Centralized heating method and system based on room temperature feedback regulation

Abstract

The invention provides a centralized heating method and system based on room temperature feedback regulation, and the method comprises the steps: collecting and verifying real-time room temperature data, and generating a room temperature feedback signal; the signals are transmitted to a central control unit for sequential processing, and a temperature-time two-dimensional data matrix is generated; a preset dynamic compensation algorithm is used for calculating a heat demand correction value of each heat supply pipe network key node; the heat supply pipeline flow is dynamically adjusted according to the correction value; and verifying the effect through a closed-loop feedback mechanism and optimizing algorithm parameters. According to the method, a network is constructed to realize multi-node synchronous acquisition and verification, and heat distribution heterogeneity and fluctuation characteristics are captured in combination with matrix processing and interpolation completion. Compared with a traditional single-point sensor, according to the scheme, through two-dimensional mapping and anomaly detection, the adjustment response time is shortened to the sampling period level, the temperature deviation calculation precision is improved to + / -0.1 DEG C level, the overheating-underheating alternating phenomenon is solved, and accurate dynamic matching of the heat demand correction value is achieved.

Description

Technical Field

[0001] This invention relates to the field of centralized heating technology, and in particular to a centralized heating method and system based on room temperature feedback regulation. Background Technology

[0002] Traditional centralized heating systems typically employ static flow distribution and fixed temperature settings. They deliver pre-set amounts of hot water or steam to various areas via a central pumping station, relying on manual experience or simple linear models to set heating parameters. Their core principle is based on a unidirectional heat transfer chain from heat source to network to user. Fixed valves or balancing valves installed at key nodes regulate network flow to maintain basic heating needs. Temperature monitoring often uses single-point sensors, resulting in low data acquisition frequency and a lack of spatiotemporal integration. This limits time-series processing capabilities and makes dynamic matching of heat loads across multiple areas difficult. The adjustment logic is driven by a simple deviation between the setpoint and actual value, lacking a closed-loop optimization mechanism based on real-time room temperature feedback, making it difficult to balance heating efficiency with user comfort.

[0003] The existing technology has two major flaws: First, the adjustment lag and accuracy are insufficient. Due to the lack of a three-dimensional room temperature sensing network and dynamic data matrix, it is impossible to capture the spatial heat distribution heterogeneity and temporal fluctuations in real time, resulting in frequent "overheating-underheating" alternation phenomena. Second, there is an imbalance between energy efficiency and safety. The static algorithm does not integrate environmental compensation factors, time compensation factors and spatial compensation factors, and cannot dynamically adapt to diurnal temperature differences, seasonal changes and user behavior patterns, resulting in energy waste and equipment overload risks. Summary of the Invention

[0004] This invention aims to at least solve the technical problems of insufficient adjustment lag and precision in the prior art, and innovatively proposes a centralized heating method and system based on room temperature feedback regulation.

[0005] To achieve the above-mentioned objectives of this invention, this invention provides a centralized heating method based on room temperature feedback regulation, the method comprising: S1. Construct a three-dimensional room temperature sensing network to collect real-time room temperature data, verify the real-time room temperature data, and generate a room temperature feedback signal; S2. The room temperature feedback signal is transmitted to the central control unit via the wireless communication module for time-series processing to generate a temperature-time two-dimensional data matrix. S3. Based on the temperature-time two-dimensional data matrix, calculate the heat demand correction value for each heating sub-region using a preset dynamic compensation algorithm; S4. Install adjustable proportional valves at key nodes of the heating network, and control the flow rate of the corresponding heating pipeline to be dynamically adjusted according to the heat demand correction value. S5. Verify the adjustment effect through a closed-loop feedback mechanism, and optimize the parameters of the dynamic compensation algorithm based on the verification results.

[0006] Furthermore, generating the room temperature feedback signal in step S1 includes: S101. Based on the spatial characteristics of the heating area, deploy a data acquisition array to form a three-dimensional room temperature sensing network; S102. Based on the three-dimensional room temperature sensing network, real-time room temperature data is collected synchronously by multiple nodes through the ZigBee wireless protocol. S103. Perform physical range verification and spatiotemporal consistency verification on the real-time room temperature data, and use the weighted average method to correct the deviation data. S104. Sort the verified valid real-time room temperature data by timestamp, generate a standardized data packet containing temperature, humidity and light values, and attach the collector ID, collection time and spatial coordinate triplet information to form a room temperature feedback signal with spatiotemporal identification.

[0007] Furthermore, generating the temperature-time two-dimensional data matrix in step S2 includes: S201. Based on the data characteristics of the room temperature feedback signal, a self-organizing network is constructed, and a star-tree hybrid topology is used to realize the aggregation and transmission of multi-node data within the heating area, and encapsulation is performed. S202, The central control unit performs time-series alignment processing and dynamic interpolation completion on the room temperature feedback signal to generate a continuous time series; S203. The continuous time series is mapped in two dimensions according to the time dimension and the spatial dimension, and a temperature-time two-dimensional data matrix is ​​constructed with the sampling period as the row index and the collector ID as the column index. The elements in the temperature-time two-dimensional data matrix are the room temperature feedback signals in the continuous time series. S204. Abnormal room temperature feedback signals in the temperature-time two-dimensional data matrix are detected by box plot method, verified by combining with the spatial heat distribution model of the heating area, and the final temperature-time two-dimensional data matrix is ​​output after correcting the abnormal data.

[0008] Furthermore, the calculation of the heat demand correction value for each heating sub-region in step S3 includes: S301. Based on the temperature-time two-dimensional data matrix, according to the heating network topology and heat load distribution characteristics, the heating area is divided into N sub-regions. Each sub-region corresponds to a set of continuous data acquisition node clusters in the temperature-time two-dimensional data matrix, forming a control unit. S302. Combining user comfort requirements and energy-saving standards, a dynamic target temperature curve is set for each of the control units. The curve parameters include the diurnal temperature difference compensation coefficient, seasonal adjustment factor and historical energy consumption benchmark, which serve as the control benchmark. S303. The actual room temperature data in the control unit is compared with the target temperature curve in real time to calculate the temperature deviation value and the deviation change rate, and to form a dynamic deviation sequence. S304. Based on the dynamic deviation sequence, a PID control algorithm is adopted, with the temperature deviation as the input, and combined with integral separation and anti-saturation mechanism, to calculate the basic heat demand correction coefficient. S305. Introduce environmental compensation factors, time compensation factors, and spatial compensation factors, and use a weighted fusion algorithm to perform multidimensional compensation on the basic heat demand correction coefficient to generate a comprehensive heat demand correction value. S306. Verify the rationality of the comprehensive heat demand correction value through the heat balance equation, and combine the hydraulic model of the heating network to simulate the adjusted flow-pressure drop relationship to ensure that the comprehensive heat demand correction value is within the safe operating threshold of the equipment, thus forming a safety constraint. S307. The comprehensive heat demand correction value is smoothed and filtered to eliminate high-frequency fluctuations and form a continuous and smooth heat demand correction sequence.

[0009] Furthermore, the expressions for calculating the temperature deviation value and the rate of change of deviation in step S303 are as follows:

[0010]

[0011]

[0012] in, Indicates time Time control unit Temperature deviation value, Indicates control unit In time The dynamic target temperature value, Indicates control unit In time The actual room temperature value, This indicates the time interval step size when calculating the rate of change of deviation. Indicates time Time control unit Temperature deviation value, This represents the minimum temperature deviation value. This indicates the maximum temperature deviation value.

[0013] Furthermore, the expression for the weighted fusion algorithm in step S305 is as follows:

[0014]

[0015]

[0016]

[0017] in, Indicates control unit In time The comprehensive heat demand correction value, Indicates control unit In time The basic heat requirement correction factor Indicates control unit Spatial weighting coefficients, Indicates control unit In time The base value of the compensation factor, Indicates the dynamic adjustment coefficient. Indicates control unit In time The compensation factor deviation value, Indicates control unit In time Environmental compensation factors Indicates time outdoor temperature, Indicates control unit In time The actual indoor temperature Indicates the design temperature difference. Indicates control unit In time Time compensation factor, Indicates the start time of the peak period. Indicates the duration of the peak period. Indicates control unit In time Spatial compensation factor, Indicates control unit Thermal inertia coefficient This represents the system's maximum thermal inertia.

