Radiation air conditioner water temperature feed-forward adjusting method and system based on multivariable robust model

Abstract

The invention discloses a radiation air conditioner water temperature feed-forward adjusting method and system based on a multivariable robust model. The method comprises the following steps that multi-source real-time data of a radiation air conditioner system are collected; preprocessing the data; constructing a multivariable robust model; on the basis of the multivariable robust model and the current system state, a feedforward control quantity used for restraining disturbance is calculated through an H-infinity robust feedforward controller; according to the deviation between the set value of the radiation outlet water temperature and the actual feedback value, the feedback control quantity used for eliminating the steady-state error is calculated; superposing the feedforward control quantity and the feedback control quantity to generate a final control instruction; and according to the final control instruction, an execution mechanism is driven to adjust the opening degree of the proportional electronic three-way water mixing valve and the frequency of the variable-frequency water pump. According to the method, a multivariable robust thermal resistance model is constructed by fusing indoor load requirements, system dynamic variables and the like in real time, and it is ensured that the radiation outlet water temperature is stabilized within the set temperature range of + / -0.3 DEG C through a double-self-correction mechanism and a cooperative control strategy.

Description

Technical Field

[0001] This invention relates to the field of radiant air conditioning control technology, and in particular to a method and system for feedforward regulation of radiant air conditioning water temperature based on a multivariable robust model. Background Technology

[0002] Radiant air conditioning systems are air conditioning technologies that use thermal radiation as the primary heat transfer method (accounting for ≥50%). By cooling / heating the inner surfaces of the building envelope (ceiling, floor, or walls), radiant surfaces are formed to directly exchange heat with the human body and indoor objects, thereby achieving temperature regulation.

[0003] The existing radiant air conditioning system has the following core problems in water temperature regulation:

[0004] 1. Insufficient accuracy: Traditional control methods (such as PID) cannot meet the high accuracy requirement of ±0.3℃, and the measured fluctuation often reaches more than ±0.7℃. This is because radiant air conditioning systems have characteristics such as large hysteresis, large inertia, and nonlinearity, making it difficult for conventional PID controllers to accurately adjust the water temperature, resulting in an uneven indoor temperature field and affecting human comfort.

[0005] 2. Poor robustness: The system does not consider dynamic factors such as indoor temperature, heat pump efficiency drift, and water pump frequency fluctuations, making it sensitive to load changes. For example, when the indoor humidity rises from 50%RH to 65%RH, the radiant water temperature fluctuation of traditional PID control can increase by more than 1%, leading to control failure.

[0006] 3. Weak Adaptability: Lacking an online self-calibration mechanism, manual intervention is required when system parameters (such as thermal resistance and heat pump performance) change. For example, pipe scaling or a decrease in heat pump efficiency can cause changes in thermal resistance parameters, which traditional PID controllers cannot automatically adapt to, requiring parameter resetting.

[0007] 4. Insufficient coordinated control: Adjusting only the three-way valve or only the water pump fails to achieve coordinated optimization of water temperature and flow rate. In traditional methods, fluctuations in water pump frequency introduce uncertainty in the mixing water temperature (e.g., the heat dissipation of the water pump is cubic with frequency, reaching 0.679kW at 60Hz, but only 0.209kW at 60Hz), but this is not coordinated with the opening of the three-way valve in a closed-loop control.

[0008] Patent CN117804035A discloses a control method, device, radiant air conditioner, and storage medium for a radiant air conditioner, relating to the field of radiant air conditioning. The method includes: acquiring the indoor temperature; comparing the indoor temperature with a first set temperature to obtain a first comparison result; and adjusting the temperature of the heat exchange medium in the liquid supply pipeline according to the current operating mode of the radiant air conditioner and the first comparison result. It can achieve automatic variable water temperature regulation, avoiding the high energy consumption problem caused by pump control, thereby reducing the energy consumption of the radiant air conditioner and ensuring user comfort. However, its control accuracy is limited, it does not consider multivariate coupling effects and parameter uncertainties, making it difficult to cope with complex actual working conditions; it lacks system modeling and parameter uncertainty handling, resulting in insufficient robustness; control accuracy, response speed, and anti-interference ability need improvement, making it difficult to achieve high-precision water temperature control.

[0009] Patent CN119934658A discloses a method for controlling a radiant air conditioning system, comprising: determining a first target mixing temperature; determining a first target opening degree of a first mixing valve based on the supply water temperature of the heat pump unit, the outlet water temperature of the floor radiant heat exchanger, and the first target mixing temperature; determining a second target opening degree of a second mixing valve based on the first target mixing temperature and the outlet water temperature of the roof radiant heat exchanger; and, when the operating mode is cooling mode, simultaneously controlling the first mixing valve to adjust to the first target opening degree and the second mixing valve to adjust to the second target opening degree to improve indoor comfort. This solution not only avoids the risk of low-temperature air accumulation and condensation caused by excessively low floor temperatures during summer cooling, but also improves indoor comfort and the operating efficiency of the radiant air conditioning system, providing users with a more comfortable indoor environment. It calculates the mixing valve opening degree and pump frequency by analyzing factors such as the radiant outlet water temperature, indoor temperature, and heat pump unit water temperature. Although multiple variables are considered, the control strategy still relies mainly on empirical parameter adjustments, lacks system modeling and parameter uncertainty handling, and has insufficient robustness. There is still considerable room for improvement in control accuracy, response speed and anti-interference capability, especially when facing multivariate coupling and parameter uncertainty, it is difficult to achieve high-precision water temperature control. Summary of the Invention

