A building heating system self-adaptive control method based on heat load prediction

Abstract

The application discloses a building heating system adaptive control method based on heat load prediction, and belongs to the technical field of intelligent monitoring and control. The method comprises the following steps: constructing a second-order lumped heat dynamic model of each heating subarea; determining the heat storage of each heating subarea through state estimation, and predicting the free temperature rising trajectory of each heating subarea and the time characteristics of reaching the target temperature; determining the optimal switching time of each heating subarea; identifying the hydraulically sensitive subarea; determining whether to trigger global collaborative scheduling based on the time distribution of the optimal switching time of each heating subarea and the distribution of the hydraulically sensitive subarea; when it is determined that the conditions for triggering global collaborative scheduling are met, the collaborative switching sequence is obtained and is updated rolling; when the subarea switching operation is performed, the current to-be-switched subarea is defined as the target switching subarea, the affected subarea is identified based on the hydraulic influence matrix, the flow deviation of the affected subarea is calculated, and the opening of the corresponding regulating valve is fed forward and compensated.

Description

Technical Field

[0001] This application relates to the field of intelligent monitoring and control technology, and more specifically, to an adaptive control method for building heating systems based on heat load prediction. Background Technology

[0002] As building scale expands and user comfort requirements increase, traditional building heating systems often rely on fixed time periods or simple temperature control strategies, resulting in large room temperature fluctuations, high energy consumption, and inconsistent zone-level regulation. Existing multi-zone heating control methods mostly employ PID or fuzzy control, but under load changes, outdoor temperature fluctuations, and hydraulic coupling effects, they often exhibit response lag, overshoot, or local temperature deviations, making it difficult to balance room temperature accuracy and system stability.

[0003] Meanwhile, existing predictive control methods are mostly designed for single-zone loads and lack the ability to coordinate and optimize hydraulic coupling and switching timing between zones. This makes it difficult to achieve feedforward compensation and rolling updates, resulting in the inability to fully realize the potential for energy saving and comfort.

[0004] In summary, how to achieve precise control of zoned room temperature based on heat load prediction in multi-zone building heating systems, while taking into account hydraulic coupling effects, switching timing coordination, and overall system stability, has become an urgent technical problem to be solved. Summary of the Invention

[0005] In view of the technical problems mentioned in the background section, an adaptive control method for building heating systems based on heat load prediction is provided.

[0006] The technical means employed in this invention are as follows:

[0007] An adaptive control method for building heating systems based on heat load prediction includes the following steps:

[0008] The building heating system is divided into zones, resulting in multiple heating zones;

[0009] Based on the real-time room temperature, supply and return water temperature difference, and branch flow rate of each heating zone, a second-order lumped thermal dynamic model of each heating zone is constructed.

[0010] The heat storage of each heating zone is determined by state estimation, and the free temperature rise trajectory of the room temperature of each heating zone and the time characteristics of reaching the target temperature are predicted based on the heat storage of each heating zone.

[0011] The optimal switching time for each heating zone is determined based on the control criterion of achieving the target room temperature without overshoot.

[0012] The hydraulic influence matrix is ​​used to predict the temperature disturbance of each heating zone to other zones at the optimal switching time, and to identify hydraulically sensitive zones.

[0013] Based on the time distribution of the optimal switching time for each heating zone and the distribution of hydraulically sensitive zones, it is determined whether to trigger global coordinated scheduling.

[0014] When it is determined that the conditions for triggering global coordinated scheduling are met, a coordinated handover timing optimization model is constructed with the optimal handover time as the initial solution, and the coordinated handover sequence is obtained and updated on a rolling basis.

[0015] When performing a zone switching operation, the zone to be switched is defined as the target switching zone. At the same time, the affected zone is identified based on the hydraulic influence matrix, and the flow deviation of the affected zone is calculated. Then, the opening of the corresponding regulating valve is adjusted by feedforward compensation based on the flow deviation.

[0016] Furthermore, constructing the second-order lumped thermal dynamics model for each heating zone includes the following steps:

[0017] Obtain historical operating data for each heating zone within a preset time span. The historical operating data includes: different outdoor temperature ranges and different heating load conditions.

[0018] The thermal dynamic model structure of each heating zone is constructed based on the equivalent thermal resistance-heat capacity network. The thermal dynamic process is equivalent to two state variables: indoor air nodes and building envelope nodes. The thermal balance equations corresponding to each state variable are established according to the thermal dynamic model structure.

[0019] The parameters to be identified are determined based on the thermal dynamic model structure. The parameters to be identified include: indoor air heat capacity, equivalent heat capacity of building envelope, equivalent thermal resistance between indoor air and building envelope, and equivalent thermal resistance between building envelope and outdoor environment.

[0020] Based on the historical operating data, the least squares identification method is used to estimate the parameters to be identified offline to obtain the initial values ​​of the model parameters.

[0021] Substituting the initial values ​​of the model parameters into the thermal dynamic model structure yields the second-order lumped thermal dynamic model for each heating zone, and the initial values ​​of the model parameters are used as the initial conditions for online parameter updates.

[0022] Furthermore, determining the heat storage capacity of each heating zone through state estimation includes the following steps:

[0023] A discrete state-space model is established based on the second-order lumped thermal dynamic model of each heating zone. The indoor air temperature and the equivalent temperature of the building envelope are used as state variables, the heat dissipation power of the heat dissipation equipment is used as the input, and the measured room temperature is used as the observation.

[0024] Based on the sensor accuracy and the statistical characteristics of historical operating data, the process noise covariance matrix and the observation noise covariance matrix are set as Kalman filter parameters;

[0025] Based on the discrete state-space model and Kalman filter parameters, the Kalman filter method is used to recursively estimate the state variables and obtain the posterior estimates of the state variables.

[0026] Based on the difference between the equivalent temperature of the building envelope and the indoor air temperature in the posterior estimate, and in conjunction with the corresponding heat capacity parameters, the heat storage capacity of each heating zone is calculated.

[0027] Furthermore, determining the optimal switching time for each heating zone includes the following steps:

[0028] Based on the free temperature rise trajectory of each heating zone, the earliest and latest switching times when the room temperature reaches the target temperature without overshoot are determined, thus obtaining the switching time interval.

[0029] Candidate switching times are determined based on the switching time interval, and the room temperature overshoot and temperature convergence time corresponding to each candidate switching time are calculated respectively.

[0030] Based on the room temperature overshoot and temperature convergence time, a comprehensive evaluation is performed on each candidate switching time to determine the optimal switching time for each heating zone.

[0031] Furthermore, the determination of whether to trigger global coordinated scheduling based on the time distribution of the optimal switching time of each heating zone and the distribution of hydraulically sensitive zones includes the following steps:

[0032] Obtain the optimal switching time for each heating zone and sort them in chronological order to obtain the optimal switching time sequence. At the same time, calculate the time interval between two adjacent optimal switching times.

