Heating and ventilation heating power constant-temperature variable-flow control method

Abstract

The invention relates to the technical field of temperature control, in particular to a heating and ventilation heating power constant-temperature variable-flow control method, and provides the following scheme that circulating pump frequency instructions and water temperature response data of multiple historical regulation and control periods are collected, and a dynamic track of the current water temperature in a temperature inertia field is constructed; calculating an attraction potential energy state between the current water temperature and a target constant temperature value in combination with the disturbance residual matrix; and outputting the frequency bandwidth interval of the circulating pump based on the state to realize dynamic adjustment. An adjacency matrix and a Laplacian matrix are constructed by introducing a hydraulic communication relation, spatial convolution and weighted updating are performed on disturbance residues in combination with a topological diffusion core, and accurate propagation and suppression of disturbance are realized. The method can effectively solve the problems of nonlinear response, delay effect and disturbance coupling of the water temperature to the pump frequency, and the temperature stability and the energy utilization efficiency of a heating system are improved.

Description

Technical Field

[0001] The present application relates to the field of temperature control technology, and in particular to a method for controlling HVAC constant temperature flow. Background Art

[0002] In existing HVAC thermal heating systems, the frequency regulation of circulating pumps mostly relies on direct feedback based on the difference between real-time water temperature and target temperature. There is a lack of in-depth analysis and modeling of the dynamic change process of water temperature, which leads to lag and instability in the regulation results. In the case of load fluctuations or frequent external disturbances, the system water temperature is prone to deviate from the target value for a long time, and even produce continuous oscillations in local pipe networks, affecting the quality of heating. At the same time, when dealing with multi-node pipe networks, existing methods do not adequately consider the correlation between water temperatures at different locations. Disturbances are easily accumulated or transmitted in the pipe network, causing local overheating or underheating, and low energy utilization efficiency. For switching control in the operation stage, such as the transition from initial temperature rise to stable heating and then to low load at night, the existing technology is insufficient in matching the sampling frequency and the adjustment amplitude, resulting in significant differences in temperature control effects in each stage, making it difficult to achieve stable constant temperature throughout the entire cycle.

[0003] In order to solve the above problems, this application designs a control method for HVAC thermal constant temperature variable flow. Summary of the Invention

[0004] The technical problem to be solved by this application is to address the deficiencies of the existing technology and provide a control method for HVAC thermal constant temperature flow control, collect the circulating pump frequency instructions and water temperature response data of multiple historical control cycles, and construct the dynamic trajectory of the current water temperature in the temperature inertia field; combine the disturbance residual matrix to calculate the attractive potential energy state between the current water temperature and the target constant temperature value; output the circulating pump frequency bandwidth interval based on this state to achieve dynamic regulation. By introducing the hydraulic connectivity relationship to construct the adjacency matrix and Laplace matrix, and combining the topological diffusion kernel to perform spatial convolution and weighted update on the disturbance residual, the accurate propagation and suppression of the disturbance can be achieved. This method can effectively deal with the nonlinear response of water temperature to pump frequency, delay effect and disturbance coupling problems, and improve the temperature stability and energy efficiency of the heating system.

[0005] To achieve the above objectives, this application provides the following technical solutions:

[0006] A control method for HVAC thermostatic flow control is applied to a heating network. The heating network includes a controller, a circulating pump, a water temperature sensor, and a network body. The controller is configured with a temperature inertia field and a disturbance residual matrix. The temperature inertia field performs virtual potential energy simulation based on a preset target constant temperature value. The disturbance residual matrix is ​​constructed based on the disturbance residual between water temperature changes and ideal changes within a historical control cycle. The control method includes:

[0007] The current water temperature is collected by the water temperature sensor, and based on the pump frequency input and water temperature output changes of the circulation pump in multiple historical control cycles, a dynamic trajectory of the current water temperature in the temperature inertia field is constructed, where the current water temperature is a water temperature sampling sequence collected in the current control cycle;

[0008] Calculating the attractive potential energy state between the current water temperature and the target constant temperature value according to the dynamic trajectory and in combination with the disturbance residual matrix;

[0009] According to the attractive potential energy state, the controller outputs a frequency bandwidth interval to control the circulating pump to perform frequency regulation within the frequency bandwidth interval, thereby obtaining the pump frequency input of the circulating pump in the current control cycle, wherein the width and center position of the frequency bandwidth interval are jointly determined according to the attractive potential energy state and the disturbance residual matrix.

[0010] The construction of the dynamic trajectory of the current water temperature in the temperature inertia field includes:

[0011] Collect the circulating pump frequency control instructions and corresponding water temperature response data within at least M previous control cycles, and construct a historical control state sequence arranged in chronological order. Each state unit in the historical control state sequence includes a timestamp, a frequency increment value, a water temperature change rate, and a sampling interval duration;

[0012] Calculating the response delay distribution characteristics of the water temperature in the heating network to the frequency control action based on the historical control state sequence, and calculating the response offset function by fitting the residual curve;

[0013] The current water temperature is fitted with the response offset function to obtain a relative offset trajectory curve with the target constant temperature value as a reference point, and a dynamic trajectory is output, wherein the relative offset trajectory curve has time as the horizontal axis and the output value of the response offset function as the vertical axis.

[0014] The construction of the disturbance residual matrix is ​​based on the historical measurement and prediction data of the previous k control cycles, where the prediction data is the water temperature change sequence calculated by the controller based on the ideal response model according to the pump frequency input of the circulation pump. The specific construction method is as follows:

[0015] After the kth control cycle ends, the actual water temperature sampling sequence within the first k control cycles is obtained and the ideal water temperature sampling sequence predicted by the controller , where j represents a single control cycle, and the specific form of the sampling sequence is as follows:

[0016] ,

[0017] Where p represents the number of sampling points;

[0018] According to the actual water temperature sampling sequence corresponding to each control cycle j and the ideal water temperature sampling sequence predicted by the controller, the residual vector is calculated by difference;

[0019] According to the hydraulic connectivity of the heating network, calculate The adjacency matrix A of order is calculated according to the adjacency matrix, and the corresponding Laplace matrix is ​​calculated. , where D represents the degree matrix of the adjacency matrix;

[0020] According to the preset diffusion coefficient , calculate the topological diffusion kernel , where e represents a matrix exponential operation, performing spatial convolution of the residual vector with the topological diffusion kernel to generate a residual field matrix;

[0021] The forgetting factor is calculated according to the pump frequency input corresponding to the control period, the residual field matrix is ​​weighted and assigned according to the forgetting factor, and the weighted residual field matrix is ​​summarized to obtain the disturbance residual matrix.

[0022] The disturbance residual matrix is ​​updated based on the pump frequency input of the previous control cycle and the current water temperature of the current control cycle. The specific update method is as follows:

[0023] Get the perturbation residual matrix output of the previous control cycle ;

[0024] The forgetting factor of the disturbance residual matrix is ​​updated according to the pump frequency input of the previous control cycle to obtain a weighted value ;

[0025] Calculate the residual field matrix of the current control cycle according to the current water temperature , calculate the updated perturbation residual matrix according to the perturbation residual matrix of the previous control cycle, the weighted value and the residual field matrix of the current control cycle , the specific calculation method is as follows:

[0026] .

[0027] According to the dynamic trajectory and in combination with the disturbance residual matrix, the attractive potential energy state between the current water temperature and the target constant temperature value is calculated, including:

[0028] According to the pump frequency input of the previous control cycle and the current water temperature of the current control cycle, the disturbance residual matrix outputted in the previous control cycle is updated to obtain an updated disturbance residual matrix;

[0029] Determine the deviation amplitude, change rate, and trend direction of the current water temperature relative to the target constant temperature value based on the dynamic trajectory, and calculate the corresponding trajectory feature vector;

[0030] The trajectory eigenvector and the updated perturbation residual matrix are coupled to calculate the attractive potential energy state.

[0031] Coupling the trajectory eigenvector and the updated perturbation residual matrix includes:

[0032] Determining a disturbance-sensitive direction factor representing a current water temperature change trend based on the trajectory characteristic vector;

[0033] Extracting a disturbance submatrix corresponding to the disturbance sensitive direction factor in the updated disturbance residual matrix;

[0034] Performing direction matching analysis on the disturbance sensitive direction factor and the disturbance sub-matrix to determine the influence of the residual disturbance on the current trajectory trend;

[0035] According to the results of the direction matching analysis, the attraction stability index of the current water temperature on the control path is calculated to obtain the attraction potential energy state.

