A heat supply joint control method and system based on heat network system load prediction

Abstract

The application provides a heat supply joint control method and system based on heat network system load prediction, and belongs to the technical field of data processing. The method comprises the following steps: performing grid division on a target heat network system heat supply coverage area according to heat supply shunt to obtain a plurality of heat supply areas; performing grade division according to heat supply demand attribute proportion conditions to obtain a first heat supply area and a second heat supply area; obtaining first demand load information; obtaining a historical load information set of the second heat supply area; performing data analysis on a plurality of historical area load information sets to generate a plurality of target feature sets; obtaining real-time feature values of the second heat supply area; inputting the load prediction model to obtain a plurality of second prediction load demands; and performing heat supply joint control on the target heat network system. The application solves the problems of long joint control period and low intelligent degree when heat supply is insufficient in the prior art, and achieves the technical effect of efficiently performing differentiated heat supply in the process of joint control of heat supply units to ensure heat supply.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, specifically relating to a heating control method and system based on load prediction of a heating network system. Background Technology

[0002] With the rapid development of economy and technology, heating network systems have become an indispensable part of people's lives. Heating through heating network systems has provided great convenience for people's work and life, and plays a very important role in promoting social development.

[0003] Currently, the demand for heating is increasing year by year. The existing heating network system's heating control mode has low efficiency in regulating the heating mode at the heating end. It cannot adjust many parameters that need to be regulated in a timely manner, and cannot meet the demand at the heating end in a timely manner, resulting in poor heating quality. At the same time, in the process of adjusting the operating parameters of the units to ensure heating, the uniform load reduction method results in the inability to meet the demand for full-load operation, thus causing delays in the production cycle.

[0004] Existing technologies suffer from problems such as long joint control cycles and low levels of intelligence when heating is insufficient. Summary of the Invention

[0005] This invention provides a heating control method and system based on heating network system load prediction, which aims to solve the technical problems of long joint control cycle and low level of intelligence in the prior art when heating is insufficient.

[0006] This invention provides a heating control method based on load prediction of a heating network system. The method is applied to an intelligent control system, which is communicatively connected to a data retrieval module. The method includes: dividing the heating coverage area of ​​the target heating network system into grids according to the heating branches to obtain multiple heating areas.

[0007] Obtain the proportion of heating demand attributes in the multiple heating areas, and classify them into first-level heating areas and second-level heating areas based on the proportion of heating demand attributes.

[0008] Obtain the primary demand load information of the primary heating area, wherein the primary heating area is a guaranteed supply area, that is, an area where the demand load must be met;

[0009] The data retrieval module obtains multiple historical load information sets for the secondary heating area, performs feature analysis on the multiple historical load information sets, and generates multiple target feature sets.

[0010] Based on the multiple target feature sets, the real-time feature values ​​of the secondary heating area are obtained, resulting in multiple sets of real-time secondary load feature values.

[0011] The multiple sets of real-time secondary load characteristic values ​​are traversed and input into the load prediction model to obtain multiple secondary predicted load demands;

[0012] The target heating network system is controlled and managed in accordance with the primary demand load information and the multiple secondary predicted load demands.

[0013] By adopting the above technical solutions, intelligent control of heating is achieved based on load forecasting of the heating network system, and the technical effect of ensuring heating requirements is guaranteed by conducting differentiated demand analysis of heating areas.

[0014] Furthermore, the fluctuation of the multiple historical load information sets is analyzed to obtain multiple historical fluctuation information, wherein the multiple historical fluctuation information includes multiple historical fluctuation amplitudes and multiple historical fluctuation correlation characteristics;

[0015] Construct a set of fluctuation level evaluation values;

[0016] Based on the set of fluctuation level evaluation values ​​and the multiple historical fluctuation amplitudes, the feature correlation coefficients are calculated to obtain multiple feature correlation values;

[0017] The association features of the multiple secondary heating areas are filtered according to the multiple feature correlation values ​​to generate the multiple target feature sets.

[0018] By adopting the above technical solutions, the demand load of primary heating areas and secondary heating areas are separated. In case of emergency supply, the load forecast can be adjusted in a timely manner, and the technical effect of more precise control can be achieved during heating joint control.

[0019] Furthermore, an emergency supply guarantee instruction is obtained, and the heating area screening module is invoked according to the emergency supply guarantee instruction;

[0020] The multiple secondary predicted load demands are input into the heating area screening module for load compression to obtain multiple secondary predicted compressed load demands.

[0021] The target heating network system is controlled and managed in conjunction with the primary demand load information and the multiple secondary predicted compressed load demands.

[0022] By adopting the above technical solution, the load of the secondary heating area can be directly reduced during emergency supply, thereby improving the efficiency of heating load adjustment and control.

[0023] Further, based on the primary demand load information and the multiple secondary predicted load demands, the total predicted load demand of the target heating network system is calculated to obtain the total predicted load information;

[0024] Obtain the real-time heat source supply status of the target heating network system and obtain the preset heating mode information, which includes the heating unit operation mode and heating unit operation parameters;

[0025] The total predicted load information is used to constrain the preset heating mode information to determine whether the preset heating mode information can meet the requirements. If it can, heating is provided according to the preset heating mode information; otherwise, the preset heating mode is used as the heating mode information to be optimized.

