Heat supply pipe network dynamic balance adjusting system based on Internet of Things

Abstract

The invention relates to the technical field of centralized heating, in particular to a heat supply pipe network dynamic balance adjusting system based on the Internet of Things, which comprises a sensing layer, a network layer, a control layer and an execution layer, and is characterized in that the sensing layer acquires parameters such as pipe network pressure, temperature and flow through a low-power-consumption sensor in a full-link manner, and transmits the parameters to the network layer through a LoRaWAN protocol; the network layer preprocesses data through an edge gateway and realizes encrypted transmission by means of 5G; the control layer integrates heat supply load prediction, particle swarm optimization and a self-learning algorithm, and accurately generates an adjustment instruction; the execution layer drives the electric adjusting ball valve to complete opening adjustment, and closed-loop control is formed. The flow deviation rate can be controlled within + / -4%, the indoor temperature fluctuation is smaller than or equal to 0.5 DEG C, energy consumption is reduced by 15%-20%, remote intelligent operation and maintenance are supported, the system is suitable for various scenes such as conventional residential quarters, large commercial complexes and old communities in cold regions, heat supply balance and energy saving performance are effectively improved, and application and popularization value is remarkable.

Description

IoT-based dynamic balance regulation system for heating networks Technical Field

[0001] This invention relates to the field of centralized heating technology, specifically to a dynamic balance adjustment system for heating networks based on the Internet of Things. Background Technology

[0002] Central heating systems are an important component of urban energy supply, with their core function being to efficiently and evenly deliver heat to end users through pipelines. However, with the acceleration of urbanization and the increasing demand for heating from residents, traditional heating network regulation technologies have gradually revealed numerous intractable defects, making them incompatible with the current requirements for refined and energy-efficient heating management.

[0003] Existing heating networks mostly employ static balancing regulation, which involves manually adjusting valve openings on-site to determine a fixed flow distribution ratio, with minimal subsequent adjustments during operation. This method has significant limitations: firstly, the adjustment accuracy heavily relies on the operator's experience, making it difficult to achieve precise hydraulic balance across the entire network. This often results in overheating near-end users and underheating far-end users, leading to large fluctuations in indoor temperature and poor comfort. Secondly, it cannot respond in real-time to dynamic factors such as changes in outdoor ambient temperature and fluctuations in end-user heating load. To ensure basic heating needs for far-end users, the heating system is often forced to adopt a crude operation mode with high flow rates and small temperature differences, resulting in significant heat energy waste and high energy consumption.

[0004] In terms of data acquisition and transmission, traditional heating systems lack end-to-end sensing capabilities, with sensors only installed at a few key nodes. This makes it impossible to comprehensively acquire core operating parameters such as pipeline pressure, temperature, and flow rate. Furthermore, data transmission often relies on wired methods, resulting in complex wiring, high costs, and significant implementation difficulties in renovation scenarios such as older residential areas. While some improvement solutions have introduced simple wireless sensing technology, they suffer from short communication distances, weak anti-interference capabilities, and high data transmission latency, making it difficult to support real-time adjustment decisions.

[0005] In terms of regulation, control, and operation and maintenance management, existing technologies are mostly decentralized and manual, lacking a centralized and intelligent decision-making platform. Managers must conduct on-site inspections to troubleshoot pipeline problems, which is not only labor-intensive and inefficient, but also fails to detect hidden issues such as pipeline leaks and valve malfunctions in a timely manner, easily leading to consequences such as heating interruptions and increased heat loss. Furthermore, existing regulation systems lack self-learning and adaptive capabilities, exhibiting poor adaptability to different building types and heating scenarios. In complex scenarios such as large commercial complexes and older residential areas in cold regions, the regulation effect is significantly reduced.

[0006] Furthermore, with the advancement of dual-carbon goals, energy conservation and emission reduction have become the core development direction of the heating industry. However, the excessive energy consumption caused by the lagging regulation and low balancing accuracy of traditional heating systems is seriously inconsistent with the requirements of low-carbon development. Therefore, developing a heating network balancing and regulation system that can achieve full-link perception, dynamic and precise regulation, and intelligent operation and maintenance management, and solving the core pain points of existing technologies such as hydraulic imbalance, excessive energy consumption, and inefficient operation and maintenance, has become a technological bottleneck that the heating industry urgently needs to overcome. Summary of the Invention