[0010] The purpose of this invention is to provide a technical solution for a feedforward regulation method and system for radiant air conditioning water temperature based on a multivariable robust model, addressing the shortcomings of existing technologies. By integrating indoor load demand and system dynamic variables in real time, a multivariable robust thermal resistance model is constructed. Through a dual self-correction mechanism and a collaborative control strategy, the opening degree of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump are adjusted in a feedforward manner to ensure that the radiant outlet water temperature is stable within the set temperature ±0.3℃ range. This invention is suitable for radiant air conditioning systems in residences, hotels, office buildings, etc.

[0011] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0012] A method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model, characterized by the following steps:

[0013] S1. Collect multi-source real-time data from the radiant air conditioning system;

[0014] S2. Preprocess the multi-source real-time data to eliminate noise interference;

[0015] S3. Construct a multivariate robust model of the dynamic characteristics of the radiant air conditioning system. The multivariate robust model is based on the state-space model and introduces a parameter uncertainty quantification method.

[0016] S4. Based on the multivariable robust model and the current system state, calculate the feedforward control quantity used to suppress disturbances through a robust feedforward controller;

[0017] S5. Based on the deviation between the set value and the actual feedback value of the radiant outlet water temperature, the feedback controller calculates the feedback control quantity used to eliminate steady-state error.

[0018] S6. Superimpose the feedforward control quantity and the feedback control quantity to generate the final control command. The control command includes the opening value of the proportional electronic three-way mixing valve and the frequency value of the variable frequency water pump.

[0019] S7. According to the final control command, drive the actuator to adjust the opening degree of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump.

[0020] This water temperature feedforward regulation method constructs a multivariate robust thermal resistance model by integrating indoor load demand and system dynamic variables in real time. Through a dual self-correction mechanism and a collaborative control strategy, it feedforward adjusts the opening of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump to ensure that the radiant outlet water temperature is stable within the set temperature ±0.3℃ range. It is suitable for radiant air conditioning systems in residences, hotels, office buildings, etc.

[0021] A radiant air conditioning water temperature feedforward regulation system based on a multivariable robust model, used to implement the radiant air conditioning water temperature feedforward regulation method described above, characterized in that it includes:

[0022] Sensor networks are used to collect multi-source real-time data from radiant air conditioning systems.

[0023] The data acquisition and preprocessing module connects to the sensor network and is used to filter and normalize multi-source real-time data.

[0024] A multivariate robust model, connected to the data acquisition and preprocessing module, is used to store and run a multivariate robust model describing the dynamic characteristics of the radiant air conditioning system, and to perform state estimation and prediction.

[0025] The feedforward compensation module is connected to the multivariate robust model and is used to calculate the feedforward control quantities of the mixing valve opening and pump frequency to suppress disturbances based on the multivariate robust model and the current system state.

[0026] The feedback controller module is used to calculate the feedback control quantity to eliminate steady-state error based on the deviation between the set value and the actual feedback value of the radiant outlet water temperature.

[0027] The control output calculation module is connected to the feedforward compensation module and the feedback controller module respectively. It is used to superimpose the feedforward control quantity and the feedback control quantity to generate the final control command.

[0028] The execution module, connected to the control output calculation module, is used to receive and execute control commands.

[0029] This system addresses the problem of insufficient water temperature control accuracy in existing radiant air conditioning systems, meeting high comfort requirements; it also solves the problem of slow response speed in existing radiant air conditioning systems, shortening the temperature adjustment cycle from 15-30 minutes to 5 minutes, improving the system's adaptability to load changes; it addresses the problem of poor anti-interference capability in existing radiant air conditioning systems, enabling the system to effectively cope with interference factors such as heat pump unit efficiency fluctuations, pipe network resistance changes, and load prediction errors through multivariate robust model design; and it addresses the problem of improper handling of multivariate coupling effects in existing radiant air conditioning systems, achieving coordinated control of multiple variables such as indoor load changes, heat pump unit water temperature, radiant outlet water temperature, radiant return temperature, and water pump frequency through a multivariate robust model.

[0030] The present invention, by adopting the above-described technical solution, has the following beneficial effects:

[0031] 1. This invention describes the dynamic characteristics of a radiant air conditioning system by using a state-space model. At the same time, it introduces a parameter uncertainty quantification method to model uncertainties such as heat pump host efficiency fluctuations and pipe network component changes as interval matrices or set matrices. Through H∞ robust control theory, a gain matrix is ​​designed so that the system can maintain good stability and control accuracy even when parameter uncertainties exist.

[0032] 2. This invention combines feedforward compensation with feedback control. Feedforward compensation is based on load change trend prediction, adjusting the mixing valve opening and pump frequency in advance. Feedback control is based on the deviation between the actual water temperature and the set value, adjusting the control quantity in real time. This collaborative control strategy can effectively improve the system's response speed and anti-interference ability to load changes, especially when facing sudden load changes, the control effect is significantly better than traditional methods.

