A digital-twin-based intelligent optimization and regulation system for a heating pipe network

Abstract

The application discloses a kind of based on digital twinning's intelligent optimization control system of heating pipe network, it is related to heating pipe network intelligent control technical field, including, digital twinning state construction module, the multi-source operation data of heating pipe network is collected, based on multi-source operation data constructs the digital twinning state data corresponding to heating pipe network;Heat inertia time delay compensation module, based on digital twinning state data calculates the dynamic heat propagation time delay in each pipe section heat transport process, according to dynamic heat propagation time delay to digital twinning state data is chronologically compensated, obtains time delay correction state data;Dynamic resistance identification module, based on time delay correction state data and heating pipe network actual operation data constructs resistance coefficient identification space, determines the real-time resistance coefficient of each pipe section by the identification mode that chaos distribution initialization and population difference self-adapting evolution are combined.The application constructs digital twinning and dynamic collaborative control closed loop, and improves the heating pipe network operation state perception precision and control optimization ability.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for heating networks, and in particular to an intelligent optimization and control system for heating networks based on digital twins. Background Technology

[0002] As a crucial component of urban centralized heating systems, the operating status of heating networks directly impacts heating efficiency, energy consumption levels, and user-side thermal comfort. With the continuous expansion of heating system scale and increasing fluctuations in user load, traditional heating networks are gradually evolving from static design and experience-based regulation to dynamic sensing and refined operation management. Current technologies typically model and analyze heating systems based on network topology, thermodynamic balance, and operational monitoring data, combining operational parameters such as flow rate, temperature, and pressure to assess and regulate network status. Furthermore, with the development of information technology and data acquisition techniques, some heating systems are beginning to introduce multi-source operational data fusion methods, leveraging data-driven approaches to analyze heating load changes and network operating status, thereby improving system adaptability and control precision.

[0003] However, in actual operation, the transmission of the heating medium in the pipeline network exhibits significant time delay characteristics, and complex thermal inertia and coupling effects exist between different pipe sections. This makes it easy for state analysis based on synchronously acquired data to produce time-series deviations, thus affecting the accurate depiction of the actual operating state. Simultaneously, the resistance characteristics of the pipeline network change dynamically due to factors such as valve regulation, local structural changes, and long-term operating state variations, making it difficult for hydraulic calculations under fixed parameter conditions to consistently reflect the actual operating characteristics. Furthermore, heating systems not only need to meet the thermal comfort needs of users during operation but also need to comprehensively consider the coordination between heat energy consumption and electrical energy consumption. Existing technologies still have room for improvement in their unified modeling capabilities for multi-objective coordinated control and reverse inference of actual user heat demand. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a digital twin-based intelligent optimization and control system for heating networks to solve the problems of inaccurate identification of operating status and difficulty in coordinated control and optimization under multi-source spatiotemporal coupling conditions in heating networks.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a digital twin-based intelligent optimization and control system for heating pipe networks, comprising: The digital twin status construction module collects multi-source operation data of the heating network and constructs digital twin status data of the heating network based on the multi-source operation data. The thermal inertia delay compensation module calculates the dynamic heat propagation delay during the heat transfer process of each pipe segment based on the digital twin state data, and performs time-series compensation on the digital twin state data according to the dynamic heat propagation delay to obtain delay-corrected state data. The dynamic resistance identification module constructs a resistance coefficient identification space based on time-delayed state data and actual operation data of the heating network, and determines the real-time resistance coefficient of each pipe section through an identification method that combines chaotic distribution initialization with population difference adaptive evolution. The thermal resistance topology reconstruction module dynamically updates the equivalent thermal resistance of each pipe segment in the heating network based on the real-time resistance coefficient, and reconstructs the thermal resistance topology of the heating network according to the updated equivalent thermal resistance relationship to obtain dynamic thermal resistance network data. The heat demand inverse inference module establishes an inverse mapping relationship between thermal comfort state and user heat demand based on dynamic thermal resistance network data and user-side temperature change data, and determines the actual heat demand distribution data of users based on the inverse mapping relationship. The control and decision-making module determines the control strategy for the heating network based on real heat demand distribution data and dynamic thermal resistance network data. The control execution feedback module uses control strategies to regulate the heating network, obtains feedback operation data after regulation, and uploads it to the database for storage.

[0007] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of collecting multi-source operational data of the heating pipe network and constructing corresponding digital twin status data of the heating pipe network based on the multi-source operational data includes the following steps: Collect multi-source operational data of the heating network, including operational data from the heat source side, network operation data, user side operation data, and ambient temperature data, and preprocess the data to generate an operational data set; Based on the physical structure of the heating network, the heat source equipment, heat exchange station and user terminal are taken as nodes, and the supply and return water pipe sections connecting each node are taken as edge relationships to construct the basic topology of the heating network. The operation data of the heat source side, the operation data of the user side, and the ambient temperature data are mapped to the corresponding nodes, and the pipeline operation data is mapped to the pipe segment to generate the operation status parameters of the nodes and the pipe segment. Based on the operating status parameters of nodes and pipe segments, digital twin status data representing the overall operating status of the heating network is generated.

[0008] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of calculating the dynamic heat propagation delay during the heat transfer process of each pipe segment based on digital twin state data, and performing time-series compensation on the digital twin state data according to the dynamic heat propagation delay to obtain delay-corrected state data includes the following steps: Based on the operating parameters of each pipe segment in the digital twin status data, the propagation time of heat in each pipe segment is determined; Based on the connection relationship between each pipe segment, the heat transmission path from the heat source to the target node is determined, and the propagation time of each pipe segment in the path is accumulated to obtain the dynamic heat propagation delay of the corresponding node. Based on dynamic heat propagation delay, time offset alignment processing is performed on temperature, flow rate and pressure parameters in digital twin state data; By integrating the state parameters after time offset alignment, delay-corrected state data is generated.

[0009] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of constructing a resistance coefficient identification space based on time-delayed state data and actual operating data of the heating pipe network includes the following steps: Based on time-delay corrected state data and actual operation data, hydraulic operation characteristic parameters and actual hydraulic response characteristic parameters of each pipe section are constructed, including temperature gradient characteristics, flow response characteristics and pressure gradient characteristics. Based on the correspondence between the hydraulic operation characteristic parameters and the actual hydraulic response characteristic parameters of each pipe section, the operation deviation characteristic quantity of each pipe section is calculated; Based on the characteristic quantity of operational deviation, a pipe section sensitivity evaluation index is constructed to identify pipe sections with abnormal hydraulic parameters, and the resistance coefficient of the corresponding pipe section is used as the resistance parameter to be identified to generate a set of resistance parameters to be identified. Based on the operational deviation characteristics and sensitivity evaluation indicators, a comprehensive influence coefficient for the pipe section is constructed, and the basic resistance coefficient of the pipe section is proportionally mapped according to the comprehensive influence coefficient to generate candidate initial values ​​for the resistance coefficient. Based on the candidate initial values ​​of the resistance coefficient and their corresponding historical operating constraints and physical structure constraints, the resistance coefficient of each pipe segment to be identified in the heating network is taken as an independent dimension in the parameter space, and a resistance parameter space structure composed of multiple dimensions is constructed. The constraint intervals for the resistance coefficient values ​​of each pipe segment are discretized to generate a set of resistance coefficient values ​​for each dimension, and a multidimensional resistance coefficient combination space is constructed based on the set of values ​​for each dimension. The operation consistency constraint judgment is performed on the combination of multidimensional resistance coefficients, and the parameter combination that does not meet the hydraulic balance condition of the heating network is eliminated; Based on the degree of deviation between the simulated operation results corresponding to each combination of resistance coefficients and the actual hydraulic response characteristics, parameter combinations with deviations exceeding a preset threshold are eliminated. The combination of resistance coefficients that simultaneously satisfies parameter boundary constraints, hydraulic operation consistency constraints, and deviation matching constraints is determined as the feasible solution domain of resistance coefficients, forming the resistance coefficient identification space.

