Regional power supply prediction model construction method and system and regional power demand prediction method and system

Abstract

The invention provides a regional power supply prediction model construction method and system and a regional power demand prediction method and system, and is applied to the technical field of regional power operation. The method comprises the following steps: fitting an internal and external temperature conversion relation of a building in each regional grid based on multi-dimensional historical meteorological data of each regional grid in a target region; based on the temperature difference between the internal temperature of the building in each regional grid and the temperature threshold value of refrigeration and heating, the cooling and heating load demand change condition of each regional grid is determined; based on the influence of the population density in each regional grid building on the temperature, carrying out fusion processing on the cooling and heating load demand change condition of each regional grid to obtain the cooling and heating load demand change condition of the target region; regional development and regional power consumption characteristics are considered, and power demand conversion is carried out based on the cooling and heating load demand change condition of the target region to obtain a power supply prediction model. According to the invention, the problems of low load prediction accuracy and low power demand prediction accuracy in emergencies are solved.

Description

Technical Field

[0001] This invention relates to the field of regional power operation technology, specifically to a method for constructing a regional power supply forecasting model, a method for forecasting regional power demand, and a system. Background Technology

[0002] For urban areas, buildings account for 30% of global energy demand and CO2 emissions, with heating accounting for about half of that. Furthermore, with global warming, population growth in tropical regions, and increasing environmental concerns, the demand for space cooling is rapidly increasing, while the efficiency of cooling equipment remains relatively low, resulting in persistently high energy demands for space cooling. Although various methods exist for simulating changes in renewable energy supply, most cannot simultaneously account for changes in energy demand from both heating and cooling, leading to low accuracy in load forecasting.

[0003] Most common load forecasting models currently rely on summarizing patterns in historical load data to predict future loads. This method cannot achieve fine-grained load forecasting within a region, resulting in low load forecasting accuracy. Furthermore, these models cannot respond quickly to sudden events that may cause abrupt load changes, leading to low load forecasting accuracy under such events, which in turn results in low accuracy of electricity demand forecasting. Summary of the Invention

[0004] To overcome the problems of low load forecasting accuracy and low power demand forecasting accuracy under sudden events, this invention provides a method for constructing a regional power supply forecasting model, a regional power demand forecasting method, and a system.

[0005] On one hand, the present invention provides a method for constructing a regional power supply prediction model, comprising:

[0006] The internal and external temperature conversion relationship of buildings within each grid is fitted based on multi-dimensional historical meteorological data of each area within the target area.

[0007] Based on the temperature difference between the internal temperature of buildings in each regional grid and the temperature thresholds for cooling and heating, the changes in cooling and heating load demand in each regional grid are determined.

[0008] The changes in cooling and heating load demand of each regional grid are fused based on the impact of population density on temperature within each grid building to obtain the changes in cooling and heating load demand of the target area.

[0009] Considering regional development and regional electricity consumption characteristics, an electricity demand conversion is performed based on the changes in cooling and heating load demand in the target region to obtain the power supply prediction model for the target region.

[0010] The regional grid is obtained by dividing the target area into spatial grids, and the internal temperature of the buildings within the regional grid is determined based on the internal and external temperature conversion relationship of the buildings within the regional grid.

[0011] Optionally, after fitting the internal and external temperature conversion relationship of buildings within each regional grid based on multi-dimensional historical meteorological data of each regional grid within the target area, the method further includes:

[0012] The initial internal and external temperature conversion relationship of buildings in each grid area is smoothed based on the thermal inertia of the buildings.

[0013] Optionally, the smoothing process for the initial internal and external temperature conversion relationship of buildings within each grid region based on the building's thermal inertia includes:

[0014] For each regional grid, the indoor temperatures of buildings within the regional grid for several consecutive historical days are weighted and fused based on the smoothing coefficient of the target region.

[0015] Optionally, after smoothing the initial internal and external temperature conversion relationship of buildings within each grid region based on the building's thermal inertia, the process also includes:

[0016] For each regional grid, based on the historical window opening behavior data of users within the buildings of that regional grid, determine the user behavior influencing factors of the buildings within that regional grid;

[0017] Analyze historical data on the impact of economic development on regional temperature in the target area to determine the temperature impact factor of economic factors in the target area.

[0018] The internal and external temperature conversion relationship of buildings within the regional grid is corrected based on the user behavior influence factor of buildings within the regional grid and the temperature influence factor of economic factors of the target area.

[0019] Optionally, the expression for the internal and external temperature conversion relationship of buildings within the regional grid is:

[0020]

[0021] In the formula, The internal temperature of the building at grid point l on date d. The internal temperature of the building at point d (date) and point l (grid area) after smoothing. τ is the user behavior influencing factor of buildings within the regional grid l, M2 is the temperature influencing factor of economic factors in the target area, and τ is the influence coefficient of user ventilation and window opening behavior on the internal temperature of buildings. The average daily window opening time for buildings within grid l is given. Let T be the initial internal temperature of the building at location d on date d and grid l in the region. 0,l Given date d, external temperature at grid l, and S 0,l W 0,l H 0,l Let be the solar radiation intensity, wind speed, and air humidity outside the building at grid l on date d; and let a, b, and c be the correlation coefficients of solar radiation intensity, wind speed, and humidity in the target area, respectively. * W * H * T * Let represent solar radiation intensity, wind speed, and humidity, which have no impact on electricity demand, respectively, and α be the smoothing coefficient for the target area. The initial internal temperature of the building at grid l on date d-1. The initial internal temperature of the building at grid l on date d-2.

[0022] Optionally, the process of fusing the changes in heating and cooling load demands of each regional grid based on the impact of population density on temperature within each regional grid building to obtain the changes in heating and cooling load demands of the target region includes:

[0023] The fusion parameters for each regional grid are determined based on the ratio between the population density of each regional grid and the population density of the target region.

[0024] The cooling and heating load demand changes of each regional grid are weighted and fused based on the fusion parameters of each regional grid to obtain the cooling and heating load demand changes of the target region.

[0025] Optionally, the step of considering regional development and regional electricity consumption characteristics, and obtaining the electricity supply forecasting model for the target region based on the changes in cooling and heating load demand in the target region, includes:

[0026] Based on the power coefficients of cooling and heating in the target area, the electricity demand is converted according to the changes in cooling and heating load demand in the target area to obtain the electricity supply model of cooling and heating load in the target area.

[0027] The power supply model for the target area is adjusted by considering the changes in power supply demand brought about by regional development and the characteristics of regional power consumption stages, so as to obtain the power supply prediction model for the target area.