[0033] 3. This invention introduces a dynamic coupling compensation mechanism for humidity and water temperature. By calculating the indoor dew point temperature, the set value of the radiant water temperature is dynamically adjusted to prevent condensation on the cold radiant surface. At the same time, the humidity deviation is used as part of the control input, and the temperature and humidity are coordinated and controlled through a multivariate robust model, which solves the problem of mutual interference between temperature and humidity control in traditional radiant air conditioning systems. Attached Figure Description

[0034] The present invention will be further described below with reference to the accompanying drawings:

[0035] Figure 1 This is a flowchart of a radiative air conditioning water temperature feedforward regulation method based on a multivariable robust model and the regulation method in the system according to the present invention.

[0036] Figure 2 This is a schematic diagram illustrating the construction process of the multivariate robust model in this invention;

[0037] Figure 3 This is a flowchart illustrating the design of the H∞ robust feedforward controller in this invention.

[0038] Figure 4 This is a flowchart illustrating the multivariate coupling effect analysis and structure matrix design in this invention.

[0039] Figure 5 This is a flowchart illustrating the dew point calculation algorithm in this invention;

[0040] Figure 6 This is a schematic diagram of the radiant air conditioning water temperature feedforward regulation system in this invention. Detailed Implementation

[0041] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0042] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. 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 skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0043] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0044] like Figure 1 As shown, this invention provides a method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model, comprising the following steps:

[0045] S1. Collect multi-source real-time data from the radiant air conditioning system.

[0046] Multi-source real-time data includes indoor environmental parameters, heat pump unit operating parameters, radiant water intake operating parameters, and variable frequency water pump operating parameters.

[0047] S2. Preprocess the multi-source real-time data to eliminate noise interference.

[0048] S3. Construct a multivariate robust model of the dynamic characteristics of the radiant air conditioning system. The multivariate robust model is based on the state-space model and introduces a parameter uncertainty quantification method.

[0049] like Figure 2 As shown, the construction of a multivariable robust model mainly includes: system variable definition, thermal equilibrium and fluid dynamics equations, and state-space model derivation.

[0050] The specific steps involved in constructing a multivariate robust model are as follows:

[0051] S3.1 System variable definition.

[0052] A radiant air conditioning system is a typical multiple-input multiple-output (MIMO) system, with complex and strongly coupled relationships between its input and output variables. Based on the system characteristics, the following state variables, input variables, and output variables are defined.

[0053] Defining system variables specifically includes the following steps:

[0054] S3.1a defines state variables, including at least radiant water temperature, indoor temperature, indoor humidity, and water pump frequency;

[0055] S3.1b defines control input variables, including the opening degree of the proportional electronic three-way mixing valve and the frequency command of the variable frequency pump;

[0056] S3.1c defines disturbance input variables, which include at least indoor load changes and heat pump unit outlet water temperature fluctuations;

[0057] S3.1d defines output variables, including radiant water temperature, indoor temperature, and indoor humidity.

[0058] State variable vector:

[0059] ;

[0060] Wherein: T radiant Radiated water temperature (°C);

[0061] T indoor Indoor temperature (°C);

[0062] H indoor Indoor humidity (%RH);

[0063] n pump Pump frequency (Hz).

[0064] Controlling input variables:

[0065] ;

[0066] Where: u valve : Proportional electronic three-way mixing valve opening (0~100%);

[0067] u pump : Variable frequency pump frequency command (Hz).

[0068] Perturbation input variables:

[0069] ;

[0070] Where: ΔQ load Indoor load variation (W);

[0071] ΔT source : Temperature fluctuation of water outlet of heat pump unit (°C).

[0072] System output variables:

[0073] ;

[0074] Where: y1: radiant water temperature (°C);

[0075] y2: Indoor temperature (°C);

[0076] y3: Indoor humidity (%RH).

[0077] S3.2 Based on the radiant terminal heat balance equation, the variable frequency pump characteristic equation, and the mixing valve characteristic equation, linearization is performed to construct the state space model of the radiant air conditioning system.

[0078] S3.3 models the uncertainty factors as interval matrices or set matrices to form a multivariate robust model.

[0079] Radiation terminal heat balance equation:

[0080] ;

[0081] in: : Water flow rate (kg / s);

[0082] c p Specific heat capacity of water (4186 J / kg·K);

[0083] T out : Water temperature at the radiant terminal (°C);

[0084] T in : Inlet water temperature at the radiant terminal (°C);

[0085] Q loss Heat loss at the radiating end (W);

[0086] ΔQ load Indoor load variation (W).

[0087] Variable frequency pump characteristic equation:

[0088] Relationship between flow rate and rotational speed: Q=k1n 2 ;

[0089] Relationship between head and speed: H = k2n 2 ;

[0090] Relationship between shaft power and rotational speed: N = k3n 2 ;

[0091] Where: n is the pump frequency in Hz;

[0092] k1, k2, and k3 are the pump characteristic coefficients.

[0093] Characteristic equation of mixing valve:

[0094] ;

[0095] ;

[0096] Wherein: T mix : The temperature (°C) of the water entering the radiation zone after mixing with the water.