[0033] Based on the distribution of hydraulically sensitive zones, identify the corresponding switching zones in the optimal switching time sequence, and determine whether adjacent switching zones belong to hydraulically sensitive zones or have a hydraulic coupling relationship with hydraulically sensitive zones;

[0034] The time interval is compared with the preset minimum safety interval. If there is an adjacent switching partition that is a hydraulically sensitive partition or has a hydraulic coupling relationship with a hydraulically sensitive partition, and the corresponding time interval is less than the minimum safety interval, it is determined that there is a switching action timing conflict, and global collaborative scheduling is triggered; otherwise, it is determined that there is no switching timing conflict, and the partition switching is executed sequentially according to the optimal switching time sequence.

[0035] Furthermore, the construction of the cooperative handover timing optimization model, obtaining the cooperative handover sequence and performing rolling updates includes the following steps:

[0036] A coordinated switching timing optimization model is established based on mixed integer programming. The switching time is a continuous decision variable, and the switching order is an integer decision variable. The optimization objective is a weighted sum of three terms, including: the weighted sum of squares of the deviations between the actual switching time and the optimal switching time of each heating zone, the integral of the expected value of the room temperature deviation of the hydraulically sensitive zone during the coordinated scheduling process, and the product of the total number of switching actions and the number of start-ups and shutdowns of the heat source equipment.

[0037] Hydraulic constraints and heat source operation constraints are applied to the coordinated switching timing optimization model; wherein, the hydraulic constraints include: flow balance, pressure difference limit and limit on the number of zones to be switched simultaneously; the heat source operation constraints include: outlet water temperature change rate, minimum operating power and continuous start-stop interval time limit;

[0038] Under the conditions of satisfying hydraulic constraints and heat source operation constraints, the coordinated switching sequence is solved to obtain the switching time and switching order of each heating zone;

[0039] A model predictive control framework is used to perform rolling updates on the coordinated switching sequence. The triggering conditions for rolling updates include timed triggering and event triggering.

[0040] After the rolling update, the old and new sequences are compared differentially. Sequence replacement is performed only when the offset exceeds the preset offset threshold or the switching order changes at the critical switching moment, and the replacement timestamp and trigger reason are recorded.

[0041] Furthermore, the method for simultaneously adjusting the opening of the corresponding regulating valve based on the flow deviation is as follows:

[0042] Determine the target switching partition and its switching action for the current execution;

[0043] Based on the hydraulic influence matrix, predict the flow deviation caused by the switching action of the target switching zone to each hydraulically sensitive zone;

[0044] Based on the flow characteristic curves of the regulating valves in each hydraulically sensitive zone, the flow deviation is converted into the corresponding valve opening compensation amount.

[0045] The current opening command of each heating zone regulating valve is superimposed with the valve opening compensation amount to obtain the feedforward compensated valve opening command.

[0046] Furthermore, the adaptive control method further includes the following steps:

[0047] Real-time collection of branch flow and room temperature data for each affected zone to determine whether the room temperature deviation has converged to the preset range;

[0048] When the room temperature deviation does not converge, the compensation command is maintained and the valve opening is dynamically corrected until the preset requirements are met; when the room temperature deviation converges to the preset range, the feedforward compensation for the target switching zone is terminated, the zone that has been switched is switched to the target temperature tracking mode, and the switching action of the next zone is executed according to the cooperative switching sequence.

[0049] Furthermore, determining whether the room temperature deviation converges to a preset range includes the following steps:

[0050] Collect the current room temperature of the target zone and continuously back from the current time. Historical room temperature data for each sampling period, and simultaneously acquire branch flow rate and valve opening data within the corresponding sampling period;

[0051] The absolute value of the difference between the current room temperature and the target temperature is determined in the following... Does it continuously remain within the allowable deviation bandwidth within each sampling period? If the condition is met, the steady-state accuracy criterion is established, and the dynamic stability criterion is then determined; otherwise, the steady-state accuracy criterion is determined to be invalid, and the room temperature deviation of the target partition is directly determined to be non-converged.

[0052] Determine the Does the rate of change of room temperature within each sampling period not exceed the preset maximum rate of change? If the dynamic stability criterion is met, the room temperature deviation of the target partition is determined to be converged; otherwise, the dynamic stability criterion is determined to be unmet, and the room temperature deviation of the partition is determined to be unconverged.

[0053] Furthermore, the target temperature tracking mode is as follows:

[0054] With the goal of maintaining room temperature within the allowable deviation range of the target temperature, the minimum heat dissipation power required to maintain stable room temperature under the current operating conditions is predicted in real time based on the second-order lumped thermal dynamic model of each heating zone.

[0055] Based on the minimum heat dissipation power, the current supply water temperature, and the supply-return water temperature difference, the required temperature-maintaining flow rate is calculated. The temperature flow is then converted into the corresponding temperature valve position.

[0056] In target temperature tracking mode, the valve position control command is updated smoothly using piecewise linear interpolation, and the valve position update step size is limited to no more than a preset percentage of valve position change per minute;

[0057] When the outdoor temperature changes abruptly, i.e., changes by more than 3°C within 15 minutes, the weather forecast data for the area where the target building is located for the next 6 hours is obtained from the meteorological data service platform, and the trend of external temperature change is predicted based on the weather forecast data.

[0058] Based on the trend of external temperature changes, the temperature flow rate is fed forward in advance to reduce the impact of load disturbance on room temperature stability.

[0059] Meanwhile, during the target temperature tracking process, the cumulative heat consumption is calculated based on the branch flow rate and the supply and return water temperature difference, and the cumulative heat consumption is monitored and compared with the theoretical prediction value obtained based on the second-order lumped thermal dynamic model. When the continuous deviation between the two exceeds the preset tolerance limit, the heat dissipation equipment is determined to be abnormal and an alarm is triggered.

[0060] Compared with the prior art, the present invention has the following advantages:

[0061] This invention determines the optimal switching time through second-order lumped thermal dynamic modeling, state estimation, and free temperature rise prediction. It also combines hydraulically sensitive zone identification and coordinated switching timing optimization to achieve feedforward compensation, rolling updates, and target temperature tracking for multi-zone switching actions. This ensures accurate room temperature control while optimizing the overall system energy efficiency and hydraulic stability. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a flowchart illustrating an adaptive control method for a building heating system based on heat load prediction, as disclosed in an embodiment of this application. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some embodiments of this invention, but not all embodiments.

[0065] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0067] likeFigure 1 As shown, an adaptive control method for a building heating system based on heat load prediction includes the following steps:

[0068] The building heating system is divided into zones, resulting in multiple heating zones;

[0069] Based on the real-time room temperature, supply and return water temperature difference, and branch flow rate of each heating zone, a second-order lumped thermal dynamic model of each heating zone is constructed.

