Multi-strategy cooperative heat supply balance control method based on virtual heat supply network grouping

Abstract

The invention discloses a multi-strategy collaborative heat supply balance control method based on virtual heat supply network grouping, which relates to the technical field of virtual heat supply networks and comprises the following steps: dividing a heat supply network into a plurality of virtual heat supply network groups according to the similarity of a first parameter set in the heat supply network; estimating the thermal load demand and the heat supply capacity of each virtual heat supply network group through a dynamic thermodynamic model and a thermal load prediction model; calculating a heat load characteristic parameter set based on the heat load demand and the heat supply capacity, and selecting a heat supply strategy of each virtual heat supply network group; and performing coordination control on the heat supply network based on the heat supply strategy. Through real-time monitoring and a dynamic grouping mechanism, abnormal nodes can be quickly identified and adjusted, and the flexibility and adaptability of a heat supply network are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual heating networks, and in particular to a multi-strategy collaborative heating balance control method based on virtual heating network grouping. Background Art

[0002] With the acceleration of urbanization and the increase in the number of buildings, the scale of the heating network has gradually expanded, and its operation and management are facing many challenges. Traditional heating systems usually adopt a centralized heating mode to transfer heat energy to each user through a trunk pipeline network. However, as the scale of the network expands, the operational complexity of the heating system has increased significantly, resulting in frequent problems such as uneven heat distribution, increased heat loss, and reduced system energy efficiency.

[0003] In order to deal with these problems, a variety of heating network optimization control strategies have been proposed in recent years. Common methods include closed-loop control based on real-time data monitoring, zoned heating strategies, and intelligent load forecasting and scheduling methods. Although these methods have improved the system's responsiveness and energy efficiency to a certain extent, they still have some shortcomings in practical applications. Summary of the invention

[0004] In view of the shortcomings of existing heating network control technology in refined management and multi-strategy coordinated regulation, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to achieve dynamic balance and efficient operation of the heating network in a large-scale complex heating network through refined virtual heating network grouping and multi-strategy collaborative control, so as to ensure the heating quality of each area and the overall energy efficiency of the system.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a multi-strategy collaborative heating balance control method based on virtual heating network grouping, which includes dividing the heating network into multiple virtual heating network groups according to the similarity of a first parameter set in the heating network; estimating the heat load demand and heating capacity of each virtual heating network group through a dynamic thermodynamic model and a heat load prediction model; calculating the heat load characteristic parameter set based on the heat load demand and heating capacity, and selecting a heating strategy for each virtual heating network group; and coordinating the heating network based on the heating strategy.

[0008] As a preferred scheme of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: dividing the heating network into multiple virtual heating network groups according to the similarity of the first parameter set in the heating network includes the following steps: collecting the first parameter set in the heating network through temperature sensors, flow meters and pressure sensors installed at the heating network nodes of the heating network; the first parameter set includes water supply temperature, return water temperature, flow rate and pressure; transmitting the first parameter set to the central controller, and preprocessing the first parameter set through the central controller; using the preprocessed first parameter set as a feature vector, clustering analysis is performed on the heating network nodes through the improved DBSCAN clustering algorithm to obtain the virtual heating network group division result; establishing a dynamic adjustment mechanism for the virtual heating network group, and dynamically adjusting the virtual heating network group division result.

[0009] As a preferred solution of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: clustering analysis is performed on the heating network nodes by an improved DBSCAN clustering algorithm to obtain the virtual heating network group division result, including the following steps: by constructing an adaptive K-nearest neighbor graph, the number of neighbors of each heating network node is dynamically adjusted to improve the local density judgment ability of the DBSCAN clustering algorithm; the DBSCAN clustering algorithm is optimized by using a particle swarm optimization algorithm to automatically search for the optimal neighborhood radius and the minimum number of samples of the DBSCAN clustering algorithm; the heating network nodes are clustered using the optimized DBSCAN clustering algorithm to obtain the initial virtual heating network group division result; the contour coefficient and Calinski-Harabasz index of the initial virtual heating network group division result are calculated to comprehensively evaluate the clustering quality; the fuzzy C-means clustering method is applied to the boundary nodes of each virtual heating network group to calculate the membership value of each boundary node belonging to different virtual heating network groups; according to the membership and physical location information of the boundary node, the virtual heating network group to which the boundary node belongs is determined to complete the virtual heating network group division of the heating network.