[0036] The width and center position of the frequency bandwidth interval are determined according to the attractive potential energy state and the disturbance residual matrix.

[0037] According to the state of the attractive potential energy, the controller outputs a frequency bandwidth interval including:

[0038] Taking the frequency center value of the previous control cycle as a reference, determining the frequency center offset of the current control cycle according to the attractive potential energy state, wherein the frequency center offset is inversely proportional to the attractive potential energy state;

[0039] Determining an adjustment coefficient of a frequency bandwidth interval according to the disturbance intensity on the current control path in the disturbance residual matrix, wherein the adjustment coefficient is used to control a width convergence rate of the frequency bandwidth;

[0040] According to the frequency center offset and the adjustment coefficient, combined with the frequency bandwidth interval of the previous cycle, the corresponding frequency bandwidth interval is output.

[0041] The method further comprises:

[0042] Determining a sampling period of the water temperature sensor according to an operating period of the heating network;

[0043] The operating period includes an initial heating stage, a stable heating stage and a nighttime load stage, and the target constant temperature value of each stage is different.

[0044] Determining a sampling period of the water temperature sensor according to an operating period of the heating network includes:

[0045] If it is in the initial temperature rise stage, set the first sampling period;

[0046] If the heating stage is stable, a second sampling period is set, wherein the first sampling period is smaller than the second sampling period;

[0047] If it is in the night load stage, setting a third sampling period, wherein the second sampling period is smaller than the third sampling period;

[0048] The first sampling period, the second sampling period and the third sampling period are adjusted according to the disturbance convergence rate in the disturbance residual matrix, the dynamic trajectory of the current water temperature in the temperature inertia field and the offset of the historical adjustment of the circulation pump frequency.

[0049] Compared with the prior art, the present invention has the following advantages:

[0050] This application achieves global perception and precise control of the water temperature status of multiple nodes in the heating network by introducing the temperature inertia field and disturbance residual matrix into the controller, combined with water temperature dynamic trajectory analysis. Not only can the water temperature offset amplitude, change rate and trend direction be considered simultaneously in frequency regulation, but the disturbance residual matrix can also be used to suppress the propagation and accumulation of local disturbances, thereby significantly improving the stability of constant temperature control. Under load fluctuations or external disturbance conditions, the water temperature can quickly return to the target constant temperature value, reducing local oscillations. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0052] Figure 1 This is a flow chart of a method for controlling HVAC constant temperature flow control according to an embodiment of the present application;

[0053] Figure 2 This is a schematic diagram of an exemplary application scenario of an embodiment of the present application;

[0054] Figure 3 This is a module diagram of the controller according to an embodiment of the present application;

[0055] Figure 4 This is a schematic diagram of the matrix calculation process of the embodiment of the present application;

[0056] Figure 5 This is a flow chart of a method for generating a dynamic trajectory according to an embodiment of the present application;

[0057] Figure 6 This is a schematic diagram of the dynamic trajectory of an embodiment of the present application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0059] References to "embodiments" herein mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It will be understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0060] In order to more clearly illustrate a control method for HVAC thermal constant temperature flow conversion of the present application, an optional specific implementation method of the present application will be described in detail below in combination with actual engineering scenarios.

[0061] In this example technology, the method is applied to a HVAC system, primarily in a secondary pipe network loop control scenario within a centrally heated building. The system architecture includes a HVAC digital circulation pump, an outlet water temperature sensor, an automatic controller, and a heating pipe network. The system's operational goal is to adjust the operating frequency of the digital circulation pump in real time during different operating periods to maintain the hot water temperature within the pipe network as stably as possible within the target constant temperature range while minimizing energy waste and the risk of overshoot.

[0062] The specific adjustment logic of the above example technology is as follows:

[0063] When the system is first started, the controller first completes the following initialization steps:

[0064] Frequency setting initialization sets the digital circulation pump's default operating frequency to 50Hz, which serves as the reference value for the controller's initial judgment logic. This frequency value is calculated by the system designer based on the building's heat load, return water flow rate, and initial heat balance, and can be adjusted appropriately based on system commissioning parameters.

[0065] Sampling points and time periods: The system divides 24 hours into fixed time periods, preferably one per hour, and sets a temperature sampling frequency within each period, for example, sampling the current outlet water temperature every 5 minutes. This time granularity allows the controller to establish temperature trend estimates within each time period and provides data support for subsequent frequency adjustment strategies.

[0066] Logical variable initialization, initialize the water temperature sampling record queue, frequency record array and historical target frequency cache table for each period to support the feedforward prediction logic and mutation compensation mechanism for frequency update.

[0067] After the system enters a stable operating state, the overall operation process of the control method is as follows:

[0068] The controller continuously collects the actual water temperature during the current period through a water temperature sensor and stores the sampled values ​​in a local cache according to a set sampling period (for example, every 5 minutes). This data is used to form a water temperature change sequence for the current period. The controller also records the actual frequency of the digital circulating pump corresponding to each sampling point, providing input for subsequent frequency adjustment decisions.

[0069] Before each new period begins, the controller uses a set of preset regulation rules to predict the target circulation pump frequency for the next period based on the last sampled water temperature data for the current period. The core principle of this prediction logic is: if the current water temperature is lower than the set value, the frequency should be appropriately increased to increase the hot water flow rate and enhance heating supply; conversely, if the current water temperature is too high, the pump speed should be appropriately reduced to prevent overshoot.

[0070] It's understandable that the regulation rule here is a simplified regulation logic based on engineering experience. Essentially, it uses the deviation between the current water temperature and the reference value to calculate the frequency target using a multiplicative factor. This calculation doesn't involve complex mathematical modeling; instead, it uses linear approximation and adjustment factor correction to determine the controller's frequency output strategy for the next stage.

[0071] In addition to feedforward predictive control, the system also features an immediate response strategy based on threshold determination. This means that during each period of operation, if the current water temperature changes by more than 3°C compared to the initial temperature for that period, the system immediately triggers a frequency update mechanism, recalculates the pump speed target frequency, and instructs the circulating pump to adjust its frequency.

[0072] In an optional embodiment, the system automatically distinguishes the operating stages according to different time periods of the day and adjusts the control strategy based on the heat load pattern:

[0073] During the initial heating phase, the system heat load rapidly increases from its nighttime low, requiring rapid indoor temperature rise. The controller initially sets the frequency to a higher range (e.g., 50 Hz) and shortens the water temperature sampling period to 2 minutes to enhance system response sensitivity and ensure rapid thermal equilibrium.

[0074] During the stable heating phase, heat load fluctuations are relatively smooth. The controller can use a predictive frequency strategy, combined with historical frequency operation trajectories, to optimize pump speed to reduce energy consumption while maintaining a stable target temperature. The sampling period is restored to 5 minutes.

[0075] During the nighttime load phase, due to the decrease in user heat load, the water temperature changes tend to be stable. The controller will set a lower frequency target value (such as below 40Hz) and extend the sampling period to 10 minutes to reduce system energy consumption and wear caused by frequent adjustments.

[0076] like Figure 1 As shown, in order to solve the problem that the response of water temperature to pump frequency is nonlinear, the present application provides a control method for HVAC thermostatic variable flow, which is applied to a heating pipe network. The heating pipe network includes a controller, a circulating pump, a water temperature sensor and a pipe network body. The controller is configured with a temperature inertia field and a disturbance residual matrix, including:

[0077] S1: The current water temperature is collected by the water temperature sensor, and a dynamic trajectory of the current water temperature in the temperature inertia field is constructed based on the pump frequency input and water temperature output changes of the circulation pump in multiple historical control cycles;

[0078] In this embodiment, a water temperature sensor acquires a water temperature sampling sequence within the current control cycle. This sampling sequence covers representative measurement points at various locations within the pipe network. This sampling sequence is combined with historical data on the circulating pump's pump frequency input and water temperature output changes over multiple control cycles to generate a dynamic trajectory of the current water temperature within the temperature inertia field. The temperature inertia field is a virtual potential energy distribution model, constructed by factoring in factors such as the target constant temperature, response delay distribution, and water temperature change rate. It is used to describe the inertial offset trend of water temperature changes to control inputs.

[0079] It is understandable that the water temperature sampling sequence avoids the drawbacks of adjustment based solely on instantaneous deviations, and can reflect the overall trend and potential stable path of water temperature changes while taking into account the thermal inertia and delay effects of the system.

[0080] Those skilled in the art will understand that the type of operation data collection can be selected according to the actual working conditions, for example, it can be single-point distribution measurement, multi-point distribution measurement, or include additional flow and pressure information. It is only necessary to ensure that the dynamic trajectory of water temperature can be derived from it to a minimum extent. This application does not impose any restrictions on this.

