A method and system for local weather prediction for a heating area

Abstract

The application discloses a kind of heating area local weather forecast method and system, belong to meteorological prediction technical field.The method includes the following steps: determining each heat branch sub-region range and corresponding position information;Obtain wide-range grid weather forecast data, and the ground height and building distribution characteristics corresponding to each grid;Get the initial local weather forecast data of each heat branch sub-region;Get update comprehensive influence weight;Compare the initial local weather forecast data of adjacent heat branch sub-region, obtain the local weather forecast data after verification;Output the local weather forecast data after verification of each heat branch sub-region.The application solves the problem of regular grid weather forecast and actual heating area boundary mispositioning in the prior art by combining the actual contour of heating area with topographic building characteristics for spatial downscaling interpolation and based on end feedback and spatial consistency verification.

Description

Technical Field

[0001] This invention relates to the field of meteorological forecasting technology, and in particular to a method and system for local meteorological forecasting in heating areas. Background Technology

[0002] During centralized heating operation, the heating load fluctuates with changes in outdoor weather conditions. Meteorological forecast data is the core input for heating dispatching decisions and heat load prediction. To achieve a precise match between the heating supply and the actual heat demand of users, heating dispatching typically calculates the setpoints for the water supply temperature and the predicted heat load for each heat exchange station based on the meteorological forecast data for the area where the heat exchange station is located, and adjusts operating parameters such as the opening degree of the primary network valves and the frequency of the circulating pump accordingly.

[0003] Existing local meteorological forecasting methods for heating areas access regular grid meteorological forecast data released by external meteorological sources and distribute the meteorological forecast data to each heat exchange station by matching the grid center point or nearest neighbor. This data serves as the basis for predicting and controlling the heat load of end users within the heating area of ​​each heat exchange station.

[0004] In existing technologies, the actual heating area of ​​a heat exchange station has an irregular shape, and its boundary is spatially misaligned with the boundary of a regular meteorological grid. Furthermore, the assumption of spatial uniformity of meteorological elements within the meteorological grid contradicts the micro-meteorological differences within the heating area caused by building density gradients, topographic elevation differences, and shading relationships. The lack of spatial downscaling of meteorological forecast data in conjunction with the actual boundary of the heating area results in the prediction results being smoothed in the spatial dimension. The end users of multiple heating branches under the jurisdiction of the same heat exchange station actually receive grid average meteorological data rather than local meteorological data reflecting the true meteorological environment of their location. The local heat load response differences between the windward and leeward sides of the end users, and between high-density areas and open areas, are masked. This creates a systematic spatial deviation between the prediction results and the actual meteorological environment at the end users. Consequently, the valve opening and pump frequency setpoints of each heating branch calculated based on the prediction results also show systematic deviations among end users. The imbalance pattern of near-end overheating and far-end undercooling is solidified or even amplified, and the improvement in prediction accuracy cannot be translated into an improvement in heating balance. Summary of the Invention

[0005] This application provides a method and system for local meteorological forecasting in heating areas, which solves the problems in the prior art where grid meteorological forecasts are misaligned with the actual heating area boundaries and lack spatial downscaling, resulting in spatially smoothed forecasts that mask local meteorological differences at the end points and cause hydraulic imbalances. It achieves the generation of refined local meteorological forecasts by combining actual boundaries and microenvironmental characteristics, eliminating spatial biases and improving heating balance.

[0006] This application provides a method for local meteorological forecasting in a heating area, including: extracting the range contours of heat exchange stations and each heating branch based on heating network distribution data, and determining the range of each heating branch sub-region and its corresponding location information;

[0007] Acquire large-scale grid-based meteorological forecast data, as well as the surface elevation and building distribution characteristics corresponding to each grid;

[0008] Based on location information, surface height, and building distribution characteristics, the comprehensive influence weight of each grid on the thermal sub-region is determined, and the initial local meteorological forecast data of each thermal sub-region is obtained.

[0009] The updated comprehensive impact weight is obtained based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each heating branch sub-region.

[0010] By comparing the initial local meteorological forecast data of adjacent thermal sub-regions, when the difference between the two exceeds the range, the one with higher forecast accuracy is used to adjust the one with lower forecast accuracy to obtain the verified local meteorological forecast data.

[0011] Output the local meteorological forecast data after verification for each thermal branch sub-region.

[0012] Further steps to determine the extent and corresponding location information of each thermal branch sub-region include:

[0013] Extract the correspondence between heat exchange stations, pipelines, and end users from the heating network distribution data;

[0014] A closed connection is made along the outer edge of the secondary pipeline to generate an irregular outline of the heating area corresponding to the heat exchange station;

[0015] When a heat exchange station contains multiple thermal branches, the irregular outline of the heating area is divided into multiple irregular outlines of thermal branch sub-areas.