[0010] As a preferred solution of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: the dynamic adjustment mechanism of the virtual heating network group includes calculating the relative deviation of the first parameter set of each heating network node, and setting the observation time window and the deviation threshold of each parameter in the first parameter set; within the observation time window, the system periodically collects the first parameter set of the heating network node, and when the relative deviation of any parameter in the first parameter set of the heating network node exceeds the preset number of times and is greater than the corresponding deviation threshold, the heating network node is marked as an abnormal node; the status of the adjacent nodes of the abnormal node is checked, and if the adjacent nodes are normal, the abnormal node is removed from its current virtual heating network group and marked as a node to be allocated; a reallocation buffer is established, and the node to be allocated is added to the reallocation buffer. When the number of nodes to be allocated in the reallocation buffer reaches the cumulative abnormal node number threshold, or the time from the last reallocation execution reaches the time interval threshold, the reallocation process is triggered and the virtual heating network group information is updated; an adjustment frequency control mechanism is established to record the time of each reallocation and the number of nodes to be allocated involved. If the number of reallocations exceeds the maximum allowed number within the evaluation period, the parameter deviation threshold is increased or the observation time window is extended.

[0011] As a preferred scheme of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: estimating the heat load demand and heating capacity of each virtual heating network group through a dynamic thermodynamic model and a heat load prediction model includes the following steps: establishing a simplified topological structure diagram of each virtual heating network group according to the virtual heating network group division result; establishing a dynamic thermodynamic model based on the simplified topological structure; collecting and preprocessing the historical operation data of each virtual heating network group, the historical operation data including water supply temperature, return water temperature, flow rate, outdoor temperature, and user heat consumption; based on the historical operation data, constructing a heat load prediction model for each virtual heating network group using a multivariate linear regression algorithm; combining the dynamic thermodynamic model and the heat load prediction model to estimate the heat load demand and heating capacity of each virtual heating network group.

[0012] As a preferred solution of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: using a multivariate linear regression algorithm to construct a heat load prediction model for each virtual heating network group includes the following steps: performing time series decomposition on the pre-processed historical operation data, and dividing the historical operation data according to multiple preset time scales according to the time series decomposition results; taking the key factors affecting the heat load as independent variables and the actual heat load data as dependent variables to construct a data set; based on the data set, applying a multivariate linear regression algorithm to calculate the regression coefficient and the intercept term to obtain an initial heat load prediction model; evaluating and optimizing the initial heat load prediction model, including calculating and testing the model residuals, and recording the test results; using the K-fold cross-validation method to evaluate the prediction performance of the initial heat load prediction model, and calculating the preset evaluation indicators; based on the evaluation results, executing a stepwise regression process, removing the independent variables whose P values ​​in the regression coefficients are greater than the preset threshold in turn, and outputting the heat load prediction model of the current virtual heating network group; for each virtual heating network group, repeating the heat load prediction model generation process; saving the heat load prediction models of all virtual heating network groups.

[0013] As a preferred scheme of the multi-strategy collaborative heating balance control method based on virtual heating network grouping described in the present invention, wherein: selecting the heating strategy for each virtual heating network group includes the following steps: using the heat load demand to estimate the daily average load of each virtual heating network group; based on historical operation data, analyzing the typical daily load pattern of each virtual heating network group, and quantifying the peak-valley difference; calculating the ratio of the heating capacity of each virtual heating network group to the daily average load, and determining the heating surplus or gap value; using the heat load prediction model to predict the heat load change trend of each virtual heating network group in the first preset period; constructing a heat load characteristic parameter set, the heat load characteristic parameter set includes the daily average load, the peak-valley difference, the heating surplus or gap value, and the heat load change trend; designing a rule-based heating strategy selection model, with the heat load characteristic parameter set as input and the corresponding heating strategy as output.

[0014] In the second aspect, an embodiment of the present invention provides a multi-strategy collaborative heating balance control system based on virtual heating network grouping, which includes a virtual heating network group division module, which is used to divide the heating network into multiple virtual heating network groups according to the similarity of the first parameter set in the heating network; a model construction module, which is used to estimate the heat load demand and heating capacity of each virtual heating network group through a dynamic thermodynamic model and a heat load prediction model; a strategy selection module, which is used to calculate the heat load characteristic parameter set based on the heat load demand and heating capacity, and select the heating strategy for each virtual heating network group; a coordination control module, which coordinates the heating network based on the heating strategy.

[0015] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the multi-strategy collaborative heating balance control method based on virtual heating network grouping as described in the first aspect of the present invention are implemented.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the multi-strategy collaborative heating balance control method based on virtual heating network grouping as described in the first aspect of the present invention are implemented.

[0017] The beneficial effects of the present invention are as follows: through real-time monitoring and dynamic grouping mechanisms, the present invention can quickly identify and adjust abnormal nodes to ensure the flexibility and adaptability of the heating network; utilize a multi-strategy collaborative control method to comprehensively consider heat load demand, heating capacity and heat exchange between virtual heating network groups to achieve accurate heat distribution and reduce energy waste; by establishing a dynamic thermodynamic model and a heat load prediction model, accurately predict heat load changes and perform pre-adjustment to effectively avoid the occurrence of insufficient or excessive heating; through virtual heating network grouping and multi-dimensional data analysis, it is possible to accurately identify the characteristics of each region and formulate corresponding heating strategies, thereby improving the overall efficiency of the heating system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 This is a framework flow chart of a multi-strategy collaborative heating balance control method based on virtual heating network grouping.

