Multi-system collaborative building energy-saving reconstruction method

Abstract

The invention relates to a multi-system collaborative building energy-saving reconstruction method, which comprises the following steps of: establishing a Revit model for a building by using modeling software, importing building Revit model data into energy consumption analysis software Energylus through data conversion, and carrying out energy consumption simulation analysis on the building Revit model data; then, energy-saving transformation is conducted on the building from the three aspects of a building envelope, a lighting system and heating and cooling, the energy-saving effect of corresponding measures is analyzed by comparing data obtained through simulation before and after transformation, and data support is provided for energy-saving transformation of the building; an electric equivalent resistance-capacitance network model of the building is further established, the indoor temperature of the building is predicted and controlled, and the control effects and energy consumption saving conditions of the model prediction control method and the PID control method are compared; the method has the advantages of being high in energy-saving control precision and good in energy-saving effect in the whole process of modeling, simulation, transformation and control of building energy-saving transformation.

Description

Technical Field

[0001] The present invention belongs to the technical field of building energy conservation, and in particular relates to a method for multi-system coordinated building energy conservation transformation. Background Art

[0002] The core carrier of building energy consumption, such as libraries, has significantly higher energy consumption per unit area than ordinary teaching buildings (about 1.5-2.3 times). This is mainly attributed to three major structural defects: First, the thermal performance of the envelope structure is poor. The heat transfer coefficient of the exterior windows of traditional libraries generally exceeds 2.5W / (m 2 ·K), resulting in high heat loss in winter. Second, the air conditioning system is inefficient, with an average energy efficiency ratio of only 2.8, far below the national energy-saving standard (≥3.5), and the aging of cooling and heating equipment is prominent. Third, the lighting system is poorly managed, with illumination in reading areas seriously exceeding standards, while intelligent control coverage is insufficient and a high proportion of fluorescent lamps are used. These high energy consumption issues in libraries make energy-saving renovations imperative for university libraries.

[0003] Traditional building renovations rely on two-dimensional drawings and empirical parameters, and the data conversion errors between BIM models and energy consumption simulation tools are very large. In addition, renovation plans often focus on a single system. For example, the upgrade of the envelope structure is not linked to HVAC dynamic adjustment, and the air conditioning load is not optimized synchronously, resulting in a significant loss of energy-saving benefits. In addition, temperature control relies on fixed set value PID control, which does not take into account the periodic fluctuations in the flow of people. For example, during the exam season, the flow of people increases by 3 times, resulting in excessive cooling / heating and large fluctuations in room temperature control. Although existing studies have proposed sub-item renovation plans, they lack a full-chain integrated approach of "modeling-diagnosis-optimization-verification", making it difficult to achieve synergistic efficiency of multiple systems.

[0004] Patent application CN113204827B discloses a BIM-based building energy-saving design method and system. By analyzing the indoor air flow direction of existing buildings and setting corresponding cooling points, a continuous air flow path is formed in the building. This additional means improves the energy efficiency of existing buildings and reduces energy consumption. The system is applicable to new or renovated projects and has strong practicality. However, it does not take into account the dynamic energy consumption changes of buildings in various usage environments. Therefore, its efficiency in achieving preset energy-saving targets is low, which in turn affects the overall energy efficiency and lifespan of the building. Summary of the Invention

[0005] In order to overcome the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a method for multi-system collaborative energy-saving renovation of buildings, which uses modeling software to establish a Revit model of the building, and then imports the building Revit model data into the energy consumption analysis software Energyplus through data conversion, and performs energy consumption simulation analysis on it; then, energy-saving renovation of the building is carried out from three aspects: the envelope structure, lighting system, and heating and cooling. By comparing the data obtained from the simulation before and after the renovation, the energy-saving effects of the corresponding measures are analyzed to provide data support for the energy-saving renovation of the building; further, an electrical equivalent resistance-capacitance network model of the building is established to predict and control the indoor temperature of the building, and compare the control effects and energy savings of the two methods of model predictive control and PID control; the technical difficulties that need to be solved in the entire process of modeling, simulation, renovation and control of the building energy-saving renovation include: data compatibility and accuracy assurance between BIM models and energy consumption simulation tools; collaborative optimization and energy-saving renovation of multiple systems of envelope structure, lighting and HVAC; high-precision parameter identification and dynamic calibration of the electrical equivalent resistance-capacitance network model; algorithm design and energy efficiency verification of multi-time scale temperature control strategy, which has the advantages of high energy-saving control accuracy and good energy-saving effect.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A multi-system collaborative building energy-saving renovation method specifically includes the following steps:

[0008] Step 1: Use the modeling software Revit to create a Revit model of the building based on the building data parameters;

[0009] Step 2: Based on EnergyPlus, perform energy consumption simulation on the building Revit model created in Step 1 to obtain the building's energy consumption, including heating energy consumption, cooling energy consumption, lighting energy consumption, and energy consumption of other equipment;

[0010] Step 3: Based on the building's energy consumption data obtained in Step 2, use EnergyPlus to design an energy-saving renovation plan for the building;

[0011] Step 4: Establish an electrical equivalent resistance-capacitance network model of the building. Based on the energy consumption simulation data output from step 2 and the renovation plan from step 3, define new parameters of the electrical equivalent resistance-capacitance network model, set energy-saving constraints for the model predictive controller (MPC), identify the parameters of the RC equivalent circuit model, and use the model predictive controller (MPC) to predict and control the room temperature to reduce building energy consumption.