[0081] S2: Calculating the attractive potential energy state between the current water temperature and the target constant temperature value based on the dynamic trajectory and the disturbance residual matrix;

[0082] In this embodiment, the perturbation residual matrix reflects the distribution of deviations between water temperature changes and the ideal response over historical regulation cycles. It is then processed using topological diffusion of the hydraulic connectivity of the heating network to reveal the spatial propagation characteristics of perturbations at different locations. In the calculation of the attractive potential energy state, the trajectory eigenvector describes the direction, magnitude, and rate of change of the current water temperature deviation, while the perturbation residual matrix is ​​used to correct the spatial and historical stability predictions of this trajectory, thereby obtaining a comprehensive and quantifiable attractive potential energy state indicator.

[0083] It is understood that this application considers not only the current deviation value but also historical disturbance patterns and spatial correlations, making it particularly suitable for complex scenarios with uneven network loads and interdependent node temperatures. The specific construction of the disturbance residual matrix can be adjusted based on the network's connectivity structure and computing resources, as long as it can reflect the coupled effects of disturbances in space and time.

[0084] S3: outputting a frequency bandwidth interval through a controller according to the state of the attractive potential energy, so as to control the circulating pump to perform frequency regulation within the frequency bandwidth interval, thereby obtaining a pump frequency input of the circulating pump in the current regulation cycle;

[0085] In this embodiment, the center position of the frequency bandwidth interval is determined by the attractive potential energy state and the perturbation residual matrix, while its width takes into account the balance between perturbation intensity and an attractive stability indicator. When the attractive potential energy state indicates a high degree of temperature stability, the bandwidth is appropriately narrowed to reduce frequency fluctuations; conversely, the bandwidth is expanded to improve response speed. Ultimately, the circulating pump performs real-time frequency adjustment within this bandwidth interval, achieving a rapid and stable approach to the target constant temperature value.

[0086] Those skilled in the art will appreciate that the bandwidth range can be implemented by the controller through software parameter configuration, or can be implemented in combination with a hardware limiting device, and this application does not impose any restrictions on this.

[0087] This application is applicable to HVAC systems with dynamic heat load fluctuations and nonlinear thermal response characteristics, and is particularly applicable to water circulation heating networks with return path coupling, significant spatial temperature inertia, frequency control lag, and distributed disturbance conduction characteristics.

[0088] It is understandable that when such systems perform constant temperature control, traditional single-point feedback strategies and static mapping models cannot effectively respond to control errors caused by pipe network structure inertia, load disturbance residuals, and historical response delays, which can easily cause water temperature fluctuations, regulation overshoot, or energy consumption redundancy.

[0089] Application scenarios include but are not limited to:

[0090] District heating systems with complex pipe layout and multi-point heat source distribution;

[0091] Floor radiant heating system with dendritic or ring-shaped hydraulic topology, and significant feedback time delay in backwater path;

[0092] Intelligent building thermal control system maintaining high response stability to small perturbations during low-load operation period (e.g. night constant temperature);

[0093] Office building or factory thermal network with concurrent regulation of multiple circulating pump frequencies and unbalanced local regulation feedback.

[0094] The control method proposed in the present application is not dependent on a single control strategy or a fixed thermal load model or static pipeline parameter configuration, but models the spatial attraction of the target constant temperature state by constructing a dynamic temperature inertia field, and describes the residual error propagation pattern of the regulation deviation in the topology structure by a disturbance residual matrix, thereby realizing variable frequency regulation control.

[0095] The applicable application scenarios need to meet at least one of the following structural or response characteristics:

[0096] The temperature control response has significant hysteresis characteristics, and real-time control is difficult to close loop accurately;

[0097] The thermal load changes show short-term severe disturbance, but the overall trend is slow;

[0098] The water temperature sampling signal is affected by the pipeline location distribution or sampling cycle configuration, resulting in unstable or delayed feedback;

[0099] The frequency adjustment of the controller is indirectly driven by the water temperature change, and there is an obvious intermediate thermal inertia link;

[0100] In the system operation scenario, it is not suitable to use large frequency mutation or high-speed response model to ensure the operation stability and pump body life.

[0101] It should be noted that the control method proposed in the present application does not rely on strong hypothesis division of the topology structure of the heating pipeline network, nor does it require absolute decoupling or feedback consistency between circulating pumps. Instead, it realizes stable guidance of the target water temperature of the system while ensuring the regulation dynamic flexibility by potential field attraction and residual disturbance suppression.

[0102] It should be noted that the disturbance residual matrix involved in the present application is not a weight or correction matrix for static compensation in the traditional sense, but a dynamic disturbance characteristic field with self-update and space-time diffusion characteristics, which can adapt to the disturbance behavior migration law in different control periods, and thus realize the sustained convergence regulation of the nonlinear thermal system under the action of disturbance.

[0103] Similarly, the frequency bandwidth adjustment described in this application is not based on a simple frequency conversion algorithm of single-point frequency modulation, but is based on the linkage analysis of the attractive potential energy state and the disturbance residual, avoiding the overshoot or oscillation problems caused by traditional frequency increase or decrease strategies when facing inertial lag.

[0104] See Figure 2 , which is a schematic diagram of an exemplary application scenario provided in an embodiment of the present application.

[0105] Figure 2 The application scenario shown includes a heat source 100 , a circulation pump 101 , multiple water temperature sensors 102 and a controller 103 , and a pipe network body 104 .

[0106] In this embodiment, the heat source 100 can be a device capable of continuously providing heat medium, such as a boiler, heat exchange unit, or centralized heating station, and is used to deliver hot water or other heat transfer medium at a preset temperature to the pipe network 104. A circulating pump 101 is disposed between the heat source 100 and the pipe network 104 and is used to continuously deliver and regulate the heat medium within the pipe network 104 under the control of a controller 103. Multiple water temperature sensors 102 are deployed at key nodes within the pipe network 104, such as on each user branch or return pipe, to collect real-time water temperature data at different locations to reflect the operating thermal distribution of the entire heating system.

[0107] Figure 2 It is further shown that the controller 103 is electrically connected to the circulation pump 101 and multiple water temperature sensors 102, and is used to receive water temperature data collected by each water temperature sensor 102, and based on the data combined with a preset temperature inertia field model and a disturbance residual matrix, calculate the attractive potential energy state between the current water temperature and the target constant temperature value, and then output the corresponding frequency bandwidth interval of the circulation pump 101 according to the state to achieve dynamic adjustment of the flow rate.

[0108] It is understood that the pipe network 104 may include a main pipe network, branch pipes, user-side heat exchange devices, etc., and its pipe layout may be a parallel structure, a ring structure, or a mixed structure to adapt to different heating load requirements. The control method of the present application is adaptable to various pipe network topologies and can achieve rapid temperature rise, stable constant temperature, and energy-saving nighttime operation of the system by adjusting the operating state and frequency range of the circulating pump 101 during different operating periods, thereby effectively reducing energy consumption and improving heating quality.

[0109] See Figure 3 , which is a module schematic diagram of the controller provided in an embodiment of the present application.

[0110] Figure 3The controller 103 comprises a water temperature data acquisition module 1031, a dynamic trajectory calculation module 1032, a disturbance residual matrix processing module 1033 and a frequency adjustment control module 1034.

[0111] The water temperature data acquisition module 1031 is configured to interact with the water temperature sensors 102 arranged at the key nodes of the heating pipe network to realize real-time acquisition and caching of water temperature sampling data in a current control period.

[0112] In this embodiment, the water temperature data acquisition module 1031 and the water temperature sensors 102 can adopt a multi-channel high-speed analog-digital conversion interface cooperating with an isolation type signal conditioning circuit to realize synchronous sampling and anti-interference processing of temperature signals of different measuring points, and ensure the accuracy and time consistency of the acquisition.

[0113] The dynamic trajectory calculation module 1032 is configured to construct a dynamic trajectory of the current water temperature in a temperature inertia field based on the water temperature sampling data and a preset temperature inertia field model.

[0114] In this embodiment, the dynamic trajectory calculation module 1032 can perform delay response modeling, residual fitting and trajectory curve generation and other operation tasks through an embedded digital signal processor or a floating-point operation unit to obtain the trajectory of the current control period.