[0020] Figure 2 The present invention is a flow chart of the dynamic adjustment mechanism of the virtual heating network group based on the multi-strategy collaborative heating balance control method of virtual heating network grouping.

[0021] Figure 3 Computer equipment diagram of a multi-strategy collaborative heating balance control method based on virtual heating network grouping. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0025] Example 1

[0026] Reference Figure 1 to Figure 3 , which is the first embodiment of the present invention, and provides a multi-strategy collaborative heat supply balance control method based on virtual heat network grouping, comprising:

[0027] S1: Divide the heating network into a plurality of virtual heating network groups according to the similarity of a first parameter set in the heating network.

[0028] Specifically, the following steps are included:

[0029] S1.1: Collect a first parameter set in the heating network through temperature sensors, flow meters and pressure sensors installed at heating network nodes of the heating network.

[0030] It should be noted that the heating network nodes include main pipe sections and key nodes. The first parameter set includes supply water temperature, return water temperature, flow rate and pressure, which together determine the operating state and efficiency of the heating network.

[0031] S1.2: Transmit the first parameter set to the central controller, and pre-process the first parameter set by the central controller.

[0032] Specifically, the first parameter set is transmitted to the central controller via industrial Ethernet or a wireless sensor network. The central controller uses a Z-score method to preliminarily screen the first parameter set and remove abnormal data points that exceed the normal range. The Kalman filter is used to smooth the first parameter set after the preliminary screening.

[0033] Preferably, the Z-score method takes into account the time series characteristics of the data and uses a sliding window to calculate the local mean and standard deviation to better adapt to the dynamic characteristics of the heating system. Kalman filtering can simultaneously consider measurement noise and system dynamic characteristics, and is more suitable for dealing with the time-varying characteristics of the heating system than the simple weighted moving average method.

[0034] S1.3: The first parameter set after preprocessing is used as a feature vector, and the heating network nodes are clustered and analyzed by the improved DBSCAN clustering algorithm to obtain the virtual heating network group division result.

[0035] It should be noted that the first parameter set after smoothing is first standardized before the improved DBSCAN clustering algorithm.

[0036] Furthermore, by constructing an adaptive K-nearest neighbor graph, the number of neighbors of each heating network node is dynamically adjusted to improve the local density judgment ability of the DBSCAN clustering algorithm; the particle swarm optimization algorithm is used to optimize the DBSCAN clustering algorithm, and the optimal neighborhood radius and minimum sample number of the DBSCAN clustering algorithm are automatically searched; the optimized DBSCAN clustering algorithm is used to perform clustering analysis on the heating network nodes to obtain the initial virtual heating network group division results; the silhouette coefficient and Calinski-Harabasz index of the initial virtual heating network group division results are calculated to comprehensively evaluate the clustering quality; the fuzzy C-means clustering method is applied to the boundary nodes of each virtual heating network group to calculate the membership value of each boundary node belonging to different virtual heating network groups; according to the membership and physical location information of the boundary nodes, the virtual heating network group to which the boundary nodes belong is determined to complete the virtual heating network group division of the heating network; and a unique group number is assigned to each virtual heating network group.

[0037] The improved DBSCAN clustering algorithm improves the comparability of features and clustering accuracy by standardizing the first parameter set. The construction of an adaptive K-nearest neighbor graph enhances the local density judgment ability and makes the algorithm more adaptable to complex heating network structures. The introduction of a particle swarm optimization algorithm to automatically search for optimal parameters reduces the influence of human factors and improves the stability of clustering results. The clustering quality is evaluated by calculating the silhouette coefficient and the Calinski-Harabasz index to ensure the rationality of the results. The fuzzy C-means clustering method is applied to the boundary nodes to achieve a more refined boundary division.

[0038] S1.4: Establish a dynamic adjustment mechanism for virtual heating network groups and dynamically adjust the division results of virtual heating network groups.

[0039] Specifically, the first parameter set of each heating network node is continuously monitored; the relative deviation of the first parameter set of each heating network node is calculated, and the specific formula is as follows:

[0040]

[0041] Among them, Ts i is the water supply temperature of node i, Tr i is the return water temperature of node i, F i is the traffic of node i, P i is the pressure at node i, Tsavg Tr avg 、F avg , P avg are the average supply water temperature, average return water temperature, average flow rate and average pressure of the virtual heating network group to which node i belongs.