[0070] The heat storage of each heating zone is determined by state estimation, and the free temperature rise trajectory of the room temperature of each heating zone and the time characteristics of reaching the target temperature are predicted based on the heat storage of each heating zone.

[0071] The optimal switching time for each heating zone is determined based on the control criterion of achieving the target room temperature without overshoot.

[0072] The hydraulic influence matrix is ​​used to predict the temperature disturbance of each heating zone to other zones at the optimal switching time, and to identify hydraulically sensitive zones.

[0073] Based on the time distribution of the optimal switching time for each heating zone and the distribution of hydraulically sensitive zones, it is determined whether to trigger global coordinated scheduling.

[0074] When it is determined that the conditions for triggering global coordinated scheduling are met, a coordinated handover timing optimization model is constructed with the optimal handover time as the initial solution, and the coordinated handover sequence is obtained and updated on a rolling basis.

[0075] When performing a zone switching operation, the zone to be switched is defined as the target switching zone. At the same time, the affected zone is identified based on the hydraulic influence matrix, and the flow deviation of the affected zone is calculated. Then, the opening of the corresponding regulating valve is adjusted by feedforward compensation based on the flow deviation.

[0076] During the feedforward compensation process, the branch flow and room temperature data of each affected zone are collected in real time to determine whether the room temperature deviation has converged to the preset range.

[0077] When the room temperature deviation does not converge, the compensation command is maintained and the valve opening is dynamically corrected until the preset requirements are met; when the room temperature deviation converges to the preset range, the feedforward compensation for the target switching zone is terminated, the zone that has been switched is switched to the target temperature tracking mode, and the switching action of the next zone is executed according to the cooperative switching sequence. The target temperature tracking mode refers to the control state that uses the target temperature as the control target and dynamically adjusts the heating supply to keep the room temperature within the allowable deviation range of the target temperature.

[0078] The adaptive control method for building heating systems based on heat load prediction described in this application collects real-time data on room temperature, supply and return water temperature difference, and branch flow rates for each heating zone. It then constructs a second-order lumped thermal dynamic model for each heating zone, enabling dynamic sensing of the thermal state of each zone. Based on state estimation, it determines the heat storage capacity of each heating zone and predicts the free temperature rise trajectory and the time characteristics for reaching the target temperature, achieving precise temperature control planning. Furthermore, using the principle of achieving the target temperature without overshoot as the control criterion, it determines the optimal switching time for each heating zone. The method also predicts the temperature disturbance of a heating zone to other zones at the optimal switching time using a hydraulic influence matrix, identifies hydraulically sensitive zones, and achieves scheduling judgment that minimizes disturbance. When the conditions for triggering global coordinated scheduling are met, a coordinated switching timing optimization model is constructed using the optimal switching time as the initial solution. The coordinated switching sequence is solved and continuously updated to achieve [the desired result]. Global optimization and dynamic adjustment of zone switching: During the switching process, the zone to be switched is identified as the target switching zone. At the same time, the affected zones are identified based on the hydraulic influence matrix and the flow deviation is calculated. Feedforward compensation commands are sent to the corresponding regulating valves to dynamically correct the valve opening and achieve rapid convergence of the room temperature deviation in the affected zones. The branch flow and room temperature data of each affected zone are collected in real time. Compensation is maintained and the valve opening is dynamically corrected until the preset requirements are met. When the room temperature deviation converges to the preset range, the feedforward compensation is terminated, the switched zones are switched to the target temperature tracking mode, and the switching action of the next zone is executed according to the collaborative switching sequence. This achieves continuous, adaptive, and precise control of the building heating system, thereby improving heating efficiency, ensuring room temperature comfort and system operation stability, and meeting the long-term management needs of building operation for efficient, low-energy, and reliable heating.

[0079] In some embodiments, the building heating system is divided into zones to obtain multiple heating zones, including the following steps:

[0080] Obtain information on building floor plan, floor functions, building envelope parameters, terminal heat dissipation equipment type, branch pipe network connection relationship and existing control loop information, and clarify the physical boundaries and hydraulic connection relationship between each room, each riser branch and each terminal loop;

[0081] Based on room orientation, floor location, degree of exterior wall exposure, usage function, design heat load and indoor temperature setting requirements, spatial units with similar thermal characteristics and similar load change patterns are classified to form several candidate zones;

[0082] Based on the branch topology of the heating pipeline network, the valve layout and flow regulation capability, branches or terminal units that can be independently controlled and are relatively identifiable by external zoning interference are designated as independent heating zones.

[0083] For each independent heating zone, a room temperature sensor, supply and return water temperature acquisition point, branch flow measurement device and corresponding regulating valve actuator are configured to establish a one-to-one correspondence between "zone number - measurement point information - actuator - pipeline branch".

[0084] The spatial scope, service targets, thermal parameters, control objectives, and related equipment of each zone are uniformly numbered and registered to form a complete list of heating zones.

[0085] The zoning method for building heating systems described in this application clarifies the physical boundaries and hydraulic relationships of rooms, riser branches, and terminal loops by acquiring information on building layout, floor functions, building envelope parameters, terminal heat dissipation equipment types, and pipe network connections. Based on spatial thermal characteristics, load variation patterns, and usage functions, units with similar thermal characteristics and load patterns are categorized into candidate zones. According to the pipe network topology and valve adjustment capabilities, independently controllable branches or terminal units that are identifiable by external zone interference are designated as independent heating zones. For each zone, room temperature sensors, supply and return water temperature acquisition points, branch flow measurement devices, and regulating valves are configured to establish a correspondence between "zone number - measuring point - actuator - pipe network branch". The spatial range, thermal parameters, control objectives, and associated equipment of each zone are uniformly registered to form a complete zone list, thereby achieving automated management and independent control of heating zones and improving adjustment accuracy and energy efficiency.

[0086] In some embodiments, the real-time room temperature, supply and return water temperature difference, and branch flow rate of each heating zone are obtained as follows:

[0087] Temperature sensors were installed on the inlet and outlet pipes of the heat dissipation equipment at the end of each heating zone, flow meters were installed at the inlet of each heating zone branch, and room temperature sensors were installed at representative indoor locations in each heating zone. The installation location, calibration time, and historical deviation information of each sensor were recorded.

[0088] The room temperature, supply water temperature, return water temperature and branch flow of each heating zone are collected according to the preset sampling period to obtain the raw data sequence;

[0089] The original data sequence is subjected to outlier removal, missing data completion, and traffic data filtering to obtain a preprocessed data sequence.

[0090] The supply and return water temperature difference of each heating zone is calculated based on the pre-processed supply and return water temperatures, and then associated and stored with the room temperature and branch flow data to obtain the real-time room temperature, supply and return water temperature difference and branch flow dataset of each heating zone.