[0010] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the method of determining the real-time resistance coefficient of each pipe segment through an identification method combining chaotic distribution initialization and population difference adaptive evolution includes the following steps: Within the resistance coefficient identification space, a set of candidate solutions is constructed based on uniform random sampling, and each candidate solution corresponds to a complete pipeline resistance parameter configuration scheme. Based on the candidate solution set, a chaotic mapping method is used to perform a linear mapping on each candidate solution to generate a new candidate solution set; Each candidate resistance coefficient from the new candidate solution set is input into the digital twin heating network. Based on the digital twin simulation results, the adaptability of each candidate resistance coefficient combination is evaluated, and the corresponding candidate resistance coefficient operation deviation value is generated. A diversity index is constructed based on the distribution characteristics of candidate solutions in the search space, and an adaptive adjustment factor is constructed based on the performance deviation evaluation value and the diversity index. The candidate resistance coefficient combination is updated based on the adaptive adjustment factor, so that the candidate solution with large running deviation evolves towards the candidate solution with small deviation. The reciprocal of the running deviation value of each candidate solution is used as the matching degree. The candidate solution set is then filtered based on the matching degree, and candidate solutions with running deviation values ​​lower than a preset threshold are selected and placed into the high matching degree candidate solution set. Identify candidate solutions whose operational deviation values ​​exceed a preset threshold and mark them as objects to be updated; Based on the constraint range of the resistance coefficient of each pipe segment, the local disturbance range of the candidate solution with high matching degree is set, and random disturbance is generated within the local disturbance range; Use random perturbation to perturb the corresponding candidate solutions in the high-matching candidate solution set to generate new candidate solutions; The new candidate solutions are used to update and replace the objects to be updated. The retained high-matching candidate solutions and the new candidate solutions are combined to form a new candidate solution set, maintaining the population size unchanged. When a candidate solution meets the convergence condition, the resistance coefficient corresponding to the optimal matching candidate solution is selected as the real-time resistance coefficient for each pipe segment.

[0011] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of dynamically updating the equivalent thermal resistance of each pipe segment in the heating pipe network according to the real-time resistance coefficient includes the following steps: Based on the real-time resistance coefficient of each pipe segment, and combined with the reference resistance coefficient and the reference equivalent thermal resistance of the pipe segment, the conversion relationship between the resistance coefficient and the equivalent thermal resistance is determined. Based on the conversion relationship, the real-time resistance coefficient of each pipe segment is mapped to the corresponding equivalent thermal resistance parameter, and the equivalent thermal resistance parameter obtained by mapping is updated segment by segment.

[0012] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of reconstructing the thermal resistance topology of the heating pipe network according to the updated equivalent thermal resistance relationship to obtain dynamic thermal resistance network data includes the following steps: Based on the updated equivalent thermal resistance parameters of each pipe segment, each pipe segment is assigned a corresponding thermal resistance weight, and the dynamic weight value of the corresponding connection edge of each pipe segment in the heating network topology data is calculated. By replacing the fixed weights of the corresponding pipe segment connection edges in the heating network topology data with dynamic weight values, the weights of each connection edge are updated in real time as the equivalent thermal resistance parameters change, thus forming a dynamic weighted connection relationship. Based on the dynamic weighted connection relationship, a dynamic thermal resistance topology structure is constructed, consisting of nodes, connecting edges, and thermal resistance weights of connecting edges, and the dynamic thermal resistance topology structure is used as dynamic thermal resistance network data.

[0013] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of establishing an inverse mapping relationship between thermal comfort state and user heat demand based on dynamic thermal resistance network data and user-side temperature change data, and determining the actual distribution data of user heat demand based on the inverse mapping relationship, includes the following steps: Based on the topological connection relationship of the dynamic thermal resistance network, the heat transfer path between the heat source node and each user node is determined. The weight values ​​of each connecting edge in the dynamic thermal resistance network are used as thermal resistance costs to evaluate each candidate transmission path, and the path with the smallest cumulative thermal resistance value is selected as the optimal transmission path for the corresponding user node. Based on user-side temperature change data, the indoor temperature monitoring values ​​of each user node are obtained, and the thermal comfort state coefficient is calculated in combination with the preset comfort temperature. Based on the cumulative thermal resistance and thermal comfort state coefficient of the optimal transmission path corresponding to each user node, an inverse mapping relationship between thermal comfort state and user thermal demand is established to obtain the thermal demand mapping value of the user node. The preset basic heating demand and the heat demand mapping value are superimposed to obtain the actual heat demand intensity of the user node; The actual heat demand intensity corresponding to each user node in the heating network is summarized to form the distribution data of actual heat demand of users.

[0014] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of determining the heating pipe network control strategy based on real heat demand distribution data and dynamic thermal resistance network data includes the following steps: Under the condition of meeting the actual heat demand of users, heat is distributed to users based on the optimal transmission path; The heating distribution results of all user nodes are integrated to generate a heating network control strategy.

[0015] As a preferred embodiment of the intelligent optimization and control system for heating pipe networks based on digital twins as described in this invention, the step of using a control strategy to regulate the heating pipe network, obtaining feedback operation data after regulation, and uploading it to a database for storage includes the following steps: Based on the control strategy, adjustment instructions are generated for the output of the heating source, the operating status of the circulating pump, and the distribution of the pipeline flow, and the adjustment instructions are sent to the corresponding execution objects of the heating pipeline network. During the execution of the regulation command, information on the changes in the operating status of the heating network after regulation is collected, including node temperature change data, pipe section flow rate change data, and pressure change data. The information on changes in the operational status after adjustment is used as feedback operational data and uploaded to the database for storage.

[0016] The beneficial effects of this invention are as follows: By constructing digital twin state data corresponding to the heating network, and calculating the dynamic heat propagation delay during the heat transfer process of each pipe segment based on the digital twin state data, time-series compensation is performed, realizing explicit correction of the time delay effect of the heating medium during actual transmission. This allows the operating states of different pipe segments to be aligned and expressed under a unified time reference, improving the time-series accuracy and dynamic consistency of the heating system state analysis. By constructing a resistance coefficient identification space based on the time-delay-corrected state data and the actual operating data of the heating network, and using a combination of chaotic distribution initialization and population difference adaptive evolution to determine the real-time resistance coefficient of each pipe segment, adaptive characterization of the dynamic changes in the hydraulic parameters of the network is achieved. This allows the resistance parameters to be adaptively updated with changes in operating state, improving the adaptability and stability of the hydraulic characteristic identification of the network. By dynamically updating the equivalent thermal resistance of each pipe segment according to the real-time resistance coefficient, and based on the updated equivalent thermal resistance... Thermal resistance relationships are used to reconstruct the thermal resistance topology of the heating network, transforming it from a static structure to a dynamic weighted structure. This allows the heat transfer path to be reconstructed according to changes in operating status, improving the dynamism and realism of the heating network structure. By establishing an inverse mapping relationship between thermal comfort and heating response based on dynamic thermal resistance network data and user-side temperature change data, and determining the actual heat demand distribution data of users based on this relationship, the reverse derivation of heating demand from actual user-side temperature changes is realized. This shifts heating demand from experience-based prediction to data-driven inversion, improving the realism and precision of heat load identification. Furthermore, by constructing a dynamic benefit constraint relationship between heat energy consumption and electricity consumption based on actual heat demand distribution data and dynamic thermal resistance network data, a synergistic trade-off between heating energy consumption and electricity consumption is achieved. This shifts regulation from single energy consumption optimization to multi-energy synergistic optimization, improving the overall economic efficiency and energy utilization coordination of the heating system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a digital twin-based intelligent optimization and control system for heating networks.