[0097] T hot T cold : Hot / cold source water temperature (°C);

[0098] Q hot Q cold Heat / cold source flow rate (m³) 3 / s);

[0099] Q mix Total flow rate after mixing (m³) 3 / s).

[0100] Linearizing the above equations, we construct a state-space model of the radiant air conditioning system, whose equations are as follows:

[0101] ;

[0102] ;

[0103] Where A is the system dynamic matrix;

[0104] B1—Control input matrix;

[0105] B2—Perturbation input matrix;

[0106] C — Output matrix;

[0107] D1 — Linear transfer matrix controlling the input to the output;

[0108] D2—The linear transfer matrix from the disturbance input to the output;

[0109] x — state variable;

[0110] u — Control input variable;

[0111] d—Perturbation input variable;

[0112] y — Output variable; —State-space model.

[0113] Derivation of the system dynamic matrix (A): The system dynamic matrix is ​​obtained by using the differential terms of the radiant terminal heat balance equation and the linearization of the variable frequency pump characteristic equation.

[0114] ;

[0115] Where: C radiant : Heat capacity at the radiating end (J / K);

[0116] C indoor Indoor air heat capacity (J / K);

[0117] G h Humidity transfer coefficient (kg / (s·%RH));

[0118] C humid Indoor humidity volume (m³) 3 );

[0119] τ pump : Pump frequency time constant (s).

[0120] Derivation of the control input matrix (B1): The control input matrix is ​​obtained by analyzing the effects of the mixing valve opening and the pump frequency on the system state.

[0121] ;

[0122] Derivation of the disturbance input matrix (B2): The disturbance input matrix is ​​obtained by examining the effects of indoor load changes and heat pump outlet water temperature fluctuations on the system state.

[0123] ;

[0124] Derivation of the output matrix (C): The output matrix is ​​obtained through the direct relationship between the system state variables and the output variables.

[0125] ;

[0126] Derivation of direct transfer matrices (D1) and (D2): The direct transfer matrices are obtained through the direct transfer relationship between the system's input and output variables.

[0127] ;

[0128] .

[0129] S4. Based on the multivariable robust model and the current system state, the feedforward control quantity used to suppress disturbances is calculated through a robust feedforward controller. The robust feedforward controller is an H∞ robust feedforward controller.

[0130] like Figure 3 As shown, the design of the H∞ robust feedforward controller mainly includes: the design steps of the H∞ robust feedforward controller, the feedforward gain matrix (K... ff ) Calculation, feedforward gain matrix (K ff Verification and correction.

[0131] The specific design steps for the H∞ robust feedforward controller include:

[0132] S4.1 converts the state-space model of the radiant air conditioning system into the standard form of H∞ control, defining the controlled output vector z and the measurement output vector y;

[0133] The standard form of H∞ control is:

[0134] ;

[0135] ;

[0136] ;

[0137] Where z: controlled output vector (radiation outlet water temperature, indoor temperature, indoor humidity);

[0138] y: Measurement output vector (same as z);

[0139] D 11 D 12 D 21D 22 : Directly pass the matrix.

[0140] For a radiant air conditioning system, assume:

[0141] C1=C, C2=C, D 11 =0, D 12 =0, D 21 =0, D 22 =0.

[0142] S4.2 By solving the algebraic Riccati equation (ARE), the state feedback gain matrix K and the feedforward gain matrix K are obtained. ff ;

[0143] The steps for solving the controller state feedback gain matrix K are as follows:

[0144] First, define the gain matrices (K) and (L):

[0145] ;

[0146] ;

[0147] Here, S and R are both positive definite matrices, which need to be solved using ARE.

[0148] Secondly, solve for ARE:

[0149] .

[0150] S4.3 Construct the controller structure as u = -Kx + K ff r, where u is the controller output (mixing valve opening degree, water pump frequency command), x is the state variable vector, and K is the state feedback gain matrix. ff : Feedforward gain matrix, r is the reference input vector including the set temperature and set humidity;

[0151] For a radiant air conditioning system, the reference input (r) is:

[0152] .

[0153] Feedforward gain matrix (K ff The design purpose is to eliminate the impact of model uncertainty on system performance.

[0154] For radiant air conditioning systems, K ff The calculation formula is:

[0155] ;

[0156] Due to the radiant air conditioning system (D) 11 =0) and (D12 =0), therefore K ff =0.

[0157] S4.4 determines the performance index γ of the radiant air conditioning system through frequency domain analysis and sensitivity optimization. The calculation formula is as follows:

[0158] ;

[0159] Where, σ max (G(jω)) is the maximum singular value of the open-loop transfer function matrix, and 1.2 is the safety factor.

[0160] The selection of performance index γ directly affects the robustness and performance of the system. For radiant air conditioning systems, the selection of γ needs to consider the following factors:

[0161] a. System dynamic characteristics: Radiant air conditioning systems are characterized by high inertia and long delay, and an appropriate γ needs to be selected to balance robustness and performance;

[0162] b. Interference characteristics: Amplitude and frequency characteristics of indoor load changes and heat pump outlet water temperature fluctuations;

[0163] c. Control accuracy requirements: The radiant water temperature must be controlled within the range of ±0.3℃ of the set temperature.