[0026] By adopting the above technical solutions, heating control can be implemented for the regional heating network supply.

[0027] Furthermore, multiple random adjustment methods are used to adjust the heating mode information to be optimized, and a first neighborhood of the heating mode information to be optimized is constructed. The first neighborhood includes multiple first adjustment heating modes, wherein the multiple random adjustment methods are used to adjust the operating mode of the heating unit and the operating parameters of the heating unit.

[0028] Obtain the fitness of the plurality of first adjusted heating modes, and obtain a plurality of first adjustment fitness;

[0029] The maximum value among the plurality of first adjustment fitness values ​​is obtained as the first fitness value, and the corresponding first adjustment heating mode is used as the first heating mode information.

[0030] By adopting the above technical solution, the heating mode is screened and optimized by using the heating mode corresponding to the maximum fitness value as the optimized heating mode information.

[0031] Furthermore, the random adjustment method for obtaining the first heating mode information is added to the taboo space, and the taboo space includes a taboo iteration number;

[0032] Continue to construct the second neighborhood of the first heating mode information and perform iterative optimization. When the number of iterations reaches the taboo iteration number, the random adjustment method for obtaining the first heating mode information is deleted from the taboo space.

[0033] When the iterative optimization reaches the preset number of iterations, the optimization is stopped, and the heating mode information corresponding to the maximum fitness value in the iterative optimization process is obtained to obtain the optimized heating mode information.

[0034] Heating control is performed based on the optimized heating mode information.

[0035] By adopting the above technical solutions, the goal of determining the optimal heating mode and improving the accuracy of heating control can be achieved.

[0036] Furthermore, the real-time maximum load of the plurality of first adjusted heating modes is calculated to obtain a plurality of real-time maximum load information, wherein the plurality of real-time maximum load information corresponds one-to-one with the plurality of first adjusted heating modes;

[0037] Based on the real-time consumption and price of heating materials for the multiple first adjusted heating modes, multiple heating costs are obtained.

[0038] The multiple real-time maximum load information and the multiple heating costs are weighted according to a preset weight ratio to obtain the multiple first adjustment fitness.

[0039] By adopting the above technical solution, the calculation method for adjusting fitness is determined, quantitative analysis is performed, and accuracy is improved.

[0040] The beneficial effects of this invention are as follows:

[0041] This invention predicts the load of the heating network system and controls the heating supply based on the prediction results. It then controls the heating supply from both the supply and receiving ends, improving both heating efficiency and the quality of heating control. This achieves the technical effect of enhancing the flexibility of heating network system regulation.

[0042] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and the drawings. Attached Figure Description

[0043] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0044] Figure 1 This is a schematic diagram of a heating control method based on load prediction of a heating network system according to an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of the process for generating multiple target feature sets in a heating control method based on heating network system load prediction according to an embodiment of the present invention.

[0046] Figure 3 This is a schematic diagram of the process for obtaining first heating mode information in a heating control method based on heating network system load prediction according to an embodiment of the present invention.

[0047] Figure 4 This is a schematic diagram of a heating control system structure based on heating network system load prediction, according to an embodiment of the present invention.

[0048] Attached reference numerals: Heating area acquisition module 11, grade classification module 12, target feature set first-level demand load acquisition module 13, historical load information acquisition module 14, real-time target feature value acquisition module 15, predicted load demand acquisition module 16, heating joint control module 17. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The same reference numerals in the drawings represent the same components. It should be noted that the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0050] Reference Figure 1 This invention proposes a heating control method based on load forecasting of a heating network system. The method is applied to an intelligent control system, which is communicatively connected to a data retrieval module. The method includes:

[0051] Step S100: Divide the target heating network system's heating coverage area into grids according to the heating branches to obtain multiple heating areas;

[0052] Specifically, the target heating network system refers to any heating network system that is subject to load forecasting and regional heating control. The heating network is a major component of a centralized heating system, responsible for heat energy transmission. The form of a heating network system depends on various factors, including the relative locations of the heat medium, heat source, and heat users, the types of heat users in the heating area, and the magnitude and nature of the heat load. By dividing the heating network coverage area into multiple grids, using the main heating pipelines as the main pipelines, and based on the distribution of the main heating pipelines within the area, the coverage area is divided into multiple heating areas. Each grid serves as a heating area, thus obtaining the multiple heating areas. These multiple heating areas refer to the division units for the heating supply of the target heating network system.

[0053] Step S200: Obtain the proportion of heating demand attributes in the multiple heating areas, and classify them into first-level heating areas and second-level heating areas according to the proportion of heating demand attributes.

[0054] Specifically, extracting the proportion of heating demand attributes across multiple heating zones involves analyzing the proportion of each unit attribute requiring heat supply within that zone. These heating demand attributes describe the intended use of the acquired heat, including commercial, industrial, and residential attributes. The primary heating zone is the area requiring guaranteed supply; due to its importance, a sufficient heat supply must be ensured. The secondary heating zone is the area where load reduction can be implemented when heat supply is insufficient, i.e., the heat supply can be reduced.