[0016] Each thermal branch sub-region has an irregular outline that corresponds to the physical heating range of a set of end users.

[0017] Obtain the center coordinates, coverage area, and heat exchange station identification mark of the irregular contour of each thermal branch sub-region, and use the center coordinates as location information;

[0018] Location information is determined before the start of each heating calculation cycle and is updated and determined again when the distribution of the heating network changes.

[0019] Furthermore, the steps to obtain large-scale grid-based weather forecast data, as well as the corresponding surface height and building distribution characteristics for each grid, include:

[0020] It receives large-scale grid weather forecast data from external meteorological sources. The grid nodes contain the predicted values ​​and location coordinates of temperature, wind speed, wind direction and humidity for different time periods.

[0021] Acquire spatial data aligned with a large-scale grid location. The spatial data includes ground elevation data, building density data, and occlusion impact values.

[0022] Surface elevation data provides the surface elevation corresponding to the grid nodes;

[0023] Building density data provides the building occupancy rate within the area of ​​a grid node;

[0024] The occlusion impact value is calculated based on the difference in building height and distance between adjacent grids, and is used to represent the occlusion situation between grids;

[0025] The data on ground height, building density, and shading impact remain unchanged throughout the heating calculation cycle, and are updated synchronously when the ground height or building is renovated.

[0026] Furthermore, the steps for determining the comprehensive influence weight of each grid on the thermal branch sub-region include:

[0027] Using location information as the target point, search for grid nodes within a specific range in the surrounding grid nodes to form a source node set;

[0028] Calculate the distance between the source node and the target node and determine the distance weight; the smaller the distance, the larger the distance weight.

[0029] The terrain weight is determined based on the difference in ground elevation between the source node and the target point. The smaller the elevation difference, the greater the terrain weight. When the elevation difference exceeds the set limit, the corresponding source node is removed.

[0030] The building weight is determined based on the difference in building density and occlusion impact values ​​between the source node and the target node; the smaller the difference, the greater the building weight.

[0031] The distance weight, terrain weight, and building weight are multiplied to obtain the comprehensive influence weight, and the comprehensive influence weight of each source node in the source node set is normalized.

[0032] Furthermore, the steps for obtaining initial local meteorological forecast data for each thermal branch sub-region include:

[0033] The normalized comprehensive impact weight is used as the weighting coefficient;

[0034] The meteorological forecast values ​​of all source nodes in the source node set for the same time period are calculated by weighted summation.

[0035] The weighted summation result is used as the local meteorological forecast value for the corresponding thermal sub-region during that time period.

[0036] Weighted summation is performed sequentially on all time periods covered by large-scale grid meteorological forecast data to obtain the local meteorological forecast data sequence corresponding to the sub-region of the thermal branch.

[0037] The weighted summation calculation process is repeated sequentially for all thermal sub-regions to obtain the local meteorological forecast data sequence corresponding to all thermal sub-regions, and the summary result is used as the initial local meteorological forecast data.

[0038] Furthermore, the steps to obtain the updated overall impact weights include:

[0039] Obtain the measured room temperature of end users within the thermal branch sub-region, and calculate the predicted room temperature based on the initial local meteorological forecast data;

[0040] The temperature prediction deviation is obtained by subtracting the measured room temperature from the predicted room temperature.

[0041] When the temperature prediction deviation is greater than the positive setpoint, the weight ratio of the source node with a larger overall influence is increased, and the weight ratio of the source node with a smaller overall influence is decreased.

[0042] When the temperature prediction deviation is less than the negative setpoint, the weight ratio of each source node is adjusted in the opposite direction.

[0043] The adjusted weights are normalized again to obtain the updated comprehensive influence weights.

[0044] Furthermore, the steps to adjust the less accurate predictions using the more accurate predictions to obtain the validated local meteorological forecast data include:

[0045] Identify combinations of adjacent thermal branch sub-regions based on the distribution of the heating network;

[0046] For each pair of adjacent thermal branch sub-region combinations, calculate the absolute value of the difference between the initial local meteorological forecast data of the two corresponding data within the same time period;

[0047] When the absolute value of the difference is not greater than the set physical limit, the original initial local meteorological forecast data shall be retained unchanged;

[0048] When the absolute value of the difference is greater than the set physical limit, the calculation reliability of the two is determined, and the value of the one with higher calculation reliability is used to adjust the value of the one with lower calculation reliability.

[0049] The reliability of the calculation is determined by comprehensively considering the number of source nodes, the average distance between the source nodes and the target point, and the fluctuation range of historical temperature prediction deviations.