[0007] The purpose of this invention is to provide a dynamic balance adjustment system for heating networks based on the Internet of Things, so as to solve the problem of excessive energy consumption caused by the existing heating systems mentioned in the background art due to adjustment lag and low balance accuracy.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a dynamic balance regulation system for heating pipe networks based on the Internet of Things, comprising a sensing layer, a network layer, a control layer, and an execution layer. The sensing layer is communicatively connected to the network layer, the network layer is communicatively connected to the control layer, and the control layer is electrically connected to the execution layer. The sensing layer is used to collect operating parameters of the entire heating pipe network, including pipe node pressure, supply and return water temperature difference, medium flow rate, outdoor ambient temperature, and building indoor temperature. The network layer is used to encrypt and transmit the operating parameters collected by the sensing layer and transmit the regulation commands output by the control layer to the execution layer. The control layer has a built-in dynamic balance regulation model. Based on the operating parameters collected by the sensing layer, the dynamic balance regulation model calculates the target heating load through a heating load prediction algorithm and generates flow regulation commands in combination with the hydraulic balance constraints of the pipe network. The calculation formula of the heating load prediction algorithm is: In the formula, For the target heating load, This is a correction factor for the heat transfer coefficient of the building envelope. The overall heat transfer coefficient of the building envelope. For the heat dissipation area of ​​the building, Set the indoor temperature. Outdoor ambient temperature This is the correction factor for heat loss in the pipeline network. For the number of pipeline branches, Let be the medium flow rate of the i-th branch. The temperature difference between the supply and return water of the i-th branch; the execution layer is used to adjust the opening of the flow regulating valves of each branch of the pipeline according to the flow regulation command output by the control layer, so as to realize the dynamic hydraulic balance of the heating pipeline network.

[0009] Preferably, the sensing layer includes pressure sensors, temperature sensors, flow sensors, and a data acquisition terminal. The pressure sensors are installed at the nodes of the main pipe, branch pipes, and end-user inlets of the pipeline network. The temperature sensors include supply and return water temperature sensors, outdoor temperature sensors, and indoor temperature sensors. The supply and return water temperature sensors are installed at the supply and return pipe ends of each branch pipe, and the indoor temperature sensors are installed in different functional areas of the building. The flow sensors are installed at the front end of the flow regulating valves of each branch pipe. All sensors are electrically connected to the data acquisition terminal.

[0010] Preferably, the network layer includes an edge gateway 5G communication module and a cloud communication module. The edge gateway is used to preprocess the operating parameters collected by the perception layer. The preprocessing includes data noise reduction and outlier removal. Data noise reduction adopts a moving average algorithm, and outlier removal adopts the 3σ criterion. The preprocessed operating parameters are transmitted to the control layer through the 5G communication module. The adjustment commands output by the control layer are transmitted to the edge gateway through the cloud communication module, and then forwarded to the execution layer by the edge gateway.

[0011] Preferably, in the preprocessing process of the edge gateway, the moving average algorithm is calculated as follows: In the formula, The data at time k is the denoised data. To adjust the sliding window size, This represents the original data collected at time kj. The value ranges from 5 to 20 and can be dynamically adjusted according to the stability of the pipeline network operation.

[0012] Preferably, the dynamic balance adjustment model of the control layer further includes a hydraulic balance deviation calculation module, which is used to calculate the deviation between the actual flow rate and the target flow rate of each branch pipe. The deviation calculation formula is: In the formula, Let be the flow deviation rate of the i-th branch pipe. Let i be the actual flow rate of the i-th branch pipe. Let i be the target flow rate of the i-th branch pipe; when At that time, the dynamic balance adjustment model generates flow adjustment commands.

[0013] Preferably, the control layer also incorporates a flow regulation command optimization algorithm. This algorithm is based on particle swarm optimization, with the objective function being minimizing the total flow deviation rate of the entire pipeline network, and the constraints being the maximum and minimum opening degrees of the flow regulating valves. The algorithm optimizes the calculation of the target opening degree of each flow regulating valve. The objective function is: The constraints are: ,in The minimum allowable flow rate for the branch pipe, This represents the maximum allowable flow rate of the branch pipe.

[0014] Preferably, the execution layer includes an electric regulating ball valve and an actuator drive module. The electric regulating ball valve is installed at key nodes of each branch pipe. The actuator drive module is electrically connected to the control layer and is used to receive the regulation command output by the control layer and drive the electric regulating ball valve to perform the opening regulation action. The opening regulation accuracy of the electric regulating ball valve is not less than 0.5%.

[0015] Preferably, the control layer further includes a data storage module and a remote interaction module. The data storage module is used to store the calculation results of the dynamic balance adjustment model of historical operating parameters collected by the perception layer and the records of adjustment instructions. The remote interaction module supports remote parameter configuration and adjustment instruction issuance through the Internet of Things and has a real-time monitoring function for operating status.

[0016] Preferably, the sensors in the sensing layer are all designed for low power consumption, supporting battery power and solar-assisted power supply. The communication protocol of the sensors adopts the LoRaWAN protocol, with a communication distance of not less than 3km, a data transmission rate of not less than 300bps, and anti-interference capability, enabling them to work stably in complex industrial electromagnetic environments.

[0017] Preferably, the dynamic equilibrium adjustment model also possesses self-learning capability, utilizing the gradient descent algorithm to... and Real-time updates are performed using the following formula: In the formula, and This is the updated correction factor. and This is the current correction factor. The learning rate is set to 0.01-0.1, ensuring that the model's prediction accuracy gradually improves over time.

