Fan-coil central control multi-zone precise temperature control method and system

Abstract

The present application relates to the field of HVAC intelligent control technology, in particular to a fan coil central control multi-region precise temperature control method and system, comprising the following steps: S1: real-time collection of environmental thermal parameters, dynamic load characteristics and fan coil operation energy efficiency state of each temperature control region, and data standardization mapping through DCS communication protocol. The present application builds a regional thermal inertia dynamic digital twin model, predicts the load trend, establishes a global energy efficiency and local comfort multi-objective game function and solves the optimal balance point, eliminates the thermal coupling effect between regions by combining the thermal decoupling algorithm, generates globally optimal and independent temperature control instructions, effectively realizes the precision of multi-region temperature control and the global energy efficiency optimization, significantly improves the system temperature control stability, response speed and control accuracy, and avoids integral saturation and steady-state error.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for HVAC systems, specifically a method and system for precise temperature control of multiple zones in a centrally controlled fan coil unit. Background Technology

[0002] Centralized control of fan coil units refers to a control mode that uses a central control system (such as a DCS central management layer) to uniformly manage and coordinate the fan coil unit equipment in multiple areas. The core of this mode is to rely on a collaborative mechanism of data acquisition, model decision-making, command execution, and closed-loop correction to achieve centralized monitoring, unified scheduling, and precise temperature control of fan coil units in each area. This breaks the limitations of independent control in a single area and takes into account the consistency of temperature control in multiple areas, optimal overall energy efficiency, and local comfort requirements.

[0003] However, in existing technologies, centralized multi-zone temperature control of fan coil units suffers from several problems, including multi-zone thermal coupling interference, low temperature control accuracy, poor system energy efficiency, delayed thermal response, easy temperature overshoot and oscillation, and susceptibility to integral saturation and steady-state errors. Furthermore, it cannot simultaneously address both overall energy efficiency and local comfort. These issues not only affect the temperature control experience in each zone, causing local temperatures to deviate from the comfortable range, but also lead to energy waste, reduced operational stability and reliability of the fan coil system, and an inability to meet the multi-zone, high-precision, and low-energy-consumption temperature control requirements of modern buildings.

[0004] Based on this, the present invention provides a method and system for precise temperature control of multiple zones in a centrally controlled fan coil unit, in order to solve the aforementioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for precise temperature control in multiple zones of a centrally controlled fan coil unit. This invention constructs a dynamic digital twin model of regional thermal inertia, predicts load trends in a rolling manner, establishes a multi-objective game function of global energy efficiency and local comfort, and solves for the optimal balance point. Combined with a thermal decoupling algorithm, it eliminates the thermal coupling effect between zones, generates globally optimal and independent temperature control commands, and effectively achieves the precision of multi-zone temperature control and the optimal global energy efficiency. Furthermore, by identifying the thermal response lag time and inertial characteristics online, it uses a predictive control algorithm for feedforward compensation, dynamically corrects the residual deviation after decoupling, and suppresses temperature overshoot oscillation through a damping control strategy. This significantly improves the system's temperature control stability, response speed, and control accuracy, and avoids integral saturation and steady-state errors.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for precise temperature control of multiple zones in a centrally controlled fan coil unit, comprising the following steps: S1: Real-time acquisition of environmental thermal parameters, dynamic load characteristics and fan coil unit operating energy efficiency status of each temperature control zone, and data standardization mapping through DCS communication protocol; S2: Based on standardized data, construct a regional thermal inertia digital twin model to predict the load evolution trend of each region, and based on a multi-objective game strategy of global energy efficiency and local comfort, dynamically decouple the thermal interference between regions and generate the optimal control command. S3: The DCS central management layer sends the global optimal control command to the field control station, driving the actuator to linearly adjust the water valve opening of the fan coil unit and perform stepless speed control of the fan speed. S4: Based on the hysteresis feature identification model, predict the thermal response delay time after S3 is executed, use the predictive control algorithm to perform feedforward compensation on the control output, and dynamically correct the residual deviation after S2 decoupling to suppress temperature overshoot oscillation.

[0007] Based on the above method, this invention also proposes a centrally controlled multi-zone precision temperature control system for fan coil units, including a data sensing and acquisition unit, a model decision optimization unit, a field execution control unit, and a closed-loop correction and stabilization unit, wherein: The data sensing and acquisition unit is used to collect relevant thermal, load and equipment energy efficiency data of each temperature control zone in real time, and to perform data standardization processing and mapping based on the DCS protocol. The model decision optimization unit is used to construct a regional thermal inertia digital twin model to predict load trends, decouple thermal disturbances through a multi-objective game strategy, and generate globally optimal temperature control commands. The field execution control unit is used to receive control commands issued by the DCS central management layer and drive the corresponding actuator to perform linear adjustment of the fan coil water valve and stepless speed control of the fan speed. The closed-loop correction and stabilization unit predicts the thermal response delay time based on the hysteresis feature identification model, uses predictive control for feedforward compensation, and dynamically corrects the residual deviation after decoupling.

[0008] The data sensing and acquisition unit includes a multi-source sensing acquisition module, a load feature extraction module, an equipment status monitoring module, and a DCS standardized mapping module, wherein: The multi-source sensor acquisition module is used to collect thermal parameters such as temperature, humidity, and inlet and outlet water temperatures of the coils in real time in various areas. The load feature extraction module is used to dynamically analyze the load features of personnel flow, equipment heat dissipation, and heat gain of the building envelope in the area. The equipment status monitoring module is used to monitor the operating energy efficiency status of the fan coil unit in real time, including its speed setting, current, and valve opening. The DCS standardization mapping module is used to convert heterogeneous data into the DCS protocol standard format and perform address mapping.