[0028] Optionally, the expression for the power supply prediction model of the target area is:

[0029] P all =Pb +P h *ΔH r +P c *ΔC r +D+M2E;

[0030] In the formula, P all P represents the predicted electricity supply and demand for the target area. b P is the reference power for the target region. h P c The power coefficients for cooling and heating are ΔH, respectively. r ΔC r These represent the changes in cooling and heating load demand in the target area, respectively. D is the electricity consumption correction amount corresponding to the characteristics of the electricity consumption stage in the area. M2 is the temperature influence factor of economic factors in the target area. E is the change in electricity demand caused by regional development in the target area.

[0031] Optionally, after obtaining the power supply prediction model for the target area, the method further includes:

[0032] Based on the historical power supply data of the target area, the model parameters of the power supply prediction model for the target area are calibrated using a genetic algorithm.

[0033] Optionally, based on historical power supply data of the target area, a genetic algorithm is used to calibrate the model parameters of the power supply prediction model for the target area, including:

[0034] A set of model parameters to be calibrated is used as individuals in a population. The fitness function is the residual standard deviation between the prediction results of the power supply prediction model for the target area and the historical power supply data of the target area. The model parameters are iteratively optimized through individual selection, crossover, and mutation operations.

[0035] Optionally, the model parameters to be calibrated include the temperature correlation coefficient, smoothing coefficient, and temperature thresholds for cooling and heating, which correspond to the meteorological data of each dimension in the internal and external temperature conversion relationship of buildings within each grid area.

[0036] On the other hand, the present invention also provides a regional electricity demand forecasting method, comprising:

[0037] Acquire current multi-dimensional meteorological data for each grid area within the target region;

[0038] Based on the current multi-dimensional meteorological data of each regional grid, the power supply prediction model constructed using any of the above implementation methods is used to make real-time predictions of the power supply demand of the target area, thereby obtaining the current predicted value of the power supply demand of the target area.

[0039] Optionally, after obtaining the current electricity supply demand forecast for the target area, the method further includes:

[0040] The power dispatching in the target area is controlled based on the changes in the current power supply and demand forecast values ​​of the target area.

[0041] On the other hand, the present invention also provides a system for constructing a regional power supply prediction model, comprising:

[0042] The relationship fitting module is used to fit the internal and external temperature conversion relationship of buildings within each grid area based on multi-dimensional historical meteorological data of each grid area within the target area.

[0043] The load change determination module is used to determine the changes in cooling and heating load demand in each area grid based on the temperature difference between the internal temperature of buildings in each area grid and the temperature thresholds for cooling and heating.

[0044] The fusion module is used to fuse the changes in cooling and heating load demand of each grid based on the impact of population density in buildings within each grid on temperature, so as to obtain the changes in cooling and heating load demand of the target area.

[0045] The model building module is used to consider regional development and regional electricity consumption characteristics, and to convert electricity demand based on the changes in cooling and heating load demand of the target area to obtain the electricity supply prediction model of the target area.

[0046] The regional grid is obtained by dividing the target area into spatial grids, and the internal temperature of the buildings within the regional grid is determined based on the internal and external temperature conversion relationship of the buildings within the regional grid.

[0047] On the other hand, the present invention also provides a regional electricity demand forecasting system, comprising:

[0048] The data acquisition module is used to acquire the current multi-dimensional meteorological data of each grid area within the target area;

[0049] The prediction module is used to perform real-time prediction of the power supply demand of the target area based on the current multi-dimensional meteorological data of each regional grid and the power supply prediction model constructed using any of the above embodiments, so as to obtain the current power supply demand prediction value of the target area.

[0050] On the other hand, the present invention also provides an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0051] The memory is used to store one or more programs;

[0052] When the one or more programs are executed by the at least one processor, the method for constructing a regional power supply forecasting model or the method for forecasting regional power demand described above are implemented.

[0053] On the other hand, the present invention also provides a readable storage medium having an executable program stored thereon, wherein when the executable program is executed, it implements the method for constructing a regional power supply forecasting model or the method for forecasting regional power demand as described in any one of the above.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0055] This invention provides a method for constructing a regional power supply forecasting model, a method and system for forecasting regional power demand. By dividing the target area into regional grids, a temperature conversion relationship between the inside and outside of the regional grid is constructed. Combined with the temperature thresholds for cooling and heating, the forecasting of cooling and heating load demand changes at the regional grid level is realized, thereby achieving fine-grained determination of cooling and heating loads within the region and improving the accuracy of load forecasting.

[0056] This invention fits the internal and external temperature relationship of buildings based on multi-dimensional historical meteorological data, then integrates the changes in heating and cooling load demands in various regional grids, taking into account the impact of population density on temperature. Finally, it constructs a power supply forecasting model for the target area by converting heating and cooling loads into electricity demand, while also considering regional development and electricity consumption characteristics. The model comprehensively considers the impact of different factors on regional load and power supply from natural, human, and economic perspectives, using the conversion between internal and external temperatures, temperature to load, and load to power supply as the main thread. This multi-faceted model significantly improves the accuracy and precision of regional load and electricity demand forecasts. In the event of emergencies, real-time monitoring of multi-dimensional meteorological data enables real-time updates to electricity demand forecasts, ensuring a rapid response of the power grid to emergencies and guaranteeing the safe and stable operation of the power system. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating a method for constructing a regional power supply prediction model according to the present invention.

[0058] Figure 2 This is a flowchart illustrating the construction of a regional power supply forecasting model and a regional power demand forecasting method according to the present invention.

[0059] Figure 3 This is a linear graph showing the fitted relationship between external wind speed and internal initial temperature in an area, representing an example of the present invention.

[0060] Figure 4 A linear graph showing the fitting relationship between the external solar radiation intensity and the internal initial temperature of a region, as an example of the present invention;

[0061] Figure 5 A fitted curve of actual heating and cooling load demand versus initial internal temperature in a region as an example of the present invention;

[0062] Figure 6 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0063] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0064] Example 1:

[0065] This invention provides a method for constructing a regional power supply prediction model, as illustrated in the schematic diagram below. Figure 1 As shown, it includes:

[0066] Step S110: Fit the internal and external temperature conversion relationship of buildings in each grid area based on the multi-dimensional historical meteorological data of each grid area within the target area.

[0067] Step S120: Based on the temperature difference between the internal temperature of buildings in each regional grid and the temperature thresholds for cooling and heating, determine the changes in cooling and heating load demand for each regional grid.

[0068] Step S130: Based on the influence of population density in buildings within each regional grid on temperature, perform fusion processing on the changes in cooling and heating load demand of each regional grid to obtain the changes in cooling and heating load demand of the target area.