[0018] Compared with the prior art, the beneficial effects of the present invention are: (1) The IoT-based dynamic balance adjustment system for heating networks of the present invention significantly improves the dynamic balance accuracy and adjustment response efficiency of heating networks, fundamentally solving the problem of hydraulic imbalance in traditional heating systems. Traditional heating networks mostly adopt static adjustment mode, relying on manual experience to set valve opening, which cannot adapt to dynamic factors such as changes in outdoor temperature and fluctuations in user heating load in real time. There are common problems of overheating at the near end and overcooling at the far end, resulting in a very poor user experience. The present invention, through the deployment of high-precision sensors across the entire sensing layer, realizes the real-time acquisition of core operating parameters such as network pressure, temperature, and flow rate. Combined with the heating load prediction algorithm and particle swarm optimization algorithm built into the control layer, it can accurately calculate the target heating load and target flow rate of each branch pipe under different operating conditions. When the flow deviation rate exceeds the threshold, the system can generate an optimization adjustment command within 30 seconds and drive the execution layer to execute, forming a closed-loop control, and stably control the flow deviation rate of the entire network within ±4%, with the indoor temperature fluctuation range not exceeding 0.5℃. Based on the data from the implementation examples, the problem of uneven heating between buildings in conventional residential communities has been completely eliminated. Large commercial complexes can accurately match the differentiated heating needs of different functional areas. The temperature fluctuation of old residential communities in cold regions, which ranged from 16℃ to 22℃, has been balanced to 19.5℃ to 20.5℃, resulting in a qualitative improvement in heating stability and comfort.

[0019] (2) The IoT-based dynamic balance regulation system for heating networks of this invention achieves precise control of heating energy consumption, significantly reducing operating costs and carbon emissions, and conforms to the industry development trend of energy conservation and emission reduction. Traditional heating systems often adopt a rough operation mode with large flow and small temperature difference to avoid insufficient heating for remote users, resulting in excessive boiler heating and serious energy waste. This invention achieves on-demand flow allocation based on real-time heating demand through intelligent regulation of the dynamic balance regulation model, avoiding ineffective heating from the source. At the same time, the model has self-learning capabilities, continuously improving the accuracy of load prediction and further optimizing energy allocation efficiency by updating correction factors α and β in real time through gradient descent algorithm. From the perspective of actual application effects, the heating energy consumption of conventional residential communities is reduced by 15%, the energy consumption of large commercial complexes is reduced by 20% due to precise adaptation to load fluctuations, and the energy consumption of old communities in cold regions is reduced by 18% while improving the uniformity of room temperature. For large-scale heating projects, this energy consumption reduction can be converted into significant economic benefits, while reducing carbon emissions from fossil energy consumption and helping to achieve dual carbon goals.

[0020] (3) The IoT-based dynamic balance adjustment system for heating networks of the present invention constructs a full-process intelligent management system, significantly improving operation and maintenance efficiency and possessing strong scenario adaptability, with wide application value. Traditional heating network management relies on manual inspection and on-site adjustment, which not only involves a large workload and high operation and maintenance costs, but also makes it difficult to detect abnormal problems such as network leakage and sensor failure in a timely manner, easily causing heating interruption losses. The present invention realizes encrypted data transmission and remote interaction through the converged communication technology of 5G and LoRaWAN. The control layer cloud platform supports real-time monitoring of operating status, remote parameter configuration and abnormal alarm functions. Managers can complete the entire process of operation and maintenance through a browser or mobile APP without on-site duty, greatly reducing labor costs. At the same time, the system has flexible configuration capabilities and can be adapted to different scenarios by adjusting sensor deployment, model parameters, etc.: for the problem of insufficient power supply in old residential areas, a lithium battery + solar auxiliary power supply mode is adopted; for the differentiated needs of multiple areas in commercial complexes, different area set temperatures can be remotely configured; for different building types, parameters such as the comprehensive heat transfer coefficient of the building envelope can be flexibly matched. In addition, the system can be compatible with existing pipe networks for renovation without the need for large-scale facility replacement, which lowers the threshold for upgrading and renovating old pipe networks and can be widely used in various heating scenarios such as residential communities, commercial complexes, and industrial plants. 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 is a system composition block diagram of the present invention; Figure 2 is a bar chart comparing the adjustment efficiency of three embodiments of the present invention; Figure 3 is a bar chart comparing the energy consumption reduction ratio of three embodiments of the present invention; Figure 4 is a bar chart comparing the cost savings of three embodiments 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] As shown in Figure 1, the IoT-based dynamic balance regulation system for heating networks provided by this invention achieves dynamic adaptive regulation of the hydraulic balance of the heating network through end-to-end data acquisition at the sensing layer, efficient encrypted transmission at the network layer, intelligent algorithm decision-making at the control layer, and precise execution regulation at the execution layer. This solves the technical problems of lagging regulation, low balancing accuracy, and excessive energy consumption in traditional heating networks. The following is a detailed description of each layer of the system: 1. The sensing layer adopts an architecture of multiple types of sensors + low-power data acquisition terminals to achieve comprehensive and real-time acquisition of heating network operating parameters. The pressure sensors are diffused silicon pressure transmitters, model PTG501, with a measurement range of 0-1.6MPa and an accuracy class of 0.2. They are installed at both ends of the main pipeline, at the connection points between each branch pipe and the main pipeline, and at the end-user's inlet pipe. One sensor is installed at each deployment point to collect pressure data from key nodes in the pipeline network in real time. The supply and return water temperature sensors are PT100 platinum resistance temperature sensors, with a measurement range of -20℃ to 120℃ and an accuracy of ±0.1℃. One sensor is installed near the supply and return water pipes of each branch pipe. One supply and return water temperature difference acquisition module is added at the building's heating inlet corresponding to the branch pipe, directly outputting the supply and return water temperature difference value; the outdoor ambient temperature sensor is a radiation-proof temperature sensor, model TR-02, installed in an unobstructed, well-ventilated high place within the heating area, the number of which is determined according to the area of ​​the heating area, one unit is set for every 100,000 square meters; the indoor temperature sensor is a wall-mounted digital temperature sensor, model DS18B20, deployed in different functional areas of the building such as living room, bedroom, and office, with at least one unit deployed in each functional area to ensure coverage of all heated spaces in the building.

