Weather process judgment and early warning method for high-load event of power grid
By building a load-meteorological database and using the XGBoost model, the problem of determining and early warning of high-load events in the power grid is solved, accurate prediction and response measures for high-load events are achieved, and the safety risk prevention and control and power generation arrangement capabilities of the power grid are improved.
Patent Information
- Application Number
- CN202410991637.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology lacks a clear method and early warning mechanism for determining high-load events in the power grid, which leads to the inability to accurately predict and respond to high-load events.
By collecting historical grid load data and meteorological data, a load-meteorological element database is constructed, four threshold intervals are set for high-load event level determination, and prediction is used using the XGBoost model, and high-load event level warning and response measures are performed in combination with meteorological data.
It has achieved clear judgments on high-load events in the power grid and warnings at different levels, supported high-load events at municipal and provincial levels, guided safety risk prevention and control, emergency response and power generation arrangements, and is easy to promote and apply.
Smart Images

Figure CN118917478A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power grids, and in particular relates to a method for determining and warning a weather process for a high-load event in a power grid. Background Art
[0002] The power load is the total amount of electric power taken from the power system by various power users' electrical equipment in industry, agriculture, enterprises, transportation, scientific research institutions, culture and entertainment, and people's lives at a certain moment or for a period of time. Changes in meteorological conditions such as temperature, humidity, air pressure, and rainfall will affect the power load and electricity demand, especially in the summer cooling and winter heating periods, which have an important impact on the power balance of the power grid. The lower the temperature in winter, the greater the heating load, and the higher the temperature in summer, the greater the cooling load. Meteorological factors dominated by temperature are important meteorological factors affecting electricity demand. Taking the highest temperature as an example, from 28°C to 35°C, as the temperature rises, the load increases sharply, and the sensitivity exceeds 4 million kilowatts / °C. The accuracy of load forecasting seriously affects power dispatching decisions. High-load events in the power grid mainly refer to situations in which the power load of the power grid increases significantly and may break historical records under conditions of continuous high temperature weather. At present, there is no clear method for determining high-load events in the power grid, and it is impossible to accurately warn of high-load events. Therefore, how to overcome the shortcomings of the existing technology is an urgent problem to be solved in the field of power grid technology. Summary of the invention
[0003] The purpose of the present invention is to solve the deficiencies of the prior art and to provide a method for determining and warning the weather process of a power grid high load event.
[0004] To achieve the above purpose, the technical solution adopted by the present invention is as follows: A method for determining and warning a weather process of a power grid high load event, characterized in that it comprises the following steps: The first step is to collect the hourly load data of various cities in the power grid of the area to be determined and the meteorological data of the corresponding time to form a sample; In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures.
[0005] Furthermore, preferably, in the first step, hourly load data of various cities in the power grid of the region to be determined and meteorological data of the corresponding time in the past five years are collected.
[0006] Furthermore, preferably, in the first step, the meteorological data includes hourly temperature, air pressure, humidity, wind speed, precipitation and irradiance data.
[0007] Furthermore, preferably, in the fourth step, K=5.
[0008] Furthermore, preferably, in the fourth step, during training, the ratio of the training set to the test set is 7:3, the training set is used for training, and the test set is used for testing.
[0009] Furthermore, preferably, in the fifth step, each city issues a warning of a corresponding level based on the respective high-load event levels obtained through prediction.
[0010] Furthermore, preferably, in the sixth step, when the regional power grid to be determined is a level one high load event, a level one high load event warning is issued, and the response measures are to continuously pay attention to changes in load and weather, and to do a good job of on-duty tracking.
[0011] Furthermore, it is preferred that in the sixth step, when the regional power grid to be determined is a level 2 high load event, a level 2 high load event warning is issued, and the response measures are to strengthen inspections of the operating status of substations, and to do a good job in safety risk prevention and control and emergency response in high temperature weather.
[0012] Furthermore, preferably, in the sixth step, when the regional power grid to be determined is a level 3 high load event, a level 3 high load event warning is issued, and the response measure is to adjust the power generation arrangement to ensure peak power consumption; the power generation arrangement includes hydropower, wind power, and photovoltaic power generation arrangements.