[0069] Step S140: Considering regional development and regional electricity consumption characteristics, power demand conversion is performed based on the changes in cooling and heating load demand of the target region to obtain the power supply prediction model of the target region.

[0070] In this example implementation, the regional grid is obtained by spatially dividing the target area. The target area can be a spatial region requiring load studies, such as one or more urban administrative regions, or a portion of an urban administrative region. Multi-dimensional historical meteorological data can include historical meteorological data related to multiple dimensions, such as indoor and outdoor air temperature, air humidity, solar irradiance angle, wind speed, precipitation, air pressure, and weather conditions. The internal temperature of buildings within the regional grid is determined based on the temperature conversion relationship between the interior and exterior of the buildings within the regional grid. Cooling and heating temperature thresholds include cooling temperature thresholds and heating temperature thresholds. The cooling temperature threshold refers to the temperature threshold at which indoor cooling is required, such as a cooling temperature threshold set to 28 degrees Celsius. The heating temperature threshold refers to the temperature threshold at which indoor heating is required, such as a heating temperature threshold set to 8 degrees Celsius. These cooling and heating temperature thresholds can be obtained based on statistical analysis of historical cooling and heating data for the region. Since different levels of regional development may affect regional temperature—for example, areas with higher urbanization levels have higher average temperatures—this factor needs to be considered. Regional electricity consumption characteristics can include both location-based and time-based characteristics. For example, office buildings and residential buildings in different locations have different electricity consumption characteristics, as do weekdays, weekends, and holidays. Therefore, this factor needs to be considered. Electricity demand can be transformed using various fitting or regression equations. The emergencies addressed in this invention can be various sudden natural disasters, such as line icing during extreme snow and ice weather or line malfunctions during typhoons. These emergencies may lead to sudden failures of distribution facilities and transmission lines, as well as substation system failures. The power supply forecasting model constructed in this example can achieve rapid response of load forecasting to real-time meteorological data, enabling the power system to make timely adjustments and responses in the face of emergencies, reducing equipment failure and maintenance rates, and ensuring the rapid restoration of safe and stable operation of the power system.

[0071] In one example implementation, after fitting the initial internal and external temperature conversion relationship of buildings within each regional grid based on multi-dimensional historical meteorological data of each regional grid and the historical internal temperature of buildings within each regional grid, the method further includes:

[0072] The initial internal and external temperature conversion relationship of buildings in each grid area is smoothed based on the thermal inertia of the buildings.

[0073] In this example implementation, since buildings exhibit thermal inertia, their internal temperature does not immediately reflect changes in external temperature. Therefore, the initial internal temperature of the building in the internal-external temperature conversion relationship is smoothed. For example, a moving average method is used to smooth the initial internal temperature over the past few days to simulate the building's thermal inertia. Exemplarily, for each regional grid, the indoor temperatures of buildings within that regional grid over several adjacent days are weighted and fused based on the smoothing coefficient of the target region to obtain a more stable and realistic internal temperature prediction. The smoothing coefficient is the temperature weight applied to the previous day, with a value between 0 and 1, which can be selected according to actual conditions. An example of the specific formula for smoothing is shown below:

[0074]

[0075] In the formula, Let d be the internal temperature of the building at point l in the grid after smoothing, and α be the smoothing coefficient for the target area. Given date d and the initial internal temperature of the building at grid l in the region, The initial internal temperature of the building at grid l on date d-1. This represents the initial internal temperature of the building at grid l on date d-2. This example uses a smoothing process to make the model more realistic.

[0076] In one example implementation, after smoothing the initial internal and external temperature conversion relationships of buildings within each grid region based on the building's thermal inertia, the method further includes:

[0077] For each regional grid, based on the historical window opening behavior data of users within the buildings of that regional grid, determine the user behavior influencing factors of the buildings within that regional grid;

[0078] Analyze historical data on the impact of economic development on regional temperature in the target area to determine the temperature impact factor of economic factors in the target area.

[0079] The internal and external temperature conversion relationship of buildings within the regional grid is corrected based on the user behavior influence factor of buildings within the regional grid and the temperature influence factor of economic factors of the target area.

[0080] In this example implementation, user behavior and economic factors can be considered to modify the indoor-outdoor temperature conversion relationship. User behavior characteristics can include user behaviors that affect the indoor temperature of a building, such as living habits and residents' occupational characteristics, like opening windows or doors for ventilation. Other characteristics can also be considered to determine the indoor-outdoor temperature conversion relationship, such as the population density of a region, the location of residents, the age composition of family members, and other humanistic characteristics, as well as regional economic level and residents' income level, and other economic characteristics. Specifically, relevant historical data of the target area can be collected first, such as multi-dimensional historical meteorological data, historical window-opening behavior data of users in each building in the region, historical development data of the region, and historical indoor and outdoor temperatures of each building. By combining multi-dimensional historical meteorological data and historical indoor and outdoor temperatures of each building, and integrating various meteorological factors (such as temperature, solar radiation intensity, wind speed, and humidity), the conversion coefficients of each meteorological factor are fitted to convert external conditions into indoor perceived temperature. For example, the expression for the initial indoor-outdoor temperature conversion relationship of buildings in each regional grid is as follows:

[0081]

[0082] In the formula, a, b, and c are the correlation coefficients of solar radiation intensity, wind speed, and humidity in the target area, respectively, which are the conversion coefficients of the fit.

[0083] By simulating the impact of user window-opening behavior on indoor temperature using historical user window-opening behavior data and historical indoor and outdoor temperature data for each building within the region, a user behavior impact factor is obtained. For example, the expression fitted to the user behavior impact factor is:

[0084]

[0085] In the formula, τ is the influence coefficient of the user's ventilation window opening behavior on the internal temperature of the building. This represents the average daily window opening time for buildings within grid l. This example, based on different regional human characteristics (primarily influenced by users' window opening and ventilation habits), compares the differences in initial internal temperature caused by users' window opening and ventilation habits in different regional grids, fitting a user behavior influencing factor. By correcting the initial internal and external temperature conversion relationship using the user behavior influencing factor, the internal and external temperature conversion relationship of buildings within the regional grid is obtained. Especially in hot weather, the original external temperature data is reintegrated into the initial internal temperature to reflect the impact of temporary ventilation measures (such as opening windows at night) on the building's internal temperature. This considers the actual situation of regional user behavior and can improve the model's prediction accuracy.