[0018] Furthermore, the formation of safety constraints in step S306 includes: S3061. Based on the comprehensive heat demand correction value, the stability of the adjusted indoor temperature is verified by the dynamic heat balance equation. S3062. Combining the aforementioned hydraulic model of the heating network, the changes in pipeline flow and pressure after adjustment are simulated using the flow-pressure drop relationship equation; S3063. Based on the dual verification results of the dynamic heat balance equation and the hydraulic model of the heating network, set the temperature safety threshold and the equipment safety threshold to form a multi-dimensional safety constraint on the comprehensive heat demand correction value. S3064. If the verification does not meet the safety constraints, the closed-loop feedback mechanism is triggered to dynamically adjust the parameters of the dynamic compensation algorithm and recalculate the heat demand correction value until all constraints are met.

[0019] Furthermore, in step S4, an adjustable proportional valve is installed at a key node of the heating network. Based on the required heat correction value, the adjustable proportional valve is used to dynamically adjust the flow rate of the corresponding heating pipeline. S401. Convert the heat demand correction value into the target flow rate adjustment amount; S402. Based on the target flow rate adjustment, calculate the valve opening adjustment using the valve flow characteristic curve and the flow-pressure drop relationship equation. S403. Based on the valve opening adjustment amount, control the corresponding adjustable proportional valve to open.

[0020] Furthermore, in step S5, the adjustment effect is verified through a closed-loop feedback mechanism, and the parameters of the dynamic compensation algorithm are optimized based on the verification results, including: S501: The room temperature sensor installed in the control unit collects the adjusted indoor temperature data in real time. S502. Compare the collected real-time indoor temperature data with the set temperature safety threshold, and calculate the temperature deviation value and the deviation change rate. S503. Based on the temperature deviation value and the rate of change of deviation, combined with historical adjustment data, the parameters of the basic heat demand correction coefficient, spatial weight coefficient and dynamic adjustment coefficient in the dynamic compensation algorithm are iteratively optimized through parameter optimization algorithm. S504. Update the optimized parameters into the dynamic compensation algorithm, recalculate the comprehensive heat demand correction value, and trigger a new round of flow regulation and safety verification process until the indoor temperature stabilizes within the set temperature safety threshold range.

[0021] In another aspect, the present invention also provides a centralized heating system based on room temperature feedback regulation, the system comprising: processor; Memory used to store processor-executable instructions; The processor is configured to implement the centralized heating method based on room temperature feedback regulation when executing the executable instructions.

[0022] The beneficial effects of this invention are as follows: This invention achieves multi-node synchronous data acquisition and spatiotemporal consistency verification by constructing a three-dimensional room temperature sensing network. Combined with the time-series processing and dynamic interpolation completion of the temperature-time two-dimensional data matrix, it can capture the spatial heat distribution heterogeneity and temporal fluctuation characteristics of the heating area in real time. Compared with the low-frequency data acquisition of traditional single-point sensors, this solution shortens the adjustment response time to the sampling cycle level (e.g., 5 minutes / time) through continuous time series two-dimensional mapping and box plot anomaly detection, and improves the temperature deviation calculation accuracy to ±0.1℃, effectively solving the alternating phenomenon of "overheating-underheating" and achieving accurate dynamic matching of the heat correction values ​​required by each control unit.

[0023] By introducing a three-dimensional compensation mechanism that incorporates environmental compensation factors (such as outdoor temperature), time compensation factors (such as peak duration), and spatial compensation factors (such as thermal inertia coefficient), combined with verification of the heat balance equation and simulation of the flow-pressure drop relationship equation, it can dynamically adapt to diurnal temperature differences, seasonal changes, and user behavior patterns. For example, by increasing the thermal inertia coefficient compensation value at night in winter, ineffective heating energy consumption can be reduced by 10%-15%. At the same time, by setting safety constraints and introducing valve opening safety adjustment factors, it ensures that pipeline pressure drop does not exceed the threshold and flow rate is not lower than the minimum allowable value, achieving a dual guarantee of energy efficiency improvement and safe equipment operation.

[0024] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0025] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a centralized heating method based on room temperature feedback regulation according to the present invention. Detailed Implementation

[0026] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0027] Example 1 As shown in the figure, a centralized heating method based on room temperature feedback regulation includes: S1. Construct a three-dimensional room temperature sensing network to collect real-time room temperature data, verify the real-time room temperature data, and generate a room temperature feedback signal; S2. The room temperature feedback signal is transmitted to the central control unit via the wireless communication module for time-series processing to generate a temperature-time two-dimensional data matrix. S3. Based on the temperature-time two-dimensional data matrix, calculate the heat demand correction value for each heating sub-region using a preset dynamic compensation algorithm; S4. Install adjustable proportional valves at key nodes of the heating network, and control the flow rate of the corresponding heating pipeline to be dynamically adjusted according to the heat demand correction value. S5. Verify the adjustment effect through a closed-loop feedback mechanism, and optimize the parameters of the dynamic compensation algorithm based on the verification results.

[0028] The principle of a centralized heating method based on room temperature feedback regulation in this embodiment is as follows: By constructing a three-dimensional room temperature sensing network, this network can cover all key nodes in the heating area, enabling multi-node synchronous acquisition of room temperature data. The acquired real-time room temperature data first undergoes spatiotemporal consistency verification to ensure the accuracy and reliability of the data, thereby generating an effective room temperature feedback signal. This feedback signal is then rapidly transmitted to the central control unit via a wireless communication module.

[0029] In the central control unit, the room temperature feedback signal undergoes time-series processing to form a two-dimensional temperature-time data matrix. This matrix not only records the room temperature data at various times but also reflects the trend of room temperature change over time.

[0030] Based on a two-dimensional temperature-time data matrix, the pre-defined dynamic compensation algorithm comprehensively considers environmental compensation factors (such as outdoor temperature), time compensation factors (such as peak period duration), and spatial compensation factors (such as thermal inertia coefficient) to calculate the heat demand correction value for each heating sub-region. This correction value can accurately reflect the actual heating demand of each region.

[0031] Adjustable proportional valves are installed at key nodes in the heating network. Based on the calculated heat demand correction value, the central control unit controls these adjustable proportional valves to dynamically adjust the flow rate of the corresponding heating pipelines. By adjusting the valve opening, the heat entering each heating sub-area can be precisely controlled, thereby achieving accurate matching of heating demand.

[0032] During the adjustment process, the closed-loop feedback mechanism continues to function. Through a room temperature sensor installed within the control unit, the system can collect real-time indoor temperature data after adjustment and compare it with the set temperature safety threshold. If the temperature deviation exceeds the allowable range, the system will immediately trigger the dynamic adjustment mechanism, recalculate the required heat correction value, and adjust the valve opening until the indoor temperature stabilizes within the set range.

[0033] Meanwhile, based on the verification results, the system will also optimize the parameters of the dynamic compensation algorithm. By continuously adjusting parameters such as the basic heat demand correction coefficient, spatial weight coefficient, and dynamic adjustment coefficient in the algorithm, the system can gradually improve the accuracy and stability of the adjustment, further reduce ineffective heating energy consumption, and improve overall heating efficiency.

[0034] As an optional embodiment of the present invention, optionally, generating the room temperature feedback signal in step S1 includes: S101. Based on the spatial characteristics of the heating area, deploy a data acquisition array to form a three-dimensional room temperature sensing network; It should be explained in detail that, in step S101, the construction of the three-dimensional room temperature sensing network needs to be deployed in layers based on the building structure, thermal coupling characteristics, and user distribution density of the heating area: In the vertical dimension, data collection points are set up in typical rooms on each floor (such as the center point of the living room); in the horizontal dimension, grids are divided according to heating sub-areas (such as unit buildings or independent buildings), and temperature collectors with three-dimensional spatial coordinate identifiers (x, y, z) are deployed at the grid intersections; for special thermally inertial areas (such as the top floor, corner rooms, and sides of large glass curtain walls), the collection points need to be densely deployed and associated with their thermal inertia intensity coefficient. All data collectors are networked in a star topology, with millisecond-level timestamp synchronization achieved by the regional gateway, ensuring that the spatial topology correlation of the collected data is strictly aligned with the timeline.