[0012] The specific method of step 2 is:

[0013] Import the building Revit model into Energyplus in gbXML / IDF format, and manually verify the envelope structure parameters including wall U-value and window-to-wall ratio; load the meteorological data of the building area; set dynamic parameters, including lighting density, equipment power schedule, occupancy density and HVAC system parameters; finally, Energyplus outputs the sub-item energy consumption data including heating energy consumption, cooling energy consumption, lighting energy consumption and equipment energy consumption.

[0014] The specific method of step 3 is: designing an energy-saving renovation plan for the building based on the sub-item energy consumption data output by EnergyPlus, including heating energy consumption, cooling energy consumption, lighting energy consumption, and equipment energy consumption;

[0015] 3.1 Carry out the renovation of the enclosure structure, including the use of wall insulation materials, setting the area and orientation of doors and windows, and using airtight structures for doors and windows;

[0016] 3.2 Intelligent transformation of lighting system

[0017] Light sensors monitor natural light intensity, and a dynamic light environment response system is used to drive the building's LED lighting system to automatically adjust fill light brightness, maintaining illumination in the reading space within a comfortable range of 300-500 lux. By monitoring the presence of people in target areas, lighting and sleep modes are switched to eliminate inefficient energy consumption. Smart socket clusters are deployed to analyze power consumption profiles of high-energy-consuming devices, including search terminals and electronic screens. The smart socket cluster supports remote start and stop, as well as abnormal power consumption warnings.

[0018] 3.3 Heating, Ventilation and Air Conditioning (HVAC) System Reconfiguration

[0019] Eliminate inefficient equipment; set air conditioning temperature and operating hours; set data for refrigeration, heat pump system and fan operation to improve the energy efficiency of the air conditioning system and reduce energy consumption.

[0020] The specific method of step 4 is:

[0021] 4.1 Establishing RC equivalent circuit model

[0022] The building thermal process is simplified into a circuit model: thermal resistance is analogous to resistors (R), and thermal capacitance is analogous to capacitors (C). Low-order structures including 1R1C or 3R2C are used to describe the heat transfer and storage process. Model inputs include outdoor temperature, solar radiation, and air conditioning power. All walls of the building are integrated and represented by a single wall with parameters of three resistors and two capacitors (3R2C). Similarly, all floors of the building are integrated and represented by a floor with parameters of three resistors and two capacitors (3R2C). All windows of the building are integrated and represented by a window with parameters of one resistor and one capacitor (1R1C). An electrically equivalent resistor-capacitor network model is established.

[0023] According to the above electrical equivalent resistance and capacitance network model, the node equations are constructed, among which the network node equations of the roof and the exterior wall are:

[0024]

[0025] The heat balance equation of the internal network nodes of the entire building is obtained from the law of conservation of energy:

[0026]

[0027] in,

[0028] Where, T out 、T d 、T in Respectively represent outdoor temperature, floor temperature and indoor temperature, the unit is ℃; T e2 、T e4 、T d2 、T d4 Respectively represent the outer surface temperature of the wall, the inner surface temperature of the wall, the outer surface temperature of the floor, and the inner surface temperature of the floor, the unit is ℃; U e1 、U e2 、U e3 They represent the concentrated convection coefficient of the outer surface of the wall, the thermal conductivity of the wall, and the concentrated convection coefficient of the inner surface of the wall, respectively, with the unit being W / K; U g1 、U g2 、U g3 Respectively represent the concentrated convection coefficient of the outer surface of the floor, the thermal conductivity of the floor, and the concentrated convection coefficient of the inner surface of the floor, the unit is W / K; U w Represents the thermal conductivity of the window, in W / K; C e1 、C e2 Represents the heat capacity of the wall, in J / K; C g1 、C g2 Represents the heat capacity of the floor, in J / K; C z represents the heat capacity of the air in the area, in J / K. ae represents the ratio of the wall area to the total area; ag represents the ratio of the floor area to the total area; Q solt Represents the amount of heat radiated by the sun through the window, in W; Q solte , Q soltd Respectively represent the heat gained by the wall and floor through solar radiation from the windows, in W; Q grad Represents the heat radiated by internal heat gain, in W; Q grade , Q gradd Respectively represent the heat gained by the wall and floor through internal heat gain radiation, the unit is W; Q soleRepresents the heat gained by the wall through solar radiation, unit is W; Q sen Represents the control input quantity, the unit is W; Q gconv represents the internal convection heat gain, in W;

[0029] 4.2 Parameter identification

[0030] Based on measured data, the least squares method or recursive least squares (RLS) method is used to identify the parameters (R, C) of the electrical equivalent RC network model to ensure that the dynamic response of the electrical equivalent RC network model is consistent with the actual thermal behavior;

[0031] 4.3 Real-time Optimization and Control Based on MPC Controller

[0032] Based on the electrical equivalent resistance-capacitance network model identified in step 4.2, an objective function for minimizing energy consumption is constructed, and constraints including the room temperature comfort zone and equipment power limit are set. At the same time, weather forecasts are introduced as feedforward disturbance variables, and the control strategy is optimized in real time through the model predictive controller (MPC) to achieve precise control of room temperature and reduce energy consumption.