[0013] Furthermore, preferably, in the sixth step, when the regional power grid to be determined is a level 4 high load event, a level 4 high load event warning is issued, and the response measures are to adopt peak shifting and air-conditioning load adjustment measures for enterprises to guide users to use electricity scientifically and reduce peak load demand.
[0014] The present invention does not limit the specific methods of the measures adopted in the above-mentioned countermeasures, and the existing technology can be adopted.
[0015] In the first step of the present invention, a sample is the load and corresponding meteorological data of the city at that hour.
[0016] In the fifth step of the present invention, it is assumed that a province has a total of n cities. If n is an odd number, n / 2 is not an integer. In this case, it is rounded down, that is, ⌊n / 2⌋ represents n / 2 rounded down; for example, if n=21, ⌊n / 2⌋ is 10; if n=201, ⌊n / 2⌋ is also 10.
[0017] Compared with the prior art, the present invention has the following beneficial effects:
[0018] Currently, there is no clear method for determining high-load events in power grids. The present invention predicts high-load events in power grids in the area to be determined, clarifies four levels of warning and judgment conditions, and can provide different levels of warnings for high-load events at the municipal to provincial levels to the dispatching department based on weather forecasts, and can support promotion and application in other provinces as well as the entire Southern Power Grid network.
[0019] The method of the present invention can guide safety risk prevention and control work, emergency response work, power generation arrangement work, etc., and is easy to promote and apply. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a flow chart of a method for determining and warning a weather process of a power grid high load event in an application example of the present invention; Figure 2 It is a specific flow chart of the construction of the high-load event prediction model of the application example of the present invention. DETAILED DESCRIPTION
[0021] The present invention is further described in detail below in conjunction with embodiments.
[0022] Those skilled in the art will appreciate that the following examples are only used to illustrate the present invention and should not be considered to limit the scope of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or the product specifications are used. If the manufacturer of the materials or equipment used is not specified, they are all conventional products that can be purchased. Example 1
[0023] A method for determining and warning a weather process of a power grid high load event comprises the following steps: The first step is to collect the hourly load data of various cities in the power grid of the area to be determined and the meteorological data of the corresponding time to form a sample; In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures. Example 2
[0024] A method for determining and warning a weather process of a power grid high load event comprises the following steps: The first step is to collect the hourly load data of various cities in the power grid of the area to be determined and the meteorological data of the corresponding time to form a sample; In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures.
[0025] In the first step, the hourly load data of each city in the power grid to be determined and the meteorological data of the corresponding time in the past five years are collected.
[0026] In the first step, the meteorological data includes hourly temperature, air pressure, humidity, wind speed, precipitation and irradiance data.
[0027] In the fourth step, K=5.
[0028] In the fourth step, during training, the ratio of the training set to the test set is 7:3, the training set is used for training, and the test set is used for testing.
[0029] In the fifth step, each city will issue a warning of the corresponding level based on the predicted high-load event level. Example 3
[0030] A method for determining and warning a weather process of a power grid high load event comprises the following steps: The first step is to collect the hourly load data of various cities in the power grid of the area to be determined and the meteorological data of the corresponding time to form a sample; In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures.
[0031] In the first step, the hourly load data of each city in the power grid to be determined and the meteorological data of the corresponding time in the past five years are collected.
[0032] In the first step, the meteorological data includes hourly temperature, air pressure, humidity, wind speed, precipitation and irradiance data.
[0033] In the fourth step, K=5.
[0034] In the fourth step, during training, the ratio of the training set to the test set is 7:3, the training set is used for training, and the test set is used for testing.
[0035] In the fifth step, each city will issue a warning of the corresponding level based on the predicted high-load event level.
[0036] In the sixth step, when the regional power grid to be determined is a Level 1 high-load event, a Level 1 high-load event warning is issued. The response measures are to continuously pay attention to load and weather changes and do a good job of on-duty tracking.
[0037] In the sixth step, when the regional power grid to be determined is a level 2 high load event, a level 2 high load event warning is issued. The response measures are to strengthen inspections of the operating status of substations and to do a good job in safety risk prevention and control and emergency response in high temperature weather.