[0086] Besides considering human factors, the impact of economic factors on temperature can also be taken into account. Based on historical regional data, a temperature impact factor for economic factors can be summarized, and the influence of economic factors on the initial internal temperature can be analyzed. Specifically, the impact of regional development on regional temperature can be statistically analyzed; areas with higher urbanization levels tend to have higher average temperatures. By statistically fitting historical urbanization level indicators or economic indicators with historical indoor temperature data, the temperature impact factor M2 of economic factors can be derived. By considering the influence of user behavior and economic factors on the initial internal temperature, the expression for the relationship between the building's internal and external temperatures can be obtained as follows:

[0087]

[0088] In the formula, The internal temperature of the building at grid point l on date d. The internal temperature of the building at point d (date) and point l (grid area) after smoothing. M1 represents the user behavior influence factor of buildings within the regional grid, and M2 represents the temperature influence factor of economic factors in the target area. The initial internal temperature after smoothing can be corrected based on the user behavior influence factor and the temperature influence factor of economic factors, further improving the accuracy of the temperature conversion relationship between the building's interior and exterior.

[0089] In some implementations, step S120 determines the changes in cooling and heating load demands for each regional grid based on the temperature difference between the internal temperature of buildings within each regional grid and the cooling and heating temperature thresholds. Specifically, it determines the changes in cooling and heating load demands of buildings based on the temperature difference between the internal temperature of buildings and the cooling or heating temperature thresholds, which can be expressed by the following formula:

[0090]

[0091] In equation (5), ΔH d,l and ΔC d,l These are the coefficients of change in cooling and heating load demand for the building, T. h T represents the heating temperature threshold. c This represents the cooling temperature threshold. A meteorological index, the load variation coefficient, is constructed to estimate building energy demand, representing the cumulative effect of temperatures falling below or exceeding a certain setpoint temperature over a given period. Heating and cooling demands are calculated based on the regional building interior temperature and the equilibrium point temperature threshold for cooling or heating.

[0092] In other examples, step S130, which involves fusing the changes in heating and cooling load demands of each regional grid based on the impact of population density within buildings on temperature, to obtain the changes in heating and cooling load demands of the target region, includes:

[0093] The fusion parameters for each regional grid are determined based on the ratio between the population density of each regional grid and the population density of the target region.

[0094] The cooling and heating load demand changes of each regional grid are weighted and fused based on the fusion parameters of each regional grid to obtain the cooling and heating load demand changes of the target region.

[0095] In this example implementation, the ratio between the population density of each regional grid and the population density of the target area can be used as the fusion parameter for the corresponding regional grid. Then, based on the fusion parameter, the changes in heating and cooling demand of each regional grid are weighted and fused to obtain the changes in heating and cooling load demand of the target area. For example, by calculating the heating and cooling load change coefficient of each grid cell in the region, adding these values, and then weighting each cell according to the population at each location, the expression for the changes in heating and cooling load demand of the target area is obtained as follows:

[0096]

[0097] In equation (6), ΔH d,r and ΔC d,r These are the cooling and heating load variation coefficients of the buildings in the target area, ρ. l ρ r Let l represent the population density of the unit area grid and the population density of the target area, respectively, and l∈r indicate that the area grid l belongs to the target area r.

[0098] In other examples, step S140, which considers regional development and regional electricity consumption characteristics, and performs electricity demand conversion based on changes in the cooling and heating load demand of the target region to obtain a power supply prediction model for the target region, includes:

[0099] Based on the power coefficients of cooling and heating in the target area, the electricity demand is converted according to the changes in cooling and heating load demand in the target area to obtain the electricity supply model of cooling and heating load in the target area.

[0100] The power supply model for the target area is adjusted by considering the changes in power supply demand brought about by regional development and the characteristics of regional power consumption stages, so as to obtain the power supply prediction model for the target area.

[0101] In this example implementation, the change in cooling and heating load demand in the target area can be the change in cooling and heating load demand in the target area. Based on the power coefficients of cooling and heating in the target area, the change in cooling and heating load demand is converted into the electricity supply demand of the target area, resulting in a cooling and heating load electricity supply model for the target area. Considering the impact of regional development and electricity consumption characteristics (such as weekend and weekday electricity consumption characteristics) on the electricity supply demand, the cooling and heating load electricity supply model is adjusted to obtain the electricity supply prediction model for the target area. For example, the expression of the final electricity supply prediction model for the target area is:

[0102] P all =P b +P h *ΔH r +P c *ΔC r +D+M2E (7)

[0103] In the formula, P all P represents the predicted electricity supply and demand for the target area. b P is the reference power for the target region. h P c The power coefficients for cooling and heating are ΔH, respectively. r ΔC r These represent the changes in cooling and heating load demand in the target area, respectively. D is the electricity consumption correction amount corresponding to the characteristics of the electricity consumption stage in the area. M2 is the temperature influence factor of economic factors in the target area. E is the change in electricity demand caused by regional development in the target area.

[0104] The above example is based on regional population weighting and electricity demand conversion. It combines population data, weights the cooling and heating load variation coefficients according to the population density of each region, and converts these weighted daily rates into the actual specific electricity demand of the region.

[0105] In some example implementations, to ensure the prediction accuracy of the power supply prediction model, after obtaining the power supply prediction model for the target area, the method further includes:

[0106] Based on the historical power supply data of the target area, the model parameters of the power supply prediction model for the target area are calibrated using a genetic algorithm.

[0107] In this example implementation, the model parameters to be calibrated include the temperature correlation coefficient and smoothing coefficient corresponding to the meteorological data of each dimension in the internal and external temperature conversion relationship of buildings within each regional grid, as well as the temperature thresholds for cooling and heating, which are a, b, c, a, and T in the above formulas (1)-(7). h T cThe model parameters are calibrated using a genetic algorithm by comparing the historical power supply data of the target area with the prediction results of the power supply prediction model, thereby improving the model's prediction accuracy. For example, a set of model parameters to be calibrated are used as individuals in the population. The fitness function is the residual standard deviation between the prediction results of the power supply prediction model of the target area and the historical power supply data of the target area. Iterative optimization of the model parameters is performed through selection, crossover, and mutation operations. For instance, a genetic algorithm is used to find model parameters that minimize the residual or residual standard deviation between the model's prediction results and the actual measured values ​​(historical power supply data). Different random seeds can be used for repeated testing to ensure that the random initial population does not substantially affect the final result. Each parameter can be randomly selected within a certain range. The specific optimization solution process is as follows: 1) Initialize the population by randomly generating 200 initial parents, each parent consisting of a set of parameters (a, b, c, α, T). h ,T c The process is as follows: 1) Formulation. 2) Fitness assessment: For each individual (parent), a regional electricity demand forecasting model is used to predict regional electricity demand, and the residual standard deviation between the predicted and actual measured electricity supply values ​​is calculated. 3) Selection: Parents are selected based on fitness, typically using roulette wheel selection or tournament selection methods. 4) Crossover: Model parameters are randomly paired from the selected parents, and new offspring are generated through crossover (i.e., model parameter swapping). 5) Mutation: The newly generated offspring are mutated (randomly changing model parameter values) to increase population diversity. 6) Replacement: Individuals with lower fitness are replaced with newly generated offspring to form a new population. 7) Iteration: Steps 2) to 6) are repeated until the maximum number of iterations is reached or the convergence condition is met. This example uses measured high-frequency electricity data for calibration to ensure the model's predictive accuracy.