[0035] S102. Based on the three-dimensional room temperature sensing network, real-time room temperature data is collected synchronously by multiple nodes through the ZigBee wireless protocol. It should be explained in detail that in step S102, based on the three-dimensional room temperature sensing network, multi-node synchronous acquisition of real-time room temperature data is performed through the ZigBee wireless protocol. Specifically, this includes: utilizing the star topology of the ZigBee network to ensure that each temperature acquisition device operates in a low-power mode, achieving millisecond-level timestamp synchronization; setting the acquisition frequency to once every 5 minutes, and avoiding wireless interference through an adaptive channel frequency hopping mechanism to ensure data transmission reliability of over 99.9%; simultaneously, combining a signal strength indicator (RSSI) for data packet verification, filtering abnormal signals in real time, and ensuring the spatiotemporal consistency and integrity of the acquired data; in addition, deploying edge computing nodes to preprocess the raw data, compressing transmission bandwidth usage, and optimizing the processing efficiency of the central control unit.

[0036] S103. Perform physical range verification and spatiotemporal consistency verification on the real-time room temperature data, and use the weighted average method to correct the deviation data. It should be explained in detail that in step S103, the physical range verification includes: setting the effective range of the room temperature sensor to -10 degrees Celsius to 10 degrees Celsius. If the collected data exceeds this range, it is determined to be a hardware malfunction and the data point is discarded. The spatiotemporal consistency verification includes: based on the spatial coordinates and timestamp of the collector, verifying whether the temperature difference between adjacent nodes is within a preset threshold (e.g., ±1.5 degrees Celsius) through a spatial correlation algorithm (e.g., K-nearest neighbor method), and checking the continuity of the time series. If the temperature change of adjacent sampling points exceeds ±2 degrees Celsius / minute, an abnormal alarm is triggered. For deviation data correction, a weighted average method combining spatial weight coefficients (e.g., inverse weighting based on the distance between the collection point and the center of the target area) and time weight coefficients (e.g., exponential decay weighting) is used to generate a corrected room temperature data series to ensure the reliability of the data.

[0037] S104. Sort the verified valid real-time room temperature data by timestamp, generate a standardized data packet containing temperature, humidity and light values, and attach the collector ID, collection time and spatial coordinate triplet information to form a room temperature feedback signal with spatiotemporal identification.

[0038] It is important to note that the valid real-time room temperature data verified in step S104 undergoes rigorous timestamp sorting to ensure the accurate temporal order of each data point. Subsequently, the system integrates this data with simultaneously collected humidity and illumination values ​​to generate a standardized data packet containing multiple environmental parameters. To enhance data traceability and spatial correlation, each data packet also includes a unique ID of the data acquisition device, precise acquisition time (accurate to the second), and a three-dimensional spatial coordinate (x, y, z) triplet. Through this structured data encapsulation, the system can generate a room temperature feedback signal with spatiotemporal identification, providing an accurate and reliable data foundation for subsequent temperature-time two-dimensional data matrix construction and dynamic compensation algorithm calculations. This processing flow not only ensures data integrity and consistency but also significantly improves the system's ability to capture the heterogeneity of spatial heat distribution in the heating area.

[0039] As an optional embodiment of the present invention, optionally, generating the temperature-time two-dimensional data matrix in step S2 includes: S201. Based on the data characteristics of the room temperature feedback signal, a self-organizing network is constructed, and a star-tree hybrid topology is used to realize the aggregation and transmission of multi-node data within the heating area, and encapsulation is performed. It is important to explain in detail that in step S201, based on the multi-node, spatiotemporally correlated data characteristics of the room temperature feedback signal, the system needs to construct a self-organizing network that balances transmission efficiency and data integrity. Specifically, a star-tree hybrid topology is adopted: at the regional level, a star network is constructed with the central control unit as the core, and each control unit directly accesses through a wireless gateway to achieve low-latency backbone transmission; at the sub-regional level, tree branches are divided according to the heating grid, and multiple temperature collectors are connected to the end of each branch to form a hierarchical data aggregation path. This structure automatically optimizes the transmission path through a dynamic routing algorithm. When a node experiences a communication failure, adjacent nodes can temporarily take over the data forwarding task to ensure network robustness. In the data encapsulation stage, the system integrates the collector ID, timestamp, spatial coordinates, and parameters such as room temperature, humidity, and light intensity into a standard data frame, and ensures data integrity through a checksum mechanism, ultimately forming a raw data stream that can be processed in a time-series manner. This design not only solves the bandwidth contention problem of large-scale node access but also improves data reliability through redundant transmission paths.

[0040] S202, The central control unit performs time-series alignment processing and dynamic interpolation completion on the room temperature feedback signal to generate a continuous time series; It should be explained in detail that in step S202, the central control unit first performs time-series alignment processing on the received room temperature feedback signal to ensure that data from different acquisition nodes with different timestamps can be arranged according to a unified time base. This step can eliminate analysis errors caused by asynchronous data acquisition times. Subsequently, for time points missing due to communication failures or data anomalies, the system adopts dynamic interpolation completion technology. This technology, based on valid data from adjacent time points and combined with the local variation trend of the time series, estimates missing values ​​through linear interpolation, spline interpolation, or more complex time series prediction models (such as the ARIMA model). This dynamic interpolation method can not only fill in data gaps but also maintain the continuity and smoothness of the time series. Through time-series alignment and dynamic interpolation completion, the system finally generates a continuous, missing-free time series that records the room temperature changes of key nodes in the heating area at different times.

[0041] S203. The continuous time series is mapped in two dimensions according to the time dimension and the spatial dimension, and a temperature-time two-dimensional data matrix is ​​constructed with the sampling period as the row index and the collector ID as the column index. The elements in the temperature-time two-dimensional data matrix are the room temperature feedback signals in the continuous time series. It is important to explain in detail that in step S203, the continuous time series, after time alignment and dynamic interpolation, is subjected to a refined two-dimensional mapping according to the time and spatial dimensions. Specifically, a preset sampling period (e.g., 5 minutes) is used as the row index to form a discrete scale on the time axis; the unique ID of each temperature collector is used as the column index to correspond to spatial nodes within the heating area. Through this mapping method, each matrix element is precisely associated with the room temperature feedback signal at a specific time and location, while preserving the correlation of auxiliary parameters such as humidity and illumination. This matrix structure not only realizes the vertical expansion of the time series but also completes the horizontal expansion of the spatial distribution through column indices, forming a data carrier with spatiotemporal dual-dimensional analytical capabilities. In practical applications, matrix elements can be dynamically updated to reflect real-time thermal state changes, providing multi-dimensional input support for subsequent dynamic compensation algorithms, ensuring that the calculation of the required heat correction value can simultaneously consider the temporal evolution pattern and spatial distribution characteristics.

[0042] S204. Abnormal room temperature feedback signals in the temperature-time two-dimensional data matrix are detected by box plot method, verified by combining with the spatial heat distribution model of the heating area, and the final temperature-time two-dimensional data matrix is ​​output after correcting the abnormal data.

[0043] It should be explained in detail that in step S204, the system first performs box plot analysis on all data from all collectors within each sampling period, calculating the first quartile, the third quartile, and the interquartile range. When the room temperature value at a certain collection point exceeds the range, the data is marked as an outlier. Subsequently, based on a pre-set spatial heat distribution model of the heating area (including building thermal inertia parameters, historical heat conduction characteristics, and the topological relationship of adjacent nodes), the spatial rationality of the outlier data points is verified: if the temperature gradient between the outlier value and its spatially associated nodes (collectors within a 5-meter radius) exceeds ±2.5℃ / meter, or the temperature deviation from the average temperature of the same elevation layer reaches ±3℃, it is determined to be a genuine anomaly. For the confirmed outlier data, the inverse distance weighted interpolation (IDW) method is used in conjunction with the valid data of adjacent nodes for correction, and finally, a spatiotemporally continuous and complete temperature-time two-dimensional data matrix is ​​output.