[0091] Specifically, the temperature sensor uses a platinum resistance temperature sensor, and the sensor accuracy is no less than [specific value missing]. The sampling period is set between 30 seconds and 2 minutes;

[0092] The flow meter is an electromagnetic flow meter, and the range is selected based on 1.2 times the design flow of the branch. A moving average filter window of not less than 1 minute is set to eliminate flow fluctuation noise.

[0093] The room temperature sensor is a wireless sensor, installed at a height of 1.2 to 1.5 meters above the ground, avoiding direct sunlight and hot or cold air vents.

[0094] The method for obtaining real-time room temperature, supply and return water temperature difference, and branch flow rate in each heating zone described in this application achieves continuous and accurate monitoring of the thermal and flow status of each heating zone by installing temperature sensors on the inlet and outlet pipe sections of the heat dissipation equipment at the end of each heating zone, flow meters at the inlet of the branch lines, and room temperature sensors at representative locations indoors. The temperature sensors are platinum resistance thermometers with an accuracy of not less than [insert accuracy here]. The sampling period is set to 30 seconds to 2 minutes. Electromagnetic flow meters are used, with the range selected based on 1.2 times the branch design flow rate. A moving average filter window of no less than 1 minute is set to eliminate flow fluctuation noise. The room temperature sensor is a wireless sensor, installed at a height of 1.2 to 1.5 meters above the ground, avoiding direct sunlight and hot / cold air vents to ensure data accuracy. According to the preset sampling period, the system automatically collects the room temperature, supply water temperature, return water temperature, and branch flow rate of each heating zone. The system performs outlier removal, missing data completion, and flow filtering on the original data sequence to obtain a pre-processed data sequence. Based on the pre-processed supply and return water temperatures, the temperature difference of each heating zone is calculated and associated with the room temperature and branch flow rate data to form a real-time and complete dataset for each heating zone. This enables precise perception and continuous monitoring of the heating system's operating status, providing a reliable data foundation for zone temperature control optimization and collaborative scheduling, and meeting the long-term needs of building heating systems for high-precision and continuous operation management.

[0095] In some embodiments, constructing a second-order lumped thermal dynamics model for each heating zone includes the following steps:

[0096] Obtain historical operating data for each heating zone within a preset time span. The historical operating data includes: different outdoor temperature ranges and different heating load conditions.

[0097] The thermal dynamic model structure of each heating zone is constructed based on the equivalent thermal resistance-heat capacity network. The thermal dynamic process is equivalent to two state variables: indoor air nodes and building envelope nodes. The thermal balance equations corresponding to each state variable are established according to the thermal dynamic model structure.

[0098] The parameters to be identified are determined based on the thermal dynamic model structure. The parameters to be identified include: indoor air heat capacity, equivalent heat capacity of building envelope, equivalent thermal resistance between indoor air and building envelope, and equivalent thermal resistance between building envelope and outdoor environment.

[0099] Based on the historical operating data, the least squares identification method is used to estimate the parameters to be identified offline to obtain the initial values ​​of the model parameters.

[0100] Substituting the initial values ​​of the model parameters into the thermal dynamic model structure yields the second-order lumped thermal dynamic model for each heating zone, and the initial values ​​of the model parameters are used as the initial conditions for online parameter updates.

[0101] The method for establishing the second-order lumped thermal dynamic model of each heating zone described in this application obtains historical operating data of each heating zone within a preset time span. This historical operating data covers different outdoor temperature ranges and different heating load conditions, providing a complete data foundation for model construction. Based on this, a thermal dynamic model structure is constructed using an equivalent thermal resistance-heat capacity network. The thermal dynamic process of each zone is equivalent to two state variables: indoor air nodes and building envelope nodes, and corresponding heat balance equations are established. Based on the model structure, parameters to be identified are determined, including indoor air heat capacity, equivalent heat capacity of the building envelope, equivalent thermal resistance between indoor air and the building envelope, and equivalent thermal resistance between the building envelope and the outdoor environment. These parameters are then estimated offline using the least squares identification method based on historical operating data to obtain initial values ​​for the model parameters. These model parameters are then substituted into the thermal dynamic model structure to form the second-order lumped thermal dynamic model of each heating zone. The obtained parameters are used as the initial conditions for online parameter updates, thereby achieving an accurate representation of the thermal dynamic characteristics of the building zones and providing a reliable model foundation for subsequent heating prediction and control.

[0102] In some embodiments, the method for determining the heat storage capacity of each heating zone through state estimation is as follows:

[0103] A discrete state-space model is established based on the second-order lumped thermal dynamic model of each heating zone. The indoor air temperature and the equivalent temperature of the building envelope are used as state variables, the heat dissipation power of the heat dissipation equipment is used as the input, and the measured room temperature is used as the observation.

[0104] Based on the sensor accuracy and the statistical characteristics of historical operating data, the process noise covariance matrix and the observation noise covariance matrix are set as Kalman filter parameters;

[0105] Based on the discrete state-space model and Kalman filter parameters, the Kalman filter method is used to recursively estimate the state variables and obtain the posterior estimates of the state variables.

[0106] Based on the difference between the equivalent temperature of the building envelope and the indoor air temperature in the posterior estimate, and in conjunction with the corresponding heat capacity parameters, the heat storage capacity of each heating zone is calculated.

[0107] The method described above for determining the heat storage capacity of each heating zone through state estimation establishes a discrete state-space model based on the second-order lumped thermal dynamic model of each heating zone. Indoor air temperature and the equivalent temperature of the building envelope are used as state variables, with the heat dissipation power of the heat dissipation equipment as the input and the measured room temperature as the observation, thus achieving a dynamic representation of the building's thermal state. Based on this, the process noise covariance matrix and the observation noise covariance matrix are set according to the sensor accuracy and the statistical characteristics of historical operating data. A Kalman filter method is then used to recursively estimate the state variables, obtaining posterior estimates. Furthermore, the heat storage capacity of each heating zone is calculated based on the difference between the equivalent temperature of the building envelope and the indoor air temperature, combined with the corresponding heat capacity parameters, achieving real-time estimation of the heat storage state of the building envelope. This provides a reliable state information basis for predicting room temperature change trends and for adaptive control of the heating system.

[0108] In some embodiments, the method for predicting the free temperature rise trajectory of each heating zone is as follows:

[0109] Obtain the state estimate of each heating zone at the current scheduling time, and use the state estimate as the initial condition for room temperature prediction;

[0110] The heat dissipation power input of the heat dissipation equipment is determined based on the current valve opening and water supply temperature, and the outdoor temperature prediction sequence is used as the external disturbance input.

[0111] Based on the initial conditions, the heat dissipation power input of the heat dissipation equipment and the external disturbance input, the identified second-order lumped thermal dynamic model is used to perform time-domain recursive calculations according to the preset prediction step size to obtain the room temperature prediction sequence of each heating zone in the prediction time domain.