[0019] Figure 2 Flowchart for real-time resistance coefficient identification and thermal resistance network reconstruction of heating pipe sections.

[0020] Figure 3 A flowchart for mapping user heat demand and generating heating network control strategies is generated. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] 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 phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0024] Reference Figures 1-3 This is the first embodiment of the present invention, which provides a digital twin-based intelligent optimization and control system for heating networks, comprising the following steps: The digital twin status construction module collects multi-source operational data of the heating network and constructs corresponding digital twin status data of the heating network based on the multi-source operational data.

[0025] Specifically, multi-source operational data of the heating network is collected, including operational data from the heat source side, network operation data, user side operation data, and ambient temperature data, and preprocessed to generate an operational data set; Based on the physical structure of the heating network, the heat source equipment, heat exchange station and user terminal are taken as nodes, and the supply and return water pipe sections connecting each node are taken as edge relationships to construct the basic topology of the heating network. Each pipe segment is assigned information including pipe segment length, pipe diameter, and connection direction to generate heating network topology data; The operation data on the heat source side is mapped to the heat source node, the temperature data on the user side is mapped to the user node, the operation data of the pipeline network is mapped to the corresponding pipe segment, and the ambient temperature data is mapped to the global environmental variable. The real-time operation status parameters of each node and each pipe segment are determined by combining the topology data of the heating pipeline network. Based on the real-time operating status parameters of each node and each pipe section, and taking the direction of heat transmission in the heating network as the basis, the state dependency relationship between nodes is established. Based on state dependency, using pipe segments as the transmission medium, the state transmission mapping relationship between different nodes is determined, and state association relationship data of the coupling relationship between each node and pipe segment in the heating network is generated. Based on the status correlation data, the operating status of each node, each pipe section and each piece of equipment is correlated and integrated to generate a global operating status set that represents the overall operating status of the heating network. The global operating status set is used as the digital twin status data corresponding to the heating network.

[0026] By collecting and preprocessing multi-source data from the heating network to form an operational data set, a unified aggregation and standardized representation of multi-source heterogeneous information of the heating system is achieved, providing a consistent data foundation for subsequent modeling. Based on the physical structure of the heating network, a basic topology is constructed with heat source equipment, heat exchange stations, and user terminals as nodes and supply and return water pipe sections as edges. This enables abstract modeling of the heating system from the "equipment level" to the "network structure level," allowing the actual connection relationships of the network to be expressed in a structured manner, supporting heat transfer path analysis and state correlation calculations. By combining data from the heat source side, user side, and ring... Ambient temperature data is mapped to nodes, and pipeline operation data is mapped to pipe segments, realizing the correspondence and binding of operation data with physical topology. This enables the spatial consistency of dispersed operation information within a unified structural framework, enhancing the interpretability and traceability of state expression. By generating digital twin state data of the heating pipeline network based on node and pipe segment operation state parameters, a unified characterization and virtual-real mapping of the overall operation state of the heating system is achieved. This provides complete data support for subsequent state analysis, parameter identification, and operation optimization, improving the system's state perception capability and enhancing the transparency of pipeline network operation.

[0027] The thermal inertia delay compensation module calculates the dynamic heat propagation delay during the heat transfer process of each pipe segment based on the digital twin state data, and performs time-series compensation on the digital twin state data according to the dynamic heat propagation delay to obtain delay-corrected state data.

[0028] Specifically, based on the supply water temperature, return water temperature, medium flow rate, pipe length, and pipe diameter data of each pipe segment in the digital twin status data, the heat transfer rate in the corresponding pipe segment is calculated using the following formula: ; in, For pipe segment index, For the first The heat transfer rate of each pipe section For the first Cross-sectional area of ​​each pipe section For the first The volumetric flow rate of the medium in each pipe section The set temperature difference influence coefficient, For the first The water supply temperature of each pipe section For the first The return water temperature of each pipe section.

[0029] Divide the length of each pipe segment by the corresponding heat transfer rate to obtain the time it takes for heat to travel in a single pipe segment. Based on the connection relationship between each pipe segment, the heat source node is taken as the starting point and the heat is transmitted to the target node through multiple continuous pipe segments to generate a heating path. The propagation time of heat through each pipe segment in the heating path is summed to obtain the dynamic heat propagation delay corresponding to the heat transfer to each node. Based on the dynamic heat propagation delay corresponding to each node, the temperature, flow, and pressure parameters in the digital twin state data are time-shifted and aligned. The current temperature, flow, and pressure parameters are mapped to the actual time position of the heat reaching the corresponding node. The formula is as follows: ; in, For time index, For node indexing, For the first Each node The original operating status data at that moment, including temperature, flow rate, and pressure parameters. For heat to transfer from the heat source to the first The cumulative propagation delay of each node, For the first Each node The running status data after time offset alignment processing.

[0030] By integrating the time-shifted temperature, flow, and pressure parameters, time-delay corrected state data is generated that matches the actual heat transfer process.

[0031] By calculating the heat propagation time in each pipe segment based on the operating parameters of each pipe segment in digital twin state data, the temporal characteristics of heat transfer in the heating system are characterized, providing a time-scale basis for dynamic thermal process modeling. By determining the transmission path from the heat source to the target node based on the connection relationships between pipe segments, and accumulating the propagation time of each pipe segment along the path, the dynamic heat propagation delay corresponding to each node is obtained, achieving a holistic characterization of the time delay effect of heat transfer in multi-level pipe networks and supporting path-level time consistency analysis of heat propagation in complex pipe networks. By performing time offset alignment processing on operating parameters such as temperature, flow rate, and pressure based on the dynamic heat propagation delay, the operating states of different nodes at different time scales can be mapped to the actual arrival time of heat, achieving temporal consistency correction of the heating system state data. By integrating the time-aligned multi-source operating parameters to generate time-delay-corrected state data, the true dynamic operating state of the heating pipe network is reconstructed, improving the temporal consistency representation capability and state restoration accuracy of the actual thermal process.

[0032] The dynamic resistance identification module constructs a resistance coefficient identification space based on time-delayed state data and actual operation data of the heating network. It determines the real-time resistance coefficient of each pipe section through an identification method that combines chaotic distribution initialization with population difference adaptive evolution.