[0033] The specific process of step 4.2 is as follows:

[0034] 4.2.1 Data Collection and Preparation

[0035] Input vector: Collect various data related to indoor temperature, including outdoor temperature, solar radiation intensity, and air conditioning set value, as the input vector of the electrical equivalent resistance-capacitance network model;

[0036] Output vector: Collect the actual measured indoor temperature and humidity data as the output vector of the electrical equivalent resistance-capacitance network model;

[0037] 4.2.2 Parameter Estimation

[0038] Objective: Use the least squares method or recursive least squares (RLS) method to estimate the value of the unknown parameter θ in the electrical equivalent resistor-capacitor network model so that the indoor temperature T predicted by the model is θ (t) is as close as possible to the actual measured indoor temperature T in ;

[0039] Constraints: Assume that the parameter θ has a value range of θ ub <θ<θ ib , that is, the parameter must be between the given upper and lower bounds;

[0040] The parameters θ are solved by minimizing the squared difference between the actual measured values ​​and the predicted values:

[0041]

[0042] Among them, T inis the actual measured indoor temperature at time t, T θ (t) is the indoor temperature predicted by the parameters θ and input data at the same time point;

[0043] 4.2.3 Parameter Fitting

[0044] Use the lsqcurvefit function to perform parameter fitting on the objective function and further optimize the parameter θ;

[0045] x=lsqcurvefit(fun,x0,x data ,y data ,lb,ub,options) (4)

[0046] in,

[0047] fun is the objective function, i.e., the error square sum function in formula (3);

[0048] x0 is the initial estimate of the parameter θ;

[0049] x data is the data of the input vector;

[0050] y data is the data of the output vector;

[0051] lb, ub are the lower and upper bounds of the parameter θ;

[0052] options are optimization options used to control the optimization process.

[0053] The specific process of step 4.3 is as follows:

[0054] 4.3.1 Prediction stage:

[0055] Input data: current building status including room temperature, equipment operating status, and future meteorological data including predicted temperature and solar radiation intensity;

[0056] Prediction method: The current building status and future meteorological data are input into the electrical equivalent resistance and capacitance network model. The electrical equivalent resistance and capacitance network model predicts and outputs the room temperature change trend and total energy consumption in the future time period based on the input data and parameter θ.

[0057] 4.3.2 Optimization phase:

[0058] Constrain room temperature range, equipment power limits and energy-saving targets to minimize total energy consumption;

[0059] Optimization method: Under the premise of meeting the above constraints, the model predictive controller (MPC) is used to control the refrigeration equipment according to the control action of the first time step each time to minimize the total energy consumption;

[0060] 4.3.3 Feedback correction stage:

[0061] Feedback data: actual measured indoor temperature and equipment operating status;

[0062] Correction method: Based on the feedback data, at the next time step, steps 4.3.1 to 4.3.2 are re-executed using the model predictive controller (MPC) based on the actual measurement data. Compared with the prior art, the present invention has the following advantages:

[0063] 1. Steps 1 to 3 of the present invention are seamlessly connected through the gbXML format of Revit and EnergyPlus (error <5%). In step 4, a correlation matrix between the thermal parameters of the envelope structure and the HVAC energy consumption is constructed to achieve full-chain data drive from geometric modeling to energy consumption diagnosis.

[0064] 2. In step 3 of the present invention, the optimal renovation plan is selected through multi-objective coordinated energy-saving renovation of the building envelope, HVAC, and lighting. The total energy consumption of the building is saved by 20.3277363 MWh, and the energy saving percentage is as high as 36.41%, achieving an effective improvement in the energy saving rate.

[0065] 3. Step 4 of the present invention is based on the dynamic parameter identification of the 3R2C equivalent model, combined with the real-time correction of the prediction deviation by Kalman filtering. In the month of June, the predicted room temperature results are as follows: the maximum error value of the Kalman filter prediction is 4.9°C, and the root mean square error is 1.783548; the maximum error value of the MPC prediction is 2.4°C, and the root mean square error is 0.545133.

[0066] 4. Compared with the KF filter prediction, the MPC prediction of the indoor temperature in step 4 of the present invention has a smaller root mean square error, indicating that the MPC prediction of the room temperature has higher accuracy. Secondly, in step 4 of the present invention, an MPC controller is designed to control the room temperature based on the MPC control principle. The results show that under MPC control, the room temperature fluctuates less, consumes less power, and is more energy-efficient than the PID control strategy, further enhancing the energy-saving effect.

