Heat supply pipe network balance regulation and optimization system based on big data

Abstract

The invention relates to the technical field of central heating, in particular to a heat supply pipe network balance regulation and optimization system based on big data, which comprises a big data acquisition module, a pipe network model construction module, a balance regulation and optimization module and an execution and monitoring feedback module. An improved adaptive genetic algorithm is adopted to solve a multi-objective optimization problem of energy consumption, heat supply uniformity and maintenance cost, optimal regulation and control parameters are output, leakage detection, heat demand pre-judgment and smart city docking functions are integrated, and closed-loop regulation and control and intelligent operation and maintenance are realized. Application shows that heat supply uniformity is larger than or equal to 0.95, the maximum energy consumption saving rate is 18.3%, leakage response is smaller than or equal to 30 s, heat supply quality, energy-saving benefits and operation and maintenance efficiency are remarkably improved, and complex pipe networks and diversified heat utilization requirements are met.

Description

Technical Field

[0001] This invention relates to the field of centralized heating technology, specifically a heating network balance control and optimization system based on big data. Background Technology

[0002] Central heating systems are a crucial component of urban infrastructure, fulfilling the core function of ensuring heating needs for residents, businesses, and public buildings. Their operational stability, heating uniformity, and energy efficiency directly impact user experience and resource utilization. With rapid urbanization, heating networks are expanding in coverage and becoming increasingly complex. Traditional heating network control and management technologies are struggling to meet the evolving heating demands, revealing numerous technical shortcomings that urgently need to be addressed.

[0003] Existing heating network regulation largely relies on fixed parameter control or manual experience-based control, lacking precise perception and dynamic response capabilities for the entire network's operational status. Traditional control systems often base parameter settings on historical experience or theoretical calculations, failing to match real-time temperature, pressure, and flow changes during network operation. They also struggle to adapt to differentiated heating demands across different areas (e.g., residential communities vs. commercial complexes), time periods (e.g., daytime vs. nighttime), and seasons (e.g., severe cold vs. transitional seasons). This results in widespread uneven heating, with nearby users experiencing excessive heat and higher temperatures, while distant users experience insufficient heat and lower temperatures, severely impacting user satisfaction. Furthermore, traditional control methods lack precise thermo-hydraulic coupling models, failing to accurately characterize the flow and heat transfer patterns within the network. This lack of scientific basis for control decisions further exacerbates the heating imbalance.

[0004] In terms of energy consumption and cost control, existing heating systems mostly operate in an extensive manner, lacking a multi-objective collaborative optimization mechanism. Traditional systems often only aim to meet basic heating needs, failing to incorporate energy conservation and maintenance cost control into the control system. This results in heat source units and circulating water pumps operating inefficiently for extended periods, leading to significant energy waste. Furthermore, existing technologies lack in-depth analysis and utilization of historical operating data, environmental parameters, and other multi-dimensional data, making it impossible to accurately predict heating demand. They can only passively respond to load changes, further increasing ineffective energy consumption. For aging pipe networks, traditional technologies struggle to adjust control strategies specifically, resulting in high maintenance costs and poor network operation economics.

[0005] In terms of operation and maintenance management and safety assurance, existing technologies rely on manual inspections and post-incident repairs, resulting in low levels of intelligence. Traditional pipeline network operation and maintenance requires a large investment of manpower for periodic inspections, which is not only inefficient and costly, but also makes it difficult to detect potential problems such as pipeline leaks and valve malfunctions in a timely manner. Often, problems are only addressed after they have escalated, resulting in significant losses of heat transfer media and economic losses. At the same time, existing leak detection technologies mostly use single detection methods, which suffer from problems such as response lag and low location accuracy, failing to meet the safe operation requirements of complex pipeline networks. In addition, existing systems mostly operate independently and are not effectively integrated with smart city management platforms, making it impossible to achieve overall coordination and collaborative management of heating resources, thus hindering the intelligent upgrading of heating systems.

[0006] In summary, existing heating network regulation and management technologies suffer from problems such as low regulation accuracy, poor energy efficiency, insufficient intelligent operation and maintenance, and weak safety assurance capabilities, making them difficult to adapt to complex network structures and diverse heating demands. Therefore, developing a heating network balance regulation and optimization system based on big data, possessing dynamic modeling and multi-objective optimization capabilities, and capable of precise regulation and intelligent operation and maintenance has become an urgent technical challenge to be solved in the current heating industry. Summary of the Invention

[0007] The purpose of this invention is to provide a heating network balance control and optimization system based on big data, so as to solve the problems mentioned in the background art, such as low control accuracy, poor energy saving and economy, insufficient intelligent operation and maintenance level, and weak safety assurance capabilities of existing heating network control and management technologies.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A big data-based heating network balance control and optimization system includes a big data acquisition module, a network model construction module, a balance control optimization module, an execution module, and a monitoring and feedback module. The big data acquisition module collects full-link operational data of the heating network. The network model construction module constructs a dynamic thermo-hydraulic coupling model based on the collected data. The balance control optimization module solves for the optimal control parameters using a multi-objective optimization algorithm. The execution module performs network balance adjustment according to the optimal control parameters. The monitoring and feedback module monitors the control effect in real time and dynamically corrects the model and parameters. The dynamic thermo-hydraulic coupling model satisfies the following innovative formula: ; The objective function of the multi-objective optimization algorithm is: ; The multi-objective optimization algorithm employs an improved adaptive genetic algorithm, which achieves global optimal solution search by dynamically adjusting the crossover and mutation probabilities. The crossover and mutation probabilities satisfy the following: ; ; In the formula, For the density of the heating medium, For the specific heat capacity of the heating medium at constant pressure, For the temperature of the heating medium, For time, For the flow rate of the heating medium, The thermal conductivity of the heating medium, To reduce power loss due to heat dissipation in the pipeline network, The heat dissipation power of the heat exchange equipment. For pipeline pressure, For the dynamic viscosity of the heating medium, This is the friction coefficient. The inner diameter of the pipe. To adjust the parameter vector, In order to regulate the energy consumption of the pipeline network, As a baseline energy consumption, The weighting coefficients are and satisfy the following conditions: , For the first The actual water supply temperature at each user terminal. Set the water supply temperature for the user. To regulate the maintenance costs of the pipeline network, Based on the baseline maintenance cost, The initial crossover probability, The initial mutation probability, This represents the individual fitness value. The average fitness value of the population. This represents the optimal fitness value for the population. It is a local constant to avoid the denominator being zero.