[0009] The model decision optimization unit includes a digital twin construction module, a load trend prediction module, a game strategy decision module, and a thermal decoupling calculation module, wherein: The digital twin construction module: constructs a dynamic digital twin model reflecting regional thermal inertia based on historical and real-time data; The load trend prediction module is used to predict the evolution trend of cold and heat loads in each region in future periods through a dynamic digital twin model. The game strategy decision module is used to establish a multi-objective game function that balances global energy efficiency and local comfort, and to solve for the optimal equilibrium point. The thermal decoupling calculation module is used to eliminate the thermal coupling effect between regions and generate independent control commands through a decoupling algorithm.

[0010] The dynamic digital twin model reflecting regional thermal inertia in the digital twin construction module is constructed using an equivalent thermal capacity-thermal resistance network, and its specific expression is as follows: ; In the formula, Let be the equivalent heat capacity of region i. Let be the equivalent thermal resistance of region i. Let i be the indoor temperature of region i. Outdoor temperature The cooling / heating provided to the fan coil units This refers to the internal thermal disturbance of personnel and equipment within the area.

[0011] The game strategy decision-making module establishes a multi-objective game function that balances global energy efficiency and local comfort, and solves for the optimal equilibrium point. The specific operations are as follows: A1: Extract real-time energy efficiency data and local comfort feedback data for each temperature control zone, determine the global energy efficiency target, local comfort target and their constraints, including the upper limit of fan coil unit operating power and the zone temperature comfort range; A2: Establish a multi-objective game function that balances global energy efficiency and local comfort, with the following specific expression: ; In the formula, Let be the local comfort function for region i. Let i be the actual temperature of region i. This refers to the regional priority weighting coefficient. The total energy consumption of the system. Energy efficiency penalty coefficient; A3: Based on the game function, a nonlinear programming algorithm is used to solve for the optimal solution, and the control parameters or temperature setpoints of the fan coil units in each region corresponding to the optimal balance point between global energy efficiency and local comfort are obtained. A4: Verify the feasibility of the optimal equilibrium point. If the constraints are met, output the decision parameters corresponding to the equilibrium point. If not, adjust the constraints and solve again until the optimal equilibrium point is obtained.

[0012] The on-site execution and control unit includes an instruction parsing and distribution module, a water valve linear drive module, a fan stepless speed regulation module, and an execution status feedback module, wherein: The instruction parsing and distribution module is used to receive and parse the global optimal control instructions issued by the DCS central management layer. The water valve linear drive module is used to continuously adjust the opening degree of the fan coil unit electric water valve according to the command to control the water flow. The fan stepless speed regulation module is used to continuously adjust the fan speed according to the command to control the air volume. The execution status feedback module is used to upload the actual execution status of valve position and fan speed to the central management layer in real time.

[0013] The closed-loop correction and stabilization unit includes a hysteresis feature identification module, a feedforward prediction and compensation module, a residual deviation correction module, and an overshoot oscillation suppression module, wherein: The hysteresis feature identification module is used to identify the hysteresis time and inertial characteristics from instruction execution to temperature response online. The feedforward prediction and compensation module: uses a predictive control algorithm based on the lag time to output compensation commands in advance; The residual deviation correction module is used to dynamically correct minor temperature deviations that still exist after decoupling. The overshoot oscillation suppression module is used to suppress temperature overshoot oscillations and accelerate system stability through a damping control strategy.

[0014] The feedforward prediction and compensation module outputs compensation commands in advance based on the lag time using a predictive control algorithm. The specific operation is as follows: B1: Based on the thermal response lag time output by the lag feature identification module, determine the prediction time domain and rolling optimization window of the predictive control algorithm; B2: Collect feedforward disturbance variables that affect the temperature of the area. The disturbance variables include the sudden change value of outdoor ambient temperature, the rate of change of the number of people in the area, and the fluctuation of fresh air load. B3: Using the state equation of the predictive control algorithm, the regional temperature trajectory at future times is calculated based on the current system state and the predicted values ​​of disturbance variables in the future time domain. B4: Calculate the pre-control quantity required to eliminate the predicted temperature deviation and convert it into a feedforward compensation command that is superimposed on the basic control command.

[0015] The specific expression for the state prediction equation of the predictive control algorithm in step B3 is as follows: ; In the formula, The temperature of the region at time k+1 is predicted based on information at time k. Let k be the measured temperature of the region. Let k be the control input for the fan coil unit at time k. Let be the feedforward disturbance variable vector detected at time k, and let A, B, and D be the system state matrix, control input matrix, and disturbance input matrix, respectively.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a regional thermal inertia dynamic digital twin model, predicts load trends in a rolling manner, establishes a multi-objective game function for global energy efficiency and local comfort, and solves for the optimal balance point. Combined with a thermal decoupling algorithm, it eliminates the thermal coupling effect between regions, generates globally optimal and independent temperature control commands, and effectively achieves the accuracy of multi-region temperature control and the optimal global energy efficiency. Furthermore, by identifying the thermal response lag time and inertial characteristics online, it uses a predictive control algorithm for feedforward compensation, dynamically corrects the residual deviation after decoupling, and suppresses temperature overshoot oscillation through a damping control strategy. This significantly improves the system's temperature control stability, response speed, and control accuracy, and avoids integral saturation and steady-state errors. Attached Figure Description

[0017] Figure 1 This is a system diagram of the centrally controlled multi-zone precision temperature control system for fan coil units according to the present invention.