[0067] In summary, the present invention uses modeling software to establish a Revit model of the building, and then imports the building Revit model data into the energy consumption analysis software Energyplus through data conversion, and performs energy consumption simulation analysis on it; then, energy-saving transformation of the building is carried out from three aspects: the envelope structure, lighting system, and heating and cooling; high-precision parameter identification and dynamic calibration of the electrical equivalent resistance-capacitance network model; algorithm design and energy efficiency verification of the multi-time scale temperature control strategy; it has the advantages of high energy-saving control precision and good energy-saving effect.

[0068] To develop an efficient energy-saving renovation plan, a building in the example was modeled in Revit 3D, and its pre-renovation energy consumption was simulated using DesignBuilder energy consumption simulation software. An energy-saving renovation plan was then proposed and the energy consumption of the building after the renovation was simulated to analyze the feasibility and energy-saving effects of the renovation plan. Compared to existing technologies, this invention achieves energy efficiency improvements of over 30%, suppressing fluctuations by ±1.3°C, and covering the entire lifecycle from design modeling to operation and maintenance control through the dual innovations of "BIM-EnergyPlus-RC model closed-loop calibration" and "multi-timescale MPC control." BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a Revit model using a certain building as an example.

[0070] Figure 2 4 is a data conversion flow chart of an embodiment.

[0071] Figure 3 The following is a comparison of the energy consumption of each item throughout the year of the embodiment (before transformation).

[0072] Figure 4 2 is a comparison of energy consumption before and after the comprehensive renovation of the building in the embodiment.

[0073] Figure 5 Schematic diagram of the building thermal dynamic RC model of the embodiment.

[0074] Figure 6 This is the flowchart of the least squares method.

[0075] Figure 7 The outdoor meteorological data of the building in the embodiment from June 1 to June 30; wherein, (a) is the solar radiation heat and the heat gain data inside the building, and (b) is the outdoor temperature data.

[0076] Figure 8 The indoor temperature under the control of the MPC of the embodiment from June 1 to June 30.

[0077] Figure 9 The energy consumption required for cooling under PID and MPC control from June 1 to June 30 of the embodiment.

[0078] Figure 10 Comparison of building energy consumption after adding model predictive control. DETAILED DESCRIPTION

[0079] The present invention will be further described in detail below with reference to the accompanying drawings and a certain architectural embodiment.

[0080] The annual energy consumption values ​​of each item of a building in a certain building embodiment (before renovation) are shown in the following table:

[0081]

[0082] A multi-system collaborative building energy-saving renovation method specifically includes the following steps:

[0083] Step 1: Use the modeling software Revit to create a Revit model based on the building data parameters. Figure 1 ;

[0084] Step 2: Based on EnergyPlus, perform energy consumption simulation on the building Revit model created in step 1, see Figure 2 , obtain the energy consumption of the building, including heating energy consumption, cooling energy consumption, lighting energy consumption and energy consumption of other equipment;

[0085] The specific method of step 2 is:

[0086] Import the building Revit model into Energyplus in gbXML / IDF format, manually verify the building envelope parameters including wall U value and window-to-wall ratio; load the meteorological data of the building area; set dynamic parameters including lighting density, equipment power schedule, occupancy density and HVAC system parameters; finally, Energyplus outputs the energy consumption data including heating energy consumption, cooling energy consumption, lighting energy consumption and equipment energy consumption. Figure 3 The specific parameters of the building's annual energy consumption are shown in the following table:

[0087]

[0088]

[0089] Step 3: Based on the building's energy consumption data obtained in Step 2, use EnergyPlus to design an energy-saving renovation plan for the building;

[0090] The specific method of step 3 is: based on the energy consumption data output by EnergyPlus, including heating energy consumption, cooling energy consumption, lighting energy consumption and equipment energy consumption,

[0091] (1) Carry out the renovation of the enclosure structure, including the use of wall insulation materials, setting the area and orientation of doors and windows, and using airtight structures for doors and windows;

[0092] (2) Intelligent transformation of lighting system

[0093] Light sensors monitor natural light intensity, and a dynamic light environment response system is used to drive the building's LED lighting system to automatically adjust fill light brightness, maintaining illumination in the reading space within a comfortable range of 300-500 lux. By monitoring the presence of people in target areas, lighting and sleep modes are switched to eliminate inefficient energy consumption. Smart socket clusters are deployed to analyze power consumption profiles of high-energy-consuming devices, including search terminals and electronic screens. The smart socket cluster supports remote start and stop, as well as abnormal power consumption warnings.

[0094] (3) Heating, ventilation and air conditioning (HVAC) system reconstruction

[0095] Eliminate inefficient equipment; set air conditioning temperature and operating hours; set data for refrigeration, heat pump system and fan operation to improve the energy efficiency of the air conditioning system and reduce energy consumption.