[0044] As an optional embodiment of the present invention, optionally, calculating the heat demand correction value for each heating sub-region in step S3 includes: S301. Based on the temperature-time two-dimensional data matrix, according to the heating network topology and heat load distribution characteristics, the heating area is divided into N sub-regions. Each sub-region corresponds to a set of continuous data acquisition node clusters in the temperature-time two-dimensional data matrix, forming a control unit. It should be explained in detail that in step S301, the division process needs to be based on the physical connection topology of the heating network (such as branch pipeline structure and valve layout) and historical heat load distribution data (such as heat consumption per unit area and user density). The collector nodes are grouped into continuous clusters according to spatial proximity and thermal similarity using a clustering algorithm (such as K-means spatial clustering). The boundary of each sub-region needs to ensure that the variance of the thermal inertia coefficient is lower than a preset threshold (such as ±0.5) and mapped to an independent control unit to achieve spatial fine segmentation of heating demand.

[0045] S302. Combining user comfort requirements and energy-saving standards, a dynamic target temperature curve is set for each of the control units. The curve parameters include the diurnal temperature difference compensation coefficient, seasonal adjustment factor and historical energy consumption benchmark, which serve as the control benchmark. It should be explained in detail that in step S302, a dynamic target temperature curve is set for each of the control units. The curve parameters include the diurnal temperature difference compensation coefficient, the seasonal adjustment factor, and the historical energy consumption benchmark, which serve as the control benchmark. The specific setting process is as follows: First, based on user comfort survey data and local building energy conservation standards, a basic target temperature range (e.g., 18-22℃) is determined. Second, the diurnal temperature difference compensation coefficient is embedded, and the target temperature is automatically lowered by 0.5-2.0℃ during nighttime hours (e.g., 23:00-6:00), with the specific value determined based on the sub-region thermal inertia intensity coefficient. The system dynamically adjusts the target temperature by ±0.3-1.0℃, incorporating a seasonal adjustment factor and combining real-time weather forecasts of outdoor temperature with historical energy consumption data for the same period. Finally, it uses the historical energy consumption benchmark of the control unit as a constraint. If the real-time calculated predicted energy consumption exceeds 105%, a smooth downward shift mechanism for the target temperature curve is automatically triggered, shifting the temperature by 0.1-0.5℃ until the predicted energy consumption returns to a reasonable range, ensuring that the overall energy consumption meets the preset energy-saving standards. This dynamic target temperature curve is continuously optimized and updated every 24 hours based on the latest user feedback, weather forecasts, and energy consumption data.

[0046] S303. The actual room temperature data in the control unit is compared with the target temperature curve in real time to calculate the temperature deviation value and the deviation change rate, and to form a dynamic deviation sequence. It should be explained in detail that, in step S303, the real-time comparison process includes: the central control unit synchronously acquires the actual room temperature data of each collector in the control unit at a preset sampling period (e.g., 5 minutes), and compares it with the corresponding value of the target temperature curve at the current moment; calculates the temperature deviation value (defined as actual temperature minus target temperature), in degrees Celsius, and simultaneously calculates the deviation change rate based on the time series difference method; all deviation values ​​and change rates are sorted by timestamp to form a dynamic deviation sequence containing deviation values, change rates, and time points, which serves as the input basis for subsequent calculation of the required heat correction value. During the calculation process, if the deviation value exceeds a preset safety threshold, or the change rate exceeds a critical threshold, the system automatically marks the abnormal state and triggers diagnostic log recording to ensure the reliability and real-time performance of the sequence data.

[0047] S304. Based on the dynamic deviation sequence, a PID control algorithm is adopted, with the temperature deviation as the input, and combined with integral separation and anti-saturation mechanism, to calculate the basic heat demand correction coefficient. It should be explained in detail that in step S304, a PID control algorithm is used based on the dynamic deviation sequence, with the temperature deviation as the input, and a basic heat demand correction coefficient is calculated by combining integral separation and anti-saturation mechanisms. Specifically, this includes: First, defining the proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd of the PID controller, where Kp is used for rapid response to temperature deviation, Ki is used to eliminate steady-state error, and Kd is used to suppress temperature fluctuations; The integral separation mechanism is set to automatically disable the integral term when the absolute value of the temperature deviation exceeds a preset threshold (e.g., ±1.0℃) to avoid overshoot due to integral accumulation; The anti-saturation mechanism constrains the controller output range (e.g., 0.8 to 1.2) through an output limiter, and when the output value approaches the limit, the integral term is dynamically adjusted using integral anti-saturation logic (e.g., back-calculation method) to prevent controller saturation; Finally, based on the real-time calculated PID output value, normalization is performed in combination with the thermal inertia coefficient of the control unit to generate a basic heat demand correction coefficient in the range of 0.9 to 1.1, which serves as the core parameter for subsequent heating regulation.

[0048] S305. Introduce environmental compensation factors, time compensation factors, and spatial compensation factors, and use a weighted fusion algorithm to perform multidimensional compensation on the basic heat demand correction coefficient to generate a comprehensive heat demand correction value. It should be explained in detail that in step S305, the environmental compensation factor mainly considers the impact of outdoor meteorological conditions on heating demand, specifically including parameters such as outdoor temperature, humidity, wind speed, and solar radiation intensity. The system acquires this data through real-time weather stations and uses a multiple linear regression model to calculate the environmental compensation factor. Its weights are determined by fitting historical data to reflect the sensitivity of different meteorological conditions to indoor heat load. For example, when the outdoor temperature is below -5℃, the environmental compensation factor will significantly increase the heat demand correction value to compensate for the extra heat lost through the building envelope. The time compensation factor is set based on the time distribution characteristics of user behavior patterns. Analysis of historical heating data reveals significant differences in heat demand between weekdays and weekends, and between daytime and nighttime. The system uses the Fourier series expansion method to model the time dimension, extracting characteristic frequencies such as daily and weekly cycles to generate a time compensation factor strongly correlated with time. For example, during the peak heating period from 7:00 to 9:00 AM on weekdays, the time compensation factor will increase the heat demand correction value in advance to ensure that the indoor temperature quickly reaches the comfortable range. The spatial compensation factor corrects for differences in building thermal characteristics at different locations within the heating area. Cluster analysis divides the control units into three categories of thermal inertia zones: high, medium, and low. For areas with high thermal inertia (such as brick-concrete structures), the spatial compensation factor appropriately reduces the heat demand correction value to avoid overheating; while for areas with low thermal inertia (such as light steel frame structures), the correction value is increased to respond quickly to temperature changes. The weighted fusion algorithm uses the analytic hierarchy process (AHP) to determine the weights of each compensation factor. First, an expert scoring method is used to construct a judgment matrix to calculate the relative importance of each factor to the comprehensive heat demand correction value. Then, a consistency check is performed to ensure the rationality of the weight allocation. Finally, the environmental compensation factor, time compensation factor, spatial compensation factor, and basic heat demand correction coefficient are linearly weighted and summed to generate a comprehensive heat demand correction value ranging from 0.7 to 1.3. This value can simultaneously reflect the multidimensional influence of meteorological conditions, temporal distribution, and spatial characteristics.

[0049] S306. Verify the rationality of the comprehensive heat demand correction value through the heat balance equation, and combine the hydraulic model of the heating network to simulate the adjusted flow-pressure drop relationship to ensure that the comprehensive heat demand correction value is within the safe operating threshold of the equipment, thus forming a safety constraint. It should be explained in detail that in step S306, the comprehensive heat demand correction value is first verified using the heat balance equation. This equation, based on the principle of energy conservation, dynamically balances the heat input (including boiler heat generation and pipeline heat transfer) and heat output (indoor heat load and building envelope heat dissipation) within the heating area. Specifically, the system substitutes the comprehensive heat demand correction value into the heat balance model to calculate the adjusted indoor temperature prediction value and compares it with the target temperature curve. If the predicted value deviates from the target range by more than ±0.5℃, a correction value backtracking mechanism is triggered, and the weight allocation of environmental, temporal, and spatial compensation factors is readjusted. Simultaneously, combined with the heating network hydraulic model, the adjusted flow-pressure drop relationship is simulated: by calculating the flow rate change of each pipe section under the corrected heat demand, the pressure drop of the pipeline is assessed to determine whether it exceeds the allowable range of the equipment (such as the head limit of the circulating pump). If the simulation results show that the pressure drop of a certain pipe section exceeds the threshold, the system automatically reduces the heat demand correction value of the corresponding control unit, prioritizing the safe operation of the network. Finally, through the dual constraints of thermal balance verification and hydraulic simulation, a comprehensive heat demand correction value is generated that satisfies both thermal comfort requirements and equipment safety operation conditions. This value serves as the final execution parameter for heating regulation.