[0042] Furthermore, an observation time window (such as 30 minutes) and deviation thresholds of various parameters in the first parameter set are set; within the observation time window, the system periodically collects the first parameter set of the heating network node, and when the relative deviation of any parameter in the first parameter set of the heating network node exceeds the preset number of times continuously and is greater than the corresponding deviation threshold, the heating network node is marked as an abnormal node; for the marked abnormal node, the status of the adjacent nodes is checked (excluding the possibility of local failure), and if the adjacent nodes are normal, the abnormal node is removed from its current virtual heating network group and marked as a node to be allocated; a reallocation buffer is established, and the node to be allocated is added to the reallocation buffer When the number of nodes in the redistribution buffer reaches the threshold of the number of accumulated abnormal nodes or the time from the last execution of the redistribution process reaches the time interval threshold, the redistribution process is triggered (S1.4) and the virtual heating network group information is updated; an adjustment frequency control mechanism is established to record the time of each redistribution and the number of nodes involved. If the number of redistributions exceeds the maximum allowed number within the evaluation period (such as 24 hours), the parameter deviation threshold is temporarily increased or the observation time window is extended; a comprehensive evaluation is conducted regularly to analyze the frequency and causes of redistribution, evaluate the overall stability of the virtual heating network group, and appropriately adjust the parameter deviation thresholds and observation time windows according to the evaluation results.

[0043] Preferably, the dynamic adjustment mechanism of the virtual heat network group realizes the timely identification and processing of abnormal nodes by continuously monitoring the first parameter set of the heating network nodes, calculating the relative deviation and setting the observation time window and deviation threshold. The mechanism introduces the redistribution buffer and trigger conditions to ensure that the system can effectively respond to network changes. By establishing an adjustment frequency control mechanism and regular comprehensive evaluation, the system can adaptively adjust the parameter threshold and observation window to balance sensitivity and stability. In addition, considering the status of adjacent nodes, the possibility of local failure is eliminated, and the accuracy of abnormal judgment is improved. This dynamic adjustment method can not only respond to abnormal changes of a single node in a timely manner, but also ensure the long-term stable operation of the entire heating network through cumulative analysis and periodic evaluation, providing strong support for the intelligent management of the heating system.

[0044] S2: Estimate the heat load demand and heating capacity of each virtual heating network group through dynamic thermodynamic model and heat load prediction model.

[0045] Specifically, the following steps are included:

[0046] S2.1: Based on the virtual heating network group division results, a simplified topology diagram of each virtual heating network group is established.

[0047] The simplified topology diagram includes pipeline parameters and key node information of each virtual heating network group.

[0048] S2.2: Based on the simplified topology, a dynamic thermodynamic model considering the thermal inertia and time delay of the pipeline network is established.

[0049] Specifically, the node-segment method is used to establish the node flow balance equation and the segment pressure drop equation. The specific formula is as follows:

[0050] ∑F i =0

[0051]

[0052] Among them, F i is the flow rate at node i, ΔH is the pressure drop of the pipe section, L is the length of the pipe section, D is the diameter of the pipe section, λ is the friction coefficient of the pipe section, F is the flow rate, and A is the cross-sectional area of ​​the pipe.

[0053] Furthermore, a one-dimensional unsteady-state heat transfer equation considering convective heat transfer and heat loss is established. The specific formula is as follows:

[0054]

[0055] Among them, ρ is the density of water, c is the specific heat capacity of water, A is the cross-sectional area of ​​the pipe, T is the water temperature, t is the time, x is the coordinate in the direction of the pipe length, F is the flow rate, U is the heat loss coefficient, T e is the ambient temperature.

[0056] Furthermore, the heat capacity of the pipe wall is added to the heat transfer equation to reflect the thermal inertia effect, that is:

[0057]

[0058] Among them, ρ w is the density of the pipe wall material, c w is the specific heat capacity of the tube wall material, A w is the cross-sectional area of ​​the tube wall.

[0059] Furthermore, an equation describing the dynamic characteristics of the heat source is added to relate the supply and return water temperatures and flow rates. The specific formula is as follows:

[0060] Q s =ρcF(Ts-Tr)

[0061] Among them, Q s is the heat source heat supply, Ts is the water supply temperature, Tr is the return water temperature, F is the flow rate, ρ is the water density, and c is the specific heat capacity of water.

[0062] Furthermore, a user-side heat balance equation is established, and the specific formula is as follows:

[0063]

[0064] Among them, Q u is the user's heat consumption, K is the comprehensive heat transfer coefficient, S is the user-side heat dissipation area, T i is the indoor temperature, T o is the outdoor temperature and C is the heat capacity of the building.

[0065] Furthermore, the above equations are combined to form a group of dynamic thermodynamic model equations describing the entire heating network; appropriate boundary conditions and initial conditions are set for the dynamic thermodynamic model according to actual operating data; the group of equations is discretized using the characteristic line method or the finite difference method, and an appropriate numerical solution algorithm is selected to solve the group of dynamic thermodynamic model equations.

[0066] S2.3: Collect and pre-process the historical operation data of each virtual heating network group.

[0067] It should be noted that the historical operating data includes water supply temperature, return water temperature, flow rate, outdoor temperature, user heat consumption, pipe network heat loss coefficient (if available), user behavior pattern indicators (if available, window opening frequency, indoor equipment usage), timestamp, etc.