[0025] The flow sensor selected is an electromagnetic flow sensor, model LDG-100, with a measurement range of 0–500 m³ / h. 3 The flow rate is measured at 0.5% per hour, with an accuracy of 0.5%. The sensors are installed at the front end of the electrically controlled regulating ball valves in each branch pipe, with a distance of at least 5 times the pipe diameter between the sensor and the regulating ball valve to avoid the impact of water flow disturbance on the flow measurement accuracy. All sensors adopt a low-power design, with core power consumption ≤10mA, supporting 3.6V lithium battery power supply. A small solar panel is also configured as an auxiliary power supply unit to ensure stable operation even during continuous rainy days. The sensor's communication module integrates a LoRaWAN chip, model SX1278, using the LoRaWAN 1.0.2 standard. The communication distance can reach 3-5km, and the data transmission rate is set to 600bps. Frequency hopping technology enhances anti-interference capabilities, making it suitable for complex electromagnetic environments such as industrial plants and residential areas.

[0026] The data acquisition terminal uses an industrial-grade low-power data acquisition unit, model DTU-600. Each acquisition terminal can connect to 8 to 16 sensor signals and connects to the sensors via an RS485 interface. The acquisition cycle can be adjusted remotely, with the default setting being 1 minute / time. After the acquisition terminal aggregates the data from each sensor, it transmits it to the edge gateway via the LoRaWAN protocol.

[0027] 2. The network layer consists of an edge gateway, a 5G communication module, a cloud communication module, and an encrypted transmission unit, realizing data preprocessing, encrypted transmission, and command forwarding. Specifically, the edge gateway uses an industrial-grade edge computing gateway, model EG-800, equipped with a quad-core ARM Cortex-A53 processor, 2GB of memory, 16GB of storage, supports multi-protocol conversion, and has a built-in LoRaWAN gateway module and 5G communication module. The LoRaWAN module receives sensor data transmitted from the data acquisition terminal, and the 5G communication module communicates with the cloud platform of the control layer through the operator's 5G network. The core function of the edge gateway is data preprocessing, including data noise reduction and outlier removal: data noise reduction uses a moving average algorithm, and the moving window size m is dynamically adjusted according to the stability of the pipeline network operation. During stable pipeline operation phases (such as nighttime), m=5 is set; during periods of greater pipeline network fluctuation (such as morning and evening peak heating times), m=20 is set. This is achieved through the formula... Calculate the denoised data, where The original data collected at time kj is used for outlier removal. The 3σ criterion is used to remove outliers. First, the mean μ and standard deviation σ of 10 consecutive sets of data collected from the same sensor are calculated. When the data x collected at a certain time satisfies |x-μ|>3σ, the data is determined to be an outlier and removed. Then, the noise-reduced data from the previous time is used to fill the outlier.

[0028] The encrypted transmission unit uses the AES-256 encryption algorithm to encrypt the transmitted data. This encryption method is used between the data acquisition terminal and the edge gateway, and between the edge gateway and the control layer cloud platform, to ensure the security and integrity of the data transmission process. The adjustment commands output by the control layer are transmitted to the edge gateway through the cloud communication module. After being decrypted, the edge gateway forwards them to the driver module of the execution layer through the LoRaWAN protocol.

[0029] 3. The control layer is the core decision-making unit of the system, consisting of a cloud platform, a dynamic balance adjustment model, a data storage module, and a remote interaction module. The cloud platform uses Alibaba Cloud ECS servers, configured with 8 cores and 16GB of memory to ensure sufficient computing and storage capabilities.