[0050] S307. The comprehensive heat demand correction value is smoothed and filtered to eliminate high-frequency fluctuations and form a continuous and smooth heat demand correction sequence.

[0051] It should be explained in detail that in step S307, the smoothing filtering process employs a low-pass filtering algorithm. By setting a cutoff frequency (e.g., 0.1Hz), high-frequency noise components in the heat demand correction value are filtered out, while the low-frequency trend of heat load changes is preserved. Specifically, the system first samples the comprehensive heat demand correction value using a sliding window. The window length is dynamically adjusted according to the thermal inertia characteristics of the heating system (e.g., 15-30 minutes) to ensure coverage of at least three sampling periods. Subsequently, a weighted moving average method is used to process the data within the window. The weight coefficients are distributed according to a Gaussian distribution, with the highest weight at the center point and decreasing towards both sides, to enhance the contribution of the data at the current moment. For boundary effect problems, the data sequence is extended using mirror extension technology to avoid distortion of the filtering result at the window edges. The filtered heat demand correction sequence must meet continuity constraints, i.e., the difference between correction values ​​at adjacent moments does not exceed a preset threshold (e.g., ±0.05). If it exceeds this threshold, linear interpolation is used for correction. The final generated continuous smooth sequence not only eliminates sensor noise and communication interference but also preserves the true trend of heat load changes. The sequence is updated every minute, matching the response time constant of the heating system, ensuring that the adjustment actions are neither too frequent and cause equipment wear, nor too infrequent and can respond to changes in heat demand in a timely manner.

[0052] As an optional embodiment of the present invention, optionally, the expression for calculating the temperature deviation value and the rate of change of deviation in step S303 is as follows:

[0053]

[0054]

[0055] in, Indicates time Time control unit Temperature deviation value, Indicates control unit In time The dynamic target temperature value, Indicates control unit In time The actual room temperature value, This indicates the time interval step size when calculating the rate of change of deviation. Indicates time Time control unit Temperature deviation value, This represents the minimum temperature deviation value. This indicates the maximum temperature deviation value.

[0056] As an optional embodiment of the present invention, optionally, the expression of the weighted fusion algorithm in step S305 is:

[0057]

[0058]

[0059]

[0060] in, Indicates control unit In time The comprehensive heat demand correction value, Indicates control unit In time The basic heat requirement correction factor Indicates control unit Spatial weighting coefficients, Indicates control unit In time The base value of the compensation factor, Indicates the dynamic adjustment coefficient. Indicates control unit In time The compensation factor deviation value, Indicates control unit In time Environmental compensation factors Indicates time outdoor temperature, Indicates control unit In time The actual indoor temperature Indicates the design temperature difference. Indicates control unit In time Time compensation factor, Indicates the start time of the peak period. Indicates the duration of the peak period. Indicates control unit In time Spatial compensation factor, Indicates control unit Thermal inertia coefficient This represents the system's maximum thermal inertia.

[0061] As an optional embodiment of the present invention, optionally, forming a safety constraint in step S306 includes: S3061. Based on the comprehensive heat demand correction value, the stability of the adjusted indoor temperature is verified by the dynamic heat balance equation. The expression for the dynamic heat balance equation is:

[0062] in, Indicates control unit In time Total input heat, Indicates the building's heat capacity. Indicates control unit In time The change in indoor temperature Indicates the time step. Indicates control unit heat transfer coefficient, Indicates heat transfer area control unit of, Indicates control unit In time Indoor temperature, Indicates time outdoor temperature, Indicates control unit In time Internal thermal gain; It should be explained in detail that in step S3061, the system first automatically matches the standard values ​​of building heat capacity C and heat transfer coefficient K according to the building type library, and dynamically collects the actual heat transfer area A, current indoor temperature, outdoor temperature, and internal heat gain through IoT sensors. The total input heat is converted from the comprehensive heat demand correction value and calculated in combination with the pipeline thermal efficiency model. The system performs a dynamic heat balance calculation every 15 minutes: substituting the above parameters into the equation, the predicted indoor temperature value after the next time step (e.g., 5 minutes) is solved. If the predicted temperature fluctuation exceeds ±0.3℃ / h for three consecutive time steps or the cumulative deviation between the predicted value and the target temperature curve exceeds 1.0℃ / h, the temperature stability is determined to be insufficient. At this time, the system automatically triggers the correction value optimization module, and according to the spatiotemporal distribution characteristics of the stability deviation, prioritizes adjusting the environmental compensation factor weight or basic heat demand correction coefficient of the corresponding control unit until the recalculated predicted temperature change rate is ≤0.2℃ / h and the cumulative deviation is ≤0.5℃ / h, satisfying the stability constraint.

[0063] S3062. Combining the aforementioned hydraulic model of the heating network, the changes in pipeline flow and pressure after adjustment are simulated using the flow-pressure drop relationship equation; The expression for the flow rate-pressure drop relationship equation is:

[0064] in, Indicates control unit In time The pipeline pressure loss value, Indicates control unit The coefficient of frictional resistance in the pipeline (determined by the roughness of the pipe and the flow velocity). Indicates control unit The length of the heating pipeline, Indicates control unit The inner diameter of the heating pipes, Indicates the first An adjustable proportional valve in time The local resistance coefficient (related to valve opening). Indicates control unit In time The density of the heating medium, Indicates control unit In time The flow velocity within the pipeline; It should be explained in detail that in step S3062, the system first automatically obtains the standard values ​​of the pipeline friction resistance coefficient f, heating pipeline length L, and heating pipeline inner diameter D of the control unit based on the pipeline network database. It also dynamically collects the heating medium density ρ, pipeline flow velocity v, and the local resistance coefficient of each adjustable proportional valve (calculated in real-time based on the valve opening) through real-time sensors. The system performs a flow-pressure drop simulation calculation every 15 minutes: substituting the above parameters into the flow-pressure drop relationship equation, it solves for the adjusted pipeline pressure loss value ΔP. If the calculated ΔP exceeds the preset maximum allowable pressure drop (e.g., 10 kPa), or the pressure drop change rate exceeds the safety threshold (e.g., ±0.5 kPa / min), it is determined that the hydraulic stability is insufficient. At this time, the system automatically triggers the valve optimization module, prioritizing the adjustment of the corresponding adjustable proportional valve opening to reduce the local resistance coefficient based on the degree of pressure drop exceeding the limit and spatial distribution characteristics, until the recalculated ΔP ≤ 8 kPa and the change rate ≤ 0.3 kPa / min, satisfying the hydraulic safety constraints.

[0065] S3063. Based on the dual verification results of the dynamic heat balance equation and the hydraulic model of the heating network, set the temperature safety threshold and the equipment safety threshold to form a multi-dimensional safety constraint on the comprehensive heat demand correction value. It should be explained in detail that in step S3063, the system first automatically sets temperature safety thresholds and equipment safety thresholds based on the dynamic heat balance verification results of step S3061 and the hydraulic model verification results of the heating network in step S3062. The temperature safety thresholds include the maximum allowable indoor temperature change rate (e.g., ±0.2℃ / h) and the maximum cumulative temperature deviation (e.g., ±0.5℃ / h); the equipment safety thresholds include the maximum allowable pipeline pressure loss (e.g., 8 kPa) and the maximum pressure change rate (e.g., 0.3 kPa / min). These thresholds are dynamically adjusted according to the building type of the control unit, the network topology, and the real-time operating status (e.g., peak and valley time weights), and integrate temperature, hydraulic, time, and spatial dimensions through a multi-dimensional constraint matrix: for example, temperature constraints are based on predicted values ​​from the dynamic heat balance equation, hydraulic constraints are based on simulated values ​​of the flow-pressure drop relationship, the time dimension introduces a peak time weighting factor, and the spatial dimension considers the thermal inertia coefficient of the control unit and the building layout. The system substitutes the comprehensive heat demand correction value into the matrix and calculates the constraint satisfaction in real time. If any dimension threshold is violated (such as the temperature change rate exceeding the limit or the pressure loss exceeding the standard), the automatic optimization module is triggered to adjust the compensation factor weight or valve opening first, until all dimensions meet the safety constraints, ensuring the heating process is stable and reliable.