[0009] Preferably, the big data acquisition module includes a temperature sensor, a pressure sensor, a flow sensor, an energy consumption meter, an environmental parameter sensor, and a user-end feedback acquisition unit. The temperature sensor is installed at the heat source outlet, main pipe nodes, branch pipe nodes, and user-end supply and return water inlets, with a measurement range of -20℃ to 120℃ and a measurement accuracy of not less than ±0.1℃. The pressure sensor is installed at key nodes and ends of the pipeline network, with a measurement range of 0 to 2.5MPa and a measurement accuracy of not less than ±0.01MPa. The flow sensor uses an electromagnetic flow meter, installed on the main pipe and major branch pipes, with a measurement range of 0 to 100m³. 3 / h, with a measurement accuracy of not less than ±0.5%.

[0010] Preferably, the pipeline network model construction module further includes a data preprocessing unit, a model training unit, and a model update unit; the data preprocessing unit uses an outlier detection algorithm to remove outliers from the collected data, and the outlier detection uses an improved Z-score algorithm that satisfies the formula: ; when When it is determined to be an outlier; in the formula, For a single data collection, For data sequences The median of For data sequences The median absolute deviation; the model training unit uses the gradient boosting tree algorithm to calibrate the parameters of the dynamic thermo-hydraulic coupling model, and the calibration error does not exceed 5%.

[0011] Preferably, the improved adaptive genetic algorithm in the balance regulation optimization module further includes a Latin hypercube sampling strategy during the population initialization phase, an elite retention mechanism during the iteration process, and a convergence criterion; the elite retention mechanism directly retains the top 10% of individuals in the population in terms of fitness value to the next generation, and the convergence criterion is that the rate of change of the optimal fitness value of the population is less than 10% over 50 consecutive generations. -4 The iteration stops when the control parameter vector is reached. This includes the heat source supply water temperature, heat source circulation flow rate, the opening degree of the regulating valves in each branch pipe, and the secondary network supply water pressure of the heat exchange station, with values ​​ranging from 60℃ to 95℃ and 50m. 3 / h~500m 3 / h, 0%~100%, 0.3MPa~1.2MPa.

[0012] Preferably, the execution module includes an electric regulating valve, a variable frequency circulating water pump, a plate heat exchanger, and a heat source regulating device; the electric regulating valve is an intelligent electric ball valve with a regulating accuracy of not less than 0.1% and a response time of not more than 2 seconds, and communicates with the balance control optimization module via the Modbus-RTU protocol; the variable frequency circulating water pump is a permanent magnet synchronous variable frequency pump with a speed range of 30Hz to 50Hz, a rated head of 20m to 100m, and a rated flow rate of 50m³ / h. 3 / h~500m 3 / h, dynamically adjust the rotation speed according to the flow command in the optimal control parameters.

[0013] Preferably, the monitoring feedback module includes a real-time data transmission unit, a control effect evaluation unit, and a model correction unit; the real-time data transmission unit adopts a 5G+edge computing architecture, with a data transmission latency of no more than 100ms and a data transmission success rate of no less than 99.9%; the control effect evaluation unit evaluates the control effect based on three dimensions: heating uniformity index, energy saving rate, and user satisfaction, with the heating uniformity index satisfying the formula: ; In the formula, As an indicator of heating uniformity, For the number of users, The average actual water supply temperature for all users is required. ≥ The model correction unit corrects the parameters of the dynamic thermo-hydraulic coupling model every 30 minutes based on the latest collected data.

[0014] Preferably, the weight coefficients of the multi-objective optimization algorithm , , The system is dynamically adjusted based on seasonal changes, user demand, and pipeline operation status, specifically as follows: Winter cold season The value ranges from 0.5 to 0.6. The value ranges from 0.3 to 0.4. The value ranges from 0.1 to 0.2; Transitional season The value ranges from 0.5 to 0.6. The value ranges from 0.3 to 0.4. The value ranges from 0.1 to 0.2; Pipeline aging period The value ranges from 0.3 to 0.4. The value ranges from 0.3 to 0.4. The value ranges from 0.2 to 0.3.

[0015] Preferably, a pipeline leak detection unit is also included. This unit, based on pressure and flow data from a big data acquisition module, combines negative pressure wave and acoustic detection methods to locate leaks using the following innovative leak judgment formula: ; In the formula, The distance from the leak point to the near-end sensor. The speed at which pressure waves propagate in the pipe. The time it takes for the pressure wave to reach the remote sensor. The value represents the change in pipeline pressure before and after the leak. The length of the pipe between the two sensors is specified, with a leak detection accuracy of no less than ±1m and a leak response time of no more than 30s.

[0016] Preferably, the pipeline network model construction module also interfaces with the smart city heating management platform to receive urban weather forecast data, regional population flow data, and building heating demand change data; the balance control optimization module integrates the above external data into a multi-objective optimization algorithm, and corrects the user-end set water supply temperature in the objective function using the following formula. : ; In the formula, The corrected water supply temperature setting. This is the difference between the forecast temperature for the day and the historical average temperature for the same period. The deviation between the actual population and the baseline population of a region. This is a meteorological correction factor, with a value ranging from 0.02 to 0.05. This is the population correction factor, with a value ranging from 0.01 to 0.03.