[0030] The dynamic balance regulation model is integrated into the cloud platform, developed using Python, and built on the TensorFlow framework. Its core components include a heating load prediction module, a hydraulic balance deviation calculation module, a flow regulation command optimization module, and a self-learning module. Its workflow is as follows: 1) Data input: Receives preprocessed data transmitted from the edge gateway, including network node pressure, supply and return water temperature difference in each branch pipe, medium flow rate, outdoor ambient temperature, and indoor temperature; 2) Heating load prediction: Predicts the heating load using the formula... Calculate the target heating load, where the comprehensive heat transfer coefficient K of the building envelope is determined according to the building type (e.g., K=1.5W / (㎡·℃) for brick-concrete structures, K=1.2W / (㎡·℃) for frame structures), the building heat dissipation area F is obtained through on-site measurement, and the indoor set temperature Tin,set is set to 22℃ by default and supports remote adjustment; 3) Calculate the hydraulic balance deviation: using the formula Calculate the flow deviation rate of each branch pipe, where the target flow rate qi,tar of the i-th branch pipe is determined based on the target heating load Qpred and the proportion of the building's heating area corresponding to that branch pipe; 4) Flow regulation command optimization: When the flow deviation rate |δqi| of a certain branch pipe is greater than 5%, the particle swarm optimization algorithm is activated to minimize the total flow deviation rate of the entire pipe network. The objective function is the minimum allowable flow rate qmin of the branch pipe (determined based on the minimum heat demand of the end user, default 0.5m). 3 The target opening degree of each electric regulating ball valve is optimized under the constraints of / h) and the maximum allowable flow rate qmax (determined according to pipe diameter and pressure rating). The population size of the particle swarm optimization algorithm is set to 50, the number of iterations is set to 100, and the inertia weight is set to 0.7; 5) Self-learning update: The correction factors α and β are updated in real time through the gradient descent algorithm, and the update formula is as follows. and The learning rate η is set to 0.05, and the values ​​of α and β are updated hourly to ensure that the model's prediction accuracy gradually improves.

[0031] The data storage module uses a MySQL database to store historical operating parameters collected by the sensing layer (storage period of 1 year), calculation results of the dynamic balance adjustment model, and adjustment command records. It supports data querying by time, region, and other dimensions. The remote interaction module is developed based on a B / S architecture. Users can log in to the cloud platform through a browser to adjust the indoor set temperature, configure adjustment precision parameters, monitor the operating status in real time, and view historical data. It also has an abnormal alarm function. When the pipeline pressure or temperature exceeds the set threshold, it automatically sends alarm information to the manager's mobile APP.

[0032] 4. The actuator layer consists of an electric regulating ball valve and an actuator drive module. The electric regulating ball valve is an intelligent electric regulating ball valve, model VQ977F-16C, with a diameter range of DN50~DN200. The selection is based on the diameter of the branch pipe. The opening adjustment range is 0~100%, and the adjustment accuracy is 0.3%, which is better than the 0.5% specified in the claims. The valve body is made of stainless steel, which has corrosion resistance and high temperature resistance characteristics, and is suitable for media temperatures ≤120℃.

[0033] The actuator drive module uses a DC brushless motor driver, model DR-2450, with an input voltage of 24VDC and an output current of 0-5A. It is electrically connected to the actuator of the electric regulating ball valve and communicates with the edge gateway through the LoRaWAN module. It receives the regulation commands forwarded by the edge gateway, drives the electric regulating ball valve to perform opening adjustment actions, and feeds back the actual opening information to the edge gateway after the adjustment is completed, forming a closed-loop control.

[0034] The overall workflow of the IoT-based dynamic balance regulation system for heating networks is as follows: 1) Start-up phase: After the system is powered on, each sensor completes initialization, the data acquisition terminal begins to collect operating parameters, and the edge gateway preprocesses the data and encrypts it before transmitting it to the control layer cloud platform; 2) Initialization regulation: The control layer dynamic balance regulation model calculates the target heating load and the target flow rate of each branch pipe based on the initial parameters, generates initialization regulation commands, and transmits them to the execution layer through the network layer to drive the electric regulating ball valve to adjust to the initial opening degree; 3) Dynamic regulation phase: After the system enters stable operation, the data acquisition terminal continuously collects operating parameters, and the control layer updates the target heating load and flow deviation rate every minute. When the flow deviation rate exceeds the threshold, an optimization regulation command is generated in real time to drive the execution layer to adjust the opening degree; 4) Anomaly handling phase: When the data collected by the sensors is abnormal, the network transmission is interrupted, or the actuator fails, the control layer issues an alarm message and simultaneously activates the backup regulation strategy to maintain the basic heating balance of the network until the fault is eliminated.

[0035] 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.

[0036] The following are the general outlines and data tables for the three sets of embodiments of the present invention: Example 1: Application of Heating Pipeline Network in Conventional Residential Communities 1) Application Scenario: A newly built residential community in a city, covering an area of ​​50,000 square meters, has 10 residential buildings, each with 6 units, totaling 600 households. The main heating pipeline has a diameter of DN300, and the branch pipelines have diameters of DN100 to DN150. The building type is a frame structure, with a comprehensive heat transfer coefficient of the building envelope K=1.2W / (㎡·℃). The heat dissipation area of ​​a single building is F=3000㎡, and the indoor set temperature T in,set =22℃, outdoor ambient temperature in winter is as low as -15℃.