[0018] Figure 2 This is a flowchart of the central control method for precise temperature control of multiple zones in fan coil units according to the present invention.

[0019] Explanation of icon numbers: 1. Data Sensing and Acquisition Unit; 11. Multi-Source Sensor Acquisition Module; 12. Load Feature Extraction Module; 13. Equipment Status Monitoring Module; 14. DCS Standardized Mapping Module; 2. Model Decision Optimization Unit; 21. Digital Twin Construction Module; 22. Load Trend Prediction Module; 23. Game Theory Strategy Decision Module; 24. Thermal Decoupling Calculation Module; 3. On-site Execution and Control Unit; 31. Command Parsing and Distribution Module; 32. Water Valve Linear Drive Module; 33. Fan Stepless Speed ​​Regulation Module; 34. Execution Status Feedback Module; 4. Closed-Loop Correction and Stabilization Unit; 41. Lag Feature Identification Module; 42. Feedforward Prediction and Compensation Module; 43. Residual Deviation Correction Module; 44. Overshoot Oscillation Suppression Module. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1: like Figure 1 As shown, this embodiment provides a centrally controlled multi-zone precision temperature control system for fan coil units, including a data sensing and acquisition unit 1, a model decision optimization unit 2, a field execution control unit 3, and a closed-loop correction and stabilization unit 4. Specifically: the data sensing and acquisition unit 1 is used to collect real-time thermal, load, and equipment energy efficiency data for each temperature control zone, and performs data standardization processing and mapping based on the DCS protocol; the model decision optimization unit 2 is used to construct a regional thermal inertia digital twin model to predict load trends, decouple thermal disturbances through a multi-objective game strategy, and generate globally optimal temperature control commands; the field execution control unit 3 is used to receive control commands issued by the DCS central management layer, drive the corresponding actuators to perform linear adjustment of the fan coil unit water valves and stepless speed control of the fan; and the closed-loop correction and stabilization unit 4 predicts the thermal response delay time based on a hysteresis feature identification model, uses predictive control for feedforward compensation, and dynamically corrects the residual deviation after decoupling.

[0022] It should be noted that the data perception and acquisition unit 1 collects and standardizes multi-region data in real time, transmits it to the model decision optimization unit 2 to construct a digital twin model and generate the optimal instruction based on multi-objective game decoupling, and then the field execution control unit 3 precisely drives the water valve and fan to execute. At the same time, the closed-loop correction and stabilization unit 4 performs feedforward compensation and residual deviation correction based on hysteresis identification.

[0023] In this embodiment, it should also be noted that the data sensing and acquisition unit 1 includes a multi-source sensor acquisition module 11, a load feature extraction module 12, an equipment status monitoring module 13, and a DCS standardization mapping module 14, wherein: the multi-source sensor acquisition module 11 is used to collect thermal parameters such as temperature, humidity, and inlet and outlet water temperature of each area in real time; the load feature extraction module 12 is used to dynamically analyze the load characteristics of personnel flow, equipment heat dissipation, and heat gain of the building envelope in the area; the equipment status monitoring module 13 is used to monitor the operating energy efficiency status of the fan coil unit in real time, including its speed, current, and valve opening; and the DCS standardization mapping module 14 is used to convert heterogeneous data into a unified DCS protocol standard format and perform address mapping.

[0024] It should be noted that the multi-source sensor acquisition module 11 collects thermal parameters of each area in real time and feeds them to the load feature extraction module 12 and the equipment status monitoring module 13 to analyze dynamic load characteristics and monitor operating energy efficiency status. The acquired heterogeneous data is uniformly integrated into the DCS standardization mapping module 14 for protocol conversion and address mapping.

[0025] Furthermore, it should be noted that the temperature sensor in the multi-source sensor acquisition module 11 has an accuracy of no less than ±0.1℃, the humidity sensor has an accuracy of no less than ±5%RH, and the coil inlet and outlet water temperature sensor has an accuracy of no less than ±0.2℃; the acquisition frequency is 1-5Hz, which can be flexibly adjusted according to the temperature control accuracy requirements; the sensor deployment locations are as follows: the temperature and humidity sensors are deployed in the core areas of each temperature control zone, and the coil inlet and outlet water temperature sensors are deployed on the inlet and outlet pipes of the fan coil unit.

[0026] The DCS standard mapping module 14 includes Modbus and OPC UA protocols for DCS communication, which can be adapted to different brands of DCS systems.

[0027] Address mapping rules: Assign a unique DCS address to each data point, and the address encoding follows the principle of "area number - module number - data type".

[0028] In this embodiment, it should also be noted that the model decision optimization unit 2 includes a digital twin construction module 21, a load trend prediction module 22, a game strategy decision module 23, and a thermal decoupling calculation module 24, wherein: the digital twin construction module 21: constructs a dynamic digital twin model reflecting the regional thermal inertia based on historical and real-time data; the dynamic digital twin model reflecting the regional thermal inertia is constructed using an equivalent thermal capacity-thermal resistance network, and the specific expression is as follows: ; In the formula, Let be the equivalent heat capacity of region i. Let be the equivalent thermal resistance of region i. Let i be the indoor temperature of region i. Outdoor temperature The cooling / heating provided to the fan coil units This refers to the internal thermal disturbance of personnel and equipment within the area. Load trend prediction module 22: used to predict the future trends of heating and cooling loads in each area using a dynamic digital twin model; Game strategy decision module 23: used to establish a multi-objective game function for global energy efficiency and local comfort and solve for the optimal balance point; specific operations are as follows: A1: Extract real-time energy efficiency data and local comfort feedback data for each temperature control area, determine the global energy efficiency target, local comfort target, and their constraints, including the upper limit of fan coil unit operating power and the area's temperature comfort range; A2: Establish a multi-objective game function for global energy efficiency and local comfort, with the specific expression as follows: ; In the formula, Let be the local comfort function for region i. Let i be the actual temperature of region i. This refers to the regional priority weighting coefficient. The total energy consumption of the system. A3: Based on the game function, a nonlinear programming algorithm is used to solve for the optimal solution, obtaining the control parameters or temperature setpoints of fan coil units in each region corresponding to the optimal balance point between global energy efficiency and local comfort; A4: Verify the feasibility of the optimal balance point. If the constraints are met, the decision parameters corresponding to the balance point are output; if not, the constraints are adjusted and the solution is re-solved until the optimal balance point is obtained. Thermal decoupling calculation module 24: Used to eliminate the thermal coupling effect between regions and generate independent control commands through a decoupling algorithm.