[0017] Preferably, the big data acquisition module also stores nearly 5 years of historical pipeline operation data, including daily hourly energy consumption data, supply and return water temperature difference data, flow data, and fault record data; the pipeline model construction module analyzes the historical data through a long short-term memory neural network to predict the pipeline heat demand for the next 24 hours, using the following prediction formula: ; In the formula, For the future Real-time forecasting of heat demand, The value ranges from 1 to 24 hours. To meet the current heating demand, This represents the length of the historical data window, ranging from 24 to 72. The current ambient temperature is used as the basis for prediction, with an error not exceeding 8%. The balance control optimization module adjusts the control parameters in advance based on the prediction results to achieve predictive balance control.

[0018] Compared with the prior art, the beneficial effects of the present invention are: (1) This invention achieves precise balance control of the heating network, significantly improving heating quality and user adaptability. Traditional heating systems generally suffer from fixed control parameters and delayed response, which easily leads to uneven heating in different areas and at different times, making it difficult to match differentiated heating needs. This invention achieves full-link data acquisition of the network through a distributed sensor network, accurately depicts the network operation status by combining a dynamic thermo-hydraulic coupling model, and then solves the optimal control parameters through an improved adaptive genetic algorithm, forming a precise closed loop of acquisition-modeling-optimization-control. From the implementation results, the heating uniformity index is stable above 0.95 in different scenarios such as the severe winter in the north, the transitional season in the south, and old urban areas, with the index reaching 0.97 in residential areas. The user-end temperature prediction error is controlled within 8%, effectively solving the temperature deviation problem of traditional systems, achieving stable heating under all operating conditions, and significantly improving user satisfaction.

[0019] (2) This invention strengthens multi-objective optimization design, significantly improving the energy-saving efficiency and economic rationality of the heating system. Traditional heating systems often adopt an extensive operation mode, resulting in pain points such as energy redundancy and high maintenance costs. This invention constructs a multi-objective optimization function that integrates energy consumption, heating uniformity, and maintenance costs, combined with a dynamic weight adjustment strategy, to maximize core benefits in different scenarios. In the transitional season commercial complex scenario, due to the priority given to energy-saving optimization, the energy saving rate reaches 18.3%; in the old urban area scenario, due to the consideration of maintenance cost control, the maintenance cost is reduced by 18%. At the same time, the system's integrated heat demand prediction function uses LSTM neural network to analyze historical data and predict the heat demand for the next 24 hours, enabling advance regulation to reduce ineffective energy consumption. Furthermore, the two-level storage architecture of edge server + cloud database ensures data security, further improving the economic efficiency and reliability of the system operation.

[0020] (3) This invention promotes the intelligent upgrading of heating network operation and maintenance, improving operational safety and management efficiency. Traditional network operation and maintenance relies on manual inspections, which suffers from problems such as delayed leak detection, untimely fault handling, and high management costs. This invention integrates core functions such as real-time monitoring feedback, accurate leak detection, and smart city integration. Through a 5G+edge computing architecture, it achieves real-time data transmission with a transmission delay controlled within 100ms and a transmission success rate of over 99.9%. The leak detection unit combines negative pressure wave method and acoustic wave detection method, with a response time of no more than 30s and a positioning accuracy of ±1m. In simulated leak tests in old urban areas, the location can be completed in 22s, effectively reducing leak losses. In addition, the system connects to the smart city heating management platform through API interface, integrates external data such as meteorology and population flow to correct heating demand parameters, achieves overall management, significantly reduces the cost of manual intervention, provides technical support for the intelligent and refined operation and maintenance of heating networks, and helps upgrade smart city infrastructure. Attached Figure Description

[0021] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are explained in detail together with the embodiments of the invention, but do not constitute a limitation thereof.

[0022] Figure 1 This is a system composition block diagram of the present invention; Figure 2 These are comparison charts showing the relationship between heating area and pipeline service life, energy saving rate, heating uniformity, and special indicators in three sets of embodiments of the present invention. Figure 3 The following are detailed indicator data charts for Embodiments 1, 2, and 3 of the present invention. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] This system mainly consists of a big data acquisition module, a pipeline network model construction module, a balance control and optimization module, an execution module, a monitoring and feedback module, and an optional pipeline network leakage detection unit. These modules work together to form a closed-loop control system. Specific implementation details are as follows: The big data acquisition module adopts a distributed sensor network layout, covering 6 types of core acquisition units: ① The temperature sensor is a PT1000 high-precision sensor, which is installed at the heat source outlet, every 500m node of the main pipe, the inlet node of the branch pipe, and the supply and return water pipes of each user. The measurement range is -20℃~120℃, the accuracy is ±0.1℃, and the data is transmitted through a 4-20mA current signal. ② The pressure sensor adopts a diffused silicon pressure transmitter, which is installed at the outlet of the pipeline pump, the end of the main pipeline, and key nodes of the branch pipeline. The measurement range is 0~2.5MPa, the accuracy is ±0.01MPa, and it has overload protection function. ③ The flow sensor selected is an electromagnetic flow meter, installed on the main pipe and branch pipes with a diameter ≥100mm, with a measurement range of 0~100m. 3 / h, accuracy ±0.5%, strong anti-interference ability, suitable for high temperature heating media; ④ The energy consumption meter uses a combination of smart energy meter and heat meter to collect energy consumption data of heat source unit and circulating water pump, as well as temperature difference and heat data of supply and return water in the pipeline network. The sampling frequency is 1 time / minute. ⑤ Environmental parameter sensors are deployed in open areas within the pipeline network coverage area to collect data on ambient temperature, wind speed, and solar radiation intensity, providing a basis for predicting heat demand; ⑥ The user feedback collection unit collects user feedback on indoor temperature and heating demand adjustment information through two methods: mobile APP and indoor thermostat, and transmits the information wirelessly.