[0037] 2) System Configuration: The sensing layer deploys 12 pressure sensors (2 on the main pipe and 10 on the branch pipe), 20 supply and return water temperature sensors (2 per building), 1 outdoor ambient temperature sensor, 600 indoor temperature sensors (1 per household), 10 flow sensors (1 per building), and 10 data acquisition terminals (1 per building); the network layer deploys 2 edge gateways (1 each in the east and west areas of the community), and the 5G communication module is Huawei ME909S-821; the control layer cloud platform is configured with 1 Alibaba Cloud ECS server, and the dynamic balance adjustment model parameters are set with an initial value of α of 0.9, an initial value of β of 0.85, a sliding window size of m=10, and a learning rate of η=0.05; the execution layer deploys 10 electric regulating ball valves (1 per building, DN125 diameter) and 10 actuator drive modules.

[0038] 3) Implementation results: After the system is in operation, the indoor temperature of each building is stable at 21.5℃~22.5℃, with a temperature fluctuation range of ≤0.5℃; the flow deviation rate of each branch pipe is controlled within ±3%, which is better than ±10% of the traditional regulation system; the overall energy consumption of the heating network is reduced by 15% compared with the traditional system, achieving a dynamic balance regulation effect.

[0039] Example 2: Application of Heating Pipeline Network in Large Commercial Complex 1) Application Scenario: A large commercial complex covers an area of ​​100,000 square meters and includes multiple functional areas such as shopping malls, office buildings, and catering areas. The main heating pipeline has a diameter of DN500, and the branch pipelines have diameters of DN150 to DN250. The building type is steel structure, and the comprehensive heat transfer coefficient of the building envelope is K=1.0W / (㎡·℃). The total heat dissipation area is F=80,000 square meters. The indoor set temperature of different functional areas is different (22℃ for shopping malls, 23℃ for office buildings, and 24℃ for catering areas). The outdoor ambient temperature is as low as -10℃ in winter, and the heat load fluctuates greatly (significant differences between weekdays and holidays, and between day and night).

[0040] 2) System Configuration: The sensing layer deploys 20 pressure sensors (2 on the main pipe and 18 on branch pipes in each functional area), 36 supply and return water temperature sensors (2 on each branch pipe), 2 outdoor ambient temperature sensors, 300 indoor temperature sensors (evenly distributed according to the functional area area), 18 flow sensors (1 on each branch pipe), and 18 data acquisition terminals; the network layer deploys 4 edge gateways (divided according to functional areas), and the 5G communication module is ZTE MC801A; the control layer cloud platform is configured with 2 Alibaba Cloud ECS servers (primary and backup architecture), the dynamic balance adjustment model parameters are set with an initial value of α 0.95 and an initial value of β 0.9, the sliding window size m is dynamically adjusted according to the operation stage (m=20 during the day on weekdays, m=10 at night, and m=15 on holidays), and the learning rate η=0.08; the execution layer deploys 18 electric regulating ball valves (diameter DN150~DN250) and 18 actuator drive modules.

[0041] 3) Implementation results: The system can accurately adjust the flow rate according to the heating demand of different functional areas, and the indoor temperature of each area is stable within the range of ±0.3℃ of the set temperature; the flow rate deviation rate is controlled within ±2%, and the response time to the fluctuation of heating load is ≤30 seconds; the energy consumption of the heating network is reduced by 20% compared with the traditional system, while reducing the amount of manual adjustment and improving management efficiency.

[0042] Example 3: Application of heating pipe network in old residential area in cold region 1) Application scenario: An old residential area in a cold region, covering an area of ​​30,000 square meters, has 8 residential buildings. The buildings were built in 1990 and are of brick-concrete structure. The comprehensive heat transfer coefficient of the building envelope is K=1.8W / (㎡·℃). The heat dissipation area of ​​a single building is F=2000㎡. The indoor set temperature Tin,set=20℃. The outdoor ambient temperature is as low as -30℃ in winter. The original pipe network has problems such as aging and leakage, and the hydraulic imbalance is serious.

[0043] 2) System Configuration: The sensing layer deploys 10 pressure sensors (2 on the main pipe and 8 on the branch pipe), 16 supply and return water temperature sensors (2 per building), 1 outdoor ambient temperature sensor, 480 indoor temperature sensors (1 per household), 8 flow sensors (1 per building), and 8 data acquisition terminals. Considering the limited power supply conditions in older residential areas, the sensors adopt a lithium battery + solar-assisted power supply mode, and the communication protocol still uses LoRaWAN. The network layer deploys 1 edge gateway, and the 5G communication module is Quectel EC200S. The control layer cloud platform is configured with 1 Alibaba Cloud ECS server, and the dynamic balance regulation model parameters are set with an initial value of α of 1.1, an initial value of β of 0.95, a sliding window size of m=15, and a learning rate of η=0.1. The execution layer deploys 8 electric regulating ball valves (DN100 diameter) and 8 actuator drive modules. At the same time, some of the original aging pipes are replaced to ensure the regulation effect.