[0029] It should be noted that the digital twin construction module 21 constructs a dynamic digital twin model reflecting the thermal inertia of the region based on historical and real-time data. The load trend prediction module 22 uses this model to predict the future load evolution trend of each region. The game strategy decision module 23 establishes a multi-objective game function based on the predicted load and real-time energy efficiency and comfort data and solves the optimal equilibrium point. The thermal decoupling calculation module 24 eliminates the thermal coupling effect between regions through the decoupling algorithm according to the control parameters corresponding to the optimal equilibrium point, and finally generates independent control instructions for each zone.

[0030] Furthermore, it should be noted that in digital twin building module 21... The value range is 1000-5000 kJ / ℃; The value range is 0.5-2.0℃·h / kW.

[0031] The value range is 0.15-0.25 for important areas (meeting rooms) and 0.05-0.1 for ordinary areas (such as corridors). The value range is 0.01-0.1.

[0032] The future time period in the load trend forecast module 22 is 1-6 hours; the rolling forecast window updates the forecast results every 10 minutes.

[0033] Local comfort function in A2 Use the following expression: ; In the formula, Set the desired temperature for region i. For comfort sensitivity coefficient, Let be the actual temperature of region i.

[0034] Total system energy consumption The specific expression for the calculation is: ; In the formula, For fan power, Let n be the change in valve opening, and n be the total number of temperature control zones. Let be the wind turbine energy consumption weighting coefficient for region i. Let be the valve energy consumption weighting coefficient for region i.

[0035] Specific operation steps of thermal decoupling calculation module 24: C1: Constructing the inter-regional thermal coupling coefficient matrix , of which elements Characterizes the intensity of thermal interference from region j to region i. ; C2: Calculate the relative gain matrix ,in This represents the Schul product operation; C3: Based on the relative gain matrix The degree of deviation of the middle element from 1 determines the strength of coupling between regions: if If region i and region j are strongly coupled, then region i and region j are strongly coupled. C4: For strongly coupled regions, a feedforward compensation decoupler is used. The decoupling compensation amount is calculated as follows: ; In the formula, The optimal instruction for region i is output by the game decision module. The optimal instruction for region j is... is the decoupling coefficient.

[0036] In this embodiment, it should also be noted that the field execution control unit 3 includes an instruction parsing and distribution module 31, a water valve linear drive module 32, a fan stepless speed regulation module 33, and an execution status feedback module 34, wherein: the instruction parsing and distribution module 31 is used to receive and parse the globally optimal control instructions issued by the DCS central management layer; the water valve linear drive module 32 is used to continuously adjust the opening of the electric water valve of the fan coil unit according to the instructions to control the water flow; the fan stepless speed regulation module 33 is used to continuously adjust the fan speed according to the instructions to control the air supply volume; and the execution status feedback module 34 is used to upload the actual execution status of the valve position and fan speed to the central management layer in real time.

[0037] It should be noted that the instruction parsing and distribution module 31 receives and parses the global optimal control instruction issued by the DCS central management layer, and sends the water valve control component and the fan control component to the water valve linear drive module 32 and the fan stepless speed regulation module 33 respectively to perform continuous adjustment. At the same time, the execution status feedback module 34 collects the valve position and speed status in real time and sends them back to the central management layer.

[0038] Furthermore, it should be noted that the water valve linear drive module 32 uses a linear drive circuit, which supports continuous adjustment from 0-100%; adjustment accuracy: not less than 1%, response time not exceeding 1 second; feedback mechanism: real-time acquisition of the actual opening degree of the water valve and transmission to the execution status feedback module 34.

[0039] The stepless speed regulation module 33 for the fan adopts frequency conversion speed regulation technology; the speed regulation range is 30%-100% of the rated speed, and the adjustment accuracy is not less than 5%.