[0055] Specifically, by weighting heating demand attributes, the preferred approach is to allocate a higher proportion of residential attributes than industrial attributes, and vice versa. Staff determine the weighting based on the actual development of each region. Based on this weighting, the attribute proportions for each heating area are weighted and calculated to obtain the final result. These results are then sorted from highest to lowest to determine the primary and secondary heating areas.

[0056] Step S300: Obtain the primary demand load information of the primary heating area, wherein the primary heating area is a guaranteed supply area, that is, an area whose demand load must be met;

[0057] Specifically, heating demand is collected based on the distribution of commercial, industrial, and residential areas within the primary heating region to obtain the primary demand load information. This primary demand load information includes total load and load urgency.

[0058] Step S400: Obtain multiple historical load information sets of the secondary heating area through the data retrieval module, perform feature analysis on the multiple historical load information sets, and generate multiple target feature sets;

[0059] Furthermore, refer to Figure 2 In this embodiment, step S400 further includes:

[0060] Step S410: Analyze the fluctuation of the multiple historical load information sets to obtain multiple historical fluctuation information, wherein the multiple historical fluctuation information includes multiple historical fluctuation amplitudes and multiple historical fluctuation correlation characteristics;

[0061] Step S420: Construct a set of fluctuation level evaluation values;

[0062] Step S430: Calculate the feature correlation coefficients based on the set of fluctuation level evaluation values ​​and the multiple historical fluctuation amplitudes to obtain multiple feature correlation values;

[0063] Step S440: Filter the associated features of the multiple heating areas according to the multiple feature correlation values, thereby generating the multiple target feature sets.

[0064] Specifically, the data retrieval module is a functional module that extracts data from the load supply database of the target heating network system. Through this module, historical load information of the secondary heating areas is extracted, resulting in multiple historical load information sets. These sets describe the load received by multiple heating areas over a historical period, including daily, weekly, and monthly supply loads, load fluctuations, etc. Retrieving historical load information from multiple heating areas provides a basis for subsequent analysis and prediction of the real-time load situation in the secondary heating areas. Furthermore, by dividing the coverage area into multiple smaller units, i.e., multiple heating areas, the analysis of load changes within the area becomes more accurate. Separating the demand loads of primary and secondary heating areas allows for timely adjustments to load forecasts in emergency supply situations, enabling more refined control during heating network management and improving the efficiency and quality of heating adjustments.

[0065] Specifically, the multiple target feature sets involve data parsing of multiple historical load information sets to analyze the main characteristics that affect the load of secondary heating areas, including periodic characteristics, quarterly characteristics, and environmental characteristics. The periodic characteristics refer to the characteristics of periodic changes in heating load based on regional distribution differences, such as daily or quarterly periodic fluctuations in heating load. The quarterly characteristics refer to the characteristics of fluctuations in heat load demand due to seasonal differences, such as a decrease in heating load in summer and an increase in heating load in winter. The environmental characteristics refer to the characteristics of load fluctuations caused by environmental factors such as weather changes, such as an increase in heating demand when temperatures suddenly drop.

[0066] Specifically, the multiple historical fluctuation information is obtained by statistically analyzing load fluctuations exceeding a preset fluctuation threshold from multiple historical load information sets. The preset fluctuation threshold is set by staff and represents the normal fluctuation range of the load. The multiple historical fluctuation amplitudes are the deviations of the load fluctuations from the preset fluctuation thresholds. The multiple historical fluctuation correlation features are the influencing characteristics that cause the multiple historical fluctuations, including periodic features, quarterly features, environmental features, etc. The fluctuation level evaluation value set refers to classifying the degree of fluctuation and assigning corresponding evaluation values, with each level corresponding to one evaluation value. Preferably, the fluctuation levels are divided into three levels: Level 1 has a deviation range of 0–30, corresponding to an evaluation value of 9; Level 2 has a deviation range of 30–45, corresponding to an evaluation value of 6; and Level 3 has a deviation range of 45–55, corresponding to an evaluation value of 3. The multiple feature correlation values ​​reflecting the correlation between the multiple historical fluctuation correlation features and the load fluctuations are calculated by multiplying the fluctuation level evaluation value set and the multiple historical fluctuation amplitudes. Since each heating area corresponds to a historical load information set, the most relevant features for load fluctuations in each heating area can be obtained by filtering multiple feature correlation values. Preferably, the top three features are selected as the multiple target feature sets by sorting the multiple feature correlation values ​​from largest to smallest.