[0033] Specifically, the temperature, flow, and pressure parameters of each pipe segment in the time-delay correction status data are standardized. Calculate the difference between the supply water temperature and the return water temperature of each pipe section to generate temperature gradient characteristics; The flow response characteristics are generated by calculating the ratio of the flow parameters of each pipe section to the design flow of the pipeline network. Calculate the pressure parameters of each pipe segment and the pressure difference between adjacent nodes to generate pressure gradient characteristics; The temperature gradient characteristics, flow response characteristics, and pressure gradient characteristics are normalized and combined to generate hydraulic operation characteristic parameters of the pipe section. The temperature, flow and pressure monitoring values ​​in the actual operation data of the heating network are mapped to the corresponding pipe sections, and the actual temperature gradient characteristics, actual flow response characteristics and actual pressure gradient characteristics of each pipe section are calculated based on the feature construction method consistent with the hydraulic operation characteristic parameters. The actual temperature gradient characteristics, actual flow response characteristics, and actual pressure gradient characteristics are normalized and combined to generate the actual hydraulic response characteristics of the pipe section under the current operating state. Based on the correspondence between the hydraulic operating characteristic parameters and the actual hydraulic response characteristics of each pipe section, the operating deviation characteristic of each pipe section is calculated using the following formula: ; in, For the first The characteristic quantity of operational deviation for each pipe section The temperature gradient characteristic is a parameter in hydraulic operation. This refers to the actual temperature gradient characteristic in the actual hydraulic response. For the flow response characteristics in the hydraulic operation characteristic parameters, This refers to the actual flow response characteristics within the actual hydraulic response characteristics. The pressure gradient characteristic is a parameter in hydraulic operation. This refers to the actual pressure gradient characteristic in the actual hydraulic response. , , These are the weighting coefficients for the set temperature deviation characteristic, flow rate deviation characteristic, and pressure deviation characteristic, respectively.

[0034] Consistency threshold judgment is performed on the operational deviation characteristics of each pipe section, deviation judgment threshold range is set, and it is found whether the hydraulic characteristics of the pipe section deviate from the normal operating state. Based on the threshold determination results, a set of pipe segments whose operational deviation characteristics exceed the threshold is selected, and the pipe segments in this set are identified as potential pipe segments with abnormal hydraulic parameters. For pipe sections with potential hydraulic parameter anomalies, a sensitivity evaluation index is constructed based on the difference amplitude between the operational deviation characteristic quantity and the hydraulic operation characteristic parameter. The formula is as follows: ; in, For the first Sensitivity evaluation indicators for each pipe section It is the maximum value among all operating deviation characteristics.

[0035] Based on the sensitivity evaluation index of each pipe section with potential hydraulic parameter anomalies, the pipe sections are sorted to determine the priority of the impact of different pipe sections on the hydraulic transmission process of the heating network. Based on the sorting results, pipe sections with sensitivity evaluation indicators exceeding the preset sensitivity threshold are selected, and the resistance coefficient of the corresponding pipe section is used as the resistance parameter to be identified. All resistance parameters to be identified are summarized to generate a set of resistance parameters to be identified; The characteristic quantities of the operating deviation of each pipe segment in the set of resistance parameters to be identified are normalized. The normalized deviation value and the corresponding sensitivity evaluation index of the pipe section are weighted and fused according to the preset weights to obtain the comprehensive influence coefficient. Based on the comprehensive influence coefficient, a proportional mapping is performed on the pipe section resistance coefficient to generate candidate initial values ​​for the resistance coefficient. The formula is as follows: ; in, For the first Candidate initial values ​​for the resistance coefficient of each pipe segment The set reference resistance coefficient value, For the first The comprehensive impact coefficient of each pipe section The set influence amplification weighting coefficient.

[0036] The maximum and minimum values ​​of the resistance coefficient for each pipe section are extracted from the historical operating data and used as the upper and lower bounds of the historical operating data, respectively. Based on the physical structural parameters of the pipe section, determine the theoretical maximum and minimum allowable values ​​of the resistance coefficient, which serve as the upper and lower bounds of the physical constraints; Centered on the candidate initial value of the drag coefficient, the intersection of the historical operating upper and lower bounds and the physical constraint upper and lower bounds is calculated to obtain the constraint range of the drag coefficient value; The range of drag coefficient values ​​is used as the input boundary condition for constructing the multidimensional drag coefficient parameter space; The resistance coefficient of each pipe segment to be identified in the heating network is used as an independent dimension in the parameter space to construct a resistance parameter space structure composed of multiple dimensions. The constraint interval of the resistance coefficient value of each pipe segment is discretized, and the continuous interval is divided into a finite number of value nodes to obtain the set of discretized resistance coefficient values ​​of each pipe segment. Based on the set of resistance coefficient values ​​after discretization of each pipe segment, Cartesian product operation is performed to generate a multidimensional resistance coefficient combination sample set. All combined samples are collected to form an overall parameter space, generating a multidimensional resistance coefficient parameter combination space for the heating network. For each set of resistance coefficient combinations in the multidimensional parameter combination space, determine whether the resistance coefficient of each pipe segment falls within the corresponding resistance coefficient value constraint range, and eliminate combinations that exceed the physical boundary. Based on the flow and pressure distribution results corresponding to each combination of resistance coefficients, it is determined whether they meet the operational consistency constraints of the heating network, and combinations that lead to hydraulic imbalance are eliminated. Calculate the degree of deviation between the simulated operating state and the actual hydraulic response characteristics under each combination of resistance coefficients, and eliminate parameter combinations whose deviation exceeds the preset tolerance threshold; The combination of drag coefficients that simultaneously satisfies parameter boundary constraints, hydraulic operation consistency constraints, and deviation matching constraints is determined as the feasible solution domain of the drag coefficients and included in the subsequent search space. Using the search space as the overall parameter solution domain, a set of multidimensional feasible combinations of heating network resistance coefficients is determined, forming a resistance coefficient identification space.

[0037] By constructing hydraulic operation characteristic parameters and actual hydraulic response characteristic parameters for each pipe section based on time-delayed state data and actual operation data, the correlation between the theoretical and actual operating states of the heating network is realized, providing a data foundation for accurately reflecting the differences in pipe section operation. Through corresponding analysis of the two types of characteristic parameters, the operating deviation characteristic quantities of each pipe section are calculated, achieving a quantitative characterization of the degree of deviation in the operating state of the pipe section, providing a basis for identifying key pipe sections affecting the network's operational performance. By constructing sensitivity evaluation indicators based on the operating deviation characteristic quantities, pipe sections with abnormal hydraulic parameters are screened, and the resistance parameters to be identified are determined, achieving precise positioning of resistance identification targets, reducing the interference of irrelevant parameters on the identification process, and improving parameter accuracy. The efficiency of resistance parameter identification is improved by combining operational deviation characteristics with sensitivity evaluation indicators to construct a comprehensive influence coefficient and generating candidate initial values ​​by proportionally mapping the basic resistance coefficient. This enables adaptive adjustment of the initial range of resistance parameters, providing initial conditions that better match the actual operating conditions for subsequent parameter optimization. By integrating historical operating constraints, physical structure constraints, and candidate initial values ​​of resistance coefficients, a multi-dimensional resistance parameter space is constructed. Parameter combinations are then gradually selected by combining parameter boundary constraints, hydraulic operation consistency constraints, and deviation matching constraints to form a resistance coefficient identification space. This effectively limits the range of feasible solutions for resistance parameters, reduces invalid search ranges, and improves the accuracy, stability, and optimization efficiency of resistance parameter identification.