[0066] S3064. If the verification does not meet the safety constraints, the closed-loop feedback mechanism is triggered to dynamically adjust the parameters of the dynamic compensation algorithm and recalculate the heat demand correction value until all constraints are met.

[0067] It should be explained in detail that in step S3064, the system first monitors in real time the satisfaction status of the multi-dimensional safety constraints (including temperature safety threshold and equipment safety threshold) set in step S3063. If any dimension threshold is violated (for example, the temperature change rate predicted by the dynamic heat balance equation exceeds ±0.2℃ / h or the pipeline pressure loss value ΔP simulated by the flow-pressure drop relationship exceeds 8 kPa), the system immediately triggers a closed-loop feedback mechanism. This mechanism includes the following steps: automatically identifying the control unit that violates the constraint and its corresponding dimension (such as temperature dimension or hydraulic dimension); dynamically adjusting the parameters of the dynamic compensation algorithm, prioritizing the modification of the weight coefficient of the environmental compensation factor or the basic heat demand correction coefficient, with the adjustment range calculated according to the degree of deviation (e.g., for every 0.1℃ increase in deviation, the weight coefficient decreases by 0.05); subsequently, recalculating the comprehensive heat demand correction value based on the adjusted parameters using the weighted fusion algorithm formula; then, re-executing the verification process of steps S3061 to S3063, including the iterative calculation of the dynamic heat balance equation and the flow-pressure drop relationship equation. The system performs a constraint check every 5 minutes and sets a maximum of 10 iterations. If all constraints are met during the iteration process (e.g., temperature change rate ≤ 0.2℃ / h, cumulative deviation ≤ 0.5℃ / h, pressure loss ΔP ≤ 8 kPa and change rate ≤ 0.3 kPa / min), the optimized correction value is output. If convergence is not achieved after exceeding the maximum number of iterations, the system automatically activates a backup control strategy (e.g., switching to a historical operating data model or manual intervention mode) and sends real-time alarms to the operator interface via the IoT platform to ensure the safety and stability of the heating process.

[0068] As an optional embodiment of the present invention, optionally, controlling the adjustable proportional valve to dynamically adjust the flow rate of the corresponding heating pipeline in step S4 includes: S401. Convert the heat demand correction value into the target flow rate adjustment amount; It should be explained in detail that in step S401, the system first converts the comprehensive heat demand correction value into the target flow adjustment amount of the corresponding control unit based on the hydraulic model of the heating network. The conversion process considers the network thermal efficiency, heat transfer coefficient, and real-time operating parameters (such as pipeline friction resistance coefficient and heating medium density), and calculates it through a flow-heat demand mapping algorithm: Target flow adjustment amount ΔQ = Heat demand correction value ΔH / (ρ × Cp × ΔT), where ρ is the heating medium density, Cp is the specific heat capacity of the medium, and ΔT is the design supply and return water temperature difference. The system updates the target flow value every 10 minutes and dynamically corrects it in conjunction with the current pipeline pressure loss ΔP (calculated in real time by the flow-pressure drop relationship equation): If ΔP exceeds the safety threshold (e.g., 8 kPa), the target flow adjustment amount is reduced proportionally (reduction coefficient = safety pressure drop / current pressure drop) to ensure that the adjustment process does not cause hydraulic imbalance. The converted target flow adjustment amount is synchronized to the adjustable proportional valve controller of the corresponding control unit through the IoT platform as the benchmark value for flow adjustment.

[0069] S402. Based on the target flow rate adjustment, calculate the valve opening adjustment using the valve flow characteristic curve and the flow-pressure drop relationship equation. The expression for calculating the valve opening adjustment is:

[0070] in, Indicates control unit In time The valve opening value that needs to be adjusted Indicates control unit In time The target flow rate value converted from the heat demand correction value. Indicates control unit The maximum flow rate of the adjustable proportional valve under design conditions. This represents the parameters of the valve flow characteristic curve. This indicates the upper limit of the physical opening of the adjustable proportional valve. Indicates control unit In time Real-time valve opening value, Indicates control unit In time The safety adjustment factor based on the verification results of the flow-pressure drop relationship equation; It should be explained in detail that in step S402, the system first obtains the adjustable proportional valve characteristic parameters of the control unit based on the valve database, including valve flow characteristic curve parameters, maximum flow rate under design conditions, and upper limit of physical opening; it then collects the real-time valve opening value of the control unit at time t via the Internet of Things; simultaneously, it extracts a safety adjustment factor from the flow-pressure drop relationship equation verification results in step S3062 (this factor automatically decays when the pipeline pressure loss ΔP exceeds a preset threshold, and its value range is 0.6-1.0). Subsequently, the target flow rate value (obtained from step S401) and the above parameters are substituted into the valve opening adjustment amount expression to calculate the current required valve opening adjustment amount. The system sets a safety threshold for a single opening change; if the calculated valve opening adjustment amount exceeds this threshold, the valve optimization module is triggered, prioritizing the reduction of the target flow rate value or dynamically correcting the safety adjustment factor, and recalculating the valve opening adjustment amount until the opening change safety constraint is met. Finally, it verifies whether the simulated pressure drop ΔP after the opening adjustment is ≤8 kPa and the rate of change is ≤0.3 kPa / min.

[0071] S403. Based on the valve opening adjustment amount, control the corresponding adjustable proportional valve to open.

[0072] It should be explained in detail that in step S403, the system automatically converts the valve opening adjustment amount calculated and verified in step S402 into a control command, which is then sent in real time to the adjustable proportional valve actuator of the corresponding control unit via the Internet of Things communication protocol. The valve opening adjustment process follows preset safety constraints: the single opening change amplitude does not exceed ±10% of the maximum opening, and the opening change rate is limited to within 5% / min. After receiving the command, the actuator drives the electric device of the proportional valve to precisely adjust the valve opening to the target value. At the same time, the system collects the actual change curve of the valve opening in real time through a high-precision displacement sensor and dynamically compares it with the command value; if the actual opening change rate exceeds the safety threshold or the opening deviation lasts for 2 minutes and is greater than the set tolerance (such as ±2% of the opening range), the actuator self-correction module is immediately triggered to perform dynamic compensation control on the valve drive. After the valve opening is adjusted, the system synchronously initiates a hydraulic state re-verification process: based on the adjusted valve opening, the updated local resistance coefficient is calculated in real time and re-substituted into the flow-pressure drop relationship equation for iterative calculation to verify whether the pipeline pressure loss ΔP meets the safety constraints of ≤8 kPa and a change rate ≤0.3 kPa / min. If the verification fails, the system automatically reverts to the previous effective valve opening state and re-triggers the valve opening optimization calculation process in step S402 until the hydraulic stability verification is passed. Finally, the verified real-time valve opening value, pipeline pressure loss value, and flow regulation result are synchronously updated to the heating system central database as historical benchmark data for the closed-loop feedback mechanism.