[0025] Meanwhile, the module is equipped with a data storage unit, which adopts a two-level storage architecture of edge server + cloud database. The edge server stores nearly 3 months of real-time operation data, and the cloud database stores nearly 5 years of historical data, including daily hourly energy consumption, supply and return water temperature difference, flow rate, fault records and other data. Data transmission adopts an encryption protocol to ensure data security.

[0026] The pipeline network model construction module builds a dynamic thermo-hydraulic coupled model based on the collected end-to-end data. The specific implementation process includes four stages: data preprocessing, model construction, model training, and dynamic updating. ① The data preprocessing unit first cleans the collected raw data, uses an improved Z-score algorithm to remove outliers, and calculates the Z-score for each data point. i Value, when |Z i Values ​​exceeding 3.5 are considered outliers and removed. Med(X) is the median of the data sequence, and MAD(X) is the absolute deviation of the median; then, missing data is filled in using linear interpolation, and finally, the data is standardized to map the data to the [0,1] interval to improve the model training efficiency; ② The model construction stage is based on the basic principles of fluid mechanics and thermodynamics to construct a dynamic thermo-hydraulic coupling model, which satisfies the following set of equations: ; In the formula, ρ is taken as the density of hot water, which is 998 kg / m³. 3 c p 4200 J / (kg·℃), λ is 0.6 W / (m·℃), μ is 0.001 Pa·s, f is determined based on pipe roughness and Reynolds number, and D is the actual inner diameter of the pipe; ③ The model training unit uses the gradient boosting tree algorithm to calibrate the model parameters, using preprocessed historical operating data as the training set, and using the error between the predicted temperature and pressure values ​​output by the model and the actual measured values ​​as the loss function. The model parameters are adjusted through iterative optimization to ensure that the calibration error does not exceed 5%; ④ The model update unit receives the latest operating data transmitted by the monitoring feedback module every 30 minutes and dynamically corrects the model parameters to ensure the consistency between the model and the actual operating state of the pipeline network.

[0027] The core of the balance regulation optimization module is an improved adaptive genetic algorithm. This algorithm solves a multi-objective optimization problem to obtain the optimal regulation parameters. The specific implementation process is as follows: ① Determine the control parameter vector X, including the heat source supply water temperature (60℃~95℃) and the heat source circulation flow rate (50m³ / h). 3 / h~500m 3 / h), the opening degree of the regulating valves of each branch pipe (0%~100%), and the secondary network water supply pressure of the heat exchange station (0.3MPa~1.2MPa); ② Construct a multi-objective optimization objective function: ; In the formula, E(X) is the energy consumption of the pipeline network after regulation, E0 is the baseline energy consumption (the average energy consumption of the same period in the past 3 months), ω1, ω2, and ω3 are weighting coefficients and satisfy ω1+ω2+ω3=1, T i (X) represents the actual water supply temperature at the i-th user terminal, T set Set the water supply temperature for the user end, M(X) is the maintenance cost of the pipeline network after adjustment, and M0 is the baseline maintenance cost; ③ Dynamic adjustment strategy for weighting coefficients: During the severe winter season (outdoor temperature ≤ -10℃), priority is given to ensuring the uniformity of heating, with ω2 set to 0.5~0.6, ω1 set to 0.3~0.4, and ω3 set to 0.1~0.2; During the transitional season (outdoor temperature 5℃~15℃), priority is given to reducing energy consumption, with ω1 set to 0.5~0.6, ω2 set to 0.3~0.4, and ω3 set to 0.1~0.2; During the aging period of the pipeline network (pipeline service life ≥ 15 years), priority is given to reducing maintenance costs, with ω3 set to 0.3~0.4, ω1 set to 0.3~0.4, and ω2 set to 0.2~0.3. ④ Improved implementation of adaptive genetic algorithm: The population initialization adopts the Latin hypercube sampling strategy to ensure population diversity, and the population size is set to 100; the crossover probability Pc and mutation probability Pm are dynamically adjusted to satisfy: ; ; In the formula, P c0 =0.8, P m0 =0.1, f is the individual fitness value, f avg f is the average fitness value of the population. max The optimal fitness value for the population is ε=10. -6 To avoid zero denominators, an elite retention mechanism is used during iteration, directly retaining the top 10% of individuals by fitness value to the next generation. The convergence criterion is that the rate of change of the optimal fitness value of the population is less than 10% over 50 consecutive generations. -4 Stop iterating when the optimal control parameters are reached, and output the optimal control parameters.

[0028] In addition, the module also integrates a heat demand forecasting function. It analyzes historical operating data from the past 5 years using a Long Short-Term Memory (LSTM) neural network to predict heat demand for the next 24 hours. The prediction formula is as follows: ; In the formula, τ = 1 to 24 hours, n = 48 (the historical data window length is 48 hours), Q(t) is the current heat demand, and T env (t) represents the current ambient temperature. The prediction error is controlled within 8%. The control parameters are adjusted in advance based on the prediction results to achieve predictive control.

[0029] The execution module receives the optimal control parameters output by the balance control and optimization module, and realizes pipeline network balance adjustment through various actuators, specifically including: ① Electric regulating valve: It adopts an intelligent electric ball valve with a regulating accuracy of 0.1% and a response time of ≤2s. It communicates with the optimization module through the Modbus-RTU protocol and dynamically adjusts the valve opening degree according to the regulating valve opening command to control the flow of each branch pipe; ② Variable frequency circulating water pump: Adopts a permanent magnet synchronous variable frequency pump with a speed range of 30Hz~50Hz, rated head of 20m~100m, and rated flow rate of 50m³ / h. 3 / h~500m 3 / h, the speed is adjusted by the frequency converter according to the flow command to achieve precise control of the circulating flow; ③ Plate heat exchanger: Equipped with an intelligent temperature control device, it adjusts the heat exchange area according to the secondary network water supply temperature command of the heat exchange station to ensure the stability of the secondary network water supply temperature; ④ Heat source regulating device: Adjusts the boiler combustion efficiency or heat pump operating power according to the heat source water supply temperature command to ensure that the heat source outlet temperature meets the requirements.