[0096] Step 4: Establish the electrical equivalent resistance and capacitance network model of the building. Based on the updated thermal performance target output from step 2 and the renovation plan from step 3, define the new parameters of the electrical equivalent resistance and capacitance network model, set energy-saving constraints for the model predictive controller (MPC), identify the parameters of the RC equivalent circuit model, and use the model predictive controller (MPC) to predict and control the room temperature to reduce the building's energy consumption. Figure 4 shown.

[0097] The specific method of step 4 is:

[0098] 4.1 Establishing RC equivalent circuit model

[0099] The building thermal process is simplified into a circuit model: thermal resistance is analogous to resistors (R), thermal capacitance is analogous to capacitors (C), and the heat transfer and storage process is described by low-order structures (such as 1R1C or 3R2C). The model input includes outdoor temperature, solar radiation and air conditioning power; all walls of the building are integrated and represented by a single wall with parameters of three resistors and two capacitors (3R2C); all floors of the building are integrated and represented by a floor with parameters of three resistors and two capacitors (3R2C); all windows of the building are integrated and represented by a window with parameters of one resistor and one capacitor (1R1C); an electrically equivalent resistor-capacitor network model is established; the dynamic thermal model of the room is as follows: Figure 5 shown.

[0100] According to the above electrical equivalent resistance and capacitance network model, the node equations are constructed, among which the network node equations of the roof and the exterior wall are:

[0101]

[0102] The heat balance equation of the internal network nodes of the entire building is obtained from the law of conservation of energy:

[0103]

[0104] in,

[0105] Where, T out 、T d 、T in Respectively represent outdoor temperature, floor temperature and indoor temperature, the unit is ℃; T e2 、T e4 、T d2 、T d4 Respectively represent the outer surface temperature of the wall, the inner surface temperature of the wall, the outer surface temperature of the floor, and the inner surface temperature of the floor, the unit is ℃; U e1 、U e2 、U e3 They represent the concentrated convection coefficient of the outer surface of the wall, the thermal conductivity of the wall, and the concentrated convection coefficient of the inner surface of the wall, respectively, with the unit being W / K; U g1 、U g2 、U g3 Respectively represent the concentrated convection coefficient of the outer surface of the floor, the thermal conductivity of the floor, and the concentrated convection coefficient of the inner surface of the floor, the unit is W / K; U w Represents the thermal conductivity of the window, in W / K; C e1 、C e2 Represents the heat capacity of the wall, in J / K; C g1 、C g2 Represents the heat capacity of the floor, in J / K; C z represents the heat capacity of the air in the area, in J / K. ae represents the ratio of the wall area to the total area; ag represents the ratio of the floor area to the total area; Q solt Represents the amount of heat radiated by the sun through the window, in W; Q solte , Q soltd Respectively represent the heat gained by the wall and floor through solar radiation from the windows, in W; Q grad Represents the heat radiated by internal heat gain, in W; Q grade , Q gradd Respectively represent the heat gained by the wall and floor through internal heat gain radiation, the unit is W; Q sole Represents the heat gained by the wall through solar radiation, unit is W; Q sen Represents the control input quantity, the unit is W; Q gconv represents the internal convection heat gain, in W;

[0106] 4.2 Parameter identification

[0107] Based on measured data (such as indoor and outdoor temperatures and energy consumption), the least squares method or recursive least squares (RLS) method is used to identify the parameters (R, C) of the electrical equivalent resistance and capacitance network model to ensure that the dynamic response of the electrical equivalent resistance and capacitance network model is consistent with the actual thermal behavior.

[0108] The specific process is:

[0109] 4.2.1 Data Collection and Preparation

[0110] Input vector: Collect various data related to indoor temperature, including outdoor temperature, solar radiation intensity, and air conditioning set value, as the input vector of the electrical equivalent resistance-capacitance network model.

[0111] Output vector: Collect the actual measured indoor temperature and humidity data as the output vector of the electrical equivalent resistance-capacitance network model;

[0112] 4.2.2 Parameter Estimation

[0113] Objective: Use recursive least squares (RLS) to estimate the value of parameter θ so that the indoor temperature T predicted by the model is θ (t) is as close as possible to the actual measured indoor temperature T in ; The least squares method flow chart is as follows Figure 6 shown.

[0114] Constraints: Assume that the parameter θ has a value range of θ ub <θ<θ ib , that is, the parameter must be between the given upper and lower bounds;

[0115] The initial values ​​of the parameters θ can be obtained by analyzing the building structure and materials. The initial values ​​of the parameters are:

[0116] U e1 =594.9630;U e2 =11.3452; U e3 =102.2593;U g1 =2.9970e+04;

[0117] U g2 =33.8040; U g3 =330.0000;U w =10.0800;C e1 =23353040;

[0118] C e2 =23353040;C g1 =3.1966e+07; C g2 =3.1966e+07

[0119] The architectural model optimizes the function by iterating the objective function within its specified input and output range according to the obtained original parameters, thereby obtaining the best optimization model for predictive control.