[0068] S2.4: Based on the preprocessed historical operation data, a heat load prediction model for each virtual heating network group is constructed using a multivariate linear regression algorithm.

[0069] Specifically, the pre-processed historical operation data is decomposed into trend items, cycle items and random items; according to the time series decomposition results, the historical operation data is divided into multiple preset time scales (hours, days, weeks); the key factors affecting the heat load are determined, including outdoor temperature, time characteristics (time period, weekday / weekend mark), historical heat load in the same period, etc.; the key factor data is used as independent variables and the actual heat load data is used as the dependent variable to construct a data set; based on the data set, the multivariate linear regression algorithm is applied to calculate the regression coefficient and the intercept term to obtain the initial heat load prediction model; the model residuals are calculated and tested, the normality, homogeneity of variance and independence of the residuals are tested, and the test results are recorded; the K-fold cross-validation method is used to evaluate the prediction performance of the initial heat load prediction model, and the preset evaluation indicators (such as root mean square error and determination coefficient) are calculated; based on the evaluation results, a stepwise regression process is performed to remove the independent variables whose P values ​​of the regression coefficients are greater than the preset threshold in turn, and the heat load prediction model of the current virtual heat network group is output; for each virtual heat network group, the heat load prediction model generation process is repeated; and the heat load prediction models of all virtual heat network groups are saved.

[0070] Preferably, the heat load forecasting model construction method deeply explores the characteristics of historical operation data through time series decomposition and multi-scale data division. This method comprehensively considers key factors such as outdoor temperature, time characteristics and historical heat load in the same period, and uses a multivariate linear regression algorithm to build an initial model. Through strict residual tests and K-fold cross validation, the statistical validity and prediction performance of the model are ensured. A stepwise regression optimization process is adopted to eliminate insignificant variables and improve the accuracy and explanatory power of the model. In order to adapt to the dynamic characteristics of the heating system, a sliding time window mechanism is introduced to regularly update the model parameters. This method can not only accurately capture the time variation law and influencing factors of the heat load, but also can dynamically adjust according to the latest data to improve the accuracy and real-time performance of the prediction.

[0071] S2.5: Combine the dynamic thermodynamic model and the heat load prediction model to estimate the heat load demand and heating capacity of each virtual heating network group.

[0072] Specifically, representative meteorological conditions are selected as typical operating conditions, including different seasons and extreme weather conditions; the heat load prediction model is used to calculate the heat load demand of each virtual heating network group under typical operating conditions; the heat load demand is input into the dynamic thermodynamic model to simulate the heating process of each virtual heating network group under typical operating conditions; based on the simulation results, the actual heating capacity of each virtual heating network group is calculated, taking into account the heat source capacity and pipeline network limitations; the actual heating capacity and heat load demand of each virtual heating network group are compared to identify potential heating surpluses or gaps; a sensitivity analysis is conducted to evaluate the impact of changes in key parameters on heating capacity and heat load demand; and the heat load demand, heating capacity and supply-demand balance data of each virtual heating network group under typical operating conditions are sorted out.

[0073] S3: Based on the heat load demand and heating capacity, calculate the heat load characteristic parameter set and select an appropriate heat strategy for each virtual heat network group.

[0074] Specifically, the following steps are included:

[0075] S3.1: Estimate the daily average load of each virtual heating network group using the estimated heat load demand under typical operating conditions.

[0076] S3.2: Based on historical operating data, analyze the typical daily load pattern of each virtual heating network group and quantify the peak-to-valley difference.

[0077] S3.3: Calculate the ratio of the heating capacity of each virtual heating network group to the daily average load and determine the heating surplus or deficit value.

[0078] S3.4: Based on the heat exchange model, estimate the available heat between adjacent virtual heat network groups.

[0079] S3.5: Construct a heat load characteristic parameter set.

[0080] It should be noted that the heat load characteristic parameter set includes daily average load, peak-to-valley difference, heat supply surplus or deficit value, and heat load change trend.

[0081] S3.6: Design a rule-based heating strategy selection model, which takes as input a set of heat load characteristic parameters and outputs the corresponding heating strategy.

[0082] Specifically, a set of heating strategies is defined, including four basic strategies: conventional heating, peak-valley regulation, demand response, and heat allocation; the threshold range of the heat load characteristic parameters is designed, and each parameter is divided into three levels: high, medium, and low; a decision rule table is constructed to clarify the selection of heating strategies corresponding to different combinations of heat load characteristic parameters; a rule matching algorithm is implemented to search the decision rule table and select an appropriate heating strategy based on the level combination of the current heat load characteristic parameters; strategy priorities are set to resolve conflicts when multiple strategies meet the conditions at the same time, and the strategy selection results and actual effects are recorded; the selected heating strategy and the corresponding parameter adjustment suggestions are output.

[0083] S4: Coordinated control of the heating network based on the heating strategy.