[0044] 3) Implementation results: After the system was put into operation, it effectively solved the problem of hydraulic imbalance in old residential areas. The indoor temperature of each building was balanced from the original 16℃~22℃ to 19.5℃~20.5℃; the flow deviation rate was controlled within ±4%; due to the improved balancing effect, the heating boiler did not need to overheat, the overall energy consumption of the pipeline network was reduced by 18%, and residents' satisfaction was significantly improved.

[0045] This invention, a dynamic balance regulation system for heating pipe networks based on the Internet of Things (IoT), has the following advantages: It significantly improves the dynamic balance accuracy and regulation response efficiency of heating pipe networks, fundamentally solving the hydraulic imbalance problem of traditional heating systems. Traditional heating pipe networks mostly adopt a static regulation mode, relying on manual experience to set valve openings. This cannot adapt to dynamic factors such as changes in outdoor temperature and fluctuations in user heating load in real time, generally resulting in problems of overheating at the near end and undercooling at the far end, leading to a poor user experience. In contrast, this invention deploys high-precision sensors across the entire sensing layer, enabling real-time acquisition of core operating parameters such as pipe network pressure, temperature, and flow rate. Combined with the heating load prediction algorithm and particle swarm optimization algorithm built into the control layer, it can accurately calculate the target heating load and target flow rate of each branch pipe under different operating conditions. When the flow deviation rate exceeds the threshold, the system can generate an optimization regulation command within 30 seconds and drive the execution layer to execute it, forming a closed-loop control. This stabilizes the flow deviation rate of the entire pipe network within ±4%, and the indoor temperature fluctuation range does not exceed 0.5℃. Based on the data from the implementation examples, the problem of uneven heating between buildings in conventional residential communities has been completely eliminated. Large commercial complexes can accurately match the differentiated heating needs of different functional areas. The temperature fluctuation of old residential communities in cold regions, which ranged from 16℃ to 22℃, has been balanced to 19.5℃ to 20.5℃, resulting in a qualitative improvement in heating stability and comfort.

[0046] This invention achieves precise control over heating energy consumption, significantly reducing operating costs and carbon emissions, aligning with the industry trend of energy conservation and emission reduction. Traditional heating systems, to avoid insufficient heating for distant users, often employ a crude operating mode of high flow rate and small temperature difference, leading to excessive boiler heating and severe energy waste. This invention, through intelligent regulation using a dynamic balance adjustment model, allocates flow rate on demand based on real-time heating needs, preventing ineffective heating at the source. Simultaneously, the model possesses self-learning capabilities, continuously improving load prediction accuracy and further optimizing energy allocation efficiency by updating correction factors α and β in real time using a gradient descent algorithm. In practical applications, heating energy consumption in conventional residential communities is reduced by 15%, while large commercial complexes see a 20% reduction due to precise adaptation to load fluctuations. Older residential communities in cold regions experience an 18% reduction in energy consumption while improving room temperature uniformity. For large-scale heating projects, this energy reduction translates into significant economic benefits while simultaneously reducing carbon emissions from fossil fuel consumption, contributing to the achievement of dual-carbon goals.

[0047] This invention constructs a fully intelligent management system, significantly improving operation and maintenance efficiency and possessing strong scenario adaptability, making it widely applicable. Traditional heating network management relies on manual inspections and on-site adjustments, which is not only labor-intensive and costly, but also makes it difficult to detect abnormalities such as network leaks and sensor malfunctions in a timely manner, easily leading to heating interruptions and losses. This invention, through the convergence of 5G and LoRaWAN communication technologies, achieves encrypted data transmission and remote interaction. The control layer cloud platform supports real-time monitoring of operating status, remote parameter configuration, and anomaly alarm functions. Managers can complete the entire operation and maintenance process through a browser or mobile APP, eliminating the need for on-site duty and significantly reducing labor costs. Simultaneously, the system has flexible configuration capabilities, adapting to different scenarios by adjusting sensor deployment and model parameters: for insufficient power supply in older residential areas, a lithium battery + solar auxiliary power supply mode is adopted; for the differentiated needs of multiple areas in commercial complexes, different area temperature settings can be remotely configured; and for different building types, parameters such as the comprehensive heat transfer coefficient of the building envelope can be flexibly matched. In addition, the system can be compatible with existing pipe networks for renovation without the need for large-scale facility replacement, which lowers the threshold for upgrading and renovating old pipe networks and can be widely used in various heating scenarios such as residential communities, commercial complexes, and industrial plants.