[0073] As an optional embodiment of the present invention, optionally, verifying the adjustment effect through a closed-loop feedback mechanism in step S5, and optimizing the parameters of the dynamic compensation algorithm based on the verification results, includes: S501: The room temperature sensor installed in the control unit collects the adjusted indoor temperature data in real time. It should be explained in detail that in step S501, the system installs high-precision room temperature sensors at key locations in each control unit (such as representative areas like the living room and bedroom). The sensors have an accuracy of ±0.1℃ and a response time of less than 30 seconds. Through an IoT communication module, the sensors upload real-time temperature data to the central control platform of the heating system every minute. During data acquisition, timestamps, control unit numbers, and sensor location information are recorded simultaneously to ensure the spatiotemporal traceability of the temperature data. The system preprocesses the collected temperature data, including outlier removal (such as data exceeding the design temperature range by ±5℃), moving average filtering (with a window length of 5 minutes), and missing value imputation (based on linear fitting of data from adjacent time periods), generating a continuous and reliable room temperature time series. Simultaneously, the system dynamically compares the real-time room temperature data with the temperature safety thresholds set in step S3063 (maximum allowable indoor temperature change rate ±0.2℃ / h, maximum cumulative temperature deviation ±0.5℃), marking potential exceedance risk points.

[0074] S502. Compare the collected real-time indoor temperature data with the set temperature safety threshold, and calculate the temperature deviation value and the deviation change rate. It should be explained in detail that in step S502, the system first extracts the real-time temperature value of the current moment from the preprocessed room temperature time series and compares it item by item with the temperature safety threshold set in step S3063. The specific calculations include: temperature deviation value ΔT = current room temperature - set temperature baseline value (e.g., 20℃), and deviation change rate ΔT. r =(Current ΔT - Previous ΔT) / Time interval (e.g., 1 minute). The system uses a sliding window algorithm (window length 15 minutes) to calculate the cumulative temperature deviation, ensuring the capture of short-term fluctuations and long-term trends. During the comparison process, if the ΔT of any control unit exceeds ±0.5℃ or ΔT... r Any deviation exceeding ±0.2℃ / h is marked as an out-of-limit event, and the duration and spatial distribution characteristics of the out-of-limit event are recorded. Simultaneously, the system generates a temperature deviation heatmap, visually displaying the temperature compliance status of each control unit. The comparison results are updated to the central database in real time, triggering different levels of response mechanisms: when the deviation value or rate of change exceeds 70% of the safety threshold, the system issues a yellow warning, alerting the operator; when it exceeds 100% of the threshold, the closed-loop feedback mechanism is automatically activated, entering the dynamic optimization process.

[0075] S503. Based on the temperature deviation value and the rate of change of deviation, combined with historical adjustment data, the parameters of the basic heat demand correction coefficient, spatial weight coefficient and dynamic adjustment coefficient in the dynamic compensation algorithm are iteratively optimized through parameter optimization algorithm. The expression for the parameter optimization algorithm is:

[0076] in, This represents the optimized parameter vector. Represents the current parameter vector. Indicates the learning rate. Represents the loss function Regarding parameter vectors The gradient; It should be explained in detail that in step S503, the system first constructs a loss function, which comprehensively considers the temperature deviation value, the rate of change of deviation, and the optimization objectives in historical adjustment data (such as minimizing the cumulative temperature deviation and stabilizing the rate of change of temperature). The loss function adopts a weighted mean square error form, where the weight of the temperature deviation term is 0.6, the weight of the rate of change of deviation term is 0.3, and the weight of the historical adjustment stability term is 0.1. The gradient of the loss function with respect to the current parameter vector (including the basic heat demand correction coefficient, spatial weight coefficient, and dynamic adjustment coefficient) is calculated using the backpropagation algorithm. The learning rate is dynamically adjusted according to the historical optimization effect: if the absolute value of the temperature deviation decreases after the previous optimization, the learning rate increases by 10%; if the deviation increases, the learning rate decreases by 20%, but not lower than 0.01. The parameter optimization process is set to a maximum of 20 iterations. After each iteration, it is verified whether the new parameters meet all safety constraints (such as temperature change rate ≤ 0.2℃ / h, pressure loss ≤ 8kPa). If the loss function does not decrease for three consecutive iterations during the iteration process, an early stopping mechanism is triggered, and the current optimal parameters are retained. The optimized parameter vector, after being normalized (scaled to the [0.8, 1.2] range), is synchronously updated to the dynamic compensation algorithm module as the benchmark parameters for the next round of adjustment. Simultaneously, the system records key metrics of this optimization process (such as initial / final loss values, number of iterations, and parameter adjustment magnitude) in the central database.

[0077] S504. Update the optimized parameters into the dynamic compensation algorithm, recalculate the comprehensive heat demand correction value, and trigger a new round of flow regulation and safety verification process until the indoor temperature stabilizes within the set temperature safety threshold range.

[0078] It should be explained in detail that in step S504, the parameter vector (basic heat demand correction coefficient, spatial weight coefficient, and dynamic adjustment coefficient) optimized in step S503 is written into the dynamic compensation algorithm, replacing the original parameters. Subsequently, the comprehensive heat demand correction value is recalculated based on the updated parameters. The calculation process integrates environmental compensation factors (dynamically adjusted according to real-time meteorological data such as outdoor temperature and wind speed) and the hydraulic state of the pipeline network (the pipeline pressure loss ΔP is fed back in real time through the flow-pressure drop relationship equation). If the deviation between the calculated comprehensive heat demand correction value and the previous adjustment result exceeds the threshold (e.g., ±5%), the safety verification module is triggered to check whether the correction value meets all constraints (temperature change rate ≤ 0.2℃ / h, pressure loss ΔP ≤ 8kPa and change rate ≤ 0.3kPa / min). After the verification is passed, the system converts the correction value into a new target flow adjustment amount and executes valve opening adjustment according to the process of S401-S403: control commands are sent to the adjustable proportional valve actuator through the IoT platform to drive the valve to adjust to the target opening, while a high-precision displacement sensor is activated to monitor the actual opening change in real time. After adjustment, the system immediately initiates the hydraulic state re-verification process. Based on the adjusted valve opening, the local resistance coefficient is recalculated, and the flow-pressure drop relationship equation is substituted to verify whether the pipeline pressure loss ΔP meets the safety constraints. If the verification fails (e.g., ΔP exceeds 8 kPa or the rate of change exceeds 0.3 kPa / min), the system automatically reverts to the previous effective valve opening state and re-triggers the parameter optimization process (S503). If the verification passes, the real-time valve opening value, pipeline pressure loss value, and flow adjustment result are synchronously updated to the central database. This process continues iteratively until the indoor temperature stabilizes within the set temperature safety threshold range (temperature deviation ΔT ≤ ±0.5℃ and deviation change rate ΔT) for three consecutive verification cycles. r (≤±0.2℃ / h), the system determines that the adjustment effect meets the standard and generates an adjustment completion report, which includes key indicators such as adjustment time, number of parameter optimizations, and final temperature deviation distribution.

[0079] Example 2 A centralized heating system based on room temperature feedback regulation includes: processor; Memory used to store processor-executable instructions; The processor is configured to implement a centralized heating method based on room temperature feedback regulation when executing executable instructions.

[0080] It should be noted that the computer device includes a processor, a memory, and may also include one or more of a multimedia component, an input / output (I / O) interface, and a communication component.

[0081] The processor controls the overall operation of the computer device to complete all or part of the steps in the above-mentioned centralized heating method based on room temperature feedback regulation.

[0082] Memory is used to store various types of data to support the operation of the computer device. This data may include, for example, instructions for any application or method used to operate on the computer device, as well as application-related data. Memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0083] The multimedia component may include a screen and an audio component, wherein the screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals; for example, the audio component may include a microphone for receiving external audio signals, the received audio signals may be further stored in memory or transmitted via a communication component; the audio component may also include at least one speaker for outputting audio signals.

[0084] I / O interfaces provide interfaces between the processor and other interface modules, such as keyboards, mice, buttons, etc.; these buttons can be virtual buttons or physical buttons.

[0085] The communication component is used for wired or wireless communication between the computer device and other devices; wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G or 5G, or one or more combinations thereof, and the corresponding communication component may include: Wi-Fi module, Bluetooth module, NFC module, mobile communication module.

[0086] As a preferred embodiment, the computer device may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-described centralized heating method based on room temperature feedback regulation.