[0084] Specifically, according to the heating strategy, the initial values ​​of the heating parameters in the virtual heating network group are set; the key operating parameters of the virtual heating network group are continuously monitored, and the heating strategy and heating parameters are dynamically adjusted according to the key operating parameters to maintain the overall balance of the heating network; the control effect is periodically evaluated, and the control process is optimized based on the evaluation results.

[0085] Furthermore, the present embodiment also provides a multi-strategy collaborative heating balance control system based on virtual heating network grouping, including a virtual heating network group division module, which is used to divide the heating network into multiple virtual heating network groups according to the similarity of the first parameter set in the heating network; a model construction module, which is used to estimate the heat load demand and heating capacity of each virtual heating network group through a dynamic thermodynamic model and a heat load prediction model; a strategy selection module, which is used to calculate the heat load characteristic parameter set based on the heat load demand and heating capacity, and select the heating strategy for each virtual heating network group; and a coordination control module, which coordinates the heating network based on the heating strategy.

[0086] This embodiment also provides a computer device, which is suitable for the case of a multi-strategy collaborative heat supply balance control method based on virtual heat network grouping, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the multi-strategy collaborative heat supply balance control method based on virtual heat network grouping as proposed in the above embodiment.

[0087] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0088] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the multi-strategy collaborative heating balance control method based on virtual heating network grouping as proposed in the above embodiment is implemented.

[0089] In summary, the present invention can quickly identify and adjust abnormal nodes through real-time monitoring and dynamic grouping mechanisms to ensure the flexibility and adaptability of the heating network; utilize multi-strategy collaborative control methods to comprehensively consider heat load demand, heating capacity and heat exchange between virtual heating network groups to achieve accurate heat distribution and reduce energy waste; by establishing a dynamic thermodynamic model and a heat load prediction model, accurately predict heat load changes and perform pre-adjustment to effectively avoid insufficient or excessive heating; through virtual heating network grouping and multi-dimensional data analysis, it can accurately identify the characteristics of each area and formulate corresponding heating strategies, thereby improving the overall efficiency of the heating system.

[0090] Example 2

[0091] Reference Figure 1 to Figure 3 , which is the second embodiment of the present invention, provides a multi-strategy collaborative heating balance control method based on virtual heating network grouping. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0092] In the actual scenario of a city's heating system, a typical heating area was selected for a one-month simulation experiment. The area covers 5 main heating branches and serves 300 heating nodes, including residential areas, commercial areas and some public facilities. During the experiment, the outdoor temperature fluctuated between -15℃ and 5℃, and the supply water temperature (70℃ to 85℃), return water temperature (50℃ to 65℃), flow rate (200m 3 / h to 800m3 / h), pressure (0.4MPa to 1.0MPa), and other key parameters. The average daily heat load of users is between 200kW and 600kW, and the load of some industrial and commercial users reaches 800kW.

[0093] Temperature sensors, flow meters and pressure sensors were installed at the main pipe sections and key nodes to collect data once a minute and transmit it to the central controller via industrial Ethernet. The Z-score method with a 10-minute sliding window was used to preliminarily screen the first parameter set, eliminating about 2% of abnormal data points. Subsequently, the Kalman filter was used to smooth the data after the preliminary screening, and the noise level was reduced by about 75%.

[0094] The smoothed data was standardized, and an adaptive K-nearest neighbor graph was constructed. A feature extraction method based on principal component analysis (PCA) was introduced to reduce the original 4-dimensional data to 3 dimensions, retaining 95% of the information. The parameters of DBSCAN were optimized by the particle swarm optimization (PSO) algorithm, and the neighborhood radius was finally determined to be 0.15 and the minimum number of samples was 5. The clustering results showed that the heating network nodes were successfully divided into 8 virtual heating network groups. The average silhouette coefficient of each group was 0.72, and the average Calinski-Harabasz index was 820, indicating that the clustering quality was good. For boundary nodes, the fuzzy C-means clustering method was applied to calculate the membership degree, which effectively solved the attribution problem of about 10% of the boundary nodes.

[0095] A dynamic adjustment mechanism for the virtual heating network group was established, with a 30-minute observation window and a 5% parameter deviation threshold. During the experiment, 15 abnormal nodes were detected, 12 of which were effectively handled through the reallocation process, and the system stability was significantly improved.

[0096] For each virtual heating network group, a dynamic thermodynamic model considering the thermal inertia and time delay of the pipe network was established. The heat capacity of the pipe wall was added to the model, and the simulation accuracy of the thermal inertia effect was improved by about 20%. At the same time, a heat load prediction model was constructed, using a multivariate linear regression algorithm, taking into account factors such as outdoor temperature, time period characteristics, and historical heat load. The root mean square error of the model is about 40kW, and the determination coefficient reaches 0.88. A 24-hour sliding window is used, and the model parameters are updated once a day, and the prediction accuracy is steadily improved.