[0050] Furthermore, the step of adjusting the less accurate prediction data using the more accurate prediction data to obtain the verified local meteorological forecast data also includes:

[0051] Using the value of the one with higher confidence as a benchmark, half of the absolute value of the difference between the two is added as an adjustment amount to calculate the corrected value of the one with lower confidence.

[0052] Based on the corrected values, re-identify whether there are cases where the absolute value of the difference in all adjacent thermal branch sub-region combinations is greater than the set physical limit;

[0053] If the absolute value of the difference is greater than the set physical limit, the numerical comparison and correction process is repeated.

[0054] The correction process stops and the verified local meteorological forecast data is output when the absolute value of the numerical difference between all adjacent thermal branch sub-region combinations is no greater than the set physical limit, or when the number of repeated corrections reaches the set maximum number of iterations.

[0055] This application provides a local meteorological forecasting system for a heating area, which is used to implement a local meteorological forecasting method for a heating area, including: a location information determination module, a data acquisition module, a comprehensive influence weight determination module, a comprehensive influence weight update module, a local meteorological forecasting data verification module, and a forecasting data output module;

[0056] The location information determination module is used to extract the range contours of heat exchange stations and each heat supply branch based on the distribution data of the heating pipeline network, and to determine the range of each heat supply branch sub-region and its corresponding location information.

[0057] The data acquisition module is used to acquire large-scale grid meteorological forecast data, as well as the surface height and building distribution characteristics corresponding to each grid.

[0058] The comprehensive influence weight determination module is used to determine the comprehensive influence weight of each grid on the thermal branch sub-region based on location information, surface height and building distribution characteristics, and to obtain the initial local meteorological forecast data of each thermal branch sub-region.

[0059] The comprehensive impact weight update module is used to obtain updated comprehensive impact weights based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each thermal branch sub-region.

[0060] The local meteorological forecast data verification module is used to compare the initial local meteorological forecast data of adjacent thermal branch sub-regions. When the difference between the two exceeds the difference range, the one with higher prediction accuracy is used to adjust the one with lower prediction accuracy to obtain the verified local meteorological forecast data.

[0061] The prediction data output module is used to output the verified local meteorological prediction data for each thermal branch sub-region.

[0062] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0063] Considering that the actual heating area of ​​the heat exchange station has an irregular shape and there are local climate differences such as building density and shading, the actual outline and location information of the thermal branch sub-regions are extracted, and the comprehensive influence weight of the grid is determined by combining the surface height and building distribution characteristics. The large-scale grid meteorological forecast data is then finely distributed to each thermal branch sub-region, breaking the spatial misalignment between the regular grid and the actual heating area. The general "grid average meteorology" is transformed into "terminal local meteorology" that reflects local environmental differences, eliminating the smoothing effect of the prediction results in the spatial dimension and reducing the systematic spatial deviation between the prediction data and the actual terminal meteorological environment.

[0064] Furthermore, in the dynamic adjustment process to reduce spatial bias, the comprehensive influence weight of the grid is updated in real time by comparing the deviation between the actual indoor temperature of the end user and the predicted indoor temperature. This achieves a closed-loop match between the prediction results and the actual heat load response characteristics of the end user, enabling the prediction data to adapt to the actual environment of different branches (such as the windward or leeward side, high-density area or open area), further improving the accuracy of local meteorological prediction and the ability to adapt to different scenarios.

[0065] Furthermore, in the multi-branch collaborative process to improve forecast accuracy, by comparing the local meteorological forecast data of adjacent thermal branch sub-regions, when the difference between the two exceeds a reasonable physical range, the one with higher forecast accuracy is used to adjust the one with lower accuracy. This performs spatial consistency constraint verification, avoiding forecast anomalies caused by local data fluctuations or excessive adjustments, and ensuring the continuity and rationality of meteorological forecast distribution throughout the heating area. This provides an accurate and consistent basis for the hydraulic balance regulation of the secondary network, and avoids systematic deviations in the valve opening and pump frequency setpoints of each branch caused by forecast distortion. Attached Figure Description

[0066] Figure 1 A flowchart of a local meteorological forecasting method for a heating area is provided in an embodiment of this application;

[0067] Figure 2 This is a schematic diagram of the structure of a local meteorological forecasting system for a heating area provided in an embodiment of this application. Detailed Implementation

[0068] This application provides a method and system for local meteorological forecasting in heating areas, which solves the problems of misalignment between regular grid meteorological forecasts and actual heating area boundaries, and the smoothing of internal micro-meteorological differences leading to systematic spatial deviations and hydraulic imbalances in the prior art. By combining the actual heating profile with terrain and building features for spatial downscaling interpolation and verifying consistency with adjacent areas based on end-point feedback, it achieves the generation of refined forecasts that reflect the real local environment at the end, thereby eliminating spatial deviations and improving heating balance.