[0108] This invention is based on high-frequency power data and fully integrates the gray box method and the daily time model. It optimizes performance by introducing statistical parameters such as meteorological variables, natural factors, and human characteristics, ensuring its rapid response and predictive accuracy under regional power supply and cooling conditions.

[0109] Example 2

[0110] Based on the same inventive concept, this invention also provides a method for regional electricity demand forecasting, comprising:

[0111] Acquire current multi-dimensional meteorological data for each grid area within the target region;

[0112] Based on the current multi-dimensional meteorological data of each regional grid, the power supply prediction model constructed in Example 1 is used to make real-time predictions of the power supply demand of the target area, and the current power supply demand prediction value of the target area is obtained.

[0113] In this example implementation, based on the multi-dimensional meteorological parameters in the power supply forecasting model, the current multi-dimensional meteorological data of each regional grid within the target area is obtained. This current multi-dimensional meteorological data is then input into the power supply forecasting model, which outputs the current predicted power supply demand value for the target area, thereby achieving real-time and accurate prediction of the power supply demand in the target area. Furthermore, power supply demand prediction for the corresponding period can be performed based on recent weather forecast data (such as weather forecast data for the next two days or the current day), thus forming a power supply demand prediction curve corresponding to the weather forecast data.

[0114] In some example implementations, after obtaining the current electricity supply demand forecast for the target area, the method further includes:

[0115] The power dispatching in the target area is controlled based on the changes in the current power supply and demand forecast values ​​of the target area.

[0116] In this example implementation, when the predicted value of power supply demand changes too much or too frequently in a short period of time, it can be assumed that the power supply demand of the target area has changed. Based on the change of the predicted value, the power dispatch of the target area can be controlled to achieve real-time and rapid response of power supply in the target area and ensure the stable and safe operation of the power system.

[0117] This invention can also analyze the characteristics of regional power supply cooling and heating load changes based on the power supply prediction model, and compare the analysis results with the actual power measurement data of the region. If the error does not meet the requirements, the influence of mixed factors such as meteorological factors, human factors, and economic factors on the regional load will be reconsidered.

[0118] The following describes the specific steps of constructing a regional power supply prediction model according to a specific embodiment of the present invention, such as... Figure 2 As shown, the following processes may be included:

[0119] (1) Collection of regional load-related information: Collect multi-dimensional meteorological data, population data and measured historical electricity demand data of the designated area;

[0120] (2) Calculate the initial internal temperature within the grid buildings in the region. Integrating various meteorological factors (such as temperature, solar radiation, wind speed, and humidity), these external meteorological data are converted into perceived indoor temperature using specific coefficients. These coefficients are derived from the collected data. Figure 3 and Figure 4 As shown, this is a fitted linear graph of the initial internal temperature of a building affected by external conditions, obtained by fitting the measured wind speed, solar radiation intensity, outdoor temperature, and initial internal temperature in a certain region. Figure 3 This is a fitted linear graph of external wind speed versus initial internal temperature. Figure 4The graph shows the fitted line graph of external solar radiation intensity and internal initial temperature. It can be seen from the graph that, under the same outdoor temperature, the higher the wind speed and the higher the solar radiation intensity, the greater the internal initial temperature.

[0121] (3) Smoothing of initial internal temperature: Due to the thermal inertia of the building, the internal temperature does not immediately reflect changes in the external temperature. Methods such as moving average are used to smooth the initial internal temperature over the past few days to simulate the thermal inertia of the building and obtain a more stable and realistic internal temperature prediction.

[0122] (4) Mixed factors to correct the smoothed internal temperature: Consider user behavior and economic factors. User behavior can be the impact of temporary ventilation measures (such as opening windows at night) on the internal temperature of the building in hot weather. Economic factors can be the impact of regional economic development on temperature.

[0123] (5) Calculate the regional grid load variation: Construct a meteorological index to estimate building energy demand, representing the cumulative effect of temperatures being lower or higher than a certain setpoint temperature over a certain period. Calculate heating and cooling demands based on the initial internal temperature of buildings within the region and relevant equilibrium point temperature thresholds. For example... Figure 5 The figure shows a fitted curve of the actual cooling and heating load demand and the initial internal temperature in a certain region. It can be seen from the figure that the cooling and heating load demand is different under different initial internal temperatures.

[0124] (6) Calculate regional load changes based on regional population proportion: Weight the cooling and heating load change coefficients according to the population density of each region;

[0125] (7) Electricity demand conversion: The regional load change obtained by weighted daily rate is converted into the actual specific electricity demand of the region to obtain the regional power supply forecast model.

[0126] (8) Model calibration is performed using measured high-frequency power data, and a real-time forecast curve of regional power supply demand is established based on real-time meteorological data and forecast meteorological data.

[0127] Current common load forecasting models face a surge in uncertainties when encountering unforeseen events (such as line icing during extreme snow and ice weather or line erratic behavior during typhoons). This makes it difficult for conventional load forecasting models to accurately measure the degree of load change and electricity demand. Furthermore, sudden natural disasters can cause temporary maintenance of distribution facilities, sudden transmission line failures, and substation system faults, leading to abrupt changes in power load and impacting the safe and stable operation of the power system, further complicating load forecasting. Particularly in disaster-prone areas, the safe and stable operation of the power system is severely threatened, necessitating the development of a dynamic and accurate method for calculating regional power supply, cooling, and heating load demand. In addition, recent events such as war and disease have made short-term energy consumption reduction a focus for many countries, thus requiring the development of a dynamic method for calculating regional power supply, cooling, and heating load demand to guide the development of regional energy systems. This invention constructs a customized regional power supply forecasting model to accurately measure the dynamic changes in regional heating and cooling loads and the power supply situation. Specifically, it considers numerous factors affecting regional power supply heating and cooling load forecasting, including natural factors, human characteristics, economic factors, and meteorological factors. Changes in natural factors can affect user electricity consumption behavior and energy-consuming equipment, leading to load changes. Human characteristics can include the population density, household location, living customs, residents' occupations, and the age composition of family members in a region, which can also influence user electricity consumption habits and load levels. Economic factors include the regional economic level and residents' income levels, which affect changes in electricity load.