[0097] Based on typical operating conditions under different meteorological conditions (such as severe winter conditions with an outdoor temperature of -10°C), the heat load demand and heating capacity of each virtual heating network group were estimated. The results show that under extremely cold weather, there is a heating gap of about 100-150kW in three virtual heating network groups, while the other five groups have a heating surplus of 20-50kW. The established heat exchange model shows that the amount of heat that can be allocated between adjacent virtual heating network groups is between 50-80kW.

[0098] According to the calculated heat load characteristic parameter set, a rule-based heating strategy selection model was designed. For virtual heating network groups with large heating gaps, the demand response strategy was mainly adopted. By adjusting the water supply temperature (increased by 3-5°C) and increasing the flow rate (10-15%), the heating gap was reduced from the initial 150kW to 30kW. For groups with heating surplus, the peak-valley regulation strategy was adopted to appropriately reduce the heating parameters during the low load period, saving about 5% of energy consumption.

[0099] Heat distribution was optimized by using the heat exchange model between virtual heat network groups. About 70kW of heat was allocated by adjusting the flow and temperature parameters between adjacent groups, further improving the balance between supply and demand. The system continuously monitors key operating parameters, conducts strategy evaluation and fine-tuning every 15 minutes, and ensures a dynamic balance of heat supply.

[0100] The experimental results show that the average heat load satisfaction rate of the heating system has increased from the initial 85% to 92%, and the energy efficiency has increased by about 8%. The heating stability during peak load periods has increased by 15%, and the user comfort score (based on indoor temperature fluctuations) has increased by 0.6 points (out of 5 points). In addition, through the coordinated regulation of the virtual heat network group, the overall response time of the system has been shortened by about 20%, providing strong support for large-scale implementation in the future.

[0101] Preferably, the comparison indicators of the present invention and the traditional method are shown in Table 1.

[0102] Table 1 Comparison indexes of the present invention and the traditional method

[0103] index The present invention Traditional methods Heating load satisfaction rate 85％ 92％ Energy efficiency Benchmark value +8％ Heating stability during peak load periods Benchmark value +15％ User comfort rating (out of 5 points) 3.8 points 4.4 points System response time Benchmark value -20％ Abnormal node processing efficiency 60％ 80％ Heat adjustment capability 30kW 70kW Coefficient of determination 0.75 0.88 Average silhouette coefficient 0.60 0.72

[0104] As shown in Table 1, the method of the present invention is significantly superior to the traditional method in multiple key indicators. In particular, there are significant improvements in heating load satisfaction rate, energy utilization efficiency, heating stability and system response time. In addition, the method also performs well in abnormal node processing and heat allocation capabilities. These improvements together lead to a significant improvement in user comfort, fully demonstrating the superiority of the multi-strategy collaborative heating balance control method based on virtual heating network grouping.

[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-strategy collaborative heat supply balance control method based on virtual heat network grouping, characterized in that: include, dividing the heating network into a plurality of virtual heating network groups according to similarities of a first parameter set in the heating network; The heat load demand and heating capacity of each virtual heating network group are estimated through dynamic thermodynamic model and heat load prediction model; Calculate a heat load characteristic parameter set based on the heat load demand and the heating capacity, and select a heating strategy for each virtual heating network group; The heating network is coordinated and controlled based on the heating strategy.

2. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 1 is characterized in that: The step of dividing the heating network into a plurality of virtual heating network groups according to the similarity of the first parameter set in the heating network comprises the following steps: The first parameter set in the heating network is collected by installing a temperature sensor, a flow meter and a pressure sensor at a heating network node of the heating network; the first parameter set includes a supply water temperature, a return water temperature, a flow rate and a pressure; transmitting the first parameter set to a central controller, and preprocessing the first parameter set by the central controller; The first parameter set after preprocessing is used as a feature vector, and cluster analysis is performed on the heating network nodes by using an improved DBSCAN clustering algorithm to obtain a virtual heating network group division result; A dynamic adjustment mechanism for the virtual heating network group is established to dynamically adjust the division result of the virtual heating network group.

3. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 2 is characterized in that: The step of performing cluster analysis on the heating network nodes by using the improved DBSCAN clustering algorithm to obtain the virtual heating network group division result comprises the following steps: By constructing an adaptive K-nearest neighbor graph, the number of neighbors of each heating network node is dynamically adjusted to improve the local density judgment ability of the DBSCAN clustering algorithm; The DBSCAN clustering algorithm is optimized by using a particle swarm optimization algorithm to automatically search for the optimal neighborhood radius and the minimum number of samples of the DBSCAN clustering algorithm; Using the optimized DBSCAN clustering algorithm to perform cluster analysis on the heating network nodes, and obtaining the initial virtual heating network group division results; Calculating the silhouette coefficient and Calinski-Harabasz index of the initial virtual heating network group division result to comprehensively evaluate the clustering quality; The fuzzy C-means clustering method is applied to the boundary nodes of each virtual heating network group to calculate the membership value of each boundary node belonging to different virtual heating network groups; According to the membership degree and physical location information of the boundary node, the virtual heating network group to which the boundary node belongs is determined, and the division of the heating network into virtual heating network groups is completed.

4. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 2 is characterized in that: The dynamic adjustment mechanism of the virtual heating network group includes: Calculate the relative deviation of the first parameter set of each heating network node, and set the observation time window and the deviation threshold of each parameter in the first parameter set; Within the observation time window, the system periodically collects the first parameter set of the heating network node, and when the relative deviation of any parameter in the first parameter set of the heating network node exceeds a preset number of times continuously and is greater than its corresponding deviation threshold, the heating network node is marked as an abnormal node; Check the status of the adjacent nodes of the abnormal node. If the adjacent nodes are normal, remove the abnormal node from its current virtual heating network group and mark it as a node to be allocated. Establish a reallocation buffer, add the nodes to be allocated to the reallocation buffer, and when the number of nodes to be allocated in the reallocation buffer reaches the threshold of the number of accumulated abnormal nodes, or the time from the last reallocation execution reaches the time interval threshold, trigger the reallocation process and update the virtual heat network group information; Establish an adjustment frequency control mechanism to record the time of each reallocation and the number of nodes to be allocated. If the number of reallocations exceeds the maximum allowed number within the evaluation period, increase the parameter deviation threshold or extend the observation time window.

5. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 1 is characterized in that: The method of estimating the heat load demand and heating capacity of each virtual heat network group by using a dynamic thermodynamic model and a heat load prediction model comprises the following steps: According to the virtual heating network group division results, a simplified topological structure diagram of each virtual heating network group is established; Establish a dynamic thermodynamic model based on a simplified topological structure; Collect and pre-process the historical operation data of each virtual heating network group, the historical operation data including supply water temperature, return water temperature, flow rate, outdoor temperature, and user heat consumption; Based on the historical operation data, a heat load prediction model of each virtual heating network group is constructed using a multivariate linear regression algorithm; The dynamic thermodynamic model and the heat load prediction model are combined to estimate the heat load demand and heating capacity of each virtual heat network group.

6. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 5 is characterized in that: The method of using a multivariate linear regression algorithm to construct a heat load prediction model for each virtual heat network group includes the following steps: Performing time series decomposition on the preprocessed historical operation data, and dividing the historical operation data into multiple preset time scales according to the time series decomposition result; The key factors affecting heat load are taken as independent variables and the actual heat load data as dependent variables to construct a data set; Based on the data set, a multivariate linear regression algorithm is applied to calculate the regression coefficient and the intercept term to obtain an initial heat load prediction model; Evaluating and optimizing the initial heat load prediction model, including: Calculate and test the model residuals and record the test results; The K-fold cross validation method is used to evaluate the prediction performance of the initial heat load prediction model and calculate the preset evaluation index; Based on the evaluation results, a stepwise regression process is performed to remove the independent variables whose P values ​​in the regression coefficients are greater than the preset threshold value in turn, and the heat load prediction model of the current virtual heating network group is output; For each virtual heating network group, a heat load prediction model generation process is performed; Save the heat load forecast model of all virtual heat network groups.

7. The multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to claim 1 is characterized in that: The selection of the heating strategy for each virtual heating network group comprises the following steps: Using the heat load demand, estimate the daily average load of each virtual heat network group; Based on historical operation data, the typical daily load pattern of each virtual heating network group is analyzed to quantify the peak-to-valley difference; Calculate the ratio of the heating capacity of each virtual heating network group to the daily average load to determine the heating surplus or deficit value; Using the heat load prediction model, predict the heat load change trend of each virtual heat network group in the first preset period; Constructing a heat load characteristic parameter set, wherein the heat load characteristic parameter set includes daily average load, peak-to-valley difference, heat supply surplus or deficit value, and heat load change trend; A rule-based heating strategy selection model is designed, which takes the heat load characteristic parameter set as input and outputs the corresponding heating strategy.

8. A multi-strategy collaborative heat supply balance control system based on virtual heat network grouping, based on the multi-strategy collaborative heat supply balance control method based on virtual heat network grouping according to any one of claims 1 to 7, characterized in that: Also includes, A virtual heating network group division module, used for dividing the heating network into a plurality of virtual heating network groups according to the similarity of the first parameter set in the heating network; A model building module, used for estimating the heat load demand and heating capacity of each virtual heat network group through a dynamic thermodynamic model and a heat load prediction model; A strategy selection module, used to calculate a heat load characteristic parameter set based on the heat load demand and the heating capacity, and select a heating strategy for each virtual heating network group; The coordination control module coordinates and controls the heating network based on the heating strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-strategy collaborative heating balance control method based on virtual heating network grouping according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-strategy collaborative heating balance control method based on virtual heating network grouping according to any one of claims 1 to 7 are implemented.

Images