[0067] Step S500: Obtain the real-time feature values ​​of the secondary heating area based on the multiple target feature sets to obtain multiple real-time secondary load feature value sets;

[0068] Step S600: Traverse the multiple sets of real-time secondary load characteristic values ​​and input them into the load prediction model to obtain multiple secondary predicted load demands;

[0069] Specifically, by using features from the multiple target feature sets as data extraction indexes, real-time feature values ​​within the secondary heating area are collected to obtain multiple sets of real-time secondary load feature values. These sets reflect the most relevant feature values ​​within the secondary heating area, further reflecting load changes and providing a basis for predicting real-time load demand in each heating area. The load prediction model is a functional model for intelligently predicting heating load demand. Input data consists of multiple sets of real-time secondary load feature values, and output data consists of multiple predicted load demands. These predicted load demands are obtained after predicting the heating load demand within the secondary heating area, including the load demand quantity. By traversing the multiple sets of real-time secondary load feature values ​​and inputting them into the load prediction model, multiple predicted secondary load demands are obtained, intelligently predicting the load demand of each heating area. The load demand of each heating area is predicted separately, reducing prediction errors and improving prediction accuracy through subdivision.

[0070] Specifically, the load prediction model, with a convolutional neural network structure, is trained using multiple sets of secondary load feature values ​​and multiple sets of historical load information as construction data. These sets are used as sample datasets, which are then divided into training and validation sets in a ratio of 2:1. The load prediction model is trained on the training set until convergence. Then, multiple sets of secondary load feature values ​​from the validation set are input into the converged load prediction model to obtain multiple sets of historical load information for validation. These historical load information sets are matched with the multiple sets of historical load information. The number of successful matches is divided by the number of historical load information sets to obtain the validation accuracy. When the validation accuracy meets the requirements, the load prediction model is output. When the validation accuracy does not meet the requirements, more construction data is acquired to incrementally train the load prediction model until the validation accuracy meets the requirements.

[0071] Step S700: Perform heating control on the target heating network system based on the primary demand load information and the multiple secondary predicted load demands.

[0072] Furthermore, step S600 in this embodiment of the application also includes:

[0073] Step S710: Calculate the total predicted load demand of the target heating network system based on the primary demand load information and the multiple secondary predicted load demands to obtain the total predicted load information;

[0074] Step S720: Obtain the real-time heat source supply status of the target heating network system and obtain the preset heating mode information, wherein the preset heating mode information includes the heating unit operation mode and the heating unit operation parameters;

[0075] Step S730: Use the total predicted load information to constrain the preset heating mode information, and determine whether the preset heating mode information can meet the requirements. If yes, then provide heating according to the preset heating mode information. If no, then use the preset heating mode as the heating mode information to be optimized.

[0076] Specifically, the total predicted load information is obtained by adding the primary demand load information and the multiple secondary predicted load demands, reflecting the load that the target heating network system needs to supply, including the total predicted load. The preset heating mode information describes the heat source used by the target heating network system when supplying load in real time, including the heating unit operating mode and operating parameters. The heating unit operating mode describes the composition of the heating units and the operating method of each unit. The heating unit operating parameters are obtained by summarizing the parameters of each piece of equipment during the operation of the heating units.

[0077] Specifically, the total predicted load information is used to constrain the preset heating mode information. That is, the total predicted load information is used as a requirement that the target heating network system must meet. The preset heating mode information is then evaluated to determine whether the current heating mode can meet the region's load demand. If yes, heating is provided according to the preset heating mode information; otherwise, the preset heating mode is designated as the heating mode to be optimized. This allows for heating network supply control within the region.

[0078] Furthermore, refer to Figure 3 In this embodiment, step S730 further includes:

[0079] Step S731: The heating mode information to be optimized is adjusted by using multiple random adjustment methods to construct a first neighborhood of the heating mode information to be optimized. The first neighborhood includes multiple first adjustment heating modes. The multiple random adjustment methods are used to adjust the operating mode of the heating unit and the operating parameters of the heating unit.

[0080] Step S732: Obtain the fitness of the plurality of first adjustment heating modes, and obtain a plurality of first adjustment fitness;

[0081] Step S733: Obtain the maximum value among the plurality of first adjustment fitness values ​​as the first fitness value, and use the corresponding first adjustment heating mode as the first heating mode information.

[0082] Furthermore, step S730 in this embodiment of the application also includes:

[0083] Step S734: Add the random adjustment method for obtaining the first heating mode information to the taboo space, wherein the taboo space includes a taboo iteration number;

[0084] Step S735: Continue to construct the second neighborhood of the first heating mode information and perform iterative optimization. When the number of iterations reaches the taboo iteration number, the random adjustment method for obtaining the first heating mode information is deleted from the taboo space.

[0085] Step S736: When the iterative optimization reaches the preset number of iterations, stop the optimization and obtain the heating mode information corresponding to the maximum fitness value in the iterative optimization process.

[0086] Step S737: Perform heating control based on the optimized heating mode information.