[0115] The disturbance residual matrix processing module 1033 is configured to, at the end of each control period, construct a residual vector according to a circulating pump frequency input and an ideal response prediction value of a water temperature sampling sequence, generate a topological diffusion kernel combining a hydraulic connection relationship of the heating pipe network, perform spatial convolution on the residual vector to form a disturbance residual matrix, and simultaneously update the disturbance residual matrix according to a forgetting factor.

[0116] In this embodiment, the disturbance residual matrix processing module 1033 can run on a multi-core microprocessor or an FPGA platform with a matrix operation optimization instruction set to realize efficient matrix calculation and real-time update.

[0117] The frequency adjustment control module 1034 is configured to calculate an attractive potential state between the current water temperature and a target constant temperature value based on a trajectory feature vector output by the dynamic trajectory calculation module 1032 and an updated disturbance residual matrix output by the disturbance residual matrix processing module 1033, and generate a frequency bandwidth interval control instruction of the circulating pump according to the attractive potential state.

[0118] In this embodiment, the frequency adjustment control module 1034 can execute a closed-loop control algorithm through a real-time operating system, and output a control signal to a circulating pump actuator through a pulse width modulation or variable frequency drive interface.

[0119] It is understood that the controller 103 in this embodiment can be an embedded control platform with industrial protection levels, such as an ARM Cortex-A series multi-core processor equipped with large-capacity DDR memory and non-volatile memory to support data buffering during long-term operation and multiple rounds of iterative calculations of historical control cycles. Furthermore, the controller 103 can also exchange data with a host computer monitoring system via Ethernet, RS-485, or a wireless communication module to facilitate remote monitoring and optimization of operating parameters, which is not specifically limited in this application.

[0120] Before developing the specific technical content corresponding to the steps, this application needs to emphasize again:

[0121] In typical heating network operation, the response of water temperature to the frequency adjustment of the circulating pump is not linear, but rather influenced by multiple factors, including network thermal inertia, fluid transport delays, and the superposition of load disturbances. Especially in conditions with long pipelines, complex branching structures, or frequent user-side load fluctuations, water temperature trends often exhibit significant lags and overshoots. This makes traditional control strategies based on immediate deviation feedback prone to over-response or hysteresis during frequency adjustment, making it difficult to maintain a constant temperature target over a long period of time.

[0122] Furthermore, due to the hydraulic connectivity between nodes in the network, load disturbances at one location can propagate along topological paths to other nodes, causing cumulative local temperature fluctuations. If the control strategy fails to identify and dynamically correct for the spatial diffusion of these disturbances, frequent adjustments at a single control point can often result in an uneven overall temperature distribution.

[0123] Based on the above problems, this application is not limited to direct PID control algorithm adjustment of a single temperature measurement point, but adopts a dual modeling approach of temperature inertia field and disturbance residual matrix:

[0124] The temperature inertia field is used to characterize the delay characteristics of water temperature in the time dimension and the attraction trend of the target constant temperature value, that is, the dynamic trajectory; the disturbance residual matrix analyzes the difference between the actual water temperature and the ideal response in the historical control cycle, and combines the hydraulic connectivity topology of the pipe network to form a dynamic data structure that can reflect the diffusion and attenuation laws of the disturbance in space.

[0125] It is understandable that when making control decisions, the dynamic trajectory and the disturbance residual matrix are calculated in parallel. By calculating the attractive potential energy state, it can not only reflect the trend information of the current temperature change, but also superimpose the spatial influence of historical disturbances, thereby achieving higher adaptability and predictability in the setting of the frequency bandwidth.

[0126] It should be noted that this application does not rely on a fixed pipe network partition structure or a single mathematical control model, but is aimed at heating thermal systems with significant response delays, strong spatial coupling, and high randomness of load disturbances. The advantage of this application is that by coupling the spatial gradient direction of the disturbance with the dynamic trajectory eigenvector, it is possible to identify potential trends that will cause overshoot or hysteresis in the control calculation, so as to compensate and limit the frequency bandwidth allocation in advance. This processing method avoids the limitation of relying solely on the size of the temperature deviation for judgment, so that in complex scenarios with multiple users, long pipelines and multiple branches, it can still maintain a high constant temperature stability and adjustment accuracy.

[0127] Next, the principle of the perturbation residual matrix of the method of the present application is further expanded.

[0128] It is easy to understand that in this application, perturbing the residual matrix includes two parts: construction and updating.

[0129] In the construction part, the operating data of the first k historical control cycles are first comprehensively processed. These data include the circulating pump frequency input within each cycle and the corresponding actual water temperature sampling sequence. At the same time, the water temperature change sequence calculated by the controller based on the ideal response model is combined with the difference between the two to obtain a residual vector reflecting the size and direction of the deviation. The residual vector not only records the temperature deviation in the time dimension, but also combines the hydraulic connectivity topology of the heating pipe network in subsequent processing to map it into a residual field matrix that reflects the propagation relationship of the disturbance on the spatial path. When constructing the residual field matrix, a topological diffusion kernel operation is introduced, so that the diffusion and attenuation laws of the disturbance influence can be exhibited between adjacent nodes or nodes with high hydraulic coupling, thereby giving the disturbance residual matrix the ability to express spatial correlation.

[0130] Among them, the ideal response model is constructed according to the factory settings of the heating network.

[0131] It is understandable that when the present application is initially put into operation or the control method is first activated, there is no disturbance residual matrix and temperature inertia field formed by the accumulation and calculation of historical data inside the controller. The reason is that these two core data structures rely on the collection, processing and modeling of the actual operating data of the heating network in different control cycles to form a representative mathematical description. The disturbance residual matrix needs to be calculated based on the difference between the actual water temperature change and the ideal predicted water temperature change in multiple historical control cycles, while the temperature inertia field needs to be established by fitting the delay characteristics, change trends and potential energy distribution between the target constant temperature value and the actual water temperature response. In the initial stage without historical operation records, this information cannot be obtained directly. Therefore, the initial version of the disturbance residual matrix and temperature inertia field must be gradually generated through the accumulation, calculation and iterative update of data from the first several control cycles, so as to provide a reliable reference basis for the subsequent attractive potential energy calculation and frequency regulation.

[0132] In this application, the specific value of k can be determined by those skilled in the art through experiments, but at least three control cycles are required, and this application does not impose too many restrictions here.

[0133] In one example, the construction of the perturbation residual matrix includes the following steps:

[0134] After the kth control cycle ends, the actual water temperature sampling sequence within the first k control cycles is obtained and the ideal water temperature sampling sequence predicted by the controller , where j represents a single control cycle, and the specific form of the sampling sequence is as follows:

[0135] ,

[0136] Where p represents the number of sampling points;

[0137] Specifically, after the kth control cycle, two types of data are required for the preceding k control cycles: one is the actual water temperature sampling sequence, and the other is the water temperature variation sequence calculated based on the ideal response model. By comparing the actual operating results with the results calculated by the ideal response model, the deviation characteristics of the control process can be quantitatively reflected, providing basic data for subsequent disturbance analysis.

[0138] In this embodiment, the sampling sequence is recorded sequentially within the control cycle, and each sampling point is timestamped to ensure consistency between subsequent residual calculations and time-series response analysis. The predicted value of the ideal response model is calculated based on the pump frequency input of the circulating pump within the corresponding control cycle using a predetermined thermodynamic transfer function and hydraulic characteristic curve, ensuring consistency between the predicted data and the actual data in terms of input conditions.

[0139] According to the actual water temperature sampling sequence corresponding to each control cycle j and the ideal water temperature sampling sequence predicted by the controller, the residual vector is calculated by difference;

[0140] In this embodiment, a residual vector is generated by calculating the difference between the two sampling sequences at each time point. This residual vector not only reflects the deviation in magnitude between the water temperature change and the ideal state, but also preserves the phase information of the fluctuation, facilitating subsequent determination of whether the disturbance is lagging or leading.

[0141] In one optional embodiment, when forming the residual vector, to reduce the interference of sampling noise, a window width smaller than the control period can be set, and the residual vector is weighted averaged. The weight distribution is centrosymmetrical, with the current sampling point as the center and the weights of adjacent points decreasing. This preserves the main trend of the disturbance while effectively suppressing high-frequency noise caused by sensor accuracy, communication delays, or transient external disturbances.