[0087] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A centralized heating method based on room temperature feedback regulation, characterized in that, The method includes: S1. Construct a three-dimensional room temperature sensing network to collect real-time room temperature data, verify the real-time room temperature data, and generate a room temperature feedback signal; S2. The room temperature feedback signal is transmitted to the central control unit via the wireless communication module for time-series processing to generate a temperature-time two-dimensional data matrix. S3. Based on the temperature-time two-dimensional data matrix, calculate the heat demand correction value for each heating sub-region using a preset dynamic compensation algorithm; S4. Install adjustable proportional valves at key nodes of the heating network, and control the flow rate of the corresponding heating pipeline to be dynamically adjusted according to the heat demand correction value. S5. Verify the adjustment effect through a closed-loop feedback mechanism, and optimize the parameters of the dynamic compensation algorithm based on the verification results.

2. The centralized heating method based on room temperature feedback regulation as described in claim 1, characterized in that, The generation of the room temperature feedback signal in step S1 includes: S101. Based on the spatial characteristics of the heating area, deploy a data acquisition array to form a three-dimensional room temperature sensing network; S102. Based on the three-dimensional room temperature sensing network, real-time room temperature data is collected synchronously by multiple nodes through the ZigBee wireless protocol. S103. Perform physical range verification and spatiotemporal consistency verification on the real-time room temperature data, and use the weighted average method to correct the deviation data. S104. Sort the verified valid real-time room temperature data by timestamp, generate a standardized data packet containing temperature, humidity and light values, and attach the collector ID, collection time and spatial coordinate triplet information to form a room temperature feedback signal with spatiotemporal identification.

3. The centralized heating method based on room temperature feedback regulation as described in claim 1, characterized in that, The generation of the temperature-time two-dimensional data matrix in step S2 includes: S201. Based on the data characteristics of the room temperature feedback signal, a self-organizing network is constructed, and a star-tree hybrid topology is used to realize the aggregation and transmission of multi-node data within the heating area, and encapsulation is performed. S202, The central control unit performs time-series alignment processing and dynamic interpolation completion on the room temperature feedback signal to generate a continuous time series; S203. The continuous time series is mapped in two dimensions according to the time dimension and the spatial dimension, and a temperature-time two-dimensional data matrix is ​​constructed with the sampling period as the row index and the collector ID as the column index. The elements in the temperature-time two-dimensional data matrix are the room temperature feedback signals in the continuous time series. S204. Abnormal room temperature feedback signals in the temperature-time two-dimensional data matrix are detected by box plot method, verified by combining with the spatial heat distribution model of the heating area, and the final temperature-time two-dimensional data matrix is ​​output after correcting the abnormal data.

4. The centralized heating method based on room temperature feedback regulation as described in claim 1, characterized in that, The calculation of the heat demand correction value for each heating sub-region in step S3 includes: S301. Based on the temperature-time two-dimensional data matrix, according to the heating network topology and heat load distribution characteristics, the heating area is divided into N sub-regions. Each sub-region corresponds to a set of continuous data acquisition node clusters in the temperature-time two-dimensional data matrix, forming a control unit. S302. Combining user comfort requirements and energy-saving standards, a dynamic target temperature curve is set for each of the control units. The curve parameters include the diurnal temperature difference compensation coefficient, seasonal adjustment factor and historical energy consumption benchmark, which serve as the control benchmark. S303. The actual room temperature data in the control unit is compared with the target temperature curve in real time to calculate the temperature deviation value and the deviation change rate, and to form a dynamic deviation sequence. S304. Based on the dynamic deviation sequence, a PID control algorithm is adopted, with the temperature deviation as the input, and combined with integral separation and anti-saturation mechanism, to calculate the basic heat demand correction coefficient. S305. Introduce environmental compensation factors, time compensation factors, and spatial compensation factors, and use a weighted fusion algorithm to perform multidimensional compensation on the basic heat demand correction coefficient to generate a comprehensive heat demand correction value. S306. Verify the rationality of the comprehensive heat demand correction value through the heat balance equation, and combine the hydraulic model of the heating network to simulate the adjusted flow-pressure drop relationship to ensure that the comprehensive heat demand correction value is within the safe operating threshold of the equipment, thus forming a safety constraint. S307. The comprehensive heat demand correction value is smoothed and filtered to eliminate high-frequency fluctuations and form a continuous and smooth heat demand correction sequence.

5. The centralized heating method based on room temperature feedback regulation as described in claim 4, characterized in that, The expressions for calculating the temperature deviation value and the rate of change of deviation in step S303 are as follows: in, Indicates time Time control unit Temperature deviation value, Indicates control unit In time The dynamic target temperature value, Indicates control unit In time The actual room temperature value, This indicates the time interval step size when calculating the rate of change of deviation. Indicates time Time control unit Temperature deviation value, This represents the minimum temperature deviation value. This indicates the maximum temperature deviation value.

6. The centralized heating method based on room temperature feedback regulation as described in claim 4, characterized in that, The expression for the weighted fusion algorithm in step S305 is: in, Indicates control unit In time The comprehensive heat demand correction value, Indicates control unit In time The basic heat requirement correction factor Indicates control unit Spatial weighting coefficients, Indicates control unit In time The base value of the compensation factor, Indicates the dynamic adjustment coefficient. Indicates control unit In time The compensation factor deviation value, Indicates control unit In time Environmental compensation factors Indicates time outdoor temperature, Indicates control unit In time The actual indoor temperature Indicates the design temperature difference. Indicates control unit In time Time compensation factor, Indicates the start time of the peak period. Indicates the duration of the peak period. Indicates control unit In time Spatial compensation factor, Indicates control unit Thermal inertia coefficient This represents the system's maximum thermal inertia.

7. The centralized heating method based on room temperature feedback regulation as described in claim 4, characterized in that, The formation of safety constraints in step S306 includes: S3061. Based on the comprehensive heat demand correction value, the stability of the adjusted indoor temperature is verified by the dynamic heat balance equation. S3062. Combining the aforementioned hydraulic model of the heating network, the changes in pipeline flow and pressure after adjustment are simulated using the flow-pressure drop relationship equation; S3063. Based on the dual verification results of the dynamic heat balance equation and the hydraulic model of the heating network, set the temperature safety threshold and the equipment safety threshold to form a multi-dimensional safety constraint on the comprehensive heat demand correction value. S3064. If the verification does not meet the safety constraints, the closed-loop feedback mechanism is triggered to dynamically adjust the parameters of the dynamic compensation algorithm and recalculate the heat demand correction value until all constraints are met.

8. The centralized heating method based on room temperature feedback regulation as described in claim 7, characterized in that, In step S4, an adjustable proportional valve is installed at a key node of the heating network. Based on the required heat correction value, the adjustable proportional valve is used to dynamically adjust the flow rate of the corresponding heating pipeline. This includes: S401. Convert the heat demand correction value into the target flow rate adjustment amount; S402. Based on the target flow rate adjustment, calculate the valve opening adjustment using the valve flow characteristic curve and the flow-pressure drop relationship equation. S403. Based on the valve opening adjustment amount, control the corresponding adjustable proportional valve to open.

9. The centralized heating method based on room temperature feedback regulation as described in claim 1, characterized in that, In step S5, the adjustment effect is verified through a closed-loop feedback mechanism, and the parameters of the dynamic compensation algorithm are optimized based on the verification results, including: S501: The room temperature sensor installed in the control unit collects the adjusted indoor temperature data in real time. S502. Compare the collected real-time indoor temperature data with the set temperature safety threshold, and calculate the temperature deviation value and the deviation change rate. S503. Based on the temperature deviation value and the rate of change of deviation, combined with historical adjustment data, the parameters of the basic heat demand correction coefficient, spatial weight coefficient and dynamic adjustment coefficient in the dynamic compensation algorithm are iteratively optimized through parameter optimization algorithm. S504. Update the optimized parameters into the dynamic compensation algorithm, recalculate the comprehensive heat demand correction value, and trigger a new round of flow regulation and safety verification process until the indoor temperature stabilizes within the set temperature safety threshold range.

10. A centralized heating system based on room temperature feedback regulation, characterized in that, The system includes: processor; Memory used to store processor-executable instructions; The processor is configured to implement the centralized heating method based on room temperature feedback regulation as described in any one of claims 1 to 9 when executing the executable instructions.

Images