[0087] Specifically, the heating mode information to be optimized is adjusted according to multiple random adjustment methods. Optionally, the operating mode and operating parameters of the heating unit are adjusted according to different adjustment ranges to adjust the heating mode information to be optimized. Based on the heating mode information to be optimized, the first adjusted heating mode obtained after random adjustment is taken as the first neighborhood. The first neighborhood is the adjustment range of the heating mode information to be optimized, including the first adjusted heating mode. Then, the multiple first adjusted heating modes are evaluated to obtain the evaluation results of the first adjusted heating mode on the regional load heating quality, and the fitness is obtained according to the level of heating quality. Thus, by evaluating the fitness of each of the multiple first adjusted heating modes, the multiple first adjustment fitnesss are obtained. From the multiple first adjustment fitnesss, the fitness with the largest value is selected as the first fitness, indicating that heating according to the first adjusted heating mode corresponding to the first fitness can achieve the optimal heating quality from an overall perspective. The first heating mode is the heating mode that is most suitable for the target heating network system in the first neighborhood.

[0088] Specifically, the random adjustment method corresponding to the first heating mode is added to the forbidden space that does not allow selection, to avoid repeatedly selecting this adjustment method and resulting in a lack of diversity in the results. The forbidden iteration number is the number of iterations that the adjustment method is prohibited from using. When the number of iterations exceeds a certain number, the influence of the adjustment method on the adjustment result will also decrease. The forbidden iteration number is set by the staff and is not limited here.

[0089] Specifically, the second neighborhood is the adjustment space corresponding to the first heating mode. Iterative optimization of the first heating mode is performed within the second neighborhood. When the number of iterations reaches the taboo iteration count, the random adjustment method of the first heating mode is removed from the taboo space to ensure the effectiveness and reliability of the iteration. Iteration cannot continue indefinitely to avoid overfitting. When the iterative optimization reaches the preset number of iterations, the heating mode corresponding to the maximum fitness value is used as the optimized heating mode information, thus achieving the goal of determining the optimal heating mode.

[0090] Furthermore, in the step of obtaining the fitness of the plurality of first adjusted heating modes and obtaining a plurality of first adjustment fitness, step S732 of this application embodiment further includes:

[0091] Step S7321: Calculate the real-time maximum load of the plurality of first adjusted heating modes to obtain a plurality of real-time maximum load information, wherein the plurality of real-time maximum load information corresponds one-to-one with the plurality of first adjusted heating modes;

[0092] Step S7322: Based on the real-time heating material consumption and real-time heating material price of the multiple first adjusted heating modes, multiple heating costs are obtained;

[0093] Step S7323: Perform weighted calculations on the multiple real-time maximum load information and the multiple heating costs according to preset weight ratios to obtain the multiple first adjustment fitness.

[0094] Specifically, the real-time maximum load achievable by multiple first-adjustment heating modes is calculated. Based on the unit operating mode and parameters corresponding to each of the multiple first-adjustment heating modes, the load at full load is used as multiple real-time maximum load information. The real-time heating material consumption is the amount of heating material consumed per unit time during the operation of the heating unit. The real-time heating material price is the price of the material during heating. The multiple heating costs are the costs incurred by the target heating network system when providing heating according to the multiple first-adjustment heating modes. The preset weight ratio is a pre-set proportion of load supply and cost in the heating mode evaluation. The multiple real-time maximum load information and the multiple heating costs are weighted according to the preset weight ratio, and the multiple first-adjustment fitness levels are obtained based on the calculation results.

[0095] Furthermore, step S700 in this embodiment of the application also includes:

[0096] Step S740: Obtain an emergency supply guarantee instruction, and retrieve the heating area screening module according to the emergency supply guarantee instruction;

[0097] Step S750: Input the multiple secondary predicted load demands into the heating area screening module for load compression to obtain multiple secondary predicted compressed load demands;

[0098] Step S760: Perform heating control on the target heating network system based on the primary demand load information and the multiple secondary predicted compressed load demands.

[0099] Specifically, the emergency supply guarantee order is issued when heating resources are scarce, ensuring heating supply to designated areas and reducing heating supply to non-designated areas. The heating area screening module analyzes the multiple secondary predicted load demands according to the heating demand attributes within each secondary heating area, and then reduces the load accordingly. By compressing the multiple secondary predicted load demands using the heating area screening module, the multiple secondary predicted compressed load demands are obtained. During emergency supply guarantee, the load in the secondary heating areas can be directly compressed, thereby improving the efficiency of heating load adjustment and control.

[0100] Furthermore, step S760 in this embodiment of the application also includes:

[0101] Step S761: Determine multiple load demands to be supported based on the multiple secondary predicted load demands and the multiple secondary predicted compressed load demands;

[0102] Step S762: Collect the distributed heating source configuration information of the secondary heating area for heating capacity analysis, wherein the distributed heating source configuration information includes heating source type and heating source specifications;

[0103] Step S763: Based on the heat supply analysis results, traverse the multiple load demands to be supported to calculate missing data, and determine whether the calculation result is greater than a preset threshold. If it is greater than the preset threshold, mark the heat supply analysis results positively.

[0104] Step S764: Determine whether the calculation result is less than the preset threshold. If it is less than the preset threshold, mark the heat supply analysis result negatively.

[0105] Step S765: Distributed heating source scheduling is performed using K-clustering, combined with positive and negative labels.