[0142] According to the hydraulic connectivity of the heating network, calculate The adjacency matrix A of order is calculated according to the adjacency matrix, and the corresponding Laplace matrix is ​​calculated. , where D represents the degree matrix of the adjacency matrix;

[0143] Specifically, the adjacency matrix A is used to capture the concept of which measuring points in the pipe network have direct hydraulic connectivity with each other in a square matrix format, facilitating the subsequent quantitative description of the spatial propagation of disturbances. During construction, the node set (i.e., measuring points or computational nodes involved in the analysis, such as main pipe nodes, branch inlets, and terminal heat exchange points) is first determined. Then, based on as-built drawings, 3D models, valve openings, check valve directions, and on-site flow direction indicators, a direct hydraulic connection between two nodes is determined. If two points are directly connected through the same pipe segment, or form a continuous path through the shell or tube side of the same equipment, the corresponding row and column positions in the adjacency matrix are marked as connected; otherwise, they are marked as disconnected.

[0144] Furthermore, for branches with obvious master-slave flow directions, the connectivity relationship can be set to direction-sensitive: upstream to downstream is marked as connected, and the reverse direction is marked as not connected if it is restricted by check or throttling.

[0145] It's understandable that the adjacency matrix A is essentially a computable connectivity table for the pipe network skeleton, providing the foundation for subsequent diffusion, coupling, and weight assignment. In practice, however, disturbance residues do not remain confined to their origin; they diffuse or decay along the actual hydraulic path. Only by explicitly specifying connectivity within the matrix can subsequent spatial processing steps follow physically accessible pathways, avoiding spurious propagation across pipes and sections, and improving the credibility and interpretability of subsequent calculations.

[0146] Further, the degree matrix D is used to record the number of each node directly connected to the outside or the comprehensive communication strength, in actual generation, only the statistical value of each row (or column) of the adjacency matrix is taken to fill in the main diagonal, and the rest is empty, which can be intuitively understood as:

[0147] A certain node is connected to several pipes, and has several direct entrances and exits, and the number is written in the corresponding dimension; if weighted communication is used (for example, the weight of thick pipes is large, the weight of long distances is small, and the weight of half-open valves is half), the weight is added and written.

[0148] Further, the Laplace matrix can be regarded as a constraint table:

[0149] The cell of each node itself fills in the total amount of communication, and the nodes directly connected to it fill in a communication mark of opposite meaning (indicating a coupling restraint relationship between the two) in the corresponding position, and the rest is empty.

[0150] It can be understood that the role of the constraint table is to fix the law of energy conservation and local coupling when subsequent disturbance diffusion and coupling calculation is performed:

[0151] In a popular understanding, a node wants to allocate residual disturbance to the periphery, and how much can be allocated and who will be allocated is completely determined by the structure of the table.

[0152] Reference Figure 4 , Figure 4 The figure is a matrix calculation process schematic diagram of the embodiment of the application.

[0153] Figure 4 In the figure, it is assumed that there is a secondary network main pipe connected to three household branches in turn, marked as N1 (upstream measuring point of the main pipe), N2 (first household branch inlet), N3 (second household branch inlet), N4 (third household branch inlet), and N5 (return side measuring point of the main pipe), N1 and N2 are directly connected to the same pipe, N2 and N3 are the same, N3 and N4 are the same, N4 and N5 are the same; each household branch inlet is connected only to the main pipe, and the internal branch does not participate in this time of communication determination.

[0154] Figure 4 Further, the adjacency matrix is shown, at this time, the meaning of the adjacency matrix can be defined as: N1 and N2 have communication, N1 and N3 / N4 / N5 have no communication; N2 and N1, N3 have communication, and N4 / N5 have no communication; N3 and N2, N4 have communication, and N1 / N5 have no communication; N4 and N3, N5 have communication, and N1 / N2 have no communication; N5 and N4 have communication, and N1 / N2 / N3 have no communication.

[0155] In an embodiment not shown in the figure, if there is an obvious unidirectional flow in the main pipe (for example, from N1 to N5), the upstream to downstream flow can be marked as strong connection, and the reverse connection can be set to weak or no connection depending on whether there is a check valve or throttling device on site.

[0156] Figure 4 The degree matrix is ​​further shown. The main diagonal of the degree matrix only needs to write the number of connections of each node: N1 is 1, N2 is 2, N3 is 2, N4 is 2, and N5 is 1;

[0157] In an embodiment not shown in the figure, if weighted connectivity is adopted, for example, the large-diameter main pipe connectivity weight is set to 1, and a certain section is only taken as 0.5 because the valve is half open, then the main diagonal of the corresponding node can be written into the accumulated weight, and this application will not go into details here.

[0158] Figure 4 The Laplace matrix is ​​further shown. The meaning of the Laplace matrix can be understood as follows:

[0159] The corresponding degree is filled in the grid corresponding to each node, and the positions directly connected to it are filled with a connectivity mark with the opposite meaning (indicating that the two are restrained from each other), and the unconnected positions remain blank.

[0160] It is easy to understand that if overheating residue appears at the N2 position, it will preferentially affect N1 and N3; if the N5 end valve position is temporarily closed, after the connection relationship is updated, N4 will no longer release the disturbance to N5, and the residue will be more transferred back and forth between N3 / N4.

[0161] According to the preset diffusion coefficient , calculate the topological diffusion kernel , where e represents a matrix exponential operation, performing spatial convolution of the residual vector with the topological diffusion kernel to generate a residual field matrix;

[0162] Specifically, the so-called topological diffusion kernel is used to propagate and attenuate the temperature residuals generated at one or several nodes during the previous control cycle along the truly accessible hydraulic connectivity path, so that the residual impact does not only remain at the source point, but is gradually distributed among adjacent nodes according to the distance, accessibility, and resistance size based on the connection relationship of the pipe network. Considering that the secondary heating network has obvious directionality and resistance differences, the calculation of the diffusion kernel is based on weighted connectivity. The weight can comprehensively consider factors such as the equivalent hydraulic resistance of the pipe section, the proportion of design flow, the valve position opening, and the historical throughput. For branches with one-way check or obvious main flow direction, a larger accessibility weight is given in the direction from upstream to downstream, and a smaller weight is given or directly blocked in the reverse direction, so that the diffusion behavior is consistent with actual hydraulics.

[0163] In this embodiment, the numerical calculation of the topological diffusion kernel is performed using scaling and rational approximation with good numerical stability, which is applicable to sparse and large-scale connectivity matrices. Specifically, it includes the following two aspects:

[0164] First, the connectivity matrix, or Laplacian matrix, is scaled according to its maximum eigenvalue and decomposed into multiple small-step equivalent diffusions. Within each step, the diffusion operator is calculated using rational approximation, ultimately resulting in a symmetric or nearly symmetric kernel matrix. To reduce online overhead, the topological diffusion kernel can be cached based on the topology version number: when valve positions, branch openings and closings, or flow direction configurations remain unchanged, the diffusion kernel is reused across multiple control cycles. When a connectivity change is detected or the diffusion coefficient is adjusted by policy, an incremental recalculation is triggered, updating only the affected rows and columns.

[0165] Secondly, the value of the diffusion coefficient is determined by two types of information: the operating period, which provides a baseline range; and online state variables (the water temperature change rate and frequency bandwidth convergence trend for the current cycle), which are used to make further corrections within the baseline range. For example, during the initial temperature rise phase, when a high temperature change rate and rapid frequency bandwidth convergence are detected, the diffusion coefficient is shifted upward to more quickly map the residual effect to downstream nodes, facilitating early convergence of the control strategy. During the nighttime load phase, when the residual disturbance decays slowly, the diffusion coefficient is shifted downward to achieve more localized diffusion, preventing oversmoothing from masking small but persistent local deviations.

[0166] It can be understood that by performing a topological diffusion on the kernel matrix through the diffusion coefficient, a set of diffused residual weights can be obtained. This set of weights represents how the previous deviation will affect the surrounding nodes along the network if no new control action is applied in this cycle;

[0167] Furthermore, based on the aforementioned residual weights and combined with the residual vectors, a residual field matrix is ​​generated. The residual field matrix reflects the residual influence of each node on itself in the diagonal positions, and reflects the coupling strength between any two nodes due to connectivity and proximity in the off-diagonal positions. The resulting residual field matrix has clear characteristics: on the one hand, it maintains the correspondence with the source residual amplitude, and on the other hand, it explicitly encodes the spatial coupling, providing a direct quantitative basis for subsequent attractive potential energy assessment and frequency bandwidth contraction. This enables earlier identification of disturbances that will be amplified or refluxed at neighboring nodes, reducing overshoot or repeated adjustments in the next cycle.

[0168] For example, the above example of connecting the main pipes N1 to N5 in series is used to illustrate the intuitive results after diffusion and convolution.