[0164] like Figure 4 As shown, it also includes multivariate coupling and decoupling processing, mainly including multivariate coupling effect analysis, relative gain matrix (RGA) calculation, and decoupling matrix design.

[0165] Radiant air conditioning systems are typical multivariable coupled systems. Changes in control variables (mixing valve opening, pump frequency) affect multiple output variables (radiant outlet water temperature, indoor temperature, indoor humidity). The formula for calculating this type of RGA is:

[0166] ;

[0167] Where: G(0): steady-state gain of the transfer function matrix at (s=0);

[0168] Hadamard product (element-wise multiplication), which is the multiplication of corresponding elements;

[0169] Φ ij Relative gain, representing the input (u) j ) for output (y i The relative gain of ).

[0170] Specifically, the steps include the following:

[0171] Based on the state-space model, calculate the steady-state gain G(0) of the system transfer function matrix.

[0172] For a radiant air conditioning system, the transfer function matrix steady-state gain .

[0173] Inverse steady-state gain matrix calculate:

[0174] ;

[0175] Among them, g ij (0) is the element in the (i)th row and (j)th column of G(0).

[0176] For example, if the steady-state gain of the transfer function matrix of a radiant air conditioning system is:

[0177] ;

[0178] Then its relative gain matrix is:

[0179] .

[0180] The relative gain matrix Φ is calculated based on the steady-state gain G(0) to quantify the coupling degree between the mixing valve opening and the pump frequency on output variables such as radiant outlet water temperature and indoor temperature.

[0181] The purpose of decoupling matrices is to reduce or eliminate the coupling effect between control variables, so that each output variable is controlled by only one input variable.

[0182] Based on the relative gain matrix Φ, design the decoupling matrix:

[0183] ;

[0184] And transform the control command u into:

[0185] ;

[0186] Where r is the decoupled reference input vector.

[0187] For a radiant air conditioning system, if the relative gain matrix is:

[0188] ;

[0189] The decoupling matrix is ​​then:

[0190] .

[0191] like Figure 5As shown, it also includes dynamic coupling processing of humidity and water temperature, including measuring basic parameters, calculating the actual water vapor pressure E_w, determining the initial guess value Td0, calculating the saturated water vapor pressure E_w (Td0), iterative solution and accuracy verification.

[0192] Specifically, the steps include the following:

[0193] Based on indoor temperature and humidity, the dew point temperature of indoor air is calculated using an iterative algorithm.

[0194] The dew point temperature is calculated using the Goff-Grech formula and solved by Newton's iteration method. The iteration ends when the interpolation between the calculated saturated vapor pressure and the actual vapor pressure is less than a preset threshold (e.g., 0.001 hPa), and the dew point temperature value is obtained.

[0195] The set value of the radiant outlet water temperature is dynamically adjusted based on the calculated dew point temperature to prevent condensation on the radiant surface.

[0196] Indoor humidity deviation is used as one of the control inputs of a multivariate robust model to achieve coordinated control of temperature and humidity.

[0197] 1. Measure basic parameters:

[0198] a. Dry bulb temperature t (°C);

[0199] b. Relative humidity (RH) (%).

[0200] 2. Calculate the actual water vapor pressure E_w

[0201] 3. Determine the initial guess value Td0

[0202] If t > 0℃, use the water surface version of the formula:

[0203]

[0204] Where: Ew: saturated water vapor pressure at the water surface (hPa);

[0205] T: Absolute temperature (K), T = t + 273.15 (t is the temperature in Celsius);

[0206] T1 = 273.15 K: Standard freezing point temperature;

[0207] Coefficient parameters: c1=-5674.5359; c2=6.3925; c3=0.9678×10 -2 c4 = 0.6222 × 10 -6 c5 = 0.2075 × 10 -18 c6 = 0.9484 × 10 -2 ;c7=4.1630;c8=-5800.2206;c9=1.3915;c10 =-0.0486; c 11 =0.4176×10 -4 ;c 12 =-0.1445×10 -17 ;c 13 =6.5460.

[0208] If t≤0℃, use the ice surface version of the formula:

[0209]

[0210] Where: Ei: saturated water vapor pressure on the ice surface (hPa).

[0211] 4. Calculate the saturated vapor pressure E_w (Td0).

[0212] Using the Goff-Grech surface version of the formula:

[0213]

[0214] Where T1 = 273.16 K, and Td0 is the absolute temperature (K).

[0215] 5. Iterative solution:

[0216] If E s (Td0)≈E w The calculation is complete;

[0217] Otherwise, adjust the Td0 value and repeat step 5.

[0218] 6. Accuracy Verification

[0219] When|E s (Td)-E w The iteration ends when | < 0.001 hPa;

[0220] Convert to Celsius temperature: Td = Td(K) - 273.15.

[0221] S5. Based on the deviation between the set value and the actual feedback value of the radiant water temperature, the feedback controller calculates the feedback control quantity used to eliminate steady-state error.

[0222] S6. The feedforward control quantity and the feedback control quantity are superimposed to generate the final control command. The control command includes the opening value of the proportional electronic three-way mixing valve and the frequency value of the variable frequency water pump.