[0040] In this embodiment, it should also be noted that the closed-loop correction and stabilization unit 4 includes a hysteresis feature identification module 41, a feedforward prediction compensation module 42, a residual deviation correction module 43, and an overshoot oscillation suppression module 44, wherein: the hysteresis feature identification module 41 is used to identify the hysteresis time and inertial characteristics from command execution to temperature response online; the feedforward prediction compensation module 42 is used to output compensation commands in advance based on the hysteresis time using a predictive control algorithm; the specific operation is as follows: B1: Based on the thermal response hysteresis time output by the hysteresis feature identification module 41, the prediction time domain and rolling optimization window of the predictive control algorithm are determined; B2: Feedforward disturbance variables affecting the temperature of the area are collected, including the sudden change value of outdoor ambient temperature, the rate of change of the number of people in the area, and the fluctuation of fresh air load; B3: Using the state equation of the predictive control algorithm, the area temperature trajectory at future times is calculated based on the current system state and the predicted value of the disturbance variables in the future time domain; the specific expression of the state prediction equation of the predictive control algorithm is: ; In the formula, The temperature of the region at time k+1 is predicted based on information at time k. Let k be the measured temperature of the region. Let k be the control input for the fan coil unit at time k. Let A, B, and D be the feedforward disturbance variable vector monitored at time k, and let A, B, and D be the system state matrix, control input matrix, and disturbance input matrix, respectively. B4: Calculates the pre-control quantity required to eliminate the predicted temperature deviation and converts it into a feedforward compensation command, which is then superimposed on the basic control command. Residual deviation correction module 43: Used to dynamically correct small temperature deviations that still exist after decoupling. Overshoot oscillation suppression module 44: Used to suppress temperature overshoot oscillations and accelerate system stability through a damping control strategy.

[0041] It should be noted that the lag feature identification module 41 identifies the lag time and inertial characteristics online and feeds them to the feedforward prediction and compensation module 42. The feedforward prediction and compensation module 42 determines the prediction time domain and collects the disturbance variables accordingly. After calculating the future temperature trajectory through the state prediction equation, it outputs the feedforward compensation command. At the same time, the residual deviation correction module 43 dynamically corrects the small deviations after decoupling, and the overshoot oscillation suppression module 44 suppresses overshoot oscillations through a damping control strategy.

[0042] Furthermore, it should be noted that the specific operation steps of the hysteresis feature identification module 41 are as follows: D1: Apply a step control signal to the fan coil unit and record the time from the issuance of the command to the temperature change reaching a steady-state value, as the initial lag time. ; D2: Use correlation analysis to update lag time online. ; In the formula, For control commands, Temperature response, where T is the length of the integration time window, obtained through a search. Make the cross-correlation function To obtain the maximum value, That is, the lag time at the current moment. ; D3: Inertial characteristics are expressed using a first-order inertial element transfer function. The time constant T in the equation is represented by the least squares fitting, and K is the gain coefficient.

[0043] Calculate the pre-control quantity in B4 The optimization objective function is: ; In the formula, To predict the time domain, To control the time domain, This is the temperature at time k+j predicted based on time k. Using the reference temperature trajectory, Q and R are the weight matrices for temperature deviation and control increment, respectively. For reference temperature trajectory, To control the increment.

[0044] The specific operating steps of the residual deviation correction module 43 are as follows: E1: The residual deviation between the actual temperature and the set value after feedforward compensation and decoupling processing. ; E2: Perform moving average filtering on the residual bias to eliminate measurement noise. ; In the formula, This represents the residual bias after filtering. M is the original residual deviation, M is the sliding window length, and j is the summation index; E3: The residual deviation is corrected using an integral separation PID algorithm, with the integral term only applied to the integral term. Enabled at specific times to avoid points saturation; E4: Adjustment amount Overlay onto the current control command.

[0045] The specific operating steps of the overshoot oscillation suppression module 44 are as follows: F1: Overshoot of real-time temperature response curve monitoring ,in Peak temperature To set the temperature; F2: When the overshoot exceeds the preset threshold of 5%, calculate the damping control correction. ; in, The damping coefficient is... The rate of temperature change; F3: Reduce the control output amplitude in advance when the rate of temperature change crosses zero and overshoot is about to occur; F4: When the temperature enters the dead zone of ±0.5℃ of the set value, gradually remove the damping correction to avoid steady-state error.