[0112] The room temperature prediction confidence interval is calculated based on the room temperature prediction sequence, and the room temperature free rise prediction trajectory for each heating zone is determined using the room temperature prediction sequence and its corresponding confidence interval.

[0113] The method for predicting the free temperature rise trajectory of each heating zone described in this application obtains the state estimate of each heating zone at the current scheduling moment and uses it as the initial condition for room temperature prediction. It then determines the heat dissipation power input of the heat dissipation equipment by combining the current valve opening and water supply temperature, while using the outdoor temperature prediction sequence as the external disturbance input. Based on this, it performs time-domain recursive calculations using an identified second-order lumped thermal dynamics model according to a preset prediction step size to obtain the room temperature prediction sequence of each heating zone in the prediction time domain, and further calculates the room temperature prediction confidence interval. Through the room temperature prediction sequence and its corresponding confidence interval, the free temperature rise prediction trajectory of each heating zone is determined, achieving continuous prediction and uncertainty assessment of room temperature change trends, thereby providing a reliable basis for zoned heating scheduling and determination of the optimal switching time.

[0114] In some embodiments, determining the optimal switching time for each heating zone includes the following steps:

[0115] Based on the free temperature rise trajectory of each heating zone, the earliest and latest switching times when the room temperature reaches the target temperature without overshoot are determined, thus obtaining the switching time interval.

[0116] Candidate switching times are determined based on the switching time interval, and the room temperature overshoot and temperature convergence time corresponding to each candidate switching time are calculated respectively.

[0117] Based on the room temperature overshoot and temperature convergence time, a comprehensive evaluation is performed on each candidate switching time to determine the optimal switching time for each heating zone.

[0118] The method for determining the optimal switching time for each heating zone described in this application analyzes the room temperature change trend based on the free temperature rise trajectory of each heating zone to determine the earliest and latest switchable times when the room temperature reaches the target temperature without overshoot, forming a switchable time interval for each heating zone. Based on this, candidate switching times are selected within the switchable time interval, and the room temperature overshoot and temperature convergence time corresponding to each candidate switching time are calculated. Furthermore, the room temperature overshoot and temperature convergence time are comprehensively evaluated to determine the optimal switching time for each heating zone. This achieves precise control of the zone heating switching process, reduces room temperature fluctuations, improves system operational stability, and provides a reliable time decision-making basis for subsequent coordinated scheduling of the heating system.

[0119] In some embodiments, the hydraulic influence matrix is ​​constructed as follows:

[0120] Before the building heating system is put into operation for the first time, the opening of the regulating valve of each heating zone is controlled one by one to conduct single-path tests. The branch flow rate and corresponding pressure difference data of each heating zone under different valve openings are collected, and the flow characteristic curve of each heating zone regulating valve is obtained by fitting based on this.

[0121] During the single-path test, the flow changes of other zone branches are recorded simultaneously to construct data on the disturbance relationship between the action of the regulating valves of each heating zone and the flow of other zones.

[0122] Based on the nodal flow balance equation and the pipe section pressure drop equation, a hydraulic model of the building heating pipe network is established using graph theory. With measured flow data as constraints, the equivalent resistance coefficient of each pipe section is solved through nonlinear optimization.

[0123] Based on the equivalent resistance coefficient and the disturbance relationship data, the flow coupling relationship between each heating zone is established, and a hydraulic influence matrix is ​​constructed accordingly.

[0124] Specifically, the elements of the hydraulic influence matrix Defined as: when the first When the zone control valve switches from fully open to fully closed, the first... The change in traffic flow of each zone branch and the first The ratio of the rated flow rates of each zone; The simulation was conducted by perturbation of the hydraulic model of the pipeline network. The simulation was performed by solving the nodal pressure equations using the Newton-Raphson iterative method. The convergence accuracy of the iteration was set to be less than 1 Pa for the pressure change at each node.

[0125] diagonal elements Indicates the first The relative change in traffic during partition switching is used to evaluate the effectiveness of the switching action itself.

[0126] The hydraulic influence matrix is ​​dynamically updated as the pipeline network operation status changes. The update trigger conditions are: the rated flow setting value of any zone changes by more than 10%, the constant pressure value is adjusted, or a parallel zone is added / removed. The updated matrix is ​​compared with the previous version. If the key elements change by more than 5%, an update confirmation notification is sent.

[0127] The method for constructing the hydraulic influence matrix described in this application involves conducting single-path tests by controlling the opening degree of the regulating valves in each heating zone before the initial commissioning of the building heating system. This process collects branch flow and corresponding pressure difference data for each heating zone under different valve opening degrees, and then fits the flow characteristic curves of the regulating valves in each heating zone. Simultaneously, the changes in the branch flow of other zones are recorded during the test to obtain data on the flow disturbance relationship between the regulating valve actions of each heating zone and other zones. Based on this, a hydraulic model of the building heating network is established using graph theory, based on the node flow balance equation and the pipe section pressure drop equation. The equivalent resistance coefficient of each pipe section is then solved through nonlinear optimization using measured flow data. Furthermore, the flow coupling relationship between each heating zone is established based on the equivalent resistance coefficient and disturbance relationship data, and a hydraulic influence matrix is ​​constructed accordingly. This achieves a quantitative expression of the hydraulic coupling characteristics of the heating network, thus providing a reliable basis for predicting and coordinating the switching interference of the heating system zones.

[0128] In some embodiments, the method for identifying hydraulically sensitive zones is as follows:

[0129] Based on the hydraulic influence matrix H and the flow change of the target switching zone, the expected flow deviation vector of each heating zone is calculated. Its formula is: ,in, Switch partitions for target Changes in flow rate; The first in the hydraulic influence matrix The column of influence coefficient vectors represents the target switching partition. The extent to which traffic changes affect all partitions;

[0130] Compare the absolute value of the flow deviation of each heating zone with its rated flow rate. If the ratio exceeds the preset sensitivity threshold, the corresponding zone is marked as a hydraulically sensitive zone.

[0131] For the marked hydraulically sensitive zones, it is further determined whether the deviation between their room temperature and the target temperature is less than a preset deviation threshold. If the threshold is met, the hydraulically sensitive zone is also marked as a thermal stability sensitive zone.

[0132] The method for identifying hydraulically sensitive zones described in this application calculates the expected flow deviation vector for each heating zone based on the hydraulic influence matrix H and the flow change of the target switching zone. Then, it compares the absolute value of the flow deviation of each heating zone with its rated flow rate. When this ratio exceeds a preset sensitivity threshold, the corresponding zone is marked as a hydraulically sensitive zone. Furthermore, it determines the deviation between the room temperature and the target temperature of the hydraulically sensitive zone. When the deviation is less than a preset threshold, it is simultaneously marked as a thermal stability sensitive zone. This method achieves accurate identification of zones in the heating system susceptible to hydraulic disturbances, providing a basis for subsequent zone switching control and coordinated scheduling, improving system operational stability, and reducing temperature fluctuations.