[0120] The final parameters after optimization are:

[0121] U e1 =594.9630;U e2 =4.85414377359824; U e3 =102.259259341067;U g1 =899145.928386081;U g2 =216.701187275775;U g3 =328.709182451339;U w =2.52002017971113; C e1 =1167652;C e2 =22589658.0138795;C g1 =2329432.93322769;C g2 =152291092.790164

[0122] Based on the meteorological parameters of the Shaw Building in January, the changes in indoor temperature and solar heat gain, the parameters were identified. The identified parameters were substituted into the discretized state space equation to obtain the discretized matrix as follows:

[0123]

[0124] C d =

[00001] T ;D d =0.

[0125] The parameters θ are solved by minimizing the squared difference between the actual measured values ​​and the predicted values:

[0126]

[0127] Among them, T in is the actual measured indoor temperature at time t, T θ (t) is the indoor temperature predicted by the parameters θ and input data at the same time point;

[0128] 4.2.3 Parameter fitting

[0129] Use the lsqcurvefit function to perform parameter fitting on the objective function and further optimize the parameter θ;

[0130] x=lsqcurvefit(fun,x0,x data ,y data ,lb,ub,options) (4)

[0131] in,

[0132] fun is the objective function, i.e., the error square sum function in formula (3);

[0133] x0 is the initial estimate of the parameter θ;

[0134] x data is the data of the input vector (such as outdoor temperature, solar radiation, air conditioning setting value, etc.);

[0135] y data is the data of the output vector (such as the actual measured indoor temperature);

[0136] lb, ub are the lower and upper bounds of the parameter θ;

[0137] options are optimization options that control the details of the optimization process.

[0138] 4.3 Real-time Optimization and Control Based on MPC Controller

[0139] Based on the electrical equivalent resistance-capacitance network model identified in step 4.2, an objective function for minimizing energy consumption is constructed. Constraints are set, including the room temperature comfort zone and the equipment power limit. Weather forecasts are introduced as feedforward disturbance variables. The MPC controller optimizes the control strategy in real time to achieve precise room temperature control and reduce energy consumption.

[0140] Specific process:

[0141] 4.3.1 Prediction stage:

[0142] Input data: current building status including room temperature, equipment operating status, and future meteorological data including predicted temperature and solar radiation intensity;

[0143] Prediction method: The current building status and future meteorological data are input into the electrical equivalent resistance and capacitance network model. The electrical equivalent resistance and capacitance network model predicts and outputs the room temperature change trend in the future time period based on the input data and parameter θ;

[0144] 4.3.2 Optimization phase:

[0145] Objective function: minimize total energy consumption;

[0146] Constraints: room temperature comfort zone (e.g., room temperature maintained at 24±1°C), equipment power limit, and energy-saving targets;

[0147] Optimization method: Under the premise of meeting the above comfort constraints, the MPC controller is used to control the refrigeration equipment to operate only according to the control action of the first time step each time to minimize the total energy consumption;

[0148] 4.3.3 Feedback correction stage:

[0149] Feedback data: actual measured indoor temperature and equipment operating status;

[0150] Correction method: Based on the feedback data, at the next time step, re-execute steps 4.3.1 to 4.3.2 according to the actual measurement data through the MPC controller to ensure the control effect.

[0151] To better demonstrate the effectiveness of MPC control, we compared the room temperature control results for a building over time periods of one day, one week, and one month. The results showed that MPC control demonstrated superior prediction and control performance compared to the PID control strategy in each of these scenarios. This is illustrated using a single month as an example.

[0152] The month of June was selected to analyze the room temperature and cooling energy consumption of the building under the two control strategies of PID and MPC. Figure 7 As shown in the figure, the room temperature and cooling energy consumption under each control strategy are as follows: Figure 8 and Figure 9 shown.

[0153] from Figure 8 As can be seen, during the month of June, the room temperature under natural conditions fluctuated around 34°C, with a minimum of 25.08°C and a maximum of 39.7°C, making the indoor temperature relatively hot. Under PID control, the indoor temperature fluctuated around 24°C. During this month, the maximum temperature under PID control reached 24.88°C, with a deviation of +0.88°C from the target temperature of 24°C. The minimum temperature under PID control was 23.09°C, with a deviation of -0.91°C. Under MPC control, however, the indoor temperature was well maintained at the target temperature of 24°C, with almost no fluctuation. Overall, the room temperature under both control strategies exhibited minimal fluctuations, effectively fulfilling their function of controlling the indoor temperature. Compared to PID control, MPC control maintained the indoor temperature at the target temperature of 24°C, demonstrating superior control effectiveness.

[0154] from Figure 9As can be seen, without a control strategy, the building's cooling energy consumption for the month of June 1st to June 30th was 998.84 kWh. During this month, the maximum daily energy consumption coincided with the highest indoor temperature of each day, indicating significant indoor temperature fluctuations. However, under PID control, building cooling energy consumption decreased to 934.56 kWh, with energy savings of 64.28 kWh. Under MPC control, building cooling energy consumption decreased to 836.85 kWh, with energy savings of 161.99 kWh. Both control strategies reduced energy consumption compared to the uncontrolled heating and cooling energy consumption. MPC control required less heating and cooling energy than PID control, demonstrating greater energy savings.