[0169] Assuming a primary overheating residual occurs at position N2, and the valve remains fully open, with the main flow direction shifting from N1 to N5, after diffusion kernel processing, N2's residual impact remains primarily its own, but a portion is distributed to N1 and N3. Due to the larger weight of the main flow direction, the share allocated to N3 is slightly higher than that to N1; N4 and N5 receive only a very small share or are largely unaffected. Using this set of diffused weights to generate the residual field matrix reveals a high coupling strength between N2 and N3, suggesting that the regulation strategy should prioritize tightening the frequency bandwidth and controlling the adjustment slope in the N2 to N3 region to prevent the disturbance from being pushed further downstream.

[0170] Furthermore, if the N5 valve position is temporarily closed at this time, the connectivity relationship is updated, the diffusion kernel changes accordingly, N4 no longer releases the residual to N5, and the high-value area of ​​the residual field matrix is ​​concentrated between N3 and N4. Based on this, the control strategy moves the center of gravity of the limit and slow adjustment forward to avoid pushing the disturbance to the closed end.

[0171] Calculating a forgetting factor according to the pump frequency input corresponding to the control cycle, weighting the residual field matrix by the forgetting factor, and summarizing the weighted residual field matrix to obtain a disturbance residual matrix;

[0172] Specifically, the forgetting factor is introduced to reflect the attenuation of the impact of historical disturbances on current control decisions. Because the operating environment and load characteristics of the heating network change over time, the residuals from different control cycles have different reference values ​​for the current cycle: the earlier the cycle, the weaker the direct impact of its disturbance on the current state; conversely, disturbances from the most recent cycle better reflect the inertia and trends of the current operating conditions.

[0173] In this embodiment, the weighting process is performed in a matrix element-by-element multiplication manner, multiplying k residual field matrices by the corresponding forgetting factors to realize the calculation of the residual distribution after attenuation. Subsequently, all the weighted residual field matrices are superimposed and summarized in chronological order to obtain a disturbance residual matrix that comprehensively reflects the multi-period disturbance distribution characteristics. The disturbance residual matrix not only retains the disturbance pattern of the latest cycle, but also integrates the residual influence of the previous cycles, and automatically suppresses the weight of the outdated information through the forgetting factor, so that the overall description is both timely and continuous. When evaluating the state of attractive potential energy, it is possible to more accurately judge whether the disturbance in a certain area is an occasional fluctuation or a continuity problem based on a disturbance map that includes both historical trends and emphasizes recent working conditions, avoiding excessive adjustments due to short-term anomalies and preventing long-term local deviations from being ignored.

[0174] In the update phase, the perturbation residual matrix is ​​not simply replaced. Instead, a weighted attenuation of the existing perturbation residual matrix is ​​performed by updating a forgetting factor related to the pump frequency input during the control cycle, gradually reducing the impact of historical perturbations over time. Simultaneously, a new residual field is calculated based on the actual water temperature sampling data of the current control cycle and superimposed on the attenuated matrix, forming an updated result that dynamically reflects the recent perturbation state.

[0175] In one example, the disturbance residual matrix is ​​updated based on the pump frequency input of the previous control cycle and the current water temperature of the current control cycle. The specific updating method is as follows:

[0176] Get the perturbation residual matrix output of the previous control cycle ;

[0177] The forgetting factor of the disturbance residual matrix is ​​updated according to the pump frequency input of the previous control cycle to obtain a weighted value ;

[0178] Calculate the residual field matrix of the current control cycle according to the current water temperature , calculate the updated perturbation residual matrix according to the perturbation residual matrix of the previous control cycle, the weighted value and the residual field matrix of the current control cycle , the specific calculation method is as follows:

[0179] .

[0180] How to calculate the residual field matrix, forgetting factor and perturbation residual matrix has been described in the previous content, and this application will not go into details here. It should be noted that the residual vector of the current water temperature here can be calculated through the target constant temperature value, and there is no need for an additional water temperature prediction step.

[0181] Next, the part of the method of this application regarding dynamic trajectory is further expanded.

[0182] For reference Figure 5 , Figure 5 Schematic diagram of the flow of the dynamic trajectory generation method according to an embodiment of the present application.

[0183] In one example, constructing the dynamic trajectory of the current water temperature in the temperature inertia field includes:

[0184] S1.1: Collect the circulating pump frequency control instructions and corresponding water temperature response data from at least the previous M control cycles, and construct a chronologically ordered historical control state sequence. Each state unit in the historical control state sequence includes a timestamp, a frequency increment value, a water temperature change rate, and a sampling interval duration.

[0185] It should be noted that the value of M here can be set by experiment, but it needs to satisfy M less than k+1;

[0186] Specifically, the purpose of constructing a historical control state sequence is to quantitatively record the dynamic relationship between the regulation behavior of the circulation pump and the water temperature changes over multiple control cycles, thereby providing computable input for subsequent response delay analysis.

[0187] In this embodiment, the historical control state sequence captures not only the absolute frequency of the circulating pump but also uses the frequency change within each cycle as the primary indicator. This eliminates fixed deviations under varying base loads and focuses on the impact of the control action itself. Furthermore, the water temperature change rate is calculated by dividing the temperature difference between adjacent sampling points by the duration of the sampling interval, providing a direct reflection of the rate of temperature rise or fall caused by the control action. The introduction of timestamps ensures the time alignment accuracy of multi-cycle data when calculating the delay distribution, avoiding offset errors caused by data asynchrony.

[0188] S1.2: Calculating the response delay distribution characteristics of the water temperature in the heating network to the frequency control action based on the historical control state sequence, and calculating the response offset function by fitting the residual curve;

[0189] Specifically, the water temperature response of the heating network exhibits a significant delay. This means that after adjusting the circulating pump frequency, the water temperature change does not occur immediately, but rather takes time to manifest. This response delay depends not only on the water flow time but also on factors such as the heat capacity of the pipeline and the heat dissipation rate at the user end. Therefore, the delay distribution can vary at different nodes and during different operational phases. Without quantitative analysis of the delay distribution, the match between control commands and water temperature responses will decrease, resulting in delayed or overshooting adjustments.

[0190] In this embodiment, the delay distribution is calculated by cross-correlating the historical frequency increment sequence with the corresponding water temperature rate of change sequence, thereby determining the degree of correlation between the two at different delay times. To eliminate measurement noise and temperature disturbances caused by factors other than frequency control, a residual curve fitting approach is used to compare the observed actual water temperature changes with the idealized changes based on the zero delay assumption. The time distribution of the residuals reflects the shape and magnitude of the delay, and the corresponding response offset function is calculated based on this time distribution of the residuals.

[0191] S1.3: Fitting the current water temperature to the response offset function to obtain a relative offset trajectory curve with the target constant temperature as a reference point, and outputting a dynamic trajectory, wherein the relative offset trajectory curve has the sampling identification point as the horizontal axis and the output value of the response offset function as the vertical axis;

[0192] Specifically, the core of the dynamic trajectory is to depict the motion of the current water temperature within the temperature inertia field. The response offset function ensures that the trajectory not only reflects temperature changes but also accounts for relative position corrections caused by delays. Using the target constant temperature as a reference point, the vertical axis of the trajectory can be converted into a relative offset.

[0193] In this embodiment, the fitting process first matches the water temperature sampling sequence within the current control cycle point by point according to the sampling identifier and the response offset function, so that each temperature sample value corresponds to a delay correction. Subsequently, the corrected temperature value is converted into a relative offset (i.e., the difference between the current value and the target constant temperature value), and a two-dimensional curve is formed with the sampling identifier axis. The direction of the two-dimensional curve in the sampling sequence, the fluctuation amplitude and trend on the vertical axis, constitute the entire information of the dynamic trajectory. The shape of the trajectory directly reflects whether the current water temperature is tending to be stable in the sampling dimension, whether there is continuous deviation or oscillation.

[0194] refer to Figure 6 , Figure 6 This is a schematic diagram of the dynamic trajectory of an embodiment of the present application.

[0195] Figure 6 The horizontal axis is shown as sampling identification points N1 to N5, which represent the sampling positions set in sequence along the water flow path of the heating network during the current control cycle. The vertical axis is the relative offset corrected based on the response offset function, and the target constant temperature value is taken as the zero reference point.

[0196] Figure 6 It is further shown that the hollow marks represent the current temperature of each sampling point, and the solid marks represent the corrected offset value corrected by the response offset function based on the current temperature and the response delay characteristics of the water temperature to the frequency control command in multiple historical control cycles.