[0128] This invention, based on the high-frequency characteristics of power data and combined with an improved time-of-day model and simulation of building interior temperature conditions, achieves rapid and accurate calculation of regional power supply cooling and heating load status. It can accurately measure the dynamic changes in regional power supply cooling and heating load and power demand. This invention's method for dynamically calculating regional power supply cooling and heating load demand is an innovative regional power demand forecasting method. Based on high-frequency power data, it comprehensively integrates the gray box method and time-of-day model, and optimizes performance by introducing additional meteorological variables and statistical parameters, ensuring rapid response and accurate prediction under regional power supply cooling and heating conditions.

[0129] Example 3

[0130] Based on the same inventive concept, this invention also provides a system for constructing a regional power supply prediction model, comprising:

[0131] The relationship fitting module is used to fit the internal and external temperature conversion relationship of buildings within each grid area based on multi-dimensional historical meteorological data of each grid area within the target area.

[0132] The load change determination module is used to determine the changes in cooling and heating load demand in each area grid based on the temperature difference between the internal temperature of buildings in each area grid and the temperature thresholds for cooling and heating.

[0133] The fusion module is used to fuse the changes in cooling and heating load demand of each grid based on the impact of population density in buildings within each grid on temperature, so as to obtain the changes in cooling and heating load demand of the target area.

[0134] The model building module is used to consider regional development and regional electricity consumption characteristics, and to convert electricity demand based on the changes in cooling and heating load demand of the target area to obtain the electricity supply prediction model of the target area.

[0135] The regional grid is obtained by dividing the target area into spatial grids; the internal temperature of the buildings within the regional grid is determined based on the internal and external temperature conversion relationship of the buildings within the regional grid.

[0136] In one possible implementation, the system further includes:

[0137] The smoothing module is used to smooth the initial internal and external temperature conversion relationship of buildings within each grid area based on the building's thermal inertia.

[0138] In one possible implementation, the smoothing module is specifically used for:

[0139] For each regional grid, the indoor temperatures of buildings within the regional grid for several consecutive historical days are weighted and fused based on the smoothing coefficient of the target region.

[0140] In one possible implementation, the system further includes a relationship correction module, the relationship correction module being used for:

[0141] For each regional grid, based on the historical window opening behavior data of users within the buildings of that regional grid, determine the user behavior influencing factors of the buildings within that regional grid;

[0142] Analyze historical data on the impact of economic development on regional temperature in the target area to determine the temperature impact factor of economic factors in the target area.

[0143] The internal and external temperature conversion relationship of buildings within the regional grid is corrected based on the user behavior influence factor of buildings within the regional grid and the temperature influence factor of economic factors of the target area.

[0144] In one possible implementation, the expression for the internal and external temperature conversion relationship of buildings within the regional grid is:

[0145]

[0146] In the formula, The internal temperature of the building at grid point l on date d. The internal temperature of the building at point d (date) and point l (grid area) after smoothing. τ is the user behavior influencing factor of buildings within the regional grid l, M2 is the temperature influencing factor of economic factors in the target area, and τ is the influence coefficient of user ventilation and window opening behavior on the internal temperature of buildings. The average daily window opening time for buildings within grid l is given. Let T be the initial internal temperature of the building at location d on date d and grid l in the region. 0,l Given date d, external temperature at grid l, and S 0,l W 0,l H 0,l denoted as d, and respectively as ... The initial internal temperature of the building at grid l on date d-1. The initial internal temperature of the building at grid l on date d-2.

[0147] In one possible implementation, the fusion module is specifically used for:

[0148] The fusion parameters for each regional grid are determined based on the ratio between the population density of each regional grid and the population density of the target region.

[0149] The cooling and heating load demand changes of each regional grid are weighted and fused based on the fusion parameters of each regional grid to obtain the cooling and heating load demand changes of the target region.

[0150] In one possible implementation, the model building module is specifically used for:

[0151] Based on the power coefficients of cooling and heating in the target area, the electricity demand is converted according to the changes in cooling and heating load demand in the target area to obtain the electricity supply model of cooling and heating load in the target area.

[0152] The power supply model for the target area is adjusted by considering the changes in power supply demand brought about by regional development and the characteristics of regional power consumption stages, so as to obtain the power supply prediction model for the target area.

[0153] In one possible implementation, the expression for the power supply prediction model of the target area is:

[0154] P all =P b +P h *ΔH r +P c *ΔC r +D+M2E;

[0155] In the formula, P all P represents the predicted electricity supply and demand for the target area. b P is the reference power for the target region. h P c The power coefficients for cooling and heating are ΔH, respectively. r ΔC r These represent the changes in cooling and heating load demand in the target area, respectively. D is the electricity consumption correction amount corresponding to the characteristics of the electricity consumption stage in the area. M2 is the temperature influence factor of economic factors in the target area. E is the change in electricity demand caused by regional development in the target area.

[0156] In one possible implementation, the system further includes:

[0157] The model calibration module is used to calibrate the model parameters of the power supply prediction model for the target area based on historical power supply data of the target area using a genetic algorithm.

[0158] In one possible implementation, the model calibration module is specifically used for:

[0159] A set of model parameters to be calibrated is used as individuals in a population. The fitness function is the residual standard deviation between the prediction results of the power supply prediction model for the target area and the historical power supply data of the target area. The model parameters are iteratively optimized through individual selection, crossover, and mutation operations.

[0160] In one possible implementation, the model parameters to be calibrated include the temperature correlation coefficient and smoothing coefficient corresponding to the meteorological data of each dimension in the internal and external temperature conversion relationship of buildings in each regional grid, as well as the temperature thresholds for cooling and heating.

[0161] Example 4

[0162] Based on the same inventive concept, the present invention also provides a regional electricity demand forecasting system, comprising:

[0163] The data acquisition module is used to acquire the current multi-dimensional meteorological data of each grid area within the target area;

[0164] The prediction module is used to predict the power supply demand of the target area in real time based on the current multi-dimensional meteorological data of each regional grid and the power supply prediction model constructed in any of the embodiments in Example 1, so as to obtain the current power supply demand prediction value of the target area.

[0165] In one possible implementation, the system further includes:

[0166] The control and scheduling module is used to control the power dispatch of the target area based on the changes in the current power supply demand forecast of the target area.

[0167] Example 5

[0168] like Figure 6 As shown, the present invention also provides an electronic device, which may be a computer device, a microcontroller device, a smart mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, processor, and transceiver component are connected via a bus; the memory can be used to store executable programs, and an exemplary executable program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, which can be accessed and / or modified when instructions are executed.