[0223] S7. According to the final control command, drive the actuator to adjust the opening degree of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump.

[0224] To verify the control effect of the feedforward regulation algorithm for water temperature in a radiant air conditioning system based on a multivariable robust model, and to compare it with the traditional PID control method, the following experiment was designed:

[0225] Experimental environment:

[0226] Standard test chamber (volume 15m×15m×3m)

[0227] Radiating terminals: Ceiling capillary network

[0228] Heat source: Ground source heat pump

[0229] Actuators: Proportional electronic three-way mixing valve, variable frequency water pump

[0230] Sensor configuration:

[0231] Indoor temperature and humidity sensor: Testo635 (accuracy ±0.2℃, sampling frequency 1Hz)

[0232] Inlet / outlet water temperature sensor at the radiant terminal: NTC thermistor (accuracy ±0.3℃, sampling frequency 1Hz)

[0233] Heat pump unit inlet / outlet water temperature sensor: PT100 platinum resistance thermometer (accuracy ±0.1℃, sampling frequency 1Hz)

[0234] Variable frequency water pump frequency sensor: Hall effect sensor (accuracy ±0.5Hz, sampling frequency 0.5Hz)

[0235] Experimental conditions:

[0236] Initial indoor temperature: 25℃

[0237] Initial indoor humidity: 50%RH

[0238] Radiation terminal type: Ceiling capillary network

[0239] Heat pump type: Ground source heat pump

[0240] Pump type: Variable frequency centrifugal pump

[0241] Adjustment target: Radiation outlet water temperature control accuracy ±0.3℃

[0242] Experimental methods:

[0243] Under the same experimental conditions, conventional PID control and the robust feedforward control of the present invention were tested respectively.

[0244] Inject a step load change (e.g., from 0 to 10 people, corresponding to a sudden increase in sensible heat load), lasting for 6 hours;

[0245] Record the system's response curve under a step load change;

[0246] Calculate the overshoot, settling time, and control accuracy for both control methods;

[0247] The energy consumption of the two control methods was measured, including the power consumption of the water pump and the input power of the heat pump.

[0248] Calculate the Overall Performance Factor (SCOP) and energy saving rate.

[0249] Experimental Data Acquisition and Processing

[0250] Data collection:

[0251] Temperature sensor: sampling frequency 1Hz, accuracy ±0.2℃

[0252] Water pump frequency sensor: sampling frequency 0.5Hz, accuracy ±0.5Hz

[0253] Load sensor: Sampling frequency 1Hz, accuracy ±5%

[0254] Data processing:

[0255] Overshoot calculation: For a step response, the overshoot is defined as:

[0256]

[0257] Among them, T peak In response to peak temperature, T set T0 is the initial temperature, used to set the temperature.

[0258] Adjustment time calculation: The time required for the system to enter the steady-state range (set value ± 0.3℃) from the start of a step disturbance.

[0259] Control accuracy calculation: The temperature fluctuation range within the steady-state interval, i.e.:

[0260] .

[0261] Energy consumption calculation:

[0262] a. Pump energy consumption: ,

[0263] in: η is the pump efficiency; ρ is the density of water; g is the acceleration due to gravity; Q(t) is the flow rate; H(t) is the head.

[0264] b. Heat pump energy consumption:

[0265] ;

[0266] ;

[0267] ;

[0268] T h This refers to the outlet water temperature of the heat pump.

[0269] c. Overall energy consumption:

[0270] d. Overall performance coefficient:

[0271] e. Energy saving rate:

[0272] The experimental results and comparative analysis are shown in Table 1: Table 1 ;

[0273] Analysis of experimental results:

[0274] 1. Comparison of control precision:

[0275] Traditional PID control: indoor temperature fluctuation range ±1~2℃, radiant water temperature control accuracy ±0.92℃;

[0276] The robust feedforward control of this invention provides: radiant water temperature control accuracy of ±0.31℃ and indoor temperature fluctuation range of ±0.3℃.

[0277] 2. Response speed comparison:

[0278] Traditional PID control: The response time for a step load change is 18.2 minutes;

[0279] The robust feedforward control of this invention reduces the response time to step load changes to approximately 3 minutes by 83.5%.

[0280] 3. Energy consumption comparison:

[0281] Traditional PID control: The total energy consumption is 1520W for water pump and 4560W for heat pump, for a total energy consumption of 6080W.

[0282] The robust feedforward control of this invention: The comprehensive energy consumption is 1280W for water pump energy consumption plus 3640W for heat pump energy consumption, for a total energy consumption of 4920W.

[0283] Energy saving effect: The overall energy efficiency ratio increased from 2.87 to 3.42, an increase of about 19.2%, with an energy saving rate of 21.6%.

[0284] 4. Robustness comparison:

[0285] Traditional PID control: Under conditions such as sudden load changes and fluctuations in heat pump efficiency, the system needs a relatively long time to recover stability;

[0286] The robust feedforward control of this invention enables the system to maintain stable control performance under disturbances and quickly recover to steady state.

[0287] Based on the above experimental results, the feedforward water temperature regulation algorithm for radiant air conditioning systems based on a multivariable robust model, as proposed in this invention, has significant advantages in control accuracy, response speed, and energy saving compared to traditional PID control. It can effectively solve the multivariable coupling and uncertainty problems in radiant air conditioning systems and provide a high-precision and highly robust control scheme for the system.