[0069] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0070] like Figure 1 The diagram shown is a flowchart of a local meteorological forecasting method for a heating area provided in an embodiment of this application. The method is applied to a local meteorological forecasting system for a heating area and includes the following steps: extracting the range contours of heat exchange stations and each heating branch according to the distribution of the heating pipeline network, and determining the range of each heating branch sub-region and its corresponding location information.

[0071] Acquire large-scale grid-based meteorological forecast data, as well as the surface elevation and building distribution characteristics corresponding to each grid;

[0072] Based on location information, surface height, and building distribution characteristics, the comprehensive influence weight of each grid on the thermal sub-region is determined, and the grid meteorological forecast data is weighted to obtain the initial local meteorological forecast data for each thermal sub-region.

[0073] Based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each heating branch sub-region, the corresponding comprehensive impact weights are adjusted to obtain the updated comprehensive impact weights.

[0074] By comparing the initial local meteorological forecast data of adjacent thermal sub-regions, when the difference between the two exceeds the range, the one with higher forecast accuracy is used to adjust the one with lower forecast accuracy to obtain the verified local meteorological forecast data.

[0075] Output the local meteorological forecast data after verification for each thermal branch sub-region.

[0076] Further steps to determine the extent and corresponding location information of each thermal branch sub-region include:

[0077] Extract the correspondence between heat exchange stations, pipelines, and end users from the distribution data of the heating network. Read the connection correspondence between heat exchange stations, secondary heating pipelines (i.e., pipelines from heat exchange stations to residential buildings), and end users (specific buildings) from the geographic information database (such as GIS system or CAD pipeline drawings) of the heating company.

[0078] The outer edge of the secondary pipeline is closed and connected to generate an irregular outline of the heating area corresponding to the heat exchange station. On the digital map, the outer edge of all the secondary pipelines under a certain heat exchange station is closed by connecting them according to a preset buffer distance (e.g., extending outward by 10 meters), thereby automatically generating an irregular polygonal outline representing the actual heating area of ​​the heat exchange station.

[0079] When a heat exchange station contains multiple thermal branches, the irregular outline of the heating area is divided into multiple thermal branch sub-area irregular outlines. Since a heat exchange station usually branches out into several main pipelines, i.e. multiple thermal branches, in order to make more accurate predictions, when a heat exchange station contains multiple thermal branches, the large irregular outline is divided into multiple smaller "thermal branch sub-area irregular outlines" according to the direction of the pipeline.

[0080] Each irregular outline of a thermal branch sub-region corresponds to a physical heating range of a group of end users. Each such sub-region corresponds to a group of end residential buildings heated by the same main pipeline.

[0081] Obtain the center coordinates, coverage area, and heat exchange station identification mark of the irregular contour of each thermal branch sub-region. Using the center coordinates as location information, and through geometric calculation methods such as the polygon centroid algorithm, obtain the center coordinates (latitude and longitude) of the irregular contour of each thermal branch sub-region, and calculate its coverage area. Save the center coordinates as the location information of the sub-region, and attach the heat exchange station identification mark, such as the heat exchange station ID.

[0082] Location information is determined before the start of each heating calculation cycle and updated when the distribution of the heating network changes. The contour extraction and center point calculation in this step are only performed once per heating calculation cycle (such as before the start of each heating season). This location information is only updated when the distribution of the heating network changes due to the renovation of old residential areas or the connection of new buildings, thus saving the amount of system calculations during daily operation.

[0083] Furthermore, the steps to obtain large-scale grid-based weather forecast data, as well as the corresponding surface height and building distribution characteristics for each grid, include:

[0084] It receives large-scale gridded weather forecast data from external meteorological sources. Each grid node contains predicted values ​​and coordinates of temperature, wind speed, wind direction, and humidity for different time periods. Through a network interface, it periodically receives large-scale gridded weather forecast data from external professional meteorological sources, such as the National Meteorological Administration's publicly available interface. This data typically divides the city into square grids (e.g., 3 km by 3 km grids), and each grid node contains predicted values ​​of temperature, wind speed, wind direction, and humidity for various time periods within the next 24 hours, as well as the coordinates of the grid center.

[0085] Acquire spatial data aligned with a large-scale grid location. The spatial data includes ground elevation data, building density data, and occlusion impact values. Simultaneously, static spatial data that is perfectly aligned with the aforementioned grid in location needs to be acquired.