[0169] The processor may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, and it is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the storage medium to realize the corresponding method flow or corresponding function, so as to realize the steps of the method for constructing a regional power supply forecasting model and the method for forecasting regional power demand in the above embodiments.

[0170] Example 6

[0171] Based on the same inventive concept, this invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory). An electronic device readable storage medium is a memory device within an electronic device used to store programs and data. It is understood that the storage medium here can include both the built-in storage medium within the electronic device and extended storage media supported by the electronic device. The storage medium provides storage space, which stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more executable programs (including program code). It should be noted that the storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. Loading and executing one or more instructions stored in the storage medium by the processor can implement the steps of the method for constructing a regional power supply forecasting model and the method for forecasting regional power demand in the above embodiments.

[0172] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0173] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0174] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0175] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the claims pending approval.

Claims

1. A method for constructing a regional power supply prediction model, characterized in that: include: Based on the multi-dimensional historical meteorological data of each regional grid in the target area, the internal and external temperature conversion relationship of the buildings in each regional grid is fitted; Based on the temperature difference between the internal temperature of the buildings in each regional grid and the temperature thresholds for cooling and heating, the change in cooling and heating load demand of each regional grid is determined; Based on the influence of population density in buildings of each regional grid on temperature, the change of cooling and heating load demand of each regional grid is integrated to obtain the change of cooling and heating load demand of the target area; Considering regional development and regional electricity consumption characteristics, converting power demand based on changes in cooling and heating load demand of the target area, and obtaining a power supply forecasting model for the target area; The regional grid is obtained by dividing the target area into spatial grids; the internal temperature of the building in the regional grid is determined based on the conversion relationship between the internal and external temperatures of the building in the regional grid.

2. The method according to claim 1, characterized in that After fitting the internal and external temperature conversion relationship of buildings in each regional grid based on the multi-dimensional historical meteorological data of each regional grid in the target area, it also includes: Based on the thermal inertia of the building, the initial internal and external temperature conversion relationship of the building in each regional grid is smoothed.

3. The method according to claim 2, characterized in that The smoothing process of the initial internal and external temperature conversion relationship of the buildings in each regional grid based on the thermal inertia of the buildings includes: For each regional grid, weighted fusion processing is performed on the indoor temperatures of the buildings in the regional grid for several consecutive historical days based on the smoothing processing coefficient of the target area.

4. The method according to claim 2, characterized in that: After smoothing the initial internal and external temperature conversion relationship of buildings in each regional grid based on the thermal inertia of the building, it also includes: For each regional grid, determining a user behavior influencing factor of the building in the regional grid based on historical window opening behavior data of users in the building in the regional grid; Collect historical data on the impact of economic development of the target area on regional temperature, and determine the temperature impact factor of the economic factors of the target area; The conversion relationship between the internal and external temperatures of the buildings in the regional grid is corrected based on the user behavior influencing factors of the buildings in the regional grid and the temperature influencing factors of the economic factors of the target area.

5. The method according to claim 4, characterized in that The expression for the conversion relationship between the internal and external temperatures of the buildings in the regional grid is: In the formula, is the internal temperature of the building at the grid cell l on date d, is the internal temperature of the building after smoothing at date d and regional grid l, is the user behavior influencing factor of the building in the regional grid l, M2 is the temperature influencing factor of the economic factor of the target area, τ is the influence coefficient of the user's ventilation and window opening behavior on the internal temperature of the building, is the average daily window opening time of buildings in the regional grid l, is the initial internal temperature of the building at the grid cell l on date d, T 0,l is the external temperature at the grid cell l on date d, S 0,l , W 0,l , H 0,l are the solar radiation intensity, wind speed, and air humidity outside the building at date d and regional grid l; a, b, and c are the correlation coefficients of solar radiation intensity, wind speed, and humidity in the target area, respectively. * , W * , H * , T * They represent the solar radiation intensity, wind speed, and humidity that have no effect on electricity demand, and α is the smoothing coefficient of the target area. is the initial internal temperature of the building at the regional grid l on date d-1, is the initial internal temperature of the building at regional grid l on date d-2.

6. The method according to claim 1, characterized in that The fusion processing of the change of the cooling and heating load demand of each regional grid based on the influence of the population density in the building of each regional grid on the temperature, and obtaining the change of the cooling and heating load demand of the target area, includes: Determining the fusion parameters of each regional grid based on the proportional relationship between the population density of each regional grid and the population density of the target area; Based on the fusion parameters of each regional grid, weighted fusion of the changes in the cooling and heating load demands of each regional grid is performed to obtain the changes in the cooling and heating load demands of the target area.

7. The method according to claim 1, characterized in that The method of taking into account regional development and regional electricity consumption characteristics, converting the power demand based on the change of the cooling and heating load demand of the target area, and obtaining the power supply prediction model of the target area includes: Based on the power coefficient of cooling and heating of the target area, convert the power demand of the target area's cooling and heating load demand changes to obtain a cooling and heating load power supply model for the target area; The electricity supply model of the target area is adjusted by taking into account the changes in electricity demand brought about by the regional development of the target area and the characteristics of regional electricity consumption stages, so as to obtain the electricity supply prediction model of the target area.

8. The method according to claim 7, characterized in that The expression of the power supply prediction model of the target area is: P all =P b +P h *ΔH r +P c *ΔC r +D+M2E; Where P all is the predicted value of power supply demand in the target area, P b is the reference power of the target area, P h , P c are the power coefficients for cooling and heating, ΔH r , ΔC r They are the changes in the cold and heat load demands of the target area, respectively, D is the electricity correction corresponding to the regional electricity consumption stage characteristics, M2 is the temperature influencing factor of the economic factors of the target area, and E is the change in electricity demand caused by the regional development of the target area.

9. The method according to any one of claims 1 to 8, characterized in that: After obtaining the power supply prediction model of the target area, the method further includes: Based on the historical electricity supply data of the target area, a genetic algorithm is used to calibrate the model parameters of the electricity supply prediction model of the target area.

10. The method according to claim 9, characterized in that: Based on the historical power supply data of the target area, the model parameters of the power supply prediction model of the target area are calibrated using a genetic algorithm, including: A set of model parameters to be calibrated is taken as population individuals, and the residual standard deviation between the prediction result of the power supply prediction model of the target area and the historical power supply data of the target area is taken as the fitness function. The model parameters are iteratively optimized through individual selection, crossover and mutation operations.

11. The method according to claim 10, characterized in that The model parameters to be calibrated include the temperature correlation coefficient corresponding to each dimension of meteorological data in the conversion relationship between the internal and external temperatures of buildings in each regional grid, the smoothing coefficient, and the temperature thresholds for cooling and heating.