[0197] Comparing the vertical axis between the hollow and solid markers reveals the impact of temperature inertia on water temperature distribution. When the corrected offset converges toward zero relative to the current offset, inertia is helping the temperature stabilize and return to the target constant. Conversely, if the corrected offset moves away from zero, inertia may be pushing the temperature further away from the target. The overall direction of the curve reveals the spatial variation of water temperature and the effects of inertia correction on trajectory morphology.

[0198] It should be noted that the corrected temperature here is not for the accuracy of the current water temperature sampling value itself. When the water temperature sensor collects the current temperature, its measurement value has an engineering acceptable accuracy and does not require error correction.

[0199] The correction described in this embodiment refers to the introduction of a response offset obtained based on the deviation relationship between the frequency control action and the water temperature change in the historical control cycle when constructing the dynamic trajectory, and superimposing it as an inertia correction value to the current water temperature sampling sequence to reflect the actual response position of the heating pipeline network in the temperature inertia field.

[0200] Specifically, in the historical control cycle, the influence of the change in the circulation pump frequency on the water temperature has a delayed effect, that is, after the control action is executed, the water temperature does not immediately reach the new equilibrium value, but goes through a transition process of a certain period of time. In the current control cycle, the actual temperature value of each sampling point collected is superimposed on the predicted offset, and the target constant temperature value is used as the reference zero point. The result is converted into a relative offset, thereby obtaining Figure 6 The relative offset trajectory curve shown not only reflects the relative position of each sampling point to the target constant temperature value, but also reflects the trend direction and inertia strength of temperature change. For example, at point N5, the current temperature is close to the target value, but the curve still maintains an upward trend, indicating that this location is still in the temperature rising inertia stage.

[0201] Next, the part of the method of the present application regarding the attractive potential energy state is further expanded.

[0202] It can be understood that the attractive potential energy state is used in this application to characterize the stability trend and regression ability of the current water temperature under the action of the target constant temperature value. Its essence is to organically combine the dynamic trajectory information of the water temperature with the spatial distribution information of the disturbance residual matrix to determine whether the water temperature change tends to be attracted by the target constant temperature value and gradually stabilize, or shows a trend of deviation or even divergence under the influence of external disturbances.

[0203] In one example, calculating the attractive potential energy state between the current water temperature and the target constant temperature value based on the dynamic trajectory and the disturbance residual matrix includes:

[0204] S2.1: Update the disturbance residual matrix output in the previous control cycle according to the pump frequency input in the previous control cycle and the current water temperature in the current control cycle to obtain an updated disturbance residual matrix;

[0205] The updating of the perturbation residual matrix has been described in detail in the foregoing content and will not be elaborated on here in this application.

[0206] S2.2: Determine the offset, rate of change, and trend direction of the current water temperature relative to the target constant temperature value based on the dynamic trajectory, and calculate the corresponding trajectory feature vector;

[0207] Specifically, this step aims to extract the dynamic trajectory, a two-dimensional description of water temperature fluctuations, into a characterization vector that can be numerically manipulated with the perturbation residual matrix. The dynamic trajectory itself only reflects the relative offset and trend of the sampling points and cannot directly contribute to the spatial distribution of the disturbance. Therefore, it is necessary to quantify the offset amplitude, rate of change, and trend direction of each sampling location in the trajectory to form a trajectory feature vector.

[0208] In this embodiment, the construction of the trajectory feature vector includes the following three dimensions:

[0209] In the first dimension, the correction offset of each sampling mark point in the current control cycle is normalized to make the values ​​at different locations comparable;

[0210] In the second dimension, the offset change rate between adjacent sampling points is calculated to obtain the local change rate, and the direction of the change trend (approaching the target constant temperature value or moving away from the target constant temperature value) is determined by the sign.

[0211] In the third dimension, the offset magnitude, rate of change, and trend direction are encoded as a triplet vector and arranged in the order of the sampling points to produce the trajectory feature vector. This constructed vector reflects the overall trend while retaining the details of local changes, making its matching analysis with the perturbation residual matrix more targeted.

[0212] S2.3: Couple the trajectory eigenvector and the updated perturbation residual matrix to calculate the attractive potential energy state;

[0213] Specifically, the core of this step is to achieve the coupled calculation of the trajectory eigenvector and the updated perturbation residual matrix. This is because a simple trajectory change can be caused by external perturbations or internal inertial adjustments, while the perturbation residual matrix records the spatial accumulation and propagation characteristics of historical perturbations. Through this coupled calculation, it is possible to determine whether the trajectory change is consistent with the distribution trend of the residual perturbation, thereby distinguishing whether it is a deviation driven by perturbations or a spontaneous stabilization process.

[0214] It can be understood that the attractive potential energy state can comprehensively reflect the return ability of water temperature under the action of the target constant temperature value and the degree of influence of external disturbances, and has a dual criterion: on the one hand, if the direction of trajectory change is consistent with the propagation direction of high-intensity disturbance, the potential energy state decreases, suggesting that the control strategy needs to increase the driving force; on the other hand, if the direction of trajectory change is opposite to the propagation direction of disturbance, the potential energy state increases, and the control strategy can appropriately converge the frequency bandwidth to reduce unnecessary adjustments.

[0215] It should be noted that the attractive potential energy state here is the aggregated overall potential energy state. In some optional implementations, the circulation pumps can be distributed to achieve the precise purpose of zoning control.

[0216] In one example, coupling the trajectory feature vector and the updated perturbation residual matrix includes:

[0217] S2.3.1: Determine a disturbance-sensitive directional factor representing the current water temperature change trend based on the trajectory characteristic vector;

[0218] In this embodiment, the directional factor is extracted by statistically analyzing the rate of change and trend direction of the trajectory feature vector to form a local directional component for each sampling location. These local components are then weighted averaged, with the weights determined based on the hydraulic importance or thermal weight distribution of each location in the pipe network structure. For example, upstream nodes close to the circulation pump can be given a higher weight due to their greater impact on the overall water temperature.

[0219] S2.3.2: Extracting a perturbation submatrix corresponding to the perturbation-sensitive directional factor from the updated perturbation residual matrix;

[0220] Specifically, since the perturbation residual matrix completely records the residual perturbation transmission relationship between each node, and the directional factor only involves the nodes on the current trend path, there is no need to operate on the entire matrix. This can not only reduce the amount of calculation, but also focus on the perturbation area directly related to the current trend.

[0221] In this embodiment, the method for extracting the perturbation submatrix is: according to the sampling node numbers involved in the directional factor, the corresponding rows and columns are selected from the perturbation residual matrix to form a smaller matrix. This submatrix specifically describes the perturbation propagation between these nodes and between them and the surrounding nodes.

[0222] S2.3.3: Perform directional matching analysis on the disturbance-sensitive directional factor and the disturbance submatrix to determine the impact of the residual disturbance on the current trajectory trend;

[0223] It is understandable that the core logic of matching analysis lies in:

[0224] If the main direction of disturbance propagation is the same as the current water temperature change trend, it can be inferred that the disturbance is pushing the temperature change in that direction; otherwise, it can be inferred that the disturbance is hindering the development of the current trend.

[0225] In this embodiment, the matching analysis is achieved by calculating the angle between the direction factor and the disturbance propagation vector in the sub-matrix. The smaller the angle, the more consistent the two are. An angle close to 180 degrees indicates opposite trends.

[0226] S2.3.4: Based on the results of the direction matching analysis, calculate the attraction stability index of the current water temperature on the control path to obtain the attraction potential energy state;

[0227] In an optional embodiment, the numerical range of the stability index can be set according to engineering experience. For example, the closer the value is to 1, the more conducive it is for the temperature to return to the target value, and the closer it is to 0, the greater the disturbance, making it difficult to stabilize near the target value.

[0228] In this embodiment, the stability index is calculated by combining directional consistency weights, disturbance intensity weights, and the node's location weight within the network's thermal structure to form a comprehensive evaluation value that combines multiple factors. Finally, this evaluation value is mapped to an attractive potential energy state for direct use in regulatory control.

[0229] The width and center position of the frequency bandwidth interval are determined according to the attractive potential energy state and the disturbance residual matrix.

[0230] In one example, according to the attractive potential energy state, the controller outputs a frequency bandwidth interval, including:

[0231] S3.1: Determine the frequency center offset of the current control cycle based on the attractive potential energy state with reference to the frequency center value of the previous control cycle, wherein the frequency center offset is inversely proportional to the attractive potential energy state;

[0232] S3.2: Determine an adjustment coefficient for a frequency bandwidth interval based on the disturbance intensity on the current control path in the disturbance residual matrix, wherein the adjustment coefficient is used to control a width convergence rate of the frequency bandwidth;

[0233] S3.3: Output the corresponding frequency bandwidth interval according to the frequency center offset and the adjustment coefficient, combined with the frequency bandwidth interval of the previous cycle.