[0106] Specifically, the multiple load demands to be supported are calculated by subtracting the multiple secondary predicted load demands from the multiple secondary predicted compressed load demands. That is, the additional heating demand required to reach full capacity in the secondary heating area. By calculating these multiple load demands, the amount of load compressed in the secondary heating area when the heating network system is insufficient can be determined, and a basis for subsequent distributed heating source allocation can also be provided.

[0107] Specifically, data is collected on the configuration of distributed heating sources within the secondary heating area to obtain information on the type of heating source, such as heat pumps and photovoltaic panels, as well as its specifications, such as rated heating capacity, hot water flow rate, rated outlet water temperature, and maximum outlet water temperature. By obtaining the type and specifications of the heating sources, the available supplementary heating capacity (i.e., supplementary heat load) for the secondary heating area is determined. Then, the heat supply analysis results of the secondary heating area are matched one-to-one with multiple load demands to be supported. That is, the heat supply analysis results are subtracted from the load demands to determine whether using the distributed heating sources within the secondary heating area can meet the demand, and whether the demand exceeds the limit and can be supplemented externally.

[0108] Specifically, the preset threshold is a load range. Within this range, the calculation results for missing data are in a normal fluctuation state. When the calculation result is greater than the preset threshold, it indicates that the heat supply from the distributed heating sources in this area can support other areas, and it is positively marked. When the calculation result is less than the preset threshold, it indicates that the heat supply from the distributed heating sources in this area cannot meet the compressed load demand, and other areas need to provide distributed heating source support, and it is negatively marked.

[0109] Specifically, using the K-clustering method, combined with the positive and negative labels, regions with similar support and demand are matched and clustered. Based on the clustering results, targeted distributed heating source scheduling is performed to meet the heating demand of secondary heating areas. Preferably, K is set to 1, and a label is randomly selected from the positive labels. Using this label as the center, negative labels are clustered. The absolute difference between the negative and positive label data is used as the clustering scale to determine the nearest negative label. This yields the negative label's heating analysis result that best matches the positive label's corresponding heating analysis result, thus enabling distributed heating source scheduling. Targeted scheduling improves the operability and accuracy of the scheduling, and enhances the efficiency of heating system control.

[0110] Reference Figure 4 As shown, this application provides a heating control system based on heating network system load forecasting. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0111] Heating area acquisition module 11 is used to divide the heating coverage area of ​​the target heating network system into grids according to the heating branches to obtain multiple heating areas;

[0112] The grading module 12 is used to obtain the proportion of heating demand attributes of the multiple heating areas, and to grade them according to the proportion of heating demand attributes to obtain first-level heating areas and second-level heating areas.

[0113] The primary demand load acquisition module 13 is used to acquire the primary demand load information of the primary heating area, wherein the primary heating area is a guaranteed supply area, that is, an area where the demand load must be met.

[0114] Historical load information acquisition module 14 is used to acquire multiple historical load information sets of the secondary heating area through the data retrieval module, perform feature analysis on the multiple historical load information sets, and generate multiple target feature sets.

[0115] Real-time target feature value acquisition module 15 is used to acquire real-time feature values ​​of the secondary heating area based on the multiple target feature sets, thereby obtaining multiple real-time secondary load feature value sets.

[0116] The load demand acquisition module 16 is used to traverse the multiple real-time secondary load characteristic value sets input into the load prediction model to obtain multiple secondary predicted load demands.

[0117] The heating control module 17 is used to control the heating of the target heating network system based on the primary demand load information and the multiple secondary predicted load demands.

[0118] Furthermore, the system also includes:

[0119] A historical fluctuation information acquisition unit is used to analyze the fluctuation of the multiple historical load information sets and obtain multiple historical fluctuation information, wherein the multiple historical fluctuation information includes multiple historical fluctuation amplitudes and multiple historical fluctuation correlation features;

[0120] An evaluation value set construction unit is used to construct a set of fluctuation level evaluation values.

[0121] A correlation value acquisition unit is used to calculate the feature correlation coefficient based on the set of fluctuation level evaluation values ​​and the multiple historical fluctuation amplitudes to obtain multiple feature correlation values.

[0122] The associated feature filtering unit is used to filter the associated features of the multiple secondary heating areas according to the multiple feature correlation values, thereby generating the multiple target feature sets.

[0123] Furthermore, the system also includes:

[0124] A supply guarantee instruction acquisition unit is used to acquire an emergency supply guarantee instruction and retrieve a heating area screening module based on the emergency supply guarantee instruction.

[0125] A load compression unit is used to input the multiple secondary predicted load demands into the heating area screening module for load compression, thereby obtaining multiple secondary predicted compressed load demands.

[0126] A heating network control unit is used to control the heating supply of the target heating network system based on the primary demand load information and the multiple secondary predicted compressed load demands.

[0127] Furthermore, the system also includes:

[0128] The total predicted load information acquisition unit is used to calculate the total predicted load demand of the target heating network system based on the primary demand load information and the multiple secondary predicted load demands, and obtain the total predicted load information.

[0129] A preset heating mode information acquisition unit is used to acquire the real-time heat source supply status of the target heating network system and acquire preset heating mode information, wherein the preset heating mode information includes the operating mode of the heating unit and the operating parameters of the heating unit.