[0038] Furthermore, within the value range of each dimension of the drag coefficient identification space, the initial value point of each dimension of the drag coefficient is selected by uniform random sampling. The resistance coefficient values ​​obtained from sampling each pipe segment are combined according to the pipe network topology to construct candidate solutions. Each candidate solution corresponds to a complete pipe network resistance parameter configuration scheme. Repeat the sampling and combination process to generate multiple candidate solutions, and then summarize all candidate solutions to form an initial candidate solution set; A chaotic sequence is generated based on a preset chaotic mapping function, resulting in a chaotic random sequence within a specified value range; Each dimension of the chaotic sequence is linearly mapped according to the constraint interval of the resistance coefficient of each pipe segment to obtain the candidate values ​​of the corresponding pipe segment resistance coefficient. The formula is as follows: ; in, For the first after chaotic mapping Candidate values ​​for the resistance coefficient of each pipe segment For the first The lower bound of the resistance coefficient value for each pipe segment is constrained. For the first The upper bound of the resistance coefficient value for each pipe section is constrained. Normalized random variables generated for chaotic sequences, with values ​​ranging from [0,1].

[0039] The corresponding elements in the initial candidate solution set are replaced with the mapped drag coefficient values ​​to form a new candidate solution set. Due to the non-periodic and sensitive characteristics of chaotic sequences, the new candidate resistance coefficients exhibit a non-uniform distribution within the global search range, thereby enhancing the search space coverage capability. Each candidate resistance coefficient in the new candidate solution set is input into the digital twin heating network to simulate the temperature distribution, flow distribution and pressure distribution under the corresponding operating conditions. Based on the evaluation method consistent with the calculation process of the characteristic quantity of pipe segment operation deviation, the operation deviation value of the corresponding candidate resistance coefficient is generated; Calculate the Euclidean distance between candidate solutions in the search space, and construct the distance relationship between candidate solutions, using the following formula: ; in, Index for candidate solutions In addition to candidate solutions The indexes of the remaining candidate solutions, For the first The candidate solution and the first Euclidean distance between candidate solutions For the number of pipe sections, For the first Among the candidate solutions, the th... The resistance coefficient of each pipe section For the first Among the candidate solutions, the th... Resistance coefficient of each pipe section.

[0040] Based on the distance relationships between candidate solutions, the average distribution distance of the candidate solution set is calculated to characterize the search diversity level. The formula is as follows: ; in, The average distribution distance of the candidate set. This represents the total number of candidate solutions.

[0041] The average distribution distance of the calculated candidate solution set is used as the search diversity index of the candidate set. The larger the value, the more dispersed the candidate solutions are in the search space and the higher the diversity. The running deviation value and the search diversity index are weighted and summed according to preset weight coefficients to construct an adaptive adjustment factor for each candidate solution. The formula is as follows: ; in, For the first An adaptive adjustment factor for each candidate solution. For the first The running deviation values ​​corresponding to each candidate solution. , These are the set operating deviation value and the weighting coefficient of the search diversity index, respectively.

[0042] Based on the adaptive adjustment factor, the update step size of the candidate solution is calculated by multiplying it by the preset default update step size, as shown in the formula: ; in, This is the preset default update step size. For the first The actual update step size for each candidate solution.

[0043] Based on the running deviation value corresponding to each candidate solution, the update direction of the candidate solution is determined, so that the candidate solution with the large running deviation value is adjusted towards the candidate solution with the small deviation value; The candidate solutions are updated based on their update step size and update direction, using the following formula: ; in, Index for iteration count, For the first The candidate solution at the th... The combination of resistance coefficients in the next iteration For the first The updated combination of drag coefficients for each candidate solution For the first The update direction vector of each candidate solution.

[0044] The reciprocal of the running deviation value of each candidate solution is used as the matching degree. The candidate solution set is then filtered based on the matching degree, and candidate solutions with running deviation values ​​lower than a preset threshold are selected and placed into the high matching degree candidate solution set. Identify candidate solutions whose operational deviation values ​​exceed a preset threshold and mark them as objects to be updated; Based on the constraint range of the resistance coefficient values ​​for each pipe segment, a local disturbance range is defined for the candidate solution with high matching degree, and a random disturbance is generated within this local disturbance range, as shown in the formula: ; in, For the high-match candidate solution set, the first one is the... The random perturbation generated by each candidate solution For the first The interval of drag coefficients corresponding to each candidate solution For the first The interval of drag coefficients corresponding to each candidate solution The disturbance coefficient is set.

[0045] A new candidate solution is generated by perturbing the corresponding candidate solution in the high-matching candidate solution set with a random perturbation amount, as shown in the formula: ; in, For the first The combination of drag coefficients for each candidate solution. For the perturbation of the first The combination of drag coefficients for each candidate solution.

[0046] The new candidate solutions are used to update and replace the objects to be updated. The retained high-matching candidate solutions and the new candidate solutions are combined to form a new candidate solution set, maintaining the population size unchanged. When a candidate solution meets the convergence condition, the resistance coefficient corresponding to the optimal matching candidate solution is selected as the real-time resistance coefficient for each pipe segment.

[0047] By constructing a candidate solution set through uniform random sampling within the drag coefficient identification space, a comprehensive initialization of drag parameter combinations is achieved, providing a broad initial search foundation for subsequent drag coefficient identification. A new candidate solution set is generated by linearly mapping the candidate solutions using chaotic mapping, achieving a non-uniform distribution of candidate solutions within the identification space. This enhances the coverage of the search space, provides richer search directions for global parameter optimization, and reduces the possibility of getting trapped in local optima. By inputting candidate drag coefficient combinations into a digital twin heating network for simulation and evaluating the adaptability of the candidate solutions based on the simulation results, a correlation analysis between drag parameters and the actual operating state of the heating system is achieved. This provides an evaluation basis that conforms to actual operating characteristics for drag coefficient identification, improving the authenticity and reliability of the identification results. Furthermore, by combining the candidate solutions… The search distribution characteristics construct a diversity index, which, together with the operational deviation, forms an adaptive adjustment factor. This achieves dynamic coordination between search capability and convergence capability, enabling candidate solutions to adaptively adjust their evolution strategy according to the current search state, thus balancing global search capability and local optimization capability. By updating the candidate resistance coefficient combination based on the adaptive adjustment factor, candidate solutions with large operational deviations continuously evolve towards a better direction, achieving gradual optimization of resistance parameters and improving the convergence efficiency of parameter identification. By selecting candidate solutions with high matching degree and retaining their excellent characteristics, while implementing local perturbation updates and replacements for candidate solutions with low matching degree, the inheritance of advantageous solutions and the exploration of new solutions are combined. This enhances population activity while maintaining a stable population size, avoiding premature convergence in the optimization process, and improving the stability of resistance coefficient identification, global optimization capability, and final identification accuracy.

[0048] The thermal resistance topology reconstruction module dynamically updates the equivalent thermal resistance of each pipe segment in the heating network based on the real-time resistance coefficient, and reconstructs the thermal resistance topology of the heating network according to the updated equivalent thermal resistance relationship to obtain dynamic thermal resistance network data.

[0049] Specifically, based on the real-time resistance coefficient of each pipe segment, combined with the reference resistance coefficient and the reference equivalent thermal resistance of the pipe segment, the conversion relationship between the resistance coefficient and the equivalent thermal resistance is determined, as shown in the formula: ; in, For the first The equivalent thermal resistance of each pipe section after the update For the first The reference equivalent thermal resistance of each pipe segment, For the first Real-time resistance coefficient of each pipe segment For the first The reference resistance coefficient of each pipe section This is the correction factor for the change in equivalent thermal resistance caused by the change in the set drag coefficient.