[0030] The monitoring and feedback module adopts a 5G+edge computing architecture to achieve real-time data transmission and dynamic evaluation of control effects. Specific implementation details are as follows: ① Real-time data transmission unit: Data transmission between various acquisition units, execution modules and core control units is realized through 5G industrial modules. The data transmission latency is ≤100ms and the transmission success rate is ≥99.9%. The edge server performs local caching and preprocessing of real-time data to reduce the computing power pressure on the cloud. ② Evaluation Unit for Control Effect: Evaluation is conducted from three dimensions: heating uniformity, energy saving rate, and user satisfaction. Heating uniformity index... The requirements are U≥0.95, energy saving rate = (E0-E(X)) / E0×100%, and user satisfaction is calculated by weighted user feedback scores; ③ Model Correction Unit: Every 30 minutes, the parameters of the dynamic thermo-hydraulic coupling model are corrected based on the latest collected operating data and evaluation results. If the evaluation index does not meet the requirements (e.g., U < 0.95), the balance control optimization module is triggered to resolve the optimal parameters to achieve closed-loop correction.

[0031] The pipeline leak detection unit uses pressure and flow data from a big data acquisition module, combined with negative pressure wave and acoustic wave detection methods, to locate leaks. Specifically, pressure sensors and acoustic wave sensors are installed every 1000m along the main pipeline. When a pressure drop Δp ≥ 0.05MPa is detected, a leak detection is initiated, and the leak location is determined using the following formula: ; Calculate the location of the leak point, where v s =1200m / s (the propagation speed of pressure wave in steel pipe), t2 is the time it takes for the pressure wave to reach the remote sensor, L0 is the distance between the two sensors, the leakage detection accuracy is ±1m, the response time is ≤30s, and after a leak is detected, an alarm is immediately sent to the monitoring feedback module, and the execution module is triggered to close the regulating valve near the leak area to reduce losses.

[0032] The pipeline network model construction module connects to the smart city heating management platform via an API interface, receiving real-time urban weather forecast data (temperature and wind speed for the next 24 hours), regional population flow data (changes in resident and transient populations), and building heating demand change data (differences in heating times for commercial and residential buildings). It then uses formulas to correct the user-set water supply temperature. ; In the formula, The corrected water supply temperature setting. This is the difference between the forecast temperature for the day and the historical average temperature for the same period. The deviation between the actual population and the baseline population of a region. This is a meteorological correction factor, with a value of 0.03. This is the population correction factor, with a value of 0.02.

[0033] The workflow of this system consists of three core stages: initialization, operation and control, and feedback and correction, as detailed below: 1. Initialization Phase: After the system starts, the big data acquisition module begins to collect operational data from the entire pipeline network, including heat source parameters, pipeline node temperature / pressure / flow rate, environmental parameters, user feedback, etc.; the pipeline network model construction module preprocesses the collected initial data, constructs a dynamic thermo-hydraulic coupling model, and completes parameter calibration; the balance control and optimization module determines the baseline energy consumption E0 and baseline maintenance cost M0 based on the initial data, and sets the initial weighting coefficients and control parameter ranges.

[0034] 2. Operation and Control Phase: The balance control and optimization module, based on real-time data acquisition and a dynamic thermo-hydraulic coupling model, solves a multi-objective optimization problem using an improved adaptive genetic algorithm, outputting optimal control parameters. The execution module adjusts the opening of the electric regulating valve, the speed of the variable frequency circulating water pump, the operating conditions of the plate heat exchanger, and the operating parameters of the heat source according to the optimal parameters to achieve balanced control of the pipeline network. At the same time, the heat demand prediction module predicts the heat demand for the next 24 hours in real time, providing a basis for predictive control.

[0035] 3. Feedback and Correction Phase: The monitoring and feedback module collects real-time pipeline operation data after regulation and control, and evaluates heating uniformity, energy saving rate, and user satisfaction; every 30 minutes, the parameters of the dynamic thermo-hydraulic coupling model are corrected based on the evaluation results; if the evaluation indicators do not meet the requirements, the balance regulation and optimization module is triggered to resolve the optimal parameters to achieve dynamic closed-loop correction; if pipeline leakage is detected, an alarm is immediately triggered and an emergency valve closure operation is executed.

[0036] To verify the effectiveness of the present invention, the following description is based on three different application scenarios. Each embodiment adopts the system structure and workflow described in the above specific embodiments, with only some parameters adjusted according to the differences in the scenarios.

[0037] Example 1: Application of heating pipe networks in residential communities during the severe winter in northern China Application Scenario: A residential community in a northern city, with a heating area of ​​100,000 square meters. The pipeline network is a branching network with an average service life of 8 years. During the severe winter, the outdoor temperature ranges from -15℃ to -5℃. The user sets the water supply temperature to T. set =80℃, requiring heating uniformity U≥0.95 and energy saving rate≥10%.

[0038] System parameter settings: weight coefficients ω1=0.35, ω2=0.55, ω3=0.1; improved adaptive genetic algorithm population size 100, maximum number of iterations 100 generations, P c0 =0.8, P m0 =0.1; the heat demand prediction window length is n=48 hours, τ=24 hours; the data transmission delay of the monitoring feedback module is ≤100ms, and the model correction cycle is 30 minutes.