[0234] In one example, the method further includes:

[0235] Determining a sampling period of the water temperature sensor according to an operating period of the heating network;

[0236] The operating period includes an initial heating stage, a stable heating stage and a nighttime load stage, and the target constant temperature value of each stage is different.

[0237] In one example, determining the sampling period of the water temperature sensor according to the operating period of the heating network includes:

[0238] If it is in the initial temperature rise stage, set the first sampling period;

[0239] If the heating stage is stable, a second sampling period is set, wherein the first sampling period is smaller than the second sampling period;

[0240] If it is in the night load stage, setting a third sampling period, wherein the second sampling period is smaller than the third sampling period;

[0241] The first sampling period, the second sampling period and the third sampling period are adjusted according to the disturbance convergence rate in the disturbance residual matrix, the dynamic trajectory of the current water temperature in the temperature inertia field and the offset of the historical adjustment of the circulation pump frequency.

[0242] It can be understood that the disturbance convergence rate is calculated based on the load change in the disturbance residual matrix, and the offset is obtained by summing and averaging the historical frequency adjustments of the circulating pump.

[0243] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for controlling HVAC constant temperature flow, applied to a heating network, characterized in that: The heating pipe network includes a controller, a circulation pump, a water temperature sensor, and a pipe network body. The controller is configured with a temperature inertia field and a disturbance residual matrix, wherein the temperature inertia field performs virtual potential energy simulation according to a preset target constant temperature value, and the disturbance residual matrix is ​​constructed according to the disturbance residual between the water temperature change and the ideal change in a historical control cycle. The control method includes: The current water temperature is collected by the water temperature sensor, and based on the pump frequency input and water temperature output changes of the circulation pump in multiple historical control cycles, a dynamic trajectory of the current water temperature in the temperature inertia field is constructed, where the current water temperature is a water temperature sampling sequence collected in the current control cycle; Calculating the attractive potential energy state between the current water temperature and the target constant temperature value according to the dynamic trajectory and in combination with the disturbance residual matrix; According to the state of the attractive potential energy, a frequency bandwidth interval is outputted by a controller to control the circulation pump to perform frequency regulation within the frequency bandwidth interval, thereby obtaining a pump frequency input of the circulation pump in the current regulation period.

2. A method for controlling HVAC constant temperature flow according to claim 1, characterized in that: The construction of the dynamic trajectory of the current water temperature in the temperature inertia field includes: Collect the circulating pump frequency control instructions and corresponding water temperature response data within at least M previous control cycles, and construct a historical control state sequence arranged in chronological order. Each state unit in the historical control state sequence includes a timestamp, a frequency increment value, a water temperature change rate, and a sampling interval duration; Calculating the response delay distribution characteristics of the water temperature in the heating network to the frequency control action based on the historical control state sequence, and calculating the response offset function by fitting the residual curve; The current water temperature is fitted with the response offset function to obtain a relative offset trajectory curve with the target constant temperature value as a reference point, and a dynamic trajectory is output, wherein the relative offset trajectory curve has the sampling identification point as the horizontal axis and the output value of the response offset function as the vertical axis.

3. The method for controlling HVAC constant temperature flow control according to claim 1, characterized in that: The construction of the disturbance residual matrix is ​​based on the historical measurement and prediction data of the previous k control cycles, where the prediction data is the water temperature change sequence calculated by the controller based on the ideal response model according to the pump frequency input of the circulation pump. The specific construction method is as follows: After the kth control cycle ends, the actual water temperature sampling sequence within the first k control cycles is obtained and the ideal water temperature sampling sequence predicted by the controller , where j represents a single control cycle, and the specific form of the sampling sequence is as follows: ， Where p represents the number of sampling points; According to the actual water temperature sampling sequence corresponding to each control cycle j and the ideal water temperature sampling sequence predicted by the controller, the residual vector is calculated by difference; According to the hydraulic connectivity of the heating network, calculate The adjacency matrix A of order is calculated according to the adjacency matrix, and the corresponding Laplace matrix is ​​calculated. , where D represents the degree matrix of the adjacency matrix; According to the preset diffusion coefficient , calculate the topological diffusion kernel , where e represents a matrix exponential operation, performing spatial convolution of the residual vector with the topological diffusion kernel to generate a residual field matrix; The forgetting factor is calculated according to the pump frequency input corresponding to the control period, the residual field matrix is ​​weighted and assigned according to the forgetting factor, and the weighted residual field matrix is ​​summarized to obtain the disturbance residual matrix.

4. A method for controlling HVAC constant temperature flow according to claim 3, characterized in that: The disturbance residual matrix is ​​updated based on the pump frequency input of the previous control cycle and the current water temperature of the current control cycle. The specific update method is as follows: Get the perturbation residual matrix output of the previous control cycle ; The forgetting factor of the disturbance residual matrix is ​​updated according to the pump frequency input of the previous control cycle to obtain a weighted value ; Calculate the residual field matrix of the current control cycle based on the current water temperature , calculate the updated perturbation residual matrix according to the perturbation residual matrix of the previous control cycle, the weighted value and the residual field matrix of the current control cycle , the specific calculation method is as follows: 。 5. The method for controlling HVAC constant temperature flow control according to claim 1, characterized in that: According to the dynamic trajectory and in combination with the disturbance residual matrix, the attractive potential energy state between the current water temperature and the target constant temperature value is calculated, including: According to the pump frequency input of the previous control cycle and the current water temperature of the current control cycle, the disturbance residual matrix outputted in the previous control cycle is updated to obtain an updated disturbance residual matrix; Determine the deviation amplitude, change rate, and trend direction of the current water temperature relative to the target constant temperature value based on the dynamic trajectory, and calculate the corresponding trajectory feature vector; The trajectory eigenvector and the updated perturbation residual matrix are coupled to calculate the attractive potential energy state.

6. A method for controlling HVAC constant temperature flow according to claim 5, characterized in that: Coupling the trajectory eigenvector and the updated perturbation residual matrix includes: Determining a disturbance-sensitive direction factor representing a current water temperature change trend based on the trajectory characteristic vector; Extracting a disturbance submatrix corresponding to the disturbance sensitive direction factor in the updated disturbance residual matrix; Performing direction matching analysis on the disturbance sensitive direction factor and the disturbance sub-matrix to determine the influence of the residual disturbance on the current trajectory trend; According to the results of the direction matching analysis, the attraction stability index of the current water temperature on the control path is calculated to obtain the attraction potential energy state.

7. The method for controlling HVAC constant temperature flow control according to claim 1, characterized in that: The width and center position of the frequency bandwidth interval are determined according to the attractive potential energy state and the disturbance residual matrix.

8. A method for controlling HVAC constant temperature flow according to claim 7, characterized in that: According to the state of the attractive potential energy, the controller outputs a frequency bandwidth interval including: Taking the frequency center value of the previous control cycle as a reference, determining the frequency center offset of the current control cycle according to the attractive potential energy state, wherein the frequency center offset is inversely proportional to the attractive potential energy state; Determining an adjustment coefficient of a frequency bandwidth interval according to the disturbance intensity on the current control path in the disturbance residual matrix, wherein the adjustment coefficient is used to control a width convergence rate of the frequency bandwidth; According to the frequency center offset and the adjustment coefficient, combined with the frequency bandwidth interval of the previous cycle, the corresponding frequency bandwidth interval is output.

9. The method for controlling HVAC constant temperature flow control according to claim 1, characterized in that: The method further comprises: Determining a sampling period of the water temperature sensor according to an operating period of the heating network; The operating period includes an initial heating stage, a stable heating stage and a nighttime load stage, and the target constant temperature value of each stage is different.

10. A method for controlling HVAC constant temperature flow control according to claim 9, characterized in that: Determining a sampling period of the water temperature sensor according to an operating period of the heating network includes: If it is in the initial temperature rise stage, set the first sampling period; If the heating stage is stable, a second sampling period is set, wherein the first sampling period is smaller than the second sampling period; If it is in the night load stage, setting a third sampling period, wherein the second sampling period is smaller than the third sampling period; The first sampling period, the second sampling period and the third sampling period are adjusted according to the disturbance convergence rate in the disturbance residual matrix, the dynamic trajectory of the current water temperature in the temperature inertia field and the offset of the historical adjustment of the circulation pump frequency.

Images