than or equal to the set threshold, then the first electricity fee condition is determined to be met; If the electricity rate coefficient is less than the set threshold, it is determined that the first electricity rate condition is not met. By combining meteorological forecasts with historical similar weather patterns, electricity consumption and power generation characteristic data for the second future cycle and related cycles are extracted. Obtain the weather forecast for the current region within the next second period to obtain the first weather condition; query the meteorological system for historical first weather conditions similar to the first weather condition, and obtain the electricity consumption for the corresponding historical second period to obtain historical second electricity consumption data; and obtain the historical first power generation data corresponding to the historical first weather condition. The matching method for the historical first weather conditions in the historical first power generation data is the same as the matching method for the historical first weather conditions in the historical second power consumption data; The number of days with power generation in the first weather condition is obtained and recorded as the first weather number; the number of days with severe weather in the first weather condition is obtained and recorded as the second weather number. Based on the first weather data and the second weather data, find the corresponding historical first weather condition; The term "power generation weather" refers to sunny and cloudy days, while "severe weather" refers to all weather conditions other than those mentioned above. If there is no historical first weather condition that matches the first weather number and the second weather number, then the first weather number is reduced by a predetermined number for matching; If a matching historical first weather condition exists, then determine whether there are multiple historical first weather conditions. If there are multiple historical first weather conditions, the average electricity consumption of the multiple historical first weather conditions shall be taken as the historical second electricity consumption data, and the average power generation of the multiple historical first weather conditions shall be taken as the historical first power generation data. Based on the electricity consumption characteristic data, the power generation characteristic data, and the first judgment result, determine whether the current region meets the load requirements in the next second cycle; If the conditions are not met, then power allocation from neighboring regions is required. If the conditions are met, daily monitoring will continue, and an early warning for electricity borrowing will be issued to the adjacent area. If the judgment is similar, it indicates that the load situation in the current region is stable.

2. The regional power grid load forecasting method based on deep learning according to claim 1, characterized in that, The step of extracting current cycle electricity consumption characteristics using a time-series analysis model based on the dynamic dataset and comparing them with historical data from the same period further includes: The electricity consumption of the current region in the first period is obtained from the historical electricity consumption database and recorded as the first electricity consumption data. The electricity consumption of the current region within the historical first cycle corresponding to the first cycle is obtained from the historical electricity consumption database and recorded as the historical first electricity consumption data. The first electricity consumption data is compared with the historical first electricity consumption data to determine whether the first electricity consumption data and the historical first electricity consumption data are similar.

3. The regional power grid load forecasting method based on deep learning according to claim 2, characterized in that, The step of comparing the first electricity consumption data with the historical first electricity consumption data to determine whether the first electricity consumption data and the historical first electricity consumption data are similar includes: Obtain the electricity consumption of multiple historical first periods prior to the first period, add them together and take the average value to obtain the historical average electricity consumption; The historical average electricity consumption is subtracted from the electricity consumption of each historical first period to obtain multiple first difference values, and the multiple first difference values ​​are compared. The second difference is obtained by subtracting the electricity consumption of the historical first period corresponding to the largest first difference from the historical average electricity consumption. Based on the second difference, it is determined whether the first electricity consumption data and the historical first electricity consumption data are similar.

4. The regional power grid load forecasting method based on deep learning according to claim 3, characterized in that, The step of determining whether the first electricity consumption data and the historical first electricity consumption data are similar based on the second difference further includes: The similarity range is defined as the range of positive and negative values ​​of the second difference, centered on the similarity threshold. If the first electricity consumption data falls within the similarity range, then it is determined to be similar; If the first electricity consumption data falls outside the similarity range, then it is determined that they are not similar.

5. The regional power grid load forecasting method based on deep learning according to claim 1, characterized in that, The step of determining whether the current region meets the load requirements in the next second cycle based on the electricity consumption characteristic data, the power generation characteristic data, and the first determination result also includes: If the second historical electricity consumption data is greater than or equal to the first historical power generation data, then regardless of whether the first electricity fee condition is met, it is determined that the load requirement is not met. If the historical second electricity consumption data is less than the historical first power generation data and the first electricity fee condition is met, then it is determined that the load requirement is not met. If the second historical electricity consumption data is less than the first historical power generation data and does not meet the first electricity fee condition, then it is determined that the load requirement is met.

6. A regional power grid load forecasting system based on deep learning, said regional power grid load forecasting system being used to implement the regional power grid load forecasting method as described in any one of claims 1-5, characterized in that, The regional power grid load forecasting system includes: The acquisition module is used to acquire the region's electricity consumption, power generation, first location information, and first weather conditions; A judgment and calculation module is used to determine whether the current region meets the load requirements. A control module, which is used to allocate and control the power supply.

Images