[0130] The mode information constraint unit is used to constrain the preset heating mode information using the total predicted load information, and to determine whether the preset heating mode information can meet the requirements. If so, heating is provided according to the preset heating mode information; otherwise, the preset heating mode is used as the heating mode information to be optimized.

[0131] Furthermore, the system also includes:

[0132] The first neighborhood construction unit is used to adjust the heating mode information to be optimized using multiple random adjustment methods to construct a first neighborhood of the heating mode information to be optimized. The first neighborhood includes multiple first adjustment heating modes, wherein the multiple random adjustment methods are to adjust the heating unit operation mode and the heating unit operation parameters.

[0133] The first adjustment fitness acquisition unit is used to acquire the fitness of the plurality of first adjustment heating modes and acquire a plurality of first adjustment fitnesss.

[0134] The first heating mode information setting unit is used to obtain the maximum value among the plurality of first adjustment fitness values ​​as the first fitness value, and to set the corresponding first adjustment heating mode as the first heating mode information.

[0135] Furthermore, the system also includes:

[0136] A taboo space addition unit is used to add the random adjustment method for obtaining the first heating mode information to the taboo space, wherein the taboo space includes a taboo iteration number;

[0137] An iterative optimization unit is used to continue constructing a second neighborhood of the first heating mode information and perform iterative optimization. When the number of iterations reaches the taboo iteration number, the random adjustment method for obtaining the first heating mode information is deleted from the taboo space.

[0138] The optimized heating mode information acquisition unit is used to stop the optimization after the iterative optimization reaches a preset number of iterations, and obtain the optimized heating mode information by taking the heating mode information corresponding to the maximum fitness value in the iterative optimization process.

[0139] An optimized heating control unit is used to perform heating control based on the optimized heating mode information.

[0140] Furthermore, the system also includes:

[0141] A real-time maximum load information acquisition unit is used to calculate the real-time maximum load of the plurality of first adjusted heating modes to obtain a plurality of real-time maximum load information, wherein the plurality of real-time maximum load information corresponds one-to-one with the plurality of first adjusted heating modes;

[0142] A heating cost acquisition unit is used to obtain multiple heating costs based on the real-time consumption of heating materials and the real-time price of heating materials in the multiple first adjusted heating modes.

[0143] Multiple first adjustment fitness acquisition units are used to perform weighted calculations on the multiple real-time maximum load information and the multiple heating costs according to preset weight ratios to obtain the multiple first adjustment fitnesss;

[0144] Furthermore, the system also includes:

[0145] The unit for determining load demand to be supported is used to determine multiple load demands to be supported based on the multiple secondary predicted load demands and the multiple secondary predicted compressed load demands.

[0146] A heat supply analysis unit is used to collect distributed heating source configuration information of the secondary heating area for heat supply analysis, wherein the distributed heating source configuration information includes heating source type and heating source specifications.

[0147] The calculation result judgment unit is used to perform missing data calculation by traversing the multiple load demands to be supported based on the heat supply analysis results, and to determine whether the calculation result is greater than a preset threshold. If it is greater than the preset threshold, the heat supply analysis result is marked positively.

[0148] A negative marking unit is used to determine whether the calculation result is less than a preset threshold. If it is less than the preset threshold, the heat supply analysis result is negatively marked.

[0149] A heating source scheduling unit is used to perform distributed heating source scheduling by using K-clustering combined with positive and negative labels.

[0150] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A heat supply joint control method based on heat network system load prediction, characterized in that, The method is applied to an intelligent control system, which is communicatively connected to a data retrieval module. The method includes: The target heating network system's heating coverage area is divided into multiple heating zones by gridding according to the heating routes. Obtain the proportion of heating demand attributes in the multiple heating areas, and classify them into first-level heating areas and second-level heating areas based on the proportion of heating demand attributes. Obtain the primary demand load information of the primary heating area, wherein the primary heating area is a guaranteed supply area, that is, an area where the demand load must be met; The data retrieval module obtains multiple historical load information sets for the secondary heating area, performs feature analysis on the multiple historical load information sets, and generates multiple target feature sets. Based on the multiple target feature sets, the real-time feature values ​​of the secondary heating area are obtained, resulting in multiple sets of real-time secondary load feature values. The multiple sets of real-time secondary load characteristic values ​​are traversed and input into the load prediction model to obtain multiple secondary predicted load demands; The target heating network system is subjected to heating control based on the primary demand load information and the multiple secondary predicted load demands. The method further includes: Obtain an emergency supply guarantee instruction, and retrieve the heating area screening module according to the emergency supply guarantee instruction; The multiple secondary predicted load demands are input into the heating area screening module for load compression to obtain multiple secondary predicted compressed load demands. The target heating network system is subjected to joint heating control based on the primary demand load information and the multiple secondary predicted compressed load demands. Based on the multiple secondary predicted load demands and the multiple secondary predicted compressed load demands, multiple load demands to be supported are determined; The distributed heating source configuration information of the secondary heating area is collected for heating capacity analysis. The distributed heating source configuration information includes the heating source type and heating source specifications. Then, the heating capacity analysis results of the secondary heating area are matched one-to-one with the demand of multiple loads to be supported. That is, the demand of loads to be supported is subtracted from the heating capacity analysis results to determine whether the demand can be met when using the distributed heating sources in the secondary heating area for heating supplementation, and whether the demand exceeds the demand and can be supported externally. Based on the heat supply analysis results, the missing data is calculated by traversing the multiple load demands to be supported. It is then determined whether the calculation result is greater than a preset threshold. If it is greater than the preset threshold, the heat supply analysis result is marked positively. Determine whether the calculation result is less than a preset threshold. If it is less than the preset threshold, mark the heat supply analysis result negatively. Distributed heating source scheduling is performed using K-clustering combined with positive and negative labels; This involves generating multiple target feature sets, including: Analyze the fluctuations of the multiple historical load information sets to obtain multiple historical fluctuation information, wherein the multiple historical fluctuation information includes multiple historical fluctuation amplitudes and multiple historical fluctuation correlation characteristics; Construct a set of fluctuation level evaluation values; Based on the set of fluctuation level evaluation values ​​and the multiple historical fluctuation amplitudes, the feature correlation coefficients are calculated to obtain multiple feature correlation values; The association features of the multiple secondary heating areas are filtered according to the multiple feature correlation values ​​to generate the multiple target feature sets.