[0133] In some embodiments, based on the time distribution of the optimal switching time for each heating zone and the distribution of hydraulically sensitive zones, it is determined whether to trigger global coordinated scheduling, including the following steps:

[0134] Obtain the optimal switching time for each heating zone and sort them in chronological order to obtain the optimal switching time sequence. At the same time, calculate the time interval between two adjacent optimal switching times.

[0135] Based on the distribution of hydraulically sensitive zones, identify the corresponding switching zones in the optimal switching time sequence, and determine whether adjacent switching zones belong to hydraulically sensitive zones or have a hydraulic coupling relationship with hydraulically sensitive zones;

[0136] The time interval is compared with the preset minimum safety interval. If there is an adjacent switching partition that is a hydraulically sensitive partition or has a hydraulic coupling relationship with a hydraulically sensitive partition, and the corresponding time interval is less than the minimum safety interval, it is determined that there is a switching action timing conflict, and global collaborative scheduling is triggered; otherwise, it is determined that there is no switching timing conflict, and the partition switching is executed sequentially according to the optimal switching time sequence.

[0137] The global coordinated scheduling triggering determination method described in this application obtains the optimal switching time of each heating zone and sorts them in chronological order to form an optimal switching time sequence. Simultaneously, it calculates the time interval between adjacent optimal switching times. Based on this, it identifies the corresponding switching zones in the sequence by considering the distribution of hydraulically sensitive zones and determines whether adjacent switching zones belong to hydraulically sensitive zones or have a hydraulic coupling relationship with them. Furthermore, it compares the time interval with a preset minimum safety interval. When there are adjacent switching zones that belong to hydraulically sensitive zones or have a hydraulic coupling relationship with them, and the corresponding time interval is less than the minimum safety interval, it determines that there is a switching action timing conflict and triggers global coordinated scheduling. This achieves coordinated control of zone switching behavior, reduces the impact of hydraulic disturbances on system operation, and improves the overall stability of the heating system.

[0138] In some embodiments, a coordinated handover timing optimization model is constructed, the coordinated handover sequence is solved, and rolling updates are performed, including the following steps:

[0139] A coordinated switching timing optimization model is established based on mixed integer programming. The switching time is a continuous decision variable, and the switching order is an integer decision variable. The optimization objective is a weighted sum of three terms, including: the weighted sum of squares of the deviations between the actual switching time and the optimal switching time of each heating zone, the integral of the expected value of the room temperature deviation of the hydraulically sensitive zone during the coordinated scheduling process, and the product of the total number of switching actions and the number of start-ups and shutdowns of the heat source equipment.

[0140] Hydraulic constraints and heat source operation constraints are applied to the coordinated switching timing optimization model; wherein, the hydraulic constraints include: flow balance, pressure difference limit and limit on the number of zones to be switched simultaneously; the heat source operation constraints include: outlet water temperature change rate, minimum operating power and continuous start-stop interval time limit;

[0141] Under the conditions of satisfying hydraulic constraints and heat source operation constraints, the coordinated switching sequence is solved to obtain the switching time and switching order of each heating zone;

[0142] A model predictive control framework is used to perform rolling updates on the coordinated switching sequence. The triggering conditions for rolling updates include timed triggering and event triggering.

[0143] After the rolling update, the old and new sequences are compared differentially. Sequence replacement is performed only when the offset exceeds the preset offset threshold or the switching order changes at the critical switching moment, and the replacement timestamp and trigger reason are recorded.

[0144] The aforementioned coordinated switching timing optimization method in this application establishes a coordinated switching timing optimization model based on mixed integer programming. The switching time is a continuous decision variable, and the switching order is an integer decision variable. The optimization objective is the weighted sum of squares of the deviations between the actual and optimal switching times for each heating zone, the integral of the expected room temperature deviation for hydraulically sensitive zones, and the weighted sum of the product of the number of switching actions and the number of start-ups and shutdowns of heat source equipment. Based on this, hydraulic constraints and heat source operation constraints are applied to the model. Hydraulic constraints include flow balance, differential pressure limits, and limits on the number of zones that can be switched simultaneously. Heat source operation constraints include the rate of change of outlet water temperature, minimum operating power, and continuous start-up / shutdown interval limits. Under the constraints, the coordinated switching sequence is solved to obtain the switching time and order for each heating zone. Furthermore, the coordinated switching sequence is continuously updated within the model predictive control framework. The update process is initiated through timed and event-triggered mechanisms. After the update, the old and new sequences are compared differentially. Sequence replacement is only performed when the critical switching time offset exceeds a preset threshold or the switching order changes, thereby achieving dynamic coordinated scheduling and stable operation of the heating system switching process.

[0145] In some embodiments, the method for simultaneously adjusting the opening of the corresponding regulating valve based on the flow deviation is as follows:

[0146] Determine the target switching partition and its switching action for the current execution;

[0147] Based on the hydraulic influence matrix, predict the flow deviation caused by the switching action of the target switching zone to each hydraulically sensitive zone;

[0148] Based on the flow characteristic curves of the regulating valves in each hydraulically sensitive zone, the flow deviation is converted into the corresponding valve opening compensation amount.

[0149] The current opening command of each heating zone regulating valve is superimposed with the valve opening compensation amount to obtain the feedforward compensated valve opening command.

[0150] The feedforward compensation command issuance method described in this application determines the target switching zone and its corresponding switching action, and predicts the flow deviation caused by the switching action to each hydraulically sensitive zone based on the hydraulic influence matrix. Based on this, the flow deviation is converted into a corresponding valve opening compensation amount by combining the flow characteristic curves of the regulating valves in each hydraulically sensitive zone. Furthermore, the current opening command of the regulating valves in each heating zone is superimposed with the valve opening compensation amount to form a feedforward compensated valve opening command, which is then issued and executed. This achieves advance compensation for switching disturbances, reduces the impact of hydraulic fluctuations on the heating status of each heating zone, and improves system operational stability.

[0151] In some embodiments, determining whether the room temperature deviation converges to a preset range includes the following steps:

[0152] Collect the current room temperature of the target zone and continuously back from the current time. Historical room temperature data for each sampling period, and simultaneously acquire branch flow rate and valve opening data within the corresponding sampling period;

[0153] The absolute value of the difference between the current room temperature and the target temperature is determined in the following... Does it continuously remain within the allowable deviation bandwidth within each sampling period? If the condition is met, the steady-state accuracy criterion is established, and the dynamic stability criterion is then determined; otherwise, the steady-state accuracy criterion is determined to be invalid, and the room temperature deviation of the target partition is directly determined to be non-converged.

[0154] Determine the Does the rate of change of room temperature within each sampling period not exceed the preset maximum rate of change? If the dynamic stability criterion is met, the room temperature deviation of the target partition is determined to be converged; otherwise, the dynamic stability criterion is determined to be unmet, and the room temperature deviation of the partition is determined to be unconverged.