[0155] The overall experimental results show that the room temperature under the PID and MPC control strategies does not fluctuate much, and both can well meet the function of controlling the indoor temperature. However, the room temperature under MPC control has almost no fluctuation, and the control effect is better. Figure 10 As shown. Figure 10 As can be seen from the data, the MPC control strategy consumes less electricity under both MPC and PID control strategies. In the simulation data for June 3rd, the MPC control strategy saved 5.64 kWh, a 17.57% energy saving. In the simulation data for January 3rd, the MPC control strategy saved 6.41 kWh, a 18.54% energy saving. In the simulation data from June 1st to June 7th, the MPC control strategy saved 34.8 kWh, a 17.06% energy saving. In the simulation data from January 1st to January 7th, the MPC control strategy saved 43.56 kWh, a 17.30% energy saving. In the simulation data for June, the MPC control strategy saved 161.99 kWh, a 16.21% energy saving. In the simulation data for January, the MPC control strategy saved 166.25 kWh, a 15.79% energy saving. This reflects the superiority of MPC model predictive control in controlling building room temperature and has a positive effect on reducing building energy consumption.

[0156] The comprehensive energy-saving effects of the energy-saving measures in the embodiment are shown in the following table:

[0157]

Claims

1. A multi-system collaborative building energy-saving transformation method, characterized in that: The specific steps include: Step 1: Use the modeling software Revit to create a Revit model of the building based on the building data parameters; Step 2: Based on EnergyPlus, perform energy consumption simulation on the building Revit model created in Step 1 to obtain the building's energy consumption, including heating energy consumption, cooling energy consumption, lighting energy consumption, and energy consumption of other equipment; Step 3: Based on the building's energy consumption data obtained in Step 2, use EnergyPlus to design an energy-saving renovation plan for the building; Step 4: Establish an electrical equivalent resistance-capacitance network model of the building. Based on the energy consumption simulation data output from step 2 and the renovation plan from step 3, define new parameters of the electrical equivalent resistance-capacitance network model, set energy-saving constraints for the model predictive controller (MPC), identify the parameters of the RC equivalent circuit model, and use the model predictive controller (MPC) to predict and control the room temperature to reduce building energy consumption.

2. The method for multi-system coordinated building energy-saving transformation according to claim 1, characterized in that: The specific method of step 2 is: Import the building Revit model into Energyplus in gbXML / IDF format, and manually verify the envelope structure parameters including wall U-value and window-to-wall ratio; load the meteorological data of the building area; set dynamic parameters, including lighting density, equipment power schedule, occupancy density and HVAC system parameters; finally, Energyplus outputs the sub-item energy consumption data including heating energy consumption, cooling energy consumption, lighting energy consumption and equipment energy consumption.

3. The method for multi-system coordinated building energy-saving transformation according to claim 1, characterized in that: The specific method of step 3 is: designing an energy-saving renovation plan for the building based on the sub-item energy consumption data output by EnergyPlus, including heating energy consumption, cooling energy consumption, lighting energy consumption, and equipment energy consumption; 3.1 Carry out the renovation of the enclosure structure, including the use of wall insulation materials, setting the area and orientation of doors and windows, and using airtight structures for doors and windows; 3.2 Intelligent transformation of lighting system By monitoring the intensity of natural light through light sensors and utilizing a dynamic response system to the light environment, the building's LED lighting system is driven to automatically adjust the fill light brightness, maintaining the illumination in the reading space within a comfortable range of 300-500 lux. By monitoring the presence of people in target areas, switching lighting and sleep modes, and eliminating ineffective energy consumption, smart socket clusters are deployed to analyze power usage of high-energy-consuming devices, including search terminals and electronic screens. Smart socket clusters support remote start and stop and abnormal power consumption warnings. 3.3 Heating, Ventilation and Air Conditioning (HVAC) System Reconfiguration Eliminate inefficient equipment; set air conditioning temperature and operating hours; set data for refrigeration, heat pump system and fan operation to improve the energy efficiency of the air conditioning system and reduce energy consumption.