[0038] In the sixth step, when the regional power grid to be determined is a level 3 high load event, a level 3 high load event warning is issued, and the response measures are to adjust the power generation arrangement to ensure peak power consumption; the power generation arrangement includes hydropower, wind power, and photovoltaic power generation arrangements.
[0039] In the sixth step, when the regional power grid to be determined is a Level 4 high-load event, a Level 4 high-load event warning is issued. The response measures are to take peak-shifting and air-conditioning load adjustment measures to guide users to use electricity scientifically and reduce peak load demand.
[0040] Application Examples
[0041] like Figure 1 and Figure 2 As shown, a method for determining and warning the weather process of a power grid high load event includes the following steps: The first step is to collect the hourly load data of various cities in the power grid in the area to be determined and the meteorological data of the corresponding time to form a sample; among them, the area to be determined is Guangdong Province; specifically, the hourly load data of various cities in the power grid in the area to be determined in the past five years (2019-2023) and the meteorological data of the corresponding time are collected.
[0042] In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures.
[0043] In the first step, the meteorological data includes hourly temperature, air pressure, humidity, wind speed, precipitation and irradiance data.
[0044] In the fourth step, K=5.
[0045] In the fourth step, during training, the ratio of the training set to the test set is 7:3, the training set is used for training, and the test set is used for testing.
[0046] In the fifth step, each city will issue a warning of the corresponding level based on the predicted high-load event level.
[0047] In the sixth step, when the regional power grid to be determined is a Level 1 high-load event, a Level 1 high-load event warning is issued. The response measures are to continuously pay attention to load and weather changes and do a good job of on-duty tracking.
[0048] In the sixth step, when the regional power grid to be determined is a level 2 high load event, a level 2 high load event warning is issued. The response measures are to strengthen inspections of the operating status of substations and to do a good job in safety risk prevention and control and emergency response in high temperature weather.
[0049] In the sixth step, when the regional power grid to be determined is a level 3 high load event, a level 3 high load event warning is issued, and the response measures are to adjust the power generation arrangement to ensure peak power consumption; the power generation arrangement includes hydropower, wind power, and photovoltaic power generation arrangements.
[0050] In the sixth step, when the regional power grid to be determined is a Level 4 high-load event, a Level 4 high-load event warning is issued. The response measures are to take peak-shifting and air-conditioning load adjustment measures to guide users to use electricity scientifically and reduce peak load demand.
[0051] From July 5 to July 16, 2023, Guangdong will experience continuous, large-scale high temperature weather, which has the characteristics of "wide range of high temperatures, continuous development, and strong extreme degree". On July 15, 93% of national stations in Guangdong experienced high temperatures above 35°C, and the highest temperature in Guangzhou reached 38.9°C. The highest load in Guangdong on July 15 reached a new high. The forecast data of various meteorological variables for the next 24 hours forecast on July 14 are input into the high-load event prediction model of the present invention, and it is calculated that a third-level high-load event warning will occur in Guangdong Province on July 15. That is, the present invention predicts that a higher level of high-load events may occur on July 15, which is consistent with the actual record high load in Guangdong on July 15. At the same time, the method of the present invention is used to conduct experiments in other provinces, such as Yunnan Province, and the prediction results are consistent with the actual situation. This shows that the method of the present invention can well judge and warn the weather process of high-load events in the power grid, and can guide safety risk prevention and control work, emergency response work, power generation arrangement work, etc.