[0046] Example 2: like Figure 2 As shown in this embodiment, the method for precise temperature control of multiple zones in a central control system for fan coil units specifically includes the following steps: S1. Multi-source data acquisition and standardization processing: S1.1: Real-time acquisition of thermal parameters for each temperature control zone, including indoor temperature, relative humidity, and inlet and outlet water temperature of fan coil units. The temperature sensor acquisition accuracy is no less than ±0.1℃, the humidity sensor acquisition accuracy is no less than ±5%RH, and the inlet and outlet water temperature sensor acquisition accuracy is no less than ±0.2℃. The acquisition frequency is set to 1-5Hz and can be flexibly adjusted according to actual temperature control accuracy requirements. Temperature and humidity sensors are deployed in the core areas of each temperature control zone, while inlet and outlet water temperature sensors for the fan coil units are deployed on the inlet and outlet pipes of the fan coil units. S1.2: Receive the transmitted thermal parameters, dynamically analyze the load characteristics of each area, specifically analyze the heat load caused by personnel movement, the heat load generated by equipment operation and the heat gain load of the building envelope, and output the dynamic load characteristic parameters of each area. S1.3: Real-time monitoring of the operating energy efficiency status of each fan coil unit, including fan operating speed, operating current and electric water valve opening, synchronous monitoring of the start-up and shutdown status and fault status of the fan coil unit, and real-time output of equipment operating data; S1.4: The three types of heterogeneous data, namely thermal parameters, load data and equipment operation data, are processed in a unified manner. They are converted into DCS protocol standard format using Modbus or OPC UA protocol, and a unique DCS address is assigned to each data point according to the principle of "area number-module number-data type". After the address mapping is completed, the standardized data is transmitted to step S2. S2, Digital Twin Modeling, Load Forecasting and Multi-Objective Game Decision Making: S2.1: Based on the real-time data transmitted in step S1, and combined with indoor and outdoor temperature and humidity, load characteristics, and equipment operation data from the past 1-3 years, a dynamic digital twin model reflecting the thermal inertia of each region is constructed using an equivalent heat capacity-thermal resistance network. The specific expression of the model is as follows: ; In the formula, Let be the equivalent heat capacity of region i. Let be the equivalent thermal resistance of region i. Let i be the indoor temperature of region i. Outdoor temperature The cooling / heating provided to the fan coil units Internal thermal disturbances to personnel and equipment within the area; S2.2: Using the dynamic digital twin model constructed in step S2.1, the evolution trend of cooling and heating load in each region in future periods is predicted in a rolling manner. The prediction time domain is set to the next 1-6 hours, and the rolling prediction window is set to update the prediction results every 10 minutes. Combined with the real-time load data and outdoor environment prediction data output in step S1, the peak value, valley value and change pattern of cooling and heating load in each region in future periods are output. S2.3: Based on the load trend prediction results of step S2.2, the real-time energy efficiency data and local comfort feedback data of step S1, establish a multi-objective game function between global energy efficiency and local comfort and solve for the optimal balance point. The specific operation is as follows: S2.3.1: Extract real-time energy efficiency data and local comfort feedback data of fan coil unit operating power and energy consumption in each temperature control zone to determine the minimum total energy consumption of the system and maintain the temperature of each zone within the comfort range. Define the constraints, including the upper limit of fan coil unit operating power (not exceeding 110% of the equipment's rated power), the comfort range of zone temperature, the water valve opening range (0-100%), and the fan speed range (30%-100% of the rated speed). S2.3.2: Establish a multi-objective game function, the specific expression of which is: ; In the formula, Let be the local comfort function for region i. Let i be the actual temperature of region i. This refers to the regional priority weighting coefficient. The total energy consumption of the system. Energy efficiency penalty coefficient; S2.3.3: Based on the above game function, a nonlinear programming algorithm is used to solve for the optimal solution. The temperature setpoint of each region, the operating power of the fan coil unit, the water valve opening, and the fan speed are used as decision variables to obtain the fan coil unit control parameters or temperature setpoint of each region corresponding to the optimal balance point between global energy efficiency and local comfort. S2.3.4: Verify the feasibility of the optimal equilibrium point, check whether each decision parameter meets the preset constraints. If the constraints are met, output the decision parameters corresponding to the equilibrium point; if the constraints are not met, appropriately relax the secondary constraints or adjust the regional priority weight coefficients. Repeat step S2.3.3 to solve the problem until the optimal equilibrium point that satisfies all constraints is obtained. S2.4: Based on the control parameters corresponding to the optimal equilibrium point output in step S2.3, the thermal coupling effect between regions is eliminated through a decoupling algorithm, generating independent temperature control commands for each region. The specific operation is as follows: S2.4.1: Constructing the inter-regional thermal coupling coefficient matrix , of which elements Characterizes the intensity of thermal interference from region j to region i, and ; S2.4.2: Calculate the relative gain matrix ,in This represents the Schul product operation; S2.4.3: Based on the relative gain matrix The degree of deviation of the element from 1 determines the strength of coupling between regions. If so, then it is determined that region i and region j are strongly coupled; S2.4.4: For regions with strong coupling, a feedforward compensation decoupler is used to calculate the decoupling compensation amount. The specific expression is as follows: In the formula, The optimal instruction for region i is output by the game decision module. The optimal instruction for region j is... As the decoupling coefficient, the decoupling compensation amount is superimposed on the optimal command to generate independent temperature control commands for each region, which are then transmitted to the DCS central management layer. S3. Control command execution and status feedback: S3.1: Receive the global optimal control command issued by the DCS central management layer, complete the command parsing within 0.5 seconds, decompose the command into water valve control component and fan control component, and send them to steps S3.2 and S3.3 respectively; S3.2: A linear drive circuit is adopted to continuously adjust the opening of the electric water valve of the fan coil unit according to the water valve control component (supporting 0-100% continuous adjustment), with an adjustment accuracy of not less than 1% and a response time of not more than 1 second. The actual opening of the water valve is collected in real time and transmitted to step S3.4. S3.3: It adopts variable frequency speed regulation technology, continuously adjusts the fan speed according to the fan control component, the speed regulation range is 30%-100% of the rated speed, the regulation accuracy is not less than 5%, and it also has overcurrent and overload protection functions. In case of abnormality, the speed regulation will stop and a fault signal will be fed back. S3.4: Collect the actual valve position of the water valve and the actual speed of the fan every 0.5 seconds. After standardizing the collected execution status data, transmit it back to the DCS central management layer in real time to provide a basis for the instruction adjustment in step S2 and the correction and optimization in step S4. S4. Closed-loop correction and stability optimization: S4.1: Online identification of lag time and inertial characteristics from command execution to temperature response. The specific operation is as follows: S4.1.1: Apply a step control signal to the fan coil unit and record the time from the issuance of the command to the temperature change reaching the steady-state value, as the initial lag time. ; S4.1.2: The lag time is updated online using correlation analysis. The correlation function expression is as follows: ; In the formula, For control commands, Temperature response, where T is the length of the integration time window, obtained through a search. Make the cross-correlation function To obtain the maximum value, That is, the lag time at the current moment. ; S4.1.3: The inertial characteristics adopt the transfer function of a first-order inertial element. The time constant T in the figure is represented by least squares fitting, and K is the gain coefficient; the hysteresis feature identification module 41 updates the hysteresis time and inertial characteristic parameters every 1-5 minutes and transmits them to step S4.2 in real time; S4.2: Based on the lag parameters output in step S4.1, a predictive control algorithm is used to output compensation commands in advance, which are then superimposed on the basic control commands. The specific operation is as follows: S4.2.1: Based on the thermal response lag time of the output, determine the prediction time domain and rolling optimization window of the predictive control algorithm. The prediction time domain is set to 1.5-2 times the lag time, and the rolling optimization window is set to 10-30 minutes. S4.2.2: Collect feedforward disturbance variables that affect the temperature of the area, including the sudden change value of outdoor ambient temperature, the rate of change of the number of people in the area, the fluctuation of fresh air load, the change of solar radiation intensity, and the change of the start-up and shutdown status of indoor equipment. All disturbance variables are transferred to this step after being standardized in step S1.4. S4.2.3: Using the state equation of the predictive control algorithm, based on the current system state and the predicted values ​​of disturbance variables in the future time domain, the regional temperature trajectory at future times is calculated. The specific expression of the state prediction equation is as follows: ; In the formula, The temperature of the region at time k+1 is predicted based on information at time k. Let k be the measured temperature of the region. Let k be the control input for the fan coil unit at time k. Let be the feedforward disturbance variable vector detected at time k, and A, B, and D be the system state matrix, control input matrix, and disturbance input matrix, respectively. S4.2.4: With the goal of "no deviation between the predicted temperature trajectory and the reference temperature trajectory", calculate the pre-control quantity required to eliminate the predicted temperature deviation. The objective function is optimized as follows: ; In the formula, To predict the time domain, To control the time domain, This is the temperature at time k+j predicted based on time k. Using the reference temperature trajectory, Q and R are the weight matrices for temperature deviation and control increment, respectively. For reference temperature trajectory, To control the increment; S4.3: Dynamically correct any minor temperature deviations that still exist after decoupling and feedforward compensation. The specific operation is as follows: S4.3.1: Collect the residual deviation between the actual temperature and the set value of each region after feedforward compensation and decoupling processing. ; S4.3.2: Perform moving average filtering on the residual deviation to remove measurement noise. The filtering expression is: ; In the formula, This represents the residual bias after filtering. M is the original residual deviation, M is the sliding window length, and j is the summation index; S4.3.3: An integral separation PID algorithm is used to correct the residual deviation, with the integral term only applied to the integral term. Enabled at specific times to avoid points saturation; S4.3.4: The calculated correction amount By superimposing this onto the current control command, the temperature control accuracy is further improved, ensuring that the final temperature deviation is maintained within ±0.2℃; S4.4: Suppress temperature overshoot oscillations and accelerate system stability through a damping control strategy. The specific operation is as follows: S4.4.1: Real-time monitoring of overshoot in temperature response curves of each region ,in Peak temperature To set the temperature; S4.4.2: When the overshoot exceeds a preset threshold of 5%, calculate the damping control correction amount, the specific expression of which is: ; in, The damping coefficient is... The rate of temperature change; S4.4.3: When the rate of temperature change crosses zero and overshoot is about to occur, reduce the control output amplitude in advance to suppress overshoot; S4.4.4: When the temperature enters the dead zone of ±0.5℃ of the set value, the damping correction is gradually removed to avoid steady-state error and ensure that the system quickly stabilizes at the set temperature.