4. The method for multi-system coordinated building energy-saving transformation according to claim 1, characterized in that: The specific method of step 4 is: 4.1 Establishing RC equivalent circuit model The building thermal process is simplified into a circuit model: thermal resistance is analogous to resistors (R), and thermal capacity is analogous to capacitors (C). Low-order structures such as 1R1C or 3R2C are used to describe the heat transfer and storage process. Model inputs include outdoor temperature, solar radiation, and air conditioning power. All walls of the building are integrated and represented by a single wall with parameters of three resistors and two capacitors (3R2C). Similarly, all floors of the building are integrated and represented by a floor with parameters of three resistors and two capacitors (3R2C). Integrate all the windows of the building and represent them through a window whose parameters are one resistor and one capacitor (1R1C); establish an electrical equivalent resistor-capacitor network model; According to the above electrical equivalent resistance and capacitance network model, the node equations are constructed, among which the network node equations of the roof and the exterior wall are: The heat balance equation of the internal network nodes of the entire building is obtained from the law of conservation of energy: in, Where, T out 、T d 、T in Respectively represent outdoor temperature, floor temperature and indoor temperature, the unit is ℃; T e2 、T e4 、T d2 、T d4 Respectively represent the outer surface temperature of the wall, the inner surface temperature of the wall, the outer surface temperature of the floor, and the inner surface temperature of the floor, the unit is ℃; U e1 、U e2 、U e3 They represent the concentrated convection coefficient of the outer surface of the wall, the thermal conductivity of the wall, and the concentrated convection coefficient of the inner surface of the wall, respectively, with the unit being W / K; U g1 、U g2 、U g3 Respectively represent the concentrated convection coefficient of the outer surface of the floor, the thermal conductivity of the floor, and the concentrated convection coefficient of the inner surface of the floor, the unit is W / K; U w Represents the thermal conductivity of the window, in W / K; C e1 、C e2 Represents the heat capacity of the wall, in J / K; C g1 、C g2 Represents the heat capacity of the floor, in J / K; C z represents the heat capacity of the air in the area, in J / K. ae represents the ratio of the wall area to the total area; ag represents the ratio of the floor area to the total area; Q solt Represents the amount of heat radiated by the sun through the window, in W; Q solte , Q soltd Respectively represent the heat gained by the wall and floor through solar radiation from the windows, in W; Q grad Represents the heat radiated by internal heat gain, in W; Q grade , Q gradd Respectively represent the heat gained by the wall and floor through internal heat gain radiation, the unit is W; Q sole Represents the heat gained by the wall through solar radiation, unit is W; Q sen Represents the control input quantity, the unit is W; Q gconv represents the internal convection heat gain, in W; 4.2 Parameter identification Based on measured data, the least squares method or recursive least squares (RLS) method is used to identify the parameters (R, C) of the electrical equivalent RC network model to ensure that the dynamic response of the electrical equivalent RC network model is consistent with the actual thermal behavior; 4.3 Real-time Optimization and Control Based on MPC Controller Based on the electrical equivalent resistance-capacitance network model identified in step 4.2, an objective function for minimizing energy consumption is constructed, and constraints including the room temperature comfort zone and equipment power limit are set. At the same time, weather forecasts are introduced as feedforward disturbance variables, and the control strategy is optimized in real time through the model predictive controller (MPC) to achieve precise control of room temperature and reduce energy consumption.

5. The method for multi-system coordinated building energy-saving transformation according to claim 5, characterized in that: The specific process of step 4.2 is as follows: 4.2.1 Data Collection and Preparation Input vector: Collect various data related to indoor temperature, including outdoor temperature, solar radiation intensity, and air conditioning set value, as the input vector of the electrical equivalent resistance-capacitance network model; Output vector: Collect the actual measured indoor temperature and humidity data as the output vector of the electrical equivalent resistance-capacitance network model; 4.2.2 Parameter Estimation Objective: Use the least squares method or recursive least squares (RLS) method to estimate the value of the unknown parameter θ in the electrical equivalent resistor-capacitor network model so that the indoor temperature T predicted by the model is θ (t) is as close as possible to the actual measured indoor temperature T in ; Constraints: Assume that the parameter θ has a value range of θ ub <θ<θ ib , that is, the parameter must be between the given upper and lower bounds; The parameters θ are solved by minimizing the squared difference between the actual measured values ​​and the predicted values: Among them, T in is the actual measured indoor temperature at time t, T θ (t) is the indoor temperature predicted by the parameters θ and input data at the same time point; 4.2.3 Parameter Fitting Use the lsqcurvefit function to perform parameter fitting on the objective function and further optimize the parameter θ; x=lsqcurvefit(fun,x0,x data ,y data ,lb,ub,options) (4) in, fun is the objective function, i.e., the error square sum function in formula (3); x0 is the initial estimate of the parameter θ; x data is the data of the input vector; y data is the data of the output vector; lb, ub are the lower and upper bounds of the parameter θ; options are optimization options used to control the optimization process.

6. The method for multi-system coordinated building energy-saving transformation according to claim 5, characterized in that: The specific process of step 4.3 is as follows: 4.3.1 Prediction stage: Input data: current building status including room temperature, equipment operating status, and future meteorological data including predicted temperature and solar radiation intensity; Prediction method: The current building status and future meteorological data are input into the electrical equivalent resistance and capacitance network model. The electrical equivalent resistance and capacitance network model predicts and outputs the room temperature change trend and total energy consumption in the future time period based on the input data and parameter θ. 4.3.2 Optimization phase: Constrain room temperature range, equipment power limits and energy-saving targets to minimize total energy consumption; Optimization method: Under the premise of meeting the above constraints, the model predictive controller (MPC) is used to control the refrigeration equipment according to the control action of the first time step each time to minimize the total energy consumption; 4.3.3 Feedback correction stage: Feedback data: actual measured indoor temperature and equipment operating status; Correction method: Based on the feedback data, at the next time step, re-execute steps 4.3.1 to 4.3.2 according to the actual measurement data through the model predictive controller (MPC).

Images