12. A method for predicting regional power demand, characterized in that: include: Obtain the current multi-dimensional meteorological data of each regional grid in the target area; Based on the current multi-dimensional meteorological data of each regional grid, the power supply forecasting model constructed by any one of claims 1-11 is used to perform real-time forecasting of the power supply demand in the target area to obtain the current power supply demand forecast value of the target area.

13. The method according to claim 12, characterized in that After obtaining the current power supply demand forecast value of the target area, the method further includes: The power dispatch of the target area is controlled based on the change of the current power supply demand forecast value of the target area.

14. A system for constructing a regional power supply prediction model, characterized in that: include: A relationship fitting module is used to fit the internal and external temperature conversion relationship of buildings in each regional grid based on the multi-dimensional historical meteorological data of each regional grid in the target area; A load change determination module, for determining a change in cooling and heating load demand of each regional grid based on a temperature difference between an internal temperature of a building in each regional grid and a temperature threshold for cooling and heating; A fusion module is used to fuse the changes in the cooling and heating load demands of each regional grid based on the impact of the population density in the buildings of each regional grid on the temperature, so as to obtain the changes in the cooling and heating load demands of the target area; A model building module is used to consider regional development and regional electricity consumption characteristics, convert electricity demand based on the change of cooling and heating load demand of the target area, and obtain an electricity supply prediction model for the target area; The regional grid is obtained by dividing the target area into spatial grids, and the internal temperature of the building in the regional grid is determined based on the conversion relationship between the internal and external temperatures of the building in the regional grid.

15. The system according to claim 14, characterized in that The system further comprises: The smoothing module is used to smooth the initial internal and external temperature conversion relationship of buildings in each regional grid based on the thermal inertia of the building.

16. The system according to claim 15, characterized in that The smoothing module is specifically used for: For each regional grid, weighted fusion processing is performed on the indoor temperatures of the buildings in the regional grid for several consecutive historical days based on the smoothing processing coefficient of the target area.

17. The system according to claim 15, characterized in that The system further comprises a relationship correction module, wherein the relationship correction module is used to: For each regional grid, determining a user behavior influencing factor of the building in the regional grid based on historical window opening behavior data of users in the building in the regional grid; Collect historical data on the impact of economic development of the target area on regional temperature, and determine the temperature impact factor of the economic factors of the target area; The conversion relationship between the internal and external temperatures of the buildings in the regional grid is corrected based on the user behavior influencing factors of the buildings in the regional grid and the temperature influencing factors of the economic factors of the target area.

18. The system according to claim 17, characterized in that The expression for the conversion relationship between the internal and external temperatures of the buildings in the regional grid is: In the formula, is the internal temperature of the building at the grid cell l on date d, is the internal temperature of the building after smoothing at date d and regional grid l, is the user behavior influencing factor of the building in the regional grid l, M2 is the temperature influencing factor of the economic factor of the target area, τ is the influence coefficient of the user's ventilation and window opening behavior on the internal temperature of the building, is the average daily window opening time of buildings in the regional grid l, is the initial internal temperature of the building at the grid cell l on date d, T 0,l is the external temperature at the grid cell l on date d, S 0,l , W 0,l , H 0,l are the solar radiation intensity, wind speed, and air humidity outside the building at date d and regional grid l; a, b, and c are the correlation coefficients of solar radiation intensity, wind speed, and humidity in the target area, respectively. * , W * , H * , T * They represent the solar radiation intensity, wind speed, and humidity that have no effect on electricity demand, and α is the smoothing coefficient of the target area. is the initial internal temperature of the building at the regional grid l on date d-1, is the initial internal temperature of the building at regional grid l on date d-2.

19. The system according to claim 14, characterized in that The fusion module is specifically used for: Determining the fusion parameters of each regional grid based on the proportional relationship between the population density of each regional grid and the population density of the target area; Based on the fusion parameters of each regional grid, weighted fusion of the changes in the cooling and heating load demands of each regional grid is performed to obtain the changes in the cooling and heating load demands of the target area.

20. The system according to claim 14, characterized in that The model building module is specifically used for: Based on the power coefficient of cooling and heating of the target area, convert the power demand of the target area's cooling and heating load demand changes to obtain a cooling and heating load power supply model for the target area; The electricity supply model of the target area is adjusted by taking into account the changes in electricity demand brought about by the regional development of the target area and the characteristics of regional electricity consumption stages, so as to obtain the electricity supply prediction model of the target area.

21. The system according to claim 20, characterized in that The expression of the power supply prediction model of the target area is: P all =P b +P h *ΔH r +P c *ΔC r +D+M2E; Where P all is the predicted value of power supply demand in the target area, P b is the reference power of the target area, P h , P c are the power coefficients for cooling and heating, ΔH r , ΔC r They are the changes in the cold and heat load demands of the target area, respectively, D is the electricity correction corresponding to the regional electricity consumption stage characteristics, M2 is the temperature influencing factor of the economic factors of the target area, and E is the change in electricity demand caused by the regional development of the target area.

22. The system according to any one of claims 14 to 21, characterized in that: The system further comprises: The model calibration module is used to calibrate the model parameters of the power supply prediction model of the target area by using a genetic algorithm based on the historical power supply data of the target area.

23. The system according to claim 22, characterized in that The model calibration module is specifically used for: A set of model parameters to be calibrated is taken as population individuals, and the residual standard deviation between the prediction result of the power supply prediction model of the target area and the historical power supply data of the target area is taken as the fitness function. The model parameters are iteratively optimized through individual selection, crossover and mutation operations.

24. The system according to claim 23, characterized in that The model parameters to be calibrated include the temperature correlation coefficient and smoothing coefficient corresponding to each dimension of meteorological data in the conversion relationship between the internal and external temperatures of buildings in each regional grid, and the temperature thresholds for cooling and heating.

25. A regional power demand forecasting system, characterized in that: include: A data acquisition module is used to obtain the current multi-dimensional meteorological data of each regional grid in the target area; A prediction module is used to perform real-time prediction of the power supply demand in the target area based on the current multi-dimensional meteorological data of each regional grid and the power supply prediction model constructed according to any one of claims 1 to 11 to obtain the current power supply demand prediction value of the target area.

26. The system according to claim 25, characterized in that The system further comprises: The control scheduling module is used to control the power scheduling of the target area based on the change of the current power supply demand forecast value of the target area.

27. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the method according to any one of claims 1 to 13 is implemented.

28. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, the method according to any one of claims 1 to 13 is implemented.

Images