[0039] Implementation Process: The big data acquisition module was deployed at 32 points, including 15 temperature sensors (1 at the heat source outlet, 6 at the main pipe nodes, 4 at the branch pipe nodes, and 4 at the user end), 8 pressure sensors, 5 flow sensors, and 2 environmental parameter sensors. The user end feedback acquisition unit covered 500 households. The pipeline network model construction module built a dynamic thermo-hydraulic coupling model based on preprocessed historical data, with a calibration error of 3.2%. The balance control and optimization module solved for the optimal control parameters using an improved adaptive genetic algorithm, outputting a heat source supply water temperature of 88℃ and a heat source circulation flow rate of 320m³.3 / h, the opening degree of the regulating valves of each branch pipe is 65%~90%, and the secondary network water supply pressure of the heat exchange station is 0.8MPa; after the execution module adjusts according to the parameters, the monitoring feedback module monitors in real time.

[0040] Implementation results: After regulation, the uniformity of heating U=0.97, which meets the requirement of U≥0.95; the energy saving rate is 12.5%, which is higher than the expected 10%; the user satisfaction rate is 96%; no pipeline leakage occurred; after model correction, the error between the predicted temperature and the actual temperature is ≤4%, achieving stable and efficient heating during the severe winter period.

[0041] Example 2: Application of Heating Pipeline Network in Commercial Complexes During Transition Season Application Scenario: A commercial complex in a southern city, with a heating area of ​​50,000 square meters. The pipeline network is a ring network with an average service life of 5 years. The outdoor temperature during the transitional season is 8℃~12℃. The commercial complex has high heating demand during the day (9:00-21:00) and low heating demand at night. The user sets the water supply temperature T. set =65℃, requiring an energy saving rate of ≥15% and a heating uniformity U≥0.95.

[0042] System parameter settings: weight coefficients ω1=0.55, ω2=0.35, ω3=0.1; improved adaptive genetic algorithm population size 100, maximum number of iterations 80 generations, P c0 =0.8, P m0 =0.1; The heat demand forecast window length is n=48 hours, τ=12 hours (forecasting in two time periods: day and night); The data transmission delay of the monitoring feedback module is ≤80ms, and the model correction cycle is 30 minutes.

[0043] Implementation Process: The big data acquisition module was deployed at 25 points, including 10 temperature sensors, 6 pressure sensors, 4 flow sensors, and 2 environmental parameter sensors. The user-end feedback acquisition unit covered 10 areas of the commercial complex. The pipeline network model construction module had a calibration error of 2.8%. Based on the predicted heat demand, the balance control and optimization module output a daytime heat source water supply temperature of 68℃ and a circulation flow rate of 280m³. 3 / h, nighttime output heat source water temperature 62℃, circulation flow rate 200m³ / h 3 / h, the opening degree of the regulating valve of each branch pipe is dynamically adjusted according to the time period; the execution module adjusts the parameters in time, and the monitoring feedback module evaluates them in real time.

[0044] Implementation results: After regulation, the uniformity of heating U=0.96, which meets the requirements; the energy saving rate is 18.3%, which is higher than the expected 15%; the average temperature in the commercial area during the day is 22℃ and at night is 18℃, which meets the heating needs of the commercial complex; the data transmission success rate is 99.95%, and the parameters after model correction have a high degree of matching with the actual operating status.

[0045] Example 3: Application of old district heating pipe networks (including leak detection) Application Scenario: An old urban area in northern China, with a heating area of ​​80,000 square meters. The pipeline network is a branching network with an average service life of 18 years, exhibiting localized aging. The outdoor temperature in winter ranges from -10℃ to 0℃. The user-side water supply temperature is set to T. set =75℃, requiring heating uniformity U≥0.95, maintenance cost reduction≥15%, and leak detection response time≤30s.

[0046] System parameter settings: weight coefficients ω1=0.35, ω2=0.25, ω3=0.4; improved adaptive genetic algorithm population size 100, maximum number of iterations 120 generations, P c0 =0.8, P m0 =0.12; The pipeline leak detection unit is equipped with pressure and acoustic sensors every 800m, the leak judgment Δp≥0.05MPa, and the leak location accuracy ±1m; The heat demand prediction window length n=72 hours, τ=24 hours.

[0047] Implementation process: The big data acquisition module deployed 30 points and added 4 sets of leak detection sensors; the pipeline network model construction module adjusted the calculation parameters of the resistance coefficient f for old pipelines, with a calibration error of 4.5%; the balance control optimization module output the optimal control parameters: heat source water supply temperature 82℃, circulation flow rate 290m³ / h. 3 / h, the opening degree of the regulating valves of each branch pipe is 70%~85%, and the secondary network water supply pressure of the heat exchange station is 0.7MPa; during operation, a main pipe leak is simulated (leakage volume 0.5m). 3 / h), triggering the leak detection unit.

[0048] Implementation results: After regulation, the uniformity of heating U=0.95, which meets the requirements; maintenance costs were reduced by 18%, higher than the expected 15%; the leak detection response time was 22s, the positioning error was 0.8m, and the emergency valve closure operation was successfully triggered, reducing leakage losses by 80%; the energy consumption saving rate was 11%, achieving safe, economical and balanced operation of the old pipeline network.