[0050] Based on the conversion relationship, the real-time resistance coefficient of each pipe segment is mapped to the corresponding equivalent thermal resistance parameter, and the equivalent thermal resistance parameter obtained by mapping is updated segment by segment.

[0051] By establishing the conversion relationship between the resistance coefficient and the equivalent thermal resistance, the correlation mapping between hydraulic parameters and thermal parameters is realized. This allows the resistance parameter identification results to be synchronously reflected in the heat transfer characteristics of the heating network, providing a unified parameter conversion mechanism for the coordinated updating of hydraulic and thermal characteristics, and enhancing the comprehensive characterization capability of the actual operating status of the heating network. By mapping the real-time resistance coefficient of each pipe segment to the corresponding equivalent thermal resistance parameter based on the conversion relationship, and updating the equivalent thermal resistance of each pipe segment segment by segment, the dynamic updating of the digital twin thermal resistance parameter is realized. This can reflect the changes in heat transfer capacity during the operation of the network in a timely manner, providing more accurate thermal parameter support for subsequent heat transfer analysis, heat demand prediction, and heating optimization and control, thereby improving the real-time performance, accuracy, and dynamic adaptability of digital twin data.

[0052] Furthermore, based on the updated equivalent thermal resistance parameters of each pipe segment, a corresponding thermal resistance weight is assigned to each pipe segment, and the dynamic weight value of the corresponding connection edge of each pipe segment in the heating network topology data is calculated. The formula is as follows: ; in, For the first The dynamic weight value of the connecting edge corresponding to each pipe segment.

[0053] By replacing the fixed weights of the corresponding pipe segment connection edges in the heating network topology data with dynamic weight values, the weights of each connection edge are updated in real time as the equivalent thermal resistance parameters change, thus forming a dynamic weighted connection relationship. Based on the dynamic weighted connection relationship, a dynamic thermal resistance topology structure is constructed, consisting of nodes, connecting edges, and thermal resistance weights of connecting edges, and the dynamic thermal resistance topology structure is used as dynamic thermal resistance network data.

[0054] By calculating the dynamic weight values ​​of the corresponding connecting edges based on the updated equivalent thermal resistance parameters of each pipe segment, a dynamic mapping from thermal resistance parameters to network topology weights is achieved. This enables the heating network topology to reflect changes in the heat transfer capacity of each pipe segment in real time, providing a data foundation that better reflects actual operating conditions for heat transfer path analysis and network status assessment. By replacing the fixed weights in the heating network topology with dynamic weight values, a dynamically weighted connection relationship that updates in real time with changes in equivalent thermal resistance is formed. This transforms the network connection relationship from a static to a dynamic description, allowing the heating network topology to continuously reflect changes in operating conditions. By constructing a dynamic thermal resistance topology structure based on the dynamic weighted connection relationship, consisting of nodes, connecting edges, and the thermal resistance weights of the connecting edges, and generating dynamic thermal resistance network data, a deep integration of the physical structure and heat transfer characteristics of the heating network is achieved. This provides dynamic network support for subsequent heat transfer path calculation, user heat demand inversion, and heating optimization and control, improving the expressive ability of the dynamic thermodynamic characteristics of the heating network and the accuracy of overall optimization analysis.

[0055] The heat demand inverse inference module establishes an inverse mapping relationship between thermal comfort state and user heat demand based on dynamic thermal resistance network data and user-side temperature change data, and determines the actual heat demand distribution data of users based on the inverse mapping relationship.

[0056] Specifically, taking the heat source node as the starting node and the user node as the ending node, according to the connection relationship in the dynamic thermal resistance network, each connection edge is traversed in turn to determine the corresponding thermal resistance transmission path between the heat source and each user node. For user nodes with multiple heat transfer paths, the dynamic weight values ​​corresponding to the connecting edges between nodes are used as the heat transfer cost. The cumulative heat transfer value of each candidate heat transfer path between any two nodes is calculated using the following formula: ; in, For heat transfer path indexing, For the first The cumulative thermal resistance value corresponding to each heat transfer path. This represents the number of connecting edges traversed by the heat transfer path.

[0057] The path with the lowest cumulative thermal resistance is taken as the optimal transmission path for the corresponding user node. Based on user-side temperature change data, the indoor temperature monitoring values ​​of each user node at multiple consecutive time points are obtained; Based on indoor temperature monitoring values ​​and a preset comfort temperature, the thermal comfort state coefficient for each user node is calculated using the following formula: ; in, Indexing for user nodes, For the first Thermal comfort coefficient of each user node For the first The current indoor temperature monitoring value of each user node. This is the preset comfort temperature value.

[0058] Based on the cumulative thermal resistance and thermal comfort coefficient of the optimal transmission path for each user node, an inverse mapping relationship between thermal comfort state and user thermal demand is established to obtain the thermal demand mapping value of the user node. The formula is as follows: ; in, For the first Hot demand mapping value for each user node For the first The cumulative thermal resistance value of the optimal thermal delivery path corresponding to each user node.

[0059] The actual heat demand intensity of user nodes is obtained by summing the preset basic heating demand and the heat demand mapping value, using the following formula: ; in, For the first The actual heat demand intensity of each user node This is based on the pre-set basic heating demand.

[0060] The actual heat demand intensity corresponding to each user node in the heating network is summarized to form the distribution data of actual heat demand of users.

[0061] By selecting the optimal heat transfer path with the minimum cumulative thermal resistance value using the thermal resistance weight of the connection edge as the thermal resistance cost, a dynamic path representation of the heat transfer process is achieved. This allows heat transfer analysis to fully reflect the real-time operating status of the heating network, providing a path basis that conforms to the actual heat transfer characteristics for user heat demand analysis. By calculating the user's thermal comfort state coefficient, a quantitative representation of the user's actual heating comfort level is achieved, introducing user-side perceived information into heat demand analysis and improving the consistency between heat demand assessment and the user's actual heating status. Furthermore, by combining the cumulative thermal resistance value of the optimal transfer path with the thermal comfort state coefficient, a thermal comfort state correlation coefficient is established. This system establishes a reverse mapping relationship between user heat demand and enables correlation analysis between heat transmission characteristics and user comfort needs. This allows for the reverse derivation of users' actual heat demand from the operating status, providing a reliable basis for on-demand heating. By superimposing the heat demand mapping value with the basic heat demand, the system obtains the actual heat demand intensity of each user node and aggregates it to form the distribution data of users' actual heat demand. This achieves a unified expression of heat demand from individual users to the spatial distribution of the entire network, providing accurate data support for heat load forecasting, optimized allocation of heat sources, and intelligent control of the heating network. This, in turn, improves the on-demand control capability and heat energy utilization efficiency of the heating system.

[0062] The control and decision-making module determines the control strategy for the heating network based on real heat demand distribution data and dynamic thermal resistance network data.

[0063] Specifically, under the condition of meeting the actual heat demand of users, heat is distributed to users based on the optimal transmission path; The heating distribution results of all user nodes are integrated to generate a heating network control strategy.