[0047] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0048] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for centralized multi-zone precise temperature control of fan coil units, characterized in that: Includes the following steps: S1: Real-time acquisition of environmental thermal parameters, dynamic load characteristics and fan coil unit operating energy efficiency status of each temperature control zone, and data standardization mapping through DCS communication protocol; S2: Based on standardized data, construct a regional thermal inertia digital twin model to predict the load evolution trend of each region, and based on a multi-objective game strategy of global energy efficiency and local comfort, dynamically decouple the thermal interference between regions and generate the optimal control command. S3: The DCS central management layer sends the global optimal control command to the field control station, driving the actuator to linearly adjust the water valve opening of the fan coil unit and perform stepless speed control of the fan speed. S4: Based on the hysteresis feature identification model, predict the thermal response delay time after S3 is executed, use the predictive control algorithm to perform feedforward compensation on the control output, and dynamically correct the residual deviation after S2 decoupling to suppress temperature overshoot oscillation.

2. A centralized multi-zone precise temperature control system for fan coil units, as described in claim 1, characterized in that, It includes a data sensing and acquisition unit (1), a model decision optimization unit (2), an on-site execution and control unit (3), and a closed-loop correction and stabilization unit (4), wherein: The data sensing and acquisition unit (1) is used to collect relevant thermal, load and equipment energy efficiency data of each temperature control zone in real time, and to perform data standardization processing and mapping based on the DCS protocol. The model decision optimization unit (2) is used to construct a regional thermal inertia digital twin model to predict load trends, decouple thermal interference through a multi-objective game strategy, and generate a globally optimal temperature control command. The field execution control unit (3) is used to receive control instructions issued by the DCS central management layer and drive the corresponding actuator to perform linear adjustment of the fan coil water valve and stepless speed control of the fan speed. The closed-loop correction and stabilization unit (4) predicts the thermal response delay time based on the hysteresis feature identification model, uses predictive control for feedforward compensation, and dynamically corrects the residual deviation after decoupling.

3. The fan coil unit central control multi-zone precision temperature control system according to claim 2, characterized in that, The data sensing and acquisition unit (1) includes a multi-source sensing acquisition module (11), a load feature extraction module (12), an equipment status monitoring module (13), and a DCS standardized mapping module (14), wherein: The multi-source sensor acquisition module (11) is used to collect thermal parameters such as temperature, humidity and inlet / outlet water temperature of each area in real time. The load feature extraction module (12) is used to dynamically analyze the load features of personnel flow, equipment heat dissipation, and heat gain of the building envelope in the area. The equipment status monitoring module (13) is used to monitor the operating energy efficiency status of the fan coil unit in real time, including its speed, current, and valve opening. The DCS standardization mapping module (14) is used to convert heterogeneous data into DCS protocol standard format and perform address mapping.