2. The method of claim 1, wherein, include: The total predicted load demand of the target heating network system is calculated based on the primary demand load information and the multiple secondary predicted load demands to obtain the total predicted load information. Obtain the real-time heat source supply status of the target heating network system and obtain the preset heating mode information, which includes the heating unit operation mode and heating unit operation parameters; The total predicted load information is used to constrain the preset heating mode information to determine whether the preset heating mode information can meet the requirements. If it can, heating is provided according to the preset heating mode information; otherwise, the preset heating mode is used as the heating mode information to be optimized.

3. The method of claim 2, wherein, include: Multiple random adjustment methods are used to adjust the heating mode information to be optimized, and a first neighborhood of the heating mode information to be optimized is constructed. The first neighborhood includes multiple first adjustment heating modes. The multiple random adjustment methods are used to adjust the operating mode and operating parameters of the heating unit. Obtain the fitness of the plurality of first adjusted heating modes, and obtain a plurality of first adjustment fitness; The maximum value among the plurality of first adjustment fitness values ​​is obtained as the first fitness value, and the corresponding first adjustment heating mode is used as the first heating mode information.

4. The method of claim 3, wherein, include: The random adjustment method for obtaining the first heating mode information is added to the taboo space, and the taboo space includes a taboo iteration number; Continue to construct the second neighborhood of the first heating mode information and perform iterative optimization. When the number of iterations reaches the taboo iteration number, the random adjustment method for obtaining the first heating mode information is deleted from the taboo space. When the iterative optimization reaches the preset number of iterations, the optimization is stopped, and the heating mode information corresponding to the maximum fitness value in the iterative optimization process is obtained to obtain the optimized heating mode information. Heating control is performed based on the optimized heating mode information.

5. The method of claim 4, wherein, The step of obtaining the fitness of the plurality of first adjusted heating modes, and obtaining the plurality of first adjustment fitness, includes: The real-time maximum load of the plurality of first adjusted heating modes is calculated to obtain a plurality of real-time maximum load information, wherein the plurality of real-time maximum load information corresponds one-to-one with the plurality of first adjusted heating modes; Based on the real-time consumption and price of heating materials for the multiple first adjusted heating modes, multiple heating costs are obtained. The multiple real-time maximum load information and the multiple heating costs are weighted according to a preset weight ratio to obtain the multiple first adjustment fitness.

6. A heat supply joint control system based on heat network system load prediction, characterized in that, The system is used to implement the heating control method based on heating network system load forecasting as described in any one of claims 1-5, including: A heating area acquisition module is used to divide the heating coverage area of ​​the target heating network system into grids according to the heating branches to obtain multiple heating areas; The grading module is used to obtain the proportion of heating demand attributes of the multiple heating areas, and to grade them according to the proportion of heating demand attributes to obtain first-level heating areas and second-level heating areas. A primary demand load acquisition module is used to acquire primary demand load information of the primary heating area, wherein the primary heating area is a guaranteed supply area, that is, an area where the demand load must be met. The historical load information acquisition module is used to acquire multiple historical load information sets of the secondary heating area through the data retrieval module, perform feature analysis on the multiple historical load information sets, and generate multiple target feature sets. A target feature set real-time target feature value acquisition module is used to obtain the real-time feature value of the secondary heating area according to the multiple target feature sets, and obtain multiple real-time secondary load feature value sets; A load demand acquisition module is used to traverse the multiple sets of real-time secondary load characteristic values ​​input into the load prediction model to obtain multiple secondary load demands. A heating control module is used to control the heating of the target heating network system based on the primary demand load information and the multiple secondary predicted load demands.

Images