[0086] The surface elevation data provides the surface elevation corresponding to each grid node, extracted from a publicly available digital elevation model (DEM), providing the surface elevation corresponding to each grid node;

[0087] Building density data provides the building occupancy ratio within the grid node's area. By reading urban building planning drawings, the total area occupied by all buildings within the grid is calculated and divided by the total area of ​​the grid, resulting in a decimal between 0 and 1, representing the building occupancy ratio.

[0088] The shading effect value is calculated based on the building height difference and distance between adjacent grids. It represents the shading situation between grids and is pre-calculated by dividing the average building height difference between two adjacent grids by their horizontal distance. The larger this value, the more severely the wind is blocked by tall buildings when it blows through these two grids.

[0089] The data on ground height, building density, and shading effects remain unchanged throughout the heating calculation cycle and are updated synchronously when ground height or buildings are updated. Considering that urban topography and buildings typically change only every few years, the above spatial data is stored in a local database and remains unchanged throughout the entire heating calculation cycle (single heating season). It is only updated synchronously when heating companies manually update urban building data, which avoids repeated and complex queries during each forecast.

[0090] Furthermore, the steps for determining the comprehensive influence weight of each grid on the thermal branch sub-region include:

[0091] Using location information as the target point, the system searches for grid nodes within a specific surrounding range to form a source node set. The goal is to calculate which surrounding meteorological grids (source nodes) will have the greatest impact on the weather of a specific cell (target point). Alternatively, using the obtained cell center location information as the target point, the system searches for all grid nodes within a set distance (e.g., 10 kilometers) of the target point and forms a "source node set."

[0092] Calculate the distance between the source node and the target node. And determine the distance weight; the smaller the distance, the larger the distance weight. The specific formula is as follows: ,in To prevent fixed minimum constants with denominators of zero;

[0093] Terrain weight is determined based on the difference in surface elevation between the source node and the target point. The smaller the elevation difference, the greater the terrain weight. When the elevation difference exceeds a set limit, the corresponding source node is removed. relative to the surface elevation of the target point The difference is determined. The smaller the height difference, the greater the terrain weight.

[0094] The formula is: .in This is the set height difference limit, such as 500 meters. It should be noted that when calculating... Greater than or equal to the set limit If the terrain difference between the two locations is too great, for example, one is on a mountaintop and the other is at the bottom of a valley, then the source node is directly removed from the source node set and will not participate in subsequent calculations.

[0095] Building weights are determined based on the differences in building density and occlusion impact values ​​between the source node and the target node; the smaller the difference, the larger the building weight. Building weights are also determined by comparing the differences in building density and occlusion impact values ​​between the source node and the corresponding values ​​in the target node's region. Smaller differences result in larger weights. A simple inverse proportional calculation is used. ,in and These are the building density values ​​for the source node and the target node, respectively.

[0096] The overall influence weight is obtained by multiplying the distance weight, terrain weight, and building weight. The comprehensive influence weights of each source node in the source node set are normalized. To ensure the correctness of the subsequent weighted summation data logic, the comprehensive influence weights of all source nodes in the set are normalized by dividing the weight of each grid by the sum of the weights of all grids, so that the sum of the normalized weights of all source nodes equals 1. This is denoted as . .

[0097] Furthermore, the steps for obtaining initial local meteorological forecast data for each thermal branch sub-region include:

[0098] After completing the weight calculation, the normalized comprehensive influence weight is used. As a weighting coefficient;

[0099] For all source nodes in the source node set, the meteorological forecast values ​​(such as temperature) for the same time period. The weighted summation is calculated using the following formula: ;

[0100] The weighted summation result is used as the local meteorological forecast value for the corresponding thermal sub-region during that time period;

[0101] Weighted summation is performed sequentially on all time periods (such as the next 24 hourly nodes) covered by the large-scale grid meteorological forecast data to obtain the local meteorological forecast data sequence corresponding to the sub-region of the thermal branch;

[0102] The weighted summation calculation process is repeated for all thermal sub-regions to obtain the local meteorological forecast data sequence corresponding to all thermal sub-regions, and the summary result is used as the initial local meteorological forecast data.

[0103] Furthermore, the steps to obtain the updated overall impact weights include:

[0104] The system acquires the measured room temperature of end users within the heating sub-region and calculates the predicted room temperature based on initial local meteorological forecast data. To prevent long-term errors, a temperature feedback mechanism based on actual measurements is introduced. The measured room temperature of end users within the heating sub-region is acquired through IoT temperature sensors (such as smart room temperature controllers) in residents' homes. Simultaneously, based on the initial local meteorological forecast data and using the heating company's existing conventional heat load forecast formula, a theoretical predicted room temperature is calculated.

[0105] The temperature prediction deviation is obtained by subtracting the measured room temperature from the predicted room temperature.