4. The fan coil unit central control multi-zone precision temperature control system according to claim 2, characterized in that, The model decision optimization unit (2) includes a digital twin construction module (21), a load trend prediction module (22), a game strategy decision module (23), and a thermal decoupling calculation module (24), wherein: The digital twin construction module (21) constructs a dynamic digital twin model reflecting regional thermal inertia based on historical and real-time data; The load trend prediction module (22) is used to predict the evolution trend of cold and heat loads in each region in future periods through a dynamic digital twin model. The game strategy decision module (23) is used to establish a multi-objective game function of global energy efficiency and local comfort and solve for the optimal equilibrium point; The thermal decoupling calculation module (24) is used to eliminate the thermal coupling effect between regions and generate independent control commands through the decoupling algorithm.

5. The fan coil unit central control multi-zone precision temperature control system according to claim 4, characterized in that, The dynamic digital twin model reflecting the regional thermal inertia in the digital twin construction module (21) is constructed using an equivalent thermal capacity-thermal resistance network, and the specific expression is as follows: ； In the formula, Let be the equivalent heat capacity of region i. Let be the equivalent thermal resistance of region i. Let i be the indoor temperature of region i. Outdoor temperature The cooling / heating provided to the fan coil units This refers to the internal thermal disturbance of personnel and equipment within the area.

6. The fan coil unit central control multi-zone precision temperature control system according to claim 4, characterized in that, The game strategy decision module (23) establishes a multi-objective game function for global energy efficiency and local comfort and solves for the optimal equilibrium point. The specific operations are as follows: A1: Extract real-time energy efficiency data and local comfort feedback data for each temperature control zone, determine the global energy efficiency target, local comfort target and their constraints, including the upper limit of fan coil unit operating power and the zone temperature comfort range; A2: Establish a multi-objective game function that balances global energy efficiency and local comfort, with the following specific expression: ； In the formula, Let be the local comfort function for region i. Let i be the actual temperature of region i. This refers to the regional priority weighting coefficient. The total energy consumption of the system. Energy efficiency penalty coefficient; A3: Based on the game function, a nonlinear programming algorithm is used to solve for the optimal solution, and the control parameters or temperature setpoints of the fan coil units in each region corresponding to the optimal balance point between global energy efficiency and local comfort are obtained. A4: Verify the feasibility of the optimal equilibrium point. If the constraints are met, output the decision parameters corresponding to the equilibrium point. If not, adjust the constraints and solve again until the optimal equilibrium point is obtained.

7. The fan coil unit central control multi-zone precision temperature control system according to claim 2, characterized in that, The on-site execution control unit (3) includes an instruction parsing and distribution module (31), a water valve linear drive module (32), a fan stepless speed regulation module (33), and an execution status feedback module (34), wherein: The instruction parsing and distribution module (31) is used to receive and parse the global optimal control instructions issued by the DCS central management layer. The water valve linear drive module (32) is used to continuously adjust the opening degree of the fan coil electric water valve according to the instruction to control the water flow rate; The fan stepless speed regulation module (33) is used to continuously adjust the fan speed according to the command to control the air volume; The execution status feedback module (34) is used to upload the actual execution status of valve position and fan speed to the central management layer in real time.

8. The fan coil unit central control multi-zone precision temperature control system according to claim 2, characterized in that, The closed-loop correction and stabilization unit (4) includes a hysteresis feature identification module (41), a feedforward prediction compensation module (42), a residual deviation correction module (43), and an overshoot oscillation suppression module (44), wherein: The hysteresis feature identification module (41) is used to identify the hysteresis time and inertial characteristics from instruction execution to temperature response online. The feedforward prediction compensation module (42) uses a predictive control algorithm to output compensation commands in advance based on the lag time. The residual deviation correction module (43) is used to dynamically correct the small temperature deviations that still exist after decoupling. The overshoot oscillation suppression module (44) is used to suppress temperature overshoot oscillations and accelerate system stability through a damping control strategy.

9. The fan coil unit central control multi-zone precision temperature control system according to claim 8, characterized in that, The feedforward prediction and compensation module (42) uses a predictive control algorithm based on the lag time to output compensation commands in advance. The specific operation is as follows: B1: Based on the thermal response lag time output by the lag feature identification module (41), determine the prediction time domain and rolling optimization window of the predictive control algorithm; B2: Collect feedforward disturbance variables that affect the temperature of the area. The disturbance variables include the sudden change value of outdoor ambient temperature, the rate of change of the number of people in the area, and the fluctuation of fresh air load. B3: Using the state equation of the predictive control algorithm, the regional temperature trajectory at future times is calculated based on the current system state and the predicted values ​​of disturbance variables in the future time domain. B4: Calculate the pre-control quantity required to eliminate the predicted temperature deviation and convert it into a feedforward compensation command that is superimposed on the basic control command.

10. The fan coil unit central control multi-zone precision temperature control system according to claim 9, characterized in that, The specific expression for the state prediction equation of the predictive control algorithm in step B3 is as follows: ； In the formula, The temperature of the region at time k+1 is predicted based on information at time k. Let k be the measured temperature of the region. Let k be the control input for the fan coil unit at time k. Let be the feedforward disturbance variable vector detected at time k, and let A, B, and D be the system state matrix, control input matrix, and disturbance input matrix, respectively.

Images