[0048] 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 dynamic balance regulation system for heating pipe networks based on the Internet of Things, characterized in that, It includes a sensing layer, a network layer, a control layer, and an execution layer. The sensing layer is communicatively connected to the network layer, the network layer is communicatively connected to the control layer, and the control layer is electrically connected to the execution layer. The sensing layer is used to collect the operating parameters of the entire heating pipeline network, including pipeline node pressure, supply and return water temperature difference, medium flow rate, outdoor ambient temperature, and building indoor temperature. The network layer is used to encrypt and transmit the operating parameters collected by the perception layer, and to transmit the adjustment instructions output by the control layer to the execution layer. The control layer incorporates a dynamic balance regulation model. This model calculates the target heating load based on operating parameters collected by the sensing layer using a heating load prediction algorithm, and generates flow regulation commands by combining these with the hydraulic balance constraints of the pipe network. The calculation formula for the heating load prediction algorithm is as follows: In the formula, For the target heating load, This is a correction factor for the heat transfer coefficient of the building envelope. The overall heat transfer coefficient of the building envelope. For the heat dissipation area of ​​the building, Set the indoor temperature. Outdoor ambient temperature This is the correction factor for heat loss in the pipeline network. For the number of pipeline branches, Let be the medium flow rate of the i-th branch. Let be the supply and return water temperature difference for the i-th branch; The execution layer is used to adjust the opening of the flow regulating valves of each branch of the pipeline according to the flow regulation command output by the control layer, so as to realize the dynamic hydraulic balance of the heating pipeline network.

2. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The sensing layer includes pressure sensors, temperature sensors, flow sensors, and a data acquisition terminal. The pressure sensors are installed at the nodes of the main pipe, branch pipes, and end-user inlets of the pipeline network. The temperature sensors include supply and return water temperature sensors, outdoor temperature sensors, and indoor temperature sensors. The supply and return water temperature sensors are installed at the supply and return pipe ends of each branch pipe, and the indoor temperature sensors are installed in different functional areas of the building. The flow sensors are installed at the front end of the flow regulating valves of each branch pipe. All sensors are electrically connected to the data acquisition terminal.

3. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The network layer includes an edge gateway 5G communication module and a cloud communication module. The edge gateway is used to preprocess the operating parameters collected by the perception layer. The preprocessing includes data noise reduction and outlier removal. Data noise reduction adopts a moving average algorithm, and outlier removal adopts the 3σ criterion. The preprocessed operating parameters are transmitted to the control layer through the 5G communication module. The adjustment commands output by the control layer are transmitted to the edge gateway through the cloud communication module, and then forwarded to the execution layer by the edge gateway.

4. The IoT-based dynamic balance adjustment system for heating networks according to claim 3, characterized in that, During the preprocessing process of the edge gateway, the formula for calculating the moving average algorithm is: In the formula, The data at time k is the denoised data. To adjust the sliding window size, This represents the original data collected at time kj. The value ranges from 5 to 20 and can be dynamically adjusted according to the stability of the pipeline network operation.

5. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The dynamic balance adjustment model of the control layer also includes a hydraulic balance deviation calculation module. This module is used to calculate the deviation between the actual flow rate and the target flow rate of each branch pipe. The deviation calculation formula is: In the formula, Let be the flow deviation rate of the i-th branch pipe. Let i be the actual flow rate of the i-th branch pipe. Let i be the target flow rate of the i-th branch pipe; when At that time, the dynamic balance adjustment model generates flow adjustment commands.

6. The IoT-based dynamic balance adjustment system for heating networks according to claim 5, characterized in that, The control layer also incorporates a flow regulation command optimization algorithm. This algorithm is based on particle swarm optimization, with the objective function being minimizing the total flow deviation rate of the entire pipeline network, and the constraints being the maximum and minimum opening degrees of the flow regulating valves. It optimizes the calculation of the target opening degree of each flow regulating valve. The objective function is: The constraints are: ,in The minimum allowable flow rate for the branch pipe, This represents the maximum allowable flow rate of the branch pipe.

7. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The execution layer includes an electric regulating ball valve and an actuator drive module. The electric regulating ball valve is installed at key nodes of each branch pipe. The actuator drive module is electrically connected to the control layer and is used to receive the regulation command output by the control layer and drive the electric regulating ball valve to perform the opening adjustment action. The opening adjustment accuracy of the electric regulating ball valve is not less than 0.5%.

8. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The control layer also includes a data storage module and a remote interaction module. The data storage module is used to store the calculation results of the dynamic balance adjustment model of historical operating parameters collected by the perception layer and the records of adjustment commands. The remote interaction module supports remote parameter configuration and adjustment command issuance through the Internet of Things and has real-time monitoring function for operating status.

9. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The sensors in the sensing layer all adopt a low-power design, supporting battery power and solar-assisted power supply. The communication protocol of the sensors adopts the LoRaWAN protocol, with a communication distance of not less than 3km, a data transmission rate of not less than 300bps, and anti-interference capability, and can work stably in complex industrial electromagnetic environments.

10. The IoT-based dynamic balance adjustment system for heating networks according to claim 1, characterized in that, The dynamic equilibrium adjustment model also possesses self-learning capabilities, utilizing the gradient descent algorithm to... and Real-time updates are performed using the following formula: ； In the formula, and This is the updated correction factor. and This is the current correction factor. The learning rate is set to 0.01-0.1, ensuring that the model's prediction accuracy gradually improves over time.

Images