[0106] When the temperature prediction deviation exceeds the positive setpoint, the weight of source nodes with a larger overall impact is increased, while the weight of source nodes with a smaller overall impact is decreased. That is, when the temperature prediction deviation exceeds the positive setpoint (e.g., exceeds...), the weight of the source nodes with a smaller overall impact is increased. If the predicted temperature is too low (i.e., the actual environment of the community is warmer than predicted, possibly due to its sheltered and sunny location), then in the set of source nodes participating in the community calculation, identify those source nodes that originally predicted higher temperatures and already had a larger overall influence weight. Forcibly increase their weight percentage by a fixed step size (e.g., increase by 5%), while simultaneously decreasing the weight percentage of source nodes that predicted lower temperatures.

[0107] When the temperature prediction deviation is less than the negative setpoint, the weight ratio of each source node is adjusted in the reverse direction. Similarly, when the temperature prediction deviation is less than the negative setpoint (e.g., less than...), the weight ratio of each source node is adjusted in the reverse direction. If the predicted temperature is too high (the actual temperature in the community is colder), then the opposite operation is performed, increasing the weight of nodes with lower predicted temperatures.

[0108] The adjusted weights are normalized again to obtain the updated comprehensive impact weights, which are then used as weighting coefficients in the next calculation cycle. After the weight adjustments are complete, the weights of all source nodes are normalized again to obtain the "updated comprehensive impact weights." These updated weights are saved and used as new weighting coefficients for rolling updates. This design allows the algorithm to have "self-correcting memory" like a human; the longer it runs, the closer the weight allocation becomes to the actual climate of the neighborhood.

[0109] Furthermore, the steps to adjust the less accurate predictions using the more accurate predictions to obtain the validated local meteorological forecast data include:

[0110] Identify adjacent heating sub-region combinations based on the heating network distribution. For example, two adjacent communities separated only by a road are unlikely to experience drastic temperature and wind speed fluctuations. This step eliminates such unreasonable data jumps. Based on the heating network distribution information, identify all geographically adjacent and contiguous "adjacent heating sub-region combinations" (e.g., community A and community B);

[0111] For each pair of adjacent thermal sub-region combinations, calculate the absolute value of the difference between the initial local meteorological forecast data of the two corresponding regions within the same time period. ;

[0112] If the absolute value of the difference is not greater than the set physical limit, the original initial local meteorological forecast data is retained unchanged. Not greater than the set physical limit (e.g., set to) If the limit is manually entered by the enterprise based on historical meteorological experience, it indicates that the data is reasonable, and the original forecast data should be retained unchanged.

[0113] When the absolute value of the difference exceeds the set physical limit, the calculation reliability of both values ​​is determined, and the value of the value with the higher calculation reliability is used to adjust the value of the value with the lower calculation reliability. If the result exceeds this limit, it indicates that at least one of the calculations is incorrect. At this point, the "calculation reliability" of cell A and cell B should be calculated separately.

[0114] Calculate credibility The evaluation formula is: .in This represents the number of source nodes in the source node set of this cell. The average distance between the source node and the target node. This represents the fluctuation range (variance) of historical temperature prediction bias. The larger, The smaller the value, the lower the historical deviation fluctuation. The smaller the value, the higher the credibility. The higher the confidence level, the better. Compare the two sides to identify the side with higher confidence (let's say A) and the side with lower confidence (let's say B).

[0115] The reliability of the calculation is determined by comprehensively considering the number of source nodes, the average distance between the source nodes and the target point, and the fluctuation range of historical temperature prediction deviations.

[0116] Furthermore, the step of adjusting the less accurate prediction data using the more accurate prediction data to obtain the verified local meteorological forecast data also includes:

[0117] Using the value of the side with higher confidence as a baseline, half the absolute value of the difference between the two is added as an adjustment amount to calculate the corrected value for the side with lower confidence. Then, using the value of A as a baseline, half the absolute value of the difference between the two (i.e., As an adjustment, B is corrected: if originally Then the corrected Conversely, add half the difference.

[0118] Based on the corrected values, re-identify whether there are any cases where the absolute value of the difference in all adjacent thermal branch sub-region combinations is greater than the set physical limit. After the correction is completed, based on these corrected values, compare the differences in all adjacent communities again to see if there are any cases where the limit is exceeded.

[0119] If the absolute value of the difference is greater than the set physical limit, the numerical comparison and correction process is repeated.

[0120] The correction process stops and the verified local meteorological forecast data is output when the absolute value of the numerical difference between all adjacent thermal sub-region combinations is no greater than the set physical limit, or when the number of repeated corrections reaches the set maximum number of iterations. To prevent the system from entering an infinite loop, a maximum number of iterations (e.g., 5 times) is set. When the numerical difference between all adjacent combinations is less than the limit, or when the number of repetitions reaches 5 times, the correction process is forcibly stopped, and this batch of final verified local meteorological forecast data is output.