[0052] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A method for determining and warning the weather process of high-load events in power grids, It is characterized in that The steps include: The first step is to collect the hourly load data of various cities in the power grid of the area to be determined and the meteorological data of the corresponding time to form a sample; In the second step, considering the important influence of high temperature on load, the samples with the highest temperature above 25°C were preliminarily screened to form a load-meteorological element database; The third step is to obtain the daily maximum load from the data collected in the first step, and then calculate the average value of the daily maximum load of each city in the regional power grid to be determined, and set 4 threshold intervals for determination: 0<Daily maximum load-the average value of daily maximum load in the city in that year≤20% of the average value of daily maximum load in the city in that year; The average value of the daily maximum load of the city in that year is 20% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ the average value of the daily maximum load of the city in that year 40%; The average value of the daily maximum load of the city in that year is 40% < the average value of the daily maximum load - the average value of the daily maximum load of the city in that year ≤ 60% of the average value of the daily maximum load of the city in that year; Daily maximum load - the average value of the daily maximum load of the city in that year > 60% of the average value of the daily maximum load of the city in that year; The four threshold intervals represent different levels of high load events, namely, level 1 high load event, level 2 high load event, level 3 high load event, and level 4 high load event; The samples in the load-meteorological element database are screened according to the four threshold intervals, and the samples that meet the judgment conditions of each threshold interval are screened, and the samples are marked with the corresponding high-load event level; The fourth step is to use the meteorological data of the samples screened in the third step as input and the corresponding high-load event level as output to train the XGBoost model; during the training, the importance ranking of all meteorological data types will be obtained, and the meteorological data types with the top K importance ranking will be selected; the meteorological data corresponding to the selected meteorological data type will be used as input again, and the corresponding high-load event level will be used as output to train the XGBoost model to obtain the high-load event prediction model for various cities; Step 5: input the collected meteorological data of various cities on the date to be predicted into the corresponding model obtained in Step 4 to obtain the current high-load event level of various cities; For the power grid in the region to be determined, assuming that there are n cities in a province, when at least ⌊n / 2⌋+1 cities experience a first-level high-load event, and the number of cities at or above the second level is less than the number of cities at the first level, the power grid in the region to be determined is a first-level high-load event; When at least ⌊n / 2⌋+1 cities experience a Level 2 high load event, and the number of Level 3 and above cities is less than the number of Level 2 cities, the regional power grid to be determined is a Level 2 high load event; When at least ⌊n / 2⌋+1 cities experience a level 3 or higher high load event, and the number of level 4 cities is less than the number of level 3 cities, the regional power grid to be determined is a level 3 high load event; When at least ⌊n / 2⌋+1 cities experience a Level 4 high load event, the regional power grid to be determined is a Level 4 high load event; The sixth step is to issue early warnings and provide response measures for the regional power grid to be determined according to the corresponding high-load event level, and then carry out corresponding control based on the response measures.
2. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the first step, the hourly load data of each city in the power grid to be determined and the meteorological data of the corresponding time in the past five years are collected.
3. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the first step, the meteorological data includes hourly temperature, air pressure, humidity, wind speed, precipitation and irradiance data.
4. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the fourth step, K=5.
5. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the fourth step, during training, the ratio of the training set to the test set is 7:3, the training set is used for training, and the test set is used for testing.
6. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the fifth step, each city will issue a warning of the corresponding level based on the predicted high-load event level.
7. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the sixth step, when the regional power grid to be determined is a Level 1 high-load event, a Level 1 high-load event warning is issued. The response measures are to continuously pay attention to load and weather changes and do a good job of on-duty tracking.
8. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the sixth step, when the regional power grid to be determined is a level 2 high load event, a level 2 high load event warning is issued. The response measures are to strengthen inspections of the operating status of substations and to do a good job in safety risk prevention and control and emergency response in high temperature weather.
9. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the sixth step, when the regional power grid to be determined is a level 3 high load event, a level 3 high load event warning is issued, and the response measures are to adjust the power generation arrangement to ensure peak power consumption; the power generation arrangement includes hydropower, wind power, and photovoltaic power generation arrangements.
10. The method for determining and warning the weather process of high-load events in a power grid according to claim 1, It is characterized in that In the sixth step, when the regional power grid to be determined is a Level 4 high-load event, a Level 4 high-load event warning is issued. The response measures are to take peak-shifting and air-conditioning load adjustment measures to guide users to use electricity scientifically and reduce peak load demand.
Citation Information
Patent Citations
Provincial power grid load zoning forecasting method and system
CN109861222A
Power grid high and low temperature identification method based on dynamic threshold
CN116244656A
Risk event processing method and device in power grid, storage medium and processor
CN118133183A
Machine-learning-enhanced distributed energy resource management system
US20230244197A1