[0288] This water temperature feedforward regulation method constructs a multivariate robust thermal resistance model by integrating indoor load demand and system dynamic variables in real time. Through a dual self-correction mechanism and a collaborative control strategy, it feedforward adjusts the opening of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump to ensure that the radiant outlet water temperature is stable within the set temperature ±0.3℃ range. It is suitable for radiant air conditioning systems in residences, hotels, office buildings, etc.

[0289] like Figure 6 As shown, this invention provides a radiant air conditioning water temperature feedforward regulation system based on a multivariable robust model, used to implement the radiant air conditioning water temperature feedforward regulation method described above, including...

[0290] The sensor network is used to collect multi-source real-time data from the radiant air conditioning system. The sensor network includes indoor temperature sensors, indoor humidity sensors, heat pump unit outlet water temperature sensors, heat pump unit inlet water temperature sensors, radiant inlet water temperature sensors, and variable frequency water pump frequency sensors, which are used to collect system operating status data.

[0291] The data acquisition and preprocessing module connects to the sensor network and is used to filter and normalize multi-source real-time data to eliminate noise interference and provide high-quality data for subsequent model calculations.

[0292] A multivariate robust model, connected to the data acquisition and preprocessing module, is used to store and run a multivariate robust model describing the dynamic characteristics of the radiant air conditioning system, and to perform state estimation and prediction.

[0293] The feedforward compensation module connects to the multivariate robust model and is used to calculate the feedforward control quantities of the mixing valve opening and pump frequency to suppress disturbances based on the multivariate robust model and the current system state. It suppresses disturbances in system variables and errors in model establishment, and provides control commands to ensure that the system remains stable and reliable under model mismatch and external disturbances.

[0294] The feedback controller module is used to calculate the feedback control quantity to eliminate steady-state error based on the deviation between the set value and the actual feedback value of the radiant outlet water temperature; it realizes adaptive optimization, eliminates dynamic error, and dynamically corrects feedforward control deviation.

[0295] The control output calculation module is connected to the feedforward compensation module and the feedback controller module respectively. It is used to superimpose the feedforward control quantity and the feedback control quantity to generate the final control command.

[0296] The execution module, connected to the control output calculation module, is used to receive and execute control commands, and adjust the opening degree of the mixing valve and the frequency of the water pump in real time according to the commands of the control output calculation module.

[0297] This system addresses the problem of insufficient water temperature control accuracy in existing radiant air conditioning systems, meeting high comfort requirements; it also solves the problem of slow response speed in existing radiant air conditioning systems, shortening the temperature adjustment cycle from 15-30 minutes to 5 minutes, improving the system's adaptability to load changes; it addresses the problem of poor anti-interference capability in existing radiant air conditioning systems, enabling the system to effectively cope with interference factors such as heat pump unit efficiency fluctuations, pipe network resistance changes, and load prediction errors through multivariate robust model design; and it addresses the problem of improper handling of multivariate coupling effects in existing radiant air conditioning systems, achieving coordinated control of multiple variables such as indoor load changes, heat pump unit water temperature, radiant outlet water temperature, radiant return temperature, and water pump frequency through a multivariate robust model.

[0298] Multi-source real-time data includes indoor environmental parameters, heat pump unit operating parameters, radiant water intake operating parameters, and variable frequency water pump operating parameters.

[0299] It also includes a monitoring module for real-time monitoring of the radiant water temperature and determining whether it is stable within the set temperature ±0.3℃ range.

[0300] The execution module includes a proportional electronic three-way mixing valve and a variable frequency water pump.

[0301] The above are merely specific embodiments of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent substitutions, or modifications made based on the present invention to achieve substantially the same technical effect are all covered within the protection scope of the present invention.

Claims

1. A feedforward regulation method for radiative air conditioning water temperature based on a multivariable robust model, characterized in that... Includes the following steps: S1. Collect multi-source real-time data from the radiant air conditioning system; S2. Preprocess the multi-source real-time data to eliminate noise interference; S3. Construct a multivariate robust model of the dynamic characteristics of the radiant air conditioning system. The multivariate robust model is based on the state-space model and introduces a parameter uncertainty quantification method. S4. Based on the multivariate robust model and the current system state, calculate the feedforward control quantity for suppressing disturbances using a robust feedforward controller; S5. Based on the deviation between the set value and the actual feedback value of the radiant outlet water temperature, the feedback controller calculates the feedback control quantity used to eliminate steady-state error. S6. The feedforward control quantity and the feedback control quantity are superimposed to generate the final control command, which includes the opening value of the proportional electronic three-way mixing valve and the frequency value of the variable frequency water pump. S7. According to the final control command, drive the actuator to adjust the opening degree of the proportional electronic three-way mixing valve and the frequency of the variable frequency water pump.

2. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 1, characterized in that: The multi-source real-time data in step S1 includes indoor environmental parameters, heat pump host operating parameters, radiant water intake operating parameters, and variable frequency water pump operating parameters.

3. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 1, characterized in that: Step S3, constructing the multivariate robust model, specifically includes the following steps: S3.1 System variable definition; S3.2 Based on the radiant terminal heat balance equation, the variable frequency pump characteristic equation, and the mixing valve characteristic equation, linearization is performed to construct a state-space model of the radiant air conditioning system, whose equations are as follows: ； ； Where A is the system dynamic matrix; B1 is the control input matrix; B2 is the disturbance input matrix; C is the output matrix; D1 is the linear transfer matrix from control input to output; D2 is the linear transfer matrix from disturbance input to output; x is the state variable; u is the control input variable; d is the disturbance input variable; and y is the output variable. —State-space model; S3.3 Models the uncertainty factors as interval matrices or set matrices to form the multivariate robust model.

4. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 3, characterized in that: Step S3.1, defining system variables, specifically includes the following steps: S3.1a defines state variables, including at least radiant water temperature, indoor temperature, indoor humidity, and water pump frequency; S3.1b defines control input variables, including the opening degree of the proportional electronic three-way mixing valve and the frequency command of the variable frequency pump; S3.1c defines disturbance input variables, which include at least indoor load changes and heat pump unit outlet water temperature fluctuations; S3.1d defines output variables, including radiant water temperature, indoor temperature, and indoor humidity.

5. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 1, characterized in that: The robust feedforward controller mentioned in step S4 is an H∞ robust feedforward controller, and its design steps specifically include: S4.1 converts the state-space model of the radiant air conditioning system into the standard form of H∞ control, defining the controlled output vector z and the measurement output vector y; S4.2 By solving the algebraic Riccati equation, the state feedback gain matrix K and the feedforward gain matrix K are obtained. ff ; S4.3 Construct the controller structure as u = -Kx + K ff r, where u is the controller output, x is the state variable vector, and r is the reference input vector including the set temperature and set humidity; S4.4 determines the performance index γ of the radiant air conditioning system through frequency domain analysis and sensitivity optimization. The calculation formula is as follows: ； Where, σ max (G(jω)) is the maximum singular value of the open-loop transfer function matrix, and 1.2 is the safety factor.

6. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 1, characterized in that: It also includes dynamic coupling processing of humidity and water temperature, specifically including the following steps: Based on indoor temperature and humidity, the dew point temperature of indoor air is calculated using an iterative algorithm. The set value of the radiant water outlet temperature is dynamically adjusted based on the calculated dew point temperature. Indoor humidity deviation is used as one of the control inputs of a multivariate robust model to achieve coordinated control of temperature and humidity.

7. A method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 6, characterized in that: The dew point temperature is calculated using the Goff-Grech formula and solved by Newton's iteration method. The iteration ends when the interpolation between the calculated saturated vapor pressure and the actual vapor pressure is less than a preset threshold, and the dew point temperature value is obtained.

8. The method for feedforward regulation of radiative air conditioning water temperature based on a multivariable robust model according to claim 3, characterized in that: It also includes multivariable coupling and decoupling processing, specifically including the following steps: Based on the state-space model, calculate the steady-state gain G(0) of the system transfer function matrix. The relative gain matrix Φ is calculated based on the steady-state gain G(0) to quantify the coupling degree of the mixing valve opening and the pump frequency to output variables such as radiant outlet water temperature and indoor temperature. Design the decoupling matrix based on the relative gain matrix Φ. and transform the control command u into , where r is the decoupled reference input vector.

9. A radiant air conditioning water temperature feedforward regulation system based on a multivariable robust model, used to implement the radiant air conditioning water temperature feedforward regulation method as described in any one of claims 1 to 8, characterized in that: include Sensor networks are used to collect multi-source real-time data from radiant air conditioning systems. The data acquisition and preprocessing module is connected to the sensor network and is used to filter and normalize multi-source real-time data. A multivariate robust model, connected to the data acquisition and preprocessing module, is used to store and run a multivariate robust model describing the dynamic characteristics of the radiant air conditioning system, and to perform state estimation and prediction. The feedforward compensation module is connected to the multivariate robust model and is used to calculate the feedforward control quantities of the mixing valve opening and pump frequency to suppress disturbances based on the multivariate robust model and the current system state. The feedback controller module is used to calculate the feedback control quantity to eliminate steady-state error based on the deviation between the set value and the actual feedback value of the radiant outlet water temperature. The control output calculation module is connected to the feedforward compensation module and the feedback controller module respectively, and is used to superimpose the feedforward control quantity and the feedback control quantity to generate the final control command. The execution module, connected to the control output calculation module, is used to receive and execute the control commands.

10. A radiative air conditioning water temperature feedforward regulation system based on a multivariable robust model according to claim 9, characterized in that: The multi-source real-time data includes indoor environmental parameters, heat pump host operating parameters, radiant water intake operating parameters, and variable frequency water pump operating parameters.

11. A radiative air conditioning water temperature feedforward regulation system based on a multivariable robust model according to claim 9, characterized in that: It also includes a monitoring module for real-time monitoring of the radiant water temperature and determining whether it is stable within the set temperature ±0.3℃ range.

12. A radiative air conditioning water temperature feedforward regulation system based on a multivariable robust model according to claim 9, characterized in that: The execution module includes a proportional electronic three-way mixing valve and a variable frequency water pump.

Images