[0121] like Figure 2 The diagram shown is a structural schematic of a local meteorological forecasting system for a heating area provided in an embodiment of this application. The local meteorological forecasting system for a heating area provided in an embodiment of this application includes: a location information determination module, a data acquisition module, a comprehensive influence weight determination module, a comprehensive influence weight update module, a local meteorological forecasting data verification module, and a forecasting data output module.

[0122] The location information determination module is used to extract the range contours of heat exchange stations and each heat supply branch based on the distribution data of the heating pipeline network, and to determine the range of each heat supply branch sub-region and its corresponding location information.

[0123] The data acquisition module is used to acquire large-scale grid meteorological forecast data, as well as the surface height and building distribution characteristics corresponding to each grid.

[0124] The comprehensive influence weight determination module is used to determine the comprehensive influence weight of each grid on the thermal branch sub-region based on location information, surface height and building distribution characteristics, and to obtain the initial local meteorological forecast data of each thermal branch sub-region.

[0125] The comprehensive impact weight update module is used to obtain updated comprehensive impact weights based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each thermal branch sub-region.

[0126] The local meteorological forecast data verification module is used to compare the initial local meteorological forecast data of adjacent thermal branch sub-regions. When the difference between the two exceeds the difference range, the one with higher prediction accuracy is used to adjust the one with lower prediction accuracy to obtain the verified local meteorological forecast data.

[0127] The prediction data output module is used to output the verified local meteorological prediction data for each thermal branch sub-region.

[0128] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0129] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0130] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0131] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0132] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0133] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for local meteorological forecasting in a heating area, characterized in that, Includes the following steps: Extract the range and outline of heat exchange stations and each heating branch based on the distribution data of the heating network, and determine the range and corresponding location information of each heating branch sub-region; Acquire large-scale grid-based meteorological forecast data, as well as the surface elevation and building distribution characteristics corresponding to each grid; Based on location information, surface height, and building distribution characteristics, the comprehensive influence weight of each grid on the thermal sub-region is determined, and the initial local meteorological forecast data of each thermal sub-region is obtained. The updated comprehensive impact weight is obtained based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each heating branch sub-region. By comparing the initial local meteorological forecast data of adjacent thermal sub-regions, when the difference between the two exceeds the range, the one with higher forecast accuracy is used to adjust the one with lower forecast accuracy to obtain the verified local meteorological forecast data. Output the local meteorological forecast data after verification for each thermal branch sub-region.

2. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps for determining the extent and corresponding location information of each thermal branch sub-region include: Extract the correspondence between heat exchange stations, pipelines, and end users from the heating network distribution data; A closed connection is made along the outer edge of the secondary pipeline to generate an irregular outline of the heating area corresponding to the heat exchange station; When a heat exchange station contains multiple thermal branches, the irregular outline of the heating area is divided into multiple irregular outlines of thermal branch sub-areas. Each thermal branch sub-region has an irregular outline that corresponds to the physical heating range of a set of end users. Obtain the center coordinates, coverage area, and heat exchange station identification mark of the irregular contour of each thermal branch sub-region, and use the center coordinates as location information; Location information is determined before the start of each heating calculation cycle and is updated and determined again when the distribution of the heating network changes.

3. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps to obtain large-scale grid-based weather forecast data, as well as the surface height and building distribution characteristics corresponding to each grid, include: It receives large-scale grid weather forecast data from external meteorological sources. The grid nodes contain the predicted values ​​and location coordinates of temperature, wind speed, wind direction and humidity for different time periods. Acquire spatial data aligned with a large-scale grid location. The spatial data includes ground elevation data, building density data, and occlusion impact values. Surface elevation data provides the surface elevation corresponding to the grid nodes; Building density data provides the building occupancy rate within the area of ​​a grid node; The occlusion impact value is calculated based on the difference in building height and distance between adjacent grids, and is used to represent the occlusion situation between grids; The data on ground height, building density, and shading impact remain unchanged throughout the heating calculation cycle, and are updated synchronously when the ground height or building is renovated.

4. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps to determine the combined influence weight of each grid on the thermal branch subregion include: Using location information as the target point, search for grid nodes within a specific range in the surrounding grid nodes to form a source node set; Calculate the distance between the source node and the target node and determine the distance weight; the smaller the distance, the larger the distance weight. The terrain weight is determined based on the difference in ground elevation between the source node and the target point. The smaller the elevation difference, the greater the terrain weight. When the elevation difference exceeds the set limit, the corresponding source node is removed. The building weight is determined based on the difference in building density and occlusion impact values ​​between the source node and the target node; the smaller the difference, the greater the building weight. The distance weight, terrain weight, and building weight are multiplied to obtain the comprehensive influence weight, and the comprehensive influence weight of each source node in the source node set is normalized.

5. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps to obtain initial local meteorological forecast data for each thermal sub-region include: The normalized comprehensive impact weight is used as the weighting coefficient; The meteorological forecast values ​​of all source nodes in the source node set for the same time period are calculated by weighted summation. The weighted summation result is used as the local meteorological forecast value for the corresponding thermal sub-region during that time period. Weighted summation is performed sequentially on all time periods covered by large-scale grid meteorological forecast data to obtain the local meteorological forecast data sequence corresponding to the sub-region of the thermal branch. The weighted summation calculation process is repeated sequentially for all thermal sub-regions to obtain the local meteorological forecast data sequence corresponding to all thermal sub-regions, and the summary result is used as the initial local meteorological forecast data.

6. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps to obtain the updated overall impact weights include: Obtain the measured room temperature of end users within the thermal branch sub-region, and calculate the predicted room temperature based on the initial local meteorological forecast data; The temperature prediction deviation is obtained by subtracting the measured room temperature from the predicted room temperature. When the temperature prediction deviation is greater than the positive setpoint, the weight of the source node with a larger overall impact is increased, and the weight of the source node with a smaller overall impact is decreased. When the temperature prediction deviation is less than the negative setpoint, the weight ratio of each source node is adjusted in the opposite direction. The adjusted weights are normalized again to obtain the updated comprehensive influence weights.

7. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps for adjusting the less accurate forecast data using the more accurate forecast data to obtain the validated local meteorological forecast data include: Identify combinations of adjacent thermal branch sub-regions based on the distribution of the heating network; For each pair of adjacent thermal branch sub-region combinations, calculate the absolute value of the difference between the initial local meteorological forecast data of the two corresponding data within the same time period; When the absolute value of the difference is not greater than the set physical limit, the original initial local meteorological forecast data shall be retained unchanged; When the absolute value of the difference is greater than the set physical limit, the calculation reliability of the two is determined, and the value of the one with higher calculation reliability is used to adjust the value of the one with lower calculation reliability. The reliability of the calculation is determined by comprehensively considering the number of source nodes, the average distance between the source nodes and the target point, and the fluctuation range of historical temperature prediction deviations.

8. The method for local meteorological forecasting in a heating area as described in claim 1, characterized in that, The steps of adjusting the less accurate forecast data using the more accurate forecast data to obtain the verified local meteorological forecast data also include: Using the value of the one with higher confidence as a benchmark, half of the absolute value of the difference between the two is added as an adjustment amount to calculate the corrected value of the one with lower confidence. Based on the corrected values, re-identify whether there are cases where the absolute value of the difference in all adjacent thermal branch sub-region combinations is greater than the set physical limit; If the absolute value of the difference is greater than the set physical limit, the numerical comparison and correction process is repeated. The correction process stops and the verified local meteorological forecast data is output when the absolute value of the numerical difference between all adjacent thermal branch sub-region combinations is no greater than the set physical limit, or when the number of repeated corrections reaches the set maximum number of iterations.

9. A local meteorological forecasting system for a heating area, used to implement the local meteorological forecasting method for a heating area as described in any one of claims 1-8, characterized in that, include: The system includes a location information determination module, a data acquisition module, a comprehensive impact weight determination module, a comprehensive impact weight update module, a local meteorological forecast data verification module, and a forecast data output module. The location information determination module is used to extract the range contours of heat exchange stations and each heat supply branch based on the distribution data of the heating pipeline network, and to determine the range of each heat supply branch sub-region and its corresponding location information. The data acquisition module is used to acquire large-scale grid meteorological forecast data, as well as the surface height and building distribution characteristics corresponding to each grid. The comprehensive influence weight determination module is used to determine the comprehensive influence weight of each grid on the thermal branch sub-region based on location information, surface height and building distribution characteristics, and to obtain the initial local meteorological forecast data of each thermal branch sub-region. The comprehensive impact weight update module is used to obtain updated comprehensive impact weights based on the deviation between the actual indoor temperature and the predicted indoor temperature of the end users in each thermal branch sub-region. The local meteorological forecast data verification module is used to compare the initial local meteorological forecast data of adjacent thermal branch sub-regions. When the difference between the two exceeds the difference range, the one with higher prediction accuracy is used to adjust the one with lower prediction accuracy to obtain the verified local meteorological forecast data. The prediction data output module is used to output the local meteorological prediction data after verification for each thermal branch sub-region.

Images