[0064] By allocating heat to each user based on the optimal heat transfer path while meeting the user's actual heat demand, a coordinated match between heating resources and user needs is achieved. This ensures that heat is transported along paths with high heat transfer efficiency, reducing ineffective losses during heat transfer and improving the utilization efficiency of heating resources while guaranteeing users' thermal comfort needs. By integrating the heat allocation results of each user node, a heating network control strategy is generated, realizing unified management of the heating system from local heating control to coordinated optimization and control of the entire network. This provides a decision-making basis for heat source output regulation, network operation optimization, and dynamic allocation of heating load, improving the overall coordination, intelligence level, and heating guarantee capability of the heating network.

[0065] The control execution feedback module uses control strategies to regulate the heating network, obtains feedback operation data after regulation, and uploads it to the database for storage.

[0066] Specifically, based on the control strategy, adjustment instructions are generated for the output of the heating source, the operating status of the circulating pump, and the distribution of the pipeline flow, and the adjustment instructions are sent to the corresponding execution objects in the heating pipeline network; During the execution of the regulation command, information on the changes in the operating status of the heating network after regulation is collected, including node temperature change data, pipe section flow rate change data, and pressure change data. The information on changes in the operational status after adjustment is used as feedback operational data and uploaded to the database for storage.

[0067] By generating adjustment commands for heat source output, circulating pump operation status, and pipeline flow distribution based on the heating network control strategy, and issuing these commands to the corresponding execution objects, the effective conversion of control strategy into on-site control actions is achieved. This enables the heating system to coordinate the adjustment of heat source, power equipment, and pipeline operation parameters based on optimization results, providing an execution foundation for intelligent operation of the heating network. By collecting real-time data on node temperature changes, pipe section flow changes, and pressure changes during the execution of adjustment commands, continuous monitoring of the control effect and changes in the heating network's operating status is achieved. This provides an objective basis for evaluating the execution effect of the control strategy and changes in operating status, improving the perceptibility and monitorability of the heating system's operating status. Furthermore, by uploading the adjusted operating status change information as feedback operating data to the database for storage, continuous accumulation and closed-loop feedback of heating network operating data are achieved. This provides reliable data support for dynamic updates of digital twin data, correction of operating parameters, and subsequent heating control, enhancing the heating system's self-learning ability, continuous optimization ability, and long-term operational stability.

[0068] In summary, this invention achieves explicit correction of the time delay effect of the heating medium during actual transmission by: constructing digital twin state data corresponding to the heating network and calculating the dynamic heat propagation delay during heat transfer in each pipe segment based on the digital twin state data, and performing time-series compensation. This enables the operating states of different pipe segments to be aligned and expressed under a unified time reference, improving the time-series accuracy and dynamic consistency of the heating system state analysis; constructing a resistance coefficient identification space based on the time-delay corrected state data and the actual operating data of the heating network, and determining the real-time resistance coefficient of each pipe segment using a combination of chaotic distribution initialization and population difference adaptive evolution, achieving adaptive characterization of the dynamic changes in the network hydraulic parameters. This allows the resistance parameters to be adaptively updated with changes in operating state, improving the adaptability and stability of the network hydraulic characteristic identification; and dynamically updating the equivalent thermal resistance of each pipe segment based on the real-time resistance coefficient, and then using the updated equivalent thermal resistance... The thermal resistance topology reconstruction of the heating network, based on the thermal resistance relationship, transforms the heating network from a static structure to a dynamic weighted structure. This allows the heat transfer path to be reconstructed according to changes in operating status, improving the dynamism and realism of the heating network structure. By establishing an inverse mapping relationship between thermal comfort and heating response based on dynamic thermal resistance network data and user-side temperature change data, and determining the actual heat demand distribution data of users based on this relationship, the reverse derivation of heating demand from actual user-side temperature changes is realized. This shifts heating demand from experience-based prediction to data-driven inversion, improving the realism and precision of heat load identification. Furthermore, by constructing a dynamic benefit constraint relationship between heat energy consumption and electricity consumption based on actual heat demand distribution data and dynamic thermal resistance network data, a synergistic trade-off between heating energy consumption and electricity consumption is achieved. This shifts regulation from single energy consumption optimization to multi-energy synergistic optimization, improving the overall economic efficiency and energy utilization coordination of the heating system.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A digital twin-based intelligent optimization and control system for heating pipe networks, characterized in that: include, The digital twin status construction module collects multi-source operation data of the heating network and constructs digital twin status data of the heating network based on the multi-source operation data. The thermal inertia delay compensation module calculates the dynamic heat propagation delay during the heat transfer process of each pipe segment based on the digital twin state data, and performs time-series compensation on the digital twin state data according to the dynamic heat propagation delay to obtain delay-corrected state data. The dynamic resistance identification module constructs a resistance coefficient identification space based on time-delayed state data and actual operation data of the heating network, and determines the real-time resistance coefficient of each pipe section through an identification method that combines chaotic distribution initialization with population difference adaptive evolution. The thermal resistance topology reconstruction module dynamically updates the equivalent thermal resistance of each pipe segment in the heating network based on the real-time resistance coefficient, and reconstructs the thermal resistance topology of the heating network according to the updated equivalent thermal resistance relationship to obtain dynamic thermal resistance network data. The heat demand inverse inference module establishes an inverse mapping relationship between thermal comfort state and user heat demand based on dynamic thermal resistance network data and user-side temperature change data, and determines the actual heat demand distribution data of users based on the inverse mapping relationship. The control and decision-making module determines the control strategy for the heating network based on real heat demand distribution data and dynamic thermal resistance network data. The control execution feedback module uses control strategies to regulate the heating network, obtains feedback operation data after regulation, and uploads it to the database for storage.

2. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 1, characterized in that: The process of collecting multi-source operational data of the heating network and constructing digital twin status data of the heating network based on the multi-source operational data includes the following steps: Collect multi-source operational data of the heating network, including operational data from the heat source side, network operation data, user side operation data, and ambient temperature data, and preprocess the data to generate an operational data set; Based on the physical structure of the heating network, the heat source equipment, heat exchange station and user terminal are taken as nodes, and the supply and return water pipe sections connecting each node are taken as edge relationships to construct the basic topology of the heating network. The operation data of the heat source side, the operation data of the user side, and the ambient temperature data are mapped to the corresponding nodes, and the pipeline operation data is mapped to the pipe segment to generate the operation status parameters of the nodes and the pipe segment. Based on the operating status parameters of nodes and pipe segments, digital twin status data representing the overall operating status of the heating network is generated.

3. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 2, characterized in that: The process of calculating the dynamic heat propagation delay during heat transfer in each pipe segment based on digital twin state data, and then performing time-series compensation on the digital twin state data according to the dynamic heat propagation delay to obtain delay-corrected state data, includes the following steps: Based on the operating parameters of each pipe segment in the digital twin status data, the propagation time of heat in each pipe segment is determined; Based on the connection relationship between each pipe segment, the heat transmission path from the heat source to the target node is determined, and the propagation time of each pipe segment in the path is accumulated to obtain the dynamic heat propagation delay of the corresponding node. Based on dynamic heat propagation delay, time offset alignment processing is performed on temperature, flow rate and pressure parameters in digital twin state data; By integrating the state parameters after time offset alignment, delay-corrected state data is generated.

4. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 3, characterized in that: The construction of the resistance coefficient identification space based on time-delay corrected state data and actual operating data of the heating network includes the following steps: Based on time-delay corrected state data and actual operation data, hydraulic operation characteristic parameters and actual hydraulic response characteristic parameters of each pipe section are constructed, including temperature gradient characteristics, flow response characteristics and pressure gradient characteristics. Based on the correspondence between the hydraulic operation characteristic parameters and the actual hydraulic response characteristic parameters of each pipe section, the operation deviation characteristic quantity of each pipe section is calculated; Based on the characteristic quantity of operational deviation, a pipe section sensitivity evaluation index is constructed to identify pipe sections with abnormal hydraulic parameters, and the resistance coefficient of the corresponding pipe section is used as the resistance parameter to be identified to generate a set of resistance parameters to be identified. Based on the operational deviation characteristics and sensitivity evaluation indicators, a comprehensive influence coefficient for the pipe section is constructed, and the basic resistance coefficient of the pipe section is proportionally mapped according to the comprehensive influence coefficient to generate candidate initial values ​​for the resistance coefficient. Based on the candidate initial values ​​of the resistance coefficient and their corresponding historical operating constraints and physical structure constraints, the resistance coefficient of each pipe segment to be identified in the heating network is taken as an independent dimension in the parameter space, and a resistance parameter space structure composed of multiple dimensions is constructed. The constraint intervals for the resistance coefficient values ​​of each pipe segment are discretized to generate a set of resistance coefficient values ​​for each dimension, and a multidimensional resistance coefficient combination space is constructed based on the set of values ​​for each dimension. The operation consistency constraint judgment is performed on the combination of multidimensional resistance coefficients, and the parameter combination that does not meet the hydraulic balance condition of the heating network is eliminated; Based on the degree of deviation between the simulated operation results corresponding to each combination of resistance coefficients and the actual hydraulic response characteristics, parameter combinations with deviations exceeding a preset threshold are eliminated. The combination of resistance coefficients that simultaneously satisfies parameter boundary constraints, hydraulic operation consistency constraints, and deviation matching constraints is determined as the feasible solution domain of resistance coefficients, forming the resistance coefficient identification space.

5. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 4, characterized in that: The identification method, which combines chaotic distribution initialization with population difference adaptive evolution, determines the real-time resistance coefficient of each pipe segment, including the following steps: Within the resistance coefficient identification space, a set of candidate solutions is constructed based on uniform random sampling, and each candidate solution corresponds to a complete pipeline resistance parameter configuration scheme. Based on the candidate solution set, a chaotic mapping method is used to perform a linear mapping on each candidate solution to generate a new candidate solution set; Each candidate resistance coefficient from the new candidate solution set is input into the digital twin heating network. Based on the digital twin simulation results, the adaptability of each candidate resistance coefficient combination is evaluated, and the corresponding candidate resistance coefficient operation deviation value is generated. A diversity index is constructed based on the distribution characteristics of candidate solutions in the search space, and an adaptive adjustment factor is constructed based on the performance deviation evaluation value and the diversity index. The candidate resistance coefficient combination is updated based on the adaptive adjustment factor, so that the candidate solution with large operating deviation evolves towards the candidate solution with small deviation. The reciprocal of the running deviation value of each candidate solution is used as the matching degree. The candidate solution set is then filtered based on the matching degree, and candidate solutions with running deviation values ​​lower than a preset threshold are selected and placed into the high matching degree candidate solution set. Identify candidate solutions whose operational deviation values ​​exceed a preset threshold and mark them as objects to be updated; Based on the constraint range of the resistance coefficient of each pipe segment, the local disturbance range of the candidate solution with high matching degree is set, and random disturbance is generated within the local disturbance range; Use random perturbation to perturb the corresponding candidate solutions in the high-matching candidate solution set to generate new candidate solutions; The new candidate solutions are used to update and replace the objects to be updated. The retained high-matching candidate solutions and the new candidate solutions are combined to form a new candidate solution set, maintaining the population size unchanged. When a candidate solution meets the convergence condition, the resistance coefficient corresponding to the optimal matching candidate solution is selected as the real-time resistance coefficient for each pipe segment.

6. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 5, characterized in that: The dynamic updating of the equivalent thermal resistance of each pipe section in the heating network based on the real-time resistance coefficient includes the following steps: Based on the real-time resistance coefficient of each pipe segment, and combined with the reference resistance coefficient and the reference equivalent thermal resistance of the pipe segment, the conversion relationship between the resistance coefficient and the equivalent thermal resistance is determined. Based on the conversion relationship, the real-time resistance coefficient of each pipe segment is mapped to the corresponding equivalent thermal resistance parameter, and the equivalent thermal resistance parameter obtained by mapping is updated segment by segment.

7. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 6, characterized in that: The step of reconstructing the thermal resistance topology of the heating network according to the updated equivalent thermal resistance relationship to obtain dynamic thermal resistance network data includes the following steps: Based on the updated equivalent thermal resistance parameters of each pipe segment, each pipe segment is assigned a corresponding thermal resistance weight, and the dynamic weight value of the corresponding connection edge of each pipe segment in the heating network topology data is calculated. By replacing the fixed weights of the corresponding pipe segment connection edges in the heating network topology data with dynamic weight values, the weights of each connection edge are updated in real time as the equivalent thermal resistance parameters change, thus forming a dynamic weighted connection relationship. Based on the dynamic weighted connection relationship, a dynamic thermal resistance topology structure is constructed, consisting of nodes, connecting edges, and thermal resistance weights of connecting edges, and the dynamic thermal resistance topology structure is used as dynamic thermal resistance network data.

8. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 7, characterized in that: The process of establishing an inverse mapping relationship between thermal comfort state and user thermal demand based on dynamic thermal resistance network data and user-side temperature change data, and determining the distribution data of actual user thermal demand based on the inverse mapping relationship, includes the following steps: Based on the topological connection relationship of the dynamic thermal resistance network, the heat transfer path between the heat source node and each user node is determined. The weight values ​​of each connecting edge in the dynamic thermal resistance network are used as thermal resistance costs to evaluate each candidate transmission path, and the path with the smallest cumulative thermal resistance value is selected as the optimal transmission path for the corresponding user node. Based on user-side temperature change data, the indoor temperature monitoring values ​​of each user node are obtained, and the thermal comfort state coefficient is calculated in combination with the preset comfort temperature. Based on the cumulative thermal resistance and thermal comfort state coefficient of the optimal transmission path corresponding to each user node, an inverse mapping relationship between thermal comfort state and user thermal demand is established to obtain the thermal demand mapping value of the user node. The preset basic heating demand and the heat demand mapping value are superimposed to obtain the actual heat demand intensity of the user node; The actual heat demand intensity corresponding to each user node in the heating network is summarized to form the distribution data of actual heat demand of users.

9. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 8, characterized in that: The method for determining the heating network control strategy based on real heat demand distribution data and dynamic thermal resistance network data includes the following steps: Under the condition of meeting the actual heat demand of users, heat is distributed to users based on the optimal transmission path; The heating distribution results of all user nodes are integrated to generate a heating network control strategy.

10. The intelligent optimization and control system for heating pipe networks based on digital twins as described in claim 9, characterized in that: The process of using control strategies to regulate the heating network, obtaining feedback operational data after regulation, and uploading it to the database for storage includes the following steps: Based on the control strategy, adjustment instructions are generated for the output of the heating source, the operating status of the circulating pump, and the distribution of the pipeline flow, and the adjustment instructions are sent to the corresponding execution objects of the heating pipeline network. During the execution of the regulation command, information on the changes in the operating status of the heating network after regulation is collected, including node temperature change data, pipe section flow rate change data, and pressure change data. The information on changes in the operational status after adjustment is used as feedback operational data and uploaded to the database for storage.

Images