[0155] The room temperature deviation convergence determination method described in this application collects the current room temperature of the target zone and historical room temperature data for K consecutive sampling periods prior to the current time, and simultaneously acquires branch flow rate and valve opening data within the corresponding periods; based on this, it determines whether the absolute value of the difference between the current room temperature and the target temperature remains continuously within the allowable deviation bandwidth within K sampling periods. Within the sampling period, the steady-state accuracy criterion is determined to be valid; simultaneously, it is determined whether the rate of change of room temperature within the sampling period does not exceed a preset maximum rate of change. This is to determine whether the dynamic stability criterion holds; if and only if the steady-state accuracy criterion and the dynamic stability criterion hold simultaneously, it is determined that the room temperature deviation of the zone has converged, otherwise it is determined that it has not converged, thereby achieving reliable identification of the temperature stability state of the heating system.

[0156] In some embodiments, the control strategy for the target temperature tracking mode includes:

[0157] With the goal of maintaining room temperature within the allowable deviation range of the target temperature, and based on the second-order lumped thermal dynamics model of each heating zone, the minimum heat dissipation power required to maintain stable room temperature under the current operating conditions is determined. To make real-time predictions;

[0158] Based on minimum heat dissipation power Based on the current water supply temperature and the temperature difference between the supply and return water, the required temperature-maintaining flow rate is calculated to keep the room temperature stable. and the temperature flow rate Converted to the corresponding temperature control valve position ;

[0159] In target temperature tracking mode, the valve position control command is updated smoothly using piecewise linear interpolation, and the valve position update step size is limited to no more than 2% of valve position change per minute;

[0160] When the outdoor temperature changes abruptly, i.e., changes exceeding 3°C within 15 minutes, the temperature maintenance flow rate should be adjusted in advance based on weather forecast data. Feedforward corrections are implemented to reduce the impact of load disturbances on room temperature stability;

[0161] Meanwhile, during the target temperature tracking process, the cumulative heat consumption is calculated based on the branch flow rate and the supply and return water temperature difference, and the cumulative heat consumption is monitored and compared with the theoretical prediction value obtained based on the second-order lumped thermal dynamic model. When the continuous deviation between the two exceeds 20%, the heat dissipation equipment is determined to be abnormal and an alarm is triggered.

[0162] The target temperature tracking mode control strategy described in this application aims to maintain room temperature within the allowable deviation range of the target temperature. Based on a second-order lumped thermal dynamic model of each heating zone, it predicts in real time the minimum heat dissipation power required to maintain stable room temperature. Combined with the current supply water temperature and the supply-return water temperature difference, it calculates the temperature-maintaining flow rate and corresponding temperature-maintaining valve position. Furthermore, it uses piecewise linear interpolation to smoothly update the valve position control command, limiting the valve position change rate to no more than 2% per minute. When the outdoor temperature changes by more than 3°C within 15 minutes, it performs feedforward correction on the temperature-maintaining flow rate Qmaint based on weather forecast data to reduce the impact of load disturbances. Simultaneously, it monitors the cumulative heat consumption during operation and compares it with the model's predicted value. When the continuous deviation exceeds 20%, it issues an alarm for abnormal heat dissipation equipment, thereby achieving stable temperature maintenance and abnormal state monitoring of the heating system.

[0163] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive control method for a building heating system based on heat load prediction, characterized in that, Includes the following steps: The building heating system is divided into zones, resulting in multiple heating zones; Based on the real-time room temperature, supply and return water temperature difference, and branch flow rate of each heating zone, a second-order lumped thermal dynamic model of each heating zone is constructed. The heat storage of each heating zone is determined by state estimation, and the free temperature rise trajectory of the room temperature of each heating zone and the time characteristics of reaching the target temperature are predicted based on the heat storage of each heating zone. The optimal switching time for each heating zone is determined based on the control criterion of achieving the target room temperature without overshoot. The hydraulic influence matrix is ​​used to predict the temperature disturbance of each heating zone to other zones at the optimal switching time, and to identify hydraulically sensitive zones. Based on the time distribution of the optimal switching time for each heating zone and the distribution of hydraulically sensitive zones, it is determined whether to trigger global coordinated scheduling. When it is determined that the conditions for triggering global coordinated scheduling are met, a coordinated handover timing optimization model is constructed with the optimal handover time as the initial solution, and the coordinated handover sequence is obtained and updated on a rolling basis. When performing a zone switching operation, the zone to be switched is defined as the target switching zone. At the same time, the affected zone is identified based on the hydraulic influence matrix, and the flow deviation of the affected zone is calculated. Then, the opening of the corresponding regulating valve is adjusted by feedforward compensation based on the flow deviation.

2. The adaptive control method for a building heating system based on heat load prediction according to claim 1, characterized in that, The construction of a second-order lumped thermal dynamic model for each heating zone includes the following steps: Obtain historical operating data for each heating zone within a preset time span. The historical operating data includes: different outdoor temperature ranges and different heating load conditions. The thermal dynamic model structure of each heating zone is constructed based on the equivalent thermal resistance-heat capacity network. The thermal dynamic process is equivalent to two state variables: indoor air nodes and building envelope nodes. The thermal balance equations corresponding to each state variable are established according to the thermal dynamic model structure. The parameters to be identified are determined based on the thermal dynamic model structure. The parameters to be identified include: indoor air heat capacity, equivalent heat capacity of building envelope, equivalent thermal resistance between indoor air and building envelope, and equivalent thermal resistance between building envelope and outdoor environment. Based on the historical operating data, the least squares identification method is used to estimate the parameters to be identified offline to obtain the initial values ​​of the model parameters. Substituting the initial values ​​of the model parameters into the thermal dynamic model structure yields the second-order lumped thermal dynamic model for each heating zone, and the initial values ​​of the model parameters are used as the initial conditions for online parameter updates.

3. The adaptive control method for a building heating system based on heat load prediction according to claim 2, characterized in that, Determining the heat storage capacity of each heating zone through state estimation includes the following steps: A discrete state-space model is established based on the second-order lumped thermal dynamic model of each heating zone. The indoor air temperature and the equivalent temperature of the building envelope are used as state variables, the heat dissipation power of the heat dissipation equipment is used as the input, and the measured room temperature is used as the observation. Based on the sensor accuracy and the statistical characteristics of historical operating data, the process noise covariance matrix and the observation noise covariance matrix are set as Kalman filter parameters; Based on the discrete state-space model and Kalman filter parameters, the Kalman filter method is used to recursively estimate the state variables and obtain the posterior estimates of the state variables. Based on the difference between the equivalent temperature of the building envelope and the indoor air temperature in the posterior estimate, and in conjunction with the corresponding heat capacity parameters, the heat storage capacity of each heating zone is calculated.