[0049] The information and parameters of the above three sets of embodiments are shown in the table below: Example one two three Application scenario core information Residential community in northern China during severe winters; heating area 100,000 square meters; branching pipe network; pipe lifespan 8 years; outdoor temperature -15℃ to -5℃; Tset=80℃ Southern transitional season commercial complex; heating area of ​​50,000 square meters; ring network; pipeline lifespan of 5 years; outdoor temperature of 8℃~12℃; Tset=65℃ Old urban area in northern China; heating area of ​​80,000 square meters; branching pipe network; pipe lifespan of 18 years; outdoor temperature -10℃ to 0℃; Tset=75℃ Key system parameters ω1=0.35, ω2=0.55, ω3=0.1; population size 100; n=48h; model correction period 30min ω1=0.55, ω2=0.35, ω3=0.1; population size 100; n=48h, τ=12h; transmission delay ≤80ms ω1=0.35, ω2=0.25, ω3=0.4; population size 100; n=72h; leakage detection Δp≥0.05MPa Key points of the core implementation process The data acquisition module has 32 data points; the model calibration error is 3.2%; the optimal parameters are: water supply temperature 88℃, flow rate 320 m³ / h, and pressure 0.8 MPa. The data acquisition module has 25 data points; the model calibration error is 2.8%; time-sharing control: daytime water supply 68℃ / flow rate 280m³ / h, nighttime 62℃ / flow rate 200m³ / h The data acquisition module has 30 data points and 4 sets of leakage sensors; the model calibration error is 4.5%; the optimal parameters are: water supply temperature 82℃, flow rate 290 m³ / h, pressure 0.7 MPa; and the simulated leakage rate is 0.5 m³ / h. Key performance indicators of implementation U=0.97; Energy saving rate 12.5%; User satisfaction 96%; Predicted temperature error ≤4% U=0.96; energy saving rate 18.3%; daytime temperature 22℃, nighttime temperature 18℃; transmission success rate 99.95% U=0.95; maintenance costs reduced by 18%; leak response time 22s, location error 0.8m; energy savings of 11%.

[0050] The heating network balance control and optimization system based on big data of this invention has the following advantages: This invention achieves precise balance control of heating networks, improving heating quality and user adaptability. Traditional heating systems generally suffer from fixed control parameters and lag response, easily leading to uneven heating and cooling in different areas and at different times, making it difficult to match differentiated heating needs. This invention achieves full-link data acquisition of the pipeline network through a distributed sensor network, accurately characterizes the network's operating status using a dynamic thermo-hydraulic coupling model, and then solves for the optimal control parameters using an improved adaptive genetic algorithm, forming a precise closed loop of acquisition-modeling-optimization-control. In terms of implementation results, under different scenarios such as the severe winter in the north, the transitional season in the south, and old urban areas, the heating uniformity index is consistently above 0.95, reaching 0.97 in residential areas. The user-end temperature prediction error is controlled within 8%, effectively solving the temperature deviation problem of traditional systems, achieving stable heating under all operating conditions, and significantly improving user satisfaction.

[0051] This invention strengthens multi-objective optimization design, significantly improving the energy efficiency and economic rationality of heating systems. Traditional heating systems often adopt an extensive operation mode, resulting in pain points such as energy redundancy and high maintenance costs. This invention constructs a multi-objective optimization function that integrates energy consumption, heating uniformity, and maintenance costs, combined with a dynamic weight adjustment strategy, to maximize core benefits in different scenarios. In the transitional season scenario of commercial complexes, due to prioritizing energy-saving optimization, the energy consumption saving rate reaches 18.3%; in the scenario of old urban areas, due to considering maintenance cost control, maintenance costs are reduced by 18%. At the same time, the system's integrated heating demand prediction function uses LSTM neural network analysis of historical data to predict heating demand for the next 24 hours, enabling advance regulation to reduce ineffective energy consumption. Furthermore, the two-level storage architecture of edge server + cloud database ensures data security, further improving the system's economic efficiency and reliability.

[0052] This invention promotes the intelligent upgrade of heating network operation and maintenance, improving operational safety and management efficiency. Traditional network operation and maintenance relies on manual inspections, which suffers from problems such as delayed leak detection, untimely fault handling, and high management costs. This invention integrates core functions such as real-time monitoring feedback, precise leak detection, and smart city integration. Through a 5G+edge computing architecture, it achieves real-time data transmission with a transmission latency controlled within 100ms and a transmission success rate exceeding 99.9%. The leak detection unit combines negative pressure wave and acoustic detection methods, with a response time of no more than 30 seconds and a positioning accuracy of ±1m. In simulated leak tests in old urban areas, it can complete the location in 22 seconds, effectively reducing leak losses. Furthermore, the system connects to the smart city heating management platform via API, integrating external data such as meteorological and population flow data to correct heating demand parameters, achieving global unified management, significantly reducing the cost of manual intervention, providing technical support for the intelligent and refined operation and maintenance of heating networks, and contributing to the upgrading of smart city infrastructure.

[0053] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A heating network balance control and optimization system based on big data, characterized in that, It includes a big data acquisition module, a pipeline network model construction module, a balance control and optimization module, an execution module, and a monitoring and feedback module; The big data acquisition module is used to collect operational data of the entire heating pipeline network. The pipeline network model construction module constructs a dynamic thermo-hydraulic coupling model based on the collected data. The balance control and optimization module solves for the optimal control parameters using a multi-objective optimization algorithm. The execution module executes pipeline balance adjustment according to the optimal control parameters. The monitoring and feedback module monitors the control effect in real time and dynamically corrects the model and parameters. The dynamic thermo-hydraulic coupling model satisfies the following formula: ； The objective function of the multi-objective optimization algorithm is: ； The multi-objective optimization algorithm employs an improved adaptive genetic algorithm, which achieves global optimal solution search by dynamically adjusting the crossover and mutation probabilities. The crossover and mutation probabilities satisfy the following: ； ； In the formula, For the density of the heating medium, For the specific heat capacity of the heating medium at constant pressure, For the temperature of the heating medium, For time, For the flow rate of the heating medium, The thermal conductivity of the heating medium, To reduce power loss due to heat dissipation in the pipeline network, The heat dissipation power of the heat exchange equipment. For pipeline pressure, For the dynamic viscosity of the heating medium, This is the friction coefficient. The inner diameter of the pipe. To adjust the parameter vector, In order to regulate the energy consumption of the pipeline network, As a baseline energy consumption, The weighting coefficients are and satisfy the following conditions: , For the first The actual water supply temperature at each user terminal. Set the water supply temperature for the user. To regulate the maintenance costs of the pipeline network, Based on the baseline maintenance cost, The initial crossover probability, The initial mutation probability, This represents the individual fitness value. The average fitness value of the population. This represents the optimal fitness value for the population. It is a local constant to avoid the denominator being zero.

2. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The big data acquisition module includes temperature sensors, pressure sensors, flow sensors, energy consumption meters, environmental parameter sensors, and user-end feedback acquisition units. The temperature sensors are installed at heat source outlets, main pipe nodes, branch pipe nodes, and user-end supply and return water inlets, with a measurement range of -20℃ to 120℃ and a measurement accuracy of no less than ±0.1℃. The pressure sensors are installed at key nodes and ends of the pipeline network, with a measurement range of 0 to 2.5MPa and a measurement accuracy of no less than ±0.01MPa. The flow sensors are electromagnetic flowmeters, installed on the main pipe and major branch pipes, with a measurement range of 0 to 100m³. 3 / h, with a measurement accuracy of not less than ±0.5%.

3. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The pipeline network model construction module also includes a data preprocessing unit, a model training unit, and a model update unit. The data preprocessing unit uses an outlier detection algorithm to remove outliers from the collected data. The outlier detection uses an improved Z-score algorithm, satisfying the formula: ； when When it is determined to be an outlier; in the formula, For a single data collection, For data sequences The median of For data sequences The median absolute deviation; the model training unit uses the gradient boosting tree algorithm to calibrate the parameters of the dynamic thermo-hydraulic coupling model, and the calibration error does not exceed 5%.

4. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The improved adaptive genetic algorithm in the balance regulation and optimization module also includes a Latin hypercube sampling strategy in the population initialization phase, an elite retention mechanism during the iteration process, and a convergence criterion. The elite retention mechanism directly retains the top 10% of individuals in the population with the highest fitness values ​​to the next generation. The convergence criterion is that the rate of change of the optimal fitness value of the population is less than 10% over 50 consecutive generations. -4 The iteration stops when the control parameter vector is reached. This includes the heat source supply water temperature, heat source circulation flow rate, the opening degree of the regulating valves in each branch pipe, and the secondary network supply water pressure of the heat exchange station, with values ​​ranging from 60℃ to 95℃ and 50℃ to 500℃ respectively. 3 / h, 0~100%, 0.3MPa~1.2MPa.

5. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The execution module includes an electric regulating valve, a variable frequency circulating water pump, a plate heat exchanger, and a heat source regulating device. The electric regulating valve is an intelligent electric ball valve with a regulating accuracy of no less than 0.1% and a response time of no more than 2 seconds. It communicates with the balance control and optimization module via the Modbus-RTU protocol. The variable frequency circulating water pump is a permanent magnet synchronous variable frequency pump with a speed range of 30Hz to 50Hz, a rated head of 20m to 100m, and a rated flow rate of 50m³ / h. 3 / h~500m 3 / h, dynamically adjust the rotation speed according to the flow command in the optimal control parameters.

6. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The monitoring feedback module includes a real-time data transmission unit, a control effect evaluation unit, and a model correction unit. The real-time data transmission unit adopts a 5G+edge computing architecture, with a data transmission latency of no more than 100ms and a data transmission success rate of no less than 99.9%. The control effect evaluation unit evaluates the performance based on three dimensions: heating uniformity index, energy saving rate, and user satisfaction. The heating uniformity index satisfies the following formula: ； In the formula, As an indicator of heating uniformity, For the number of users, The average actual water supply temperature for all users is required. ≥ The model correction unit corrects the parameters of the dynamic thermo-hydraulic coupling model every 30 minutes based on the latest collected data.

7. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The weight coefficients of the multi-objective optimization algorithm , , The system is dynamically adjusted based on seasonal changes, user demand, and pipeline operation status, specifically as follows: Winter cold season The value ranges from 0.5 to 0.

6. The value ranges from 0.3 to 0.

4. The value ranges from 0.1 to 0.2; Transitional season The value ranges from 0.5 to 0.

6. The value ranges from 0.3 to 0.

4. The value ranges from 0.1 to 0.2; Pipeline aging period The value ranges from 0.3 to 0.

4. The value ranges from 0.3 to 0.

4. The value ranges from 0.2 to 0.

3.

8. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, It also includes a pipeline leak detection unit, which, based on pressure and flow data from a big data acquisition module, combines negative pressure wave and acoustic detection methods to locate leaks using the following innovative leak judgment formula: ； In the formula, The distance from the leak point to the near-end sensor. The speed at which pressure waves propagate in the pipe. The time it takes for the pressure wave to reach the remote sensor. The value represents the change in pipeline pressure before and after the leak. The length of the pipe between the two sensors is specified, with a leak detection accuracy of no less than ±1m and a leak response time of no more than 30s.

9. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The pipeline network model construction module also interfaces with the smart city heating management platform to receive urban weather forecast data, regional population flow data, and building heating demand change data. The balance control and optimization module integrates the above external data into a multi-objective optimization algorithm, and corrects the user-end set water supply temperature in the objective function using the following formula. : ； In the formula, The corrected water supply temperature setting. This is the difference between the forecast temperature for the day and the historical average temperature for the same period. The deviation between the actual population and the baseline population of a region. This is a meteorological correction factor, with a value ranging from 0.02 to 0.

05. This is the population correction factor, with a value ranging from 0.01 to 0.

03.

10. The heating network balance control and optimization system based on big data according to claim 1, characterized in that, The big data acquisition module also stores nearly 5 years of historical pipeline operation data, including daily hourly energy consumption data, supply and return water temperature difference data, flow data, and fault record data; the pipeline model construction module analyzes the historical data through a long short-term memory neural network to predict the pipeline heat demand for the next 24 hours, using the following prediction formula: ； In the formula, For the future Real-time forecasting of heat demand, The value ranges from 1 to 24 hours. To meet the current heating demand, This represents the length of the historical data window, ranging from 24 to 72. The current ambient temperature is used for prediction, with an error not exceeding 8%. The balance control optimization module adjusts the control parameters in advance based on the prediction results to achieve predictive balance control.

Images