4. The adaptive control method for a building heating system based on heat load prediction according to claim 1, characterized in that, Determining the optimal switching time for each heating zone includes the following steps: Based on the free temperature rise trajectory of each heating zone, the earliest and latest switching times when the room temperature reaches the target temperature without overshoot are determined, thus obtaining the switching time interval. Candidate switching times are determined based on the switching time interval, and the room temperature overshoot and temperature convergence time corresponding to each candidate switching time are calculated respectively. Based on the room temperature overshoot and temperature convergence time, a comprehensive evaluation is performed on each candidate switching time to determine the optimal switching time for each heating zone.

5. The adaptive control method for a building heating system based on heat load prediction according to claim 1, characterized in that, The determination of whether to trigger global coordinated scheduling based on the time distribution of the optimal switching time of each heating zone and the distribution of hydraulically sensitive zones includes the following steps: Obtain the optimal switching time for each heating zone and sort them in chronological order to obtain the optimal switching time sequence. At the same time, calculate the time interval between two adjacent optimal switching times. Based on the distribution of hydraulically sensitive zones, identify the corresponding switching zones in the optimal switching time sequence, and determine whether adjacent switching zones belong to hydraulically sensitive zones or have a hydraulic coupling relationship with hydraulically sensitive zones; The time interval is compared with the preset minimum safety interval. If there is an adjacent switching partition that is a hydraulically sensitive partition or has a hydraulic coupling relationship with a hydraulically sensitive partition, and the corresponding time interval is less than the minimum safety interval, it is determined that there is a switching action timing conflict, and global collaborative scheduling is triggered; otherwise, it is determined that there is no switching timing conflict, and the partition switching is executed sequentially according to the optimal switching time sequence.

6. The adaptive control method for a building heating system based on heat load prediction according to claim 1, characterized in that, The process of constructing a coordinated handover timing optimization model, obtaining the coordinated handover sequence, and performing rolling updates includes the following steps: A coordinated switching timing optimization model is established based on mixed integer programming. The switching time is a continuous decision variable, and the switching order is an integer decision variable. The optimization objective is a weighted sum of three terms, including: the weighted sum of squares of the deviations between the actual switching time and the optimal switching time of each heating zone, the integral of the expected value of the room temperature deviation of the hydraulically sensitive zone during the coordinated scheduling process, and the product of the total number of switching actions and the number of start-ups and shutdowns of the heat source equipment. Hydraulic constraints and heat source operation constraints are applied to the coordinated switching timing optimization model; wherein, the hydraulic constraints include: flow balance, pressure difference limit and limit on the number of zones to be switched simultaneously; the heat source operation constraints include: outlet water temperature change rate, minimum operating power and continuous start-stop interval time limit; Under the conditions of satisfying hydraulic constraints and heat source operation constraints, the coordinated switching sequence is solved to obtain the switching time and switching order of each heating zone; A model predictive control framework is used to perform rolling updates on the coordinated switching sequence. The triggering conditions for rolling updates include timed triggering and event triggering. After the rolling update, the old and new sequences are compared differentially. Sequence replacement is performed only when the offset exceeds the preset offset threshold or the switching order changes at the critical switching moment, and the replacement timestamp and trigger reason are recorded.

7. An adaptive control method for a building heating system based on heat load prediction according to any one of claims 1-6, characterized in that, The method for simultaneously adjusting the opening of the corresponding regulating valve based on the flow deviation is as follows: Determine the target switching partition and its switching action for the current execution; Based on the hydraulic influence matrix, predict the flow deviation caused by the switching action of the target switching zone to each hydraulically sensitive zone; Based on the flow characteristic curves of the regulating valves in each hydraulically sensitive zone, the flow deviation is converted into the corresponding valve opening compensation amount. The current opening command of each heating zone regulating valve is superimposed with the valve opening compensation amount to obtain the feedforward compensated valve opening command.

8. The adaptive control method for a building heating system based on heat load prediction according to claim 7, characterized in that, The adaptive control method further includes the following steps: Real-time collection of branch flow and room temperature data for each affected zone to determine whether the room temperature deviation has converged to the preset range; When the room temperature deviation does not converge, the compensation command is maintained and the valve opening is dynamically corrected until the preset requirements are met; when the room temperature deviation converges to the preset range, the feedforward compensation for the target switching zone is terminated, the zone that has been switched is switched to the target temperature tracking mode, and the switching action of the next zone is executed according to the cooperative switching sequence.

9. The adaptive control method for a building heating system based on heat load prediction according to claim 8, characterized in that, The process of determining whether the room temperature deviation has converged to a preset range includes the following steps: Collect the current room temperature of the target zone and continuously back from the current time. Historical room temperature data for each sampling period, and simultaneously acquire branch flow rate and valve opening data within the corresponding sampling period; The absolute value of the difference between the current room temperature and the target temperature is determined in the following... Does it continuously remain within the allowable deviation bandwidth within each sampling period? If the condition is met, the steady-state accuracy criterion is established, and the dynamic stability criterion is then determined; otherwise, the steady-state accuracy criterion is determined to be invalid, and the room temperature deviation of the target partition is directly determined to be non-converged. Determine the Does the rate of change of room temperature within each sampling period not exceed the preset maximum rate of change? If the dynamic stability criterion is met, the room temperature deviation of the target partition is determined to be converged; otherwise, the dynamic stability criterion is determined to be unmet, and the room temperature deviation of the partition is determined to be unconverged.

10. The adaptive control method for a building heating system based on heat load prediction according to claim 8, characterized in that, The target temperature tracking mode is as follows: With the goal of maintaining room temperature within the allowable deviation range of the target temperature, the minimum heat dissipation power required to maintain stable room temperature under the current operating conditions is predicted in real time based on the second-order lumped thermal dynamic model of each heating zone. Based on the minimum heat dissipation power, the current supply water temperature, and the supply-return water temperature difference, the required temperature-maintaining flow rate is calculated. The temperature flow rate is then converted into the corresponding temperature valve position. In target temperature tracking mode, the valve position control command is updated smoothly using piecewise linear interpolation, and the valve position update step size is limited to no more than a preset percentage of valve position change per minute; When the outdoor temperature changes abruptly, i.e., changes by more than 3°C within 15 minutes, the weather forecast data for the area where the target building is located for the next 6 hours is obtained from the meteorological data service platform, and the trend of external temperature change is predicted based on the weather forecast data. Based on the trend of external temperature changes, the temperature flow rate is fed forward in advance to reduce the impact of load disturbance on room temperature stability. Meanwhile, during the target temperature tracking process, the cumulative heat consumption is calculated based on the branch flow rate and the supply and return water temperature difference, and the cumulative heat consumption is monitored and compared with the theoretical prediction value obtained based on the second-order lumped thermal dynamic model. When the continuous deviation between the two exceeds the preset tolerance limit, the heat dissipation equipment is determined to be abnormal and an alarm is triggered.

Images