Hardware-in-the-loop simulation control method and management platform for building air-conditioning systems based on PLM system

Abstract

The present invention discloses a semi-physical simulation control method and management platform for a building air-conditioning system based on a PLM system. The management platform includes: a simulation optimization module for establishing a system simulation model and obtaining the system's equipment control optimization parameters based on the system's real-time operating data; a system equipment module for obtaining the equipment attributes and operating parameters of the building's air-conditioning system equipment and performing equipment control; a database module including a building information database, an equipment operation database, and a simulation optimization database; and a visualization module for dynamically displaying the building BIM model, measured data, and simulation optimization data in the building air-conditioning system based on the data in the database module. The semi-physical simulation control and management platform for a building air-conditioning system based on a full lifecycle management PLM system proposed by the present invention ensures that the entire system is always operating in an optimal state, achieving optimal control of the building air-conditioning system.

Description

Technical Field

[0001] The present invention relates to energy-saving control technology, and in particular to a PLM system-based semi-physical simulation control management method and platform for building air-conditioning systems. Background Art

[0002] In public buildings, the operating energy consumption of the HVAC system accounts for 30% to 60% of the total building energy consumption. Reducing the operating energy consumption of HVAC is the key to reducing the total energy consumption of the building. Accurate and reasonable operation and maintenance adjustment and optimization control of the HVAC system are a strong guarantee for reducing the operating energy consumption of equipment, and are also a major test faced by operation and maintenance management.

[0003] With the development of cloud computing, the improvement and widespread application of AI computing power, problems such as inaccurate prediction results and slow optimization speed in the use of model prediction-based optimization control have been effectively solved. The feasibility of deploying model prediction-based optimization control for the entire building has also been greatly improved.

[0004] Existing management platforms, such as the patent (publication number: CN108281176A) that announced a BIM-based hospital building intelligent operation and maintenance management system and method, and the patent (publication number: CN115563688A) that announced a BIM model in a BIM-based operation and maintenance management system, do not consider combining the system's measured data to dynamically and in real time display the equipment's operating status. It only serves as a static display, which is insufficient to bring into play the full role of the BIM model. At the same time, it does not consider expanding the model's attributes, and the model has no logical meaning. In addition, in terms of control strategy, it does not consider making full use of the building air-conditioning system's historical operating data to train the control algorithm to optimize the equipment's control to achieve the purpose of reducing operating energy consumption and saving energy and reducing emissions. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a semi-physical simulation control management method and platform for building air-conditioning systems based on a PLM system in response to the defects in the prior art.

[0006] The technical solution adopted by the present invention to solve the technical problem is: a semi-physical simulation control management platform for building air-conditioning systems based on the PLM system, comprising:

[0007] Visualization module, used for dynamic display of building BIM models, measured data, and simulation optimization data in building air-conditioning systems;

[0008] The visualization module includes a building BIM model dynamic display unit, a measured data display unit, and a simulation optimization data display unit;

[0009] The BIM model dynamic display unit is used to display the building information model and equipment attribute information obtained from the database module, as well as the real-time operation data obtained from the sensors; the real-time operation data obtained from the sensors includes: valve opening, fan speed, and pressure gauge data; the relevant equipment components in the BIM model are in dynamic change according to the changes in the real-time operation data, such as the valve opening, fan speed, and pressure gauge indication change with the real-time operation data.

[0010] The BIM model dynamic display unit can interact with the measured data display unit and the simulation data display unit. By clicking on the device node, the measured data display unit and the simulation data display unit will display the data of the corresponding device respectively. By clicking on the device in the measured data display unit and the simulation data display unit, you can locate and select the specified device in the BIM model dynamic display unit;

[0011] The measured data display unit is used to display the historical operation data of the equipment and the real-time collected data, and to evaluate the equipment operation status based on the real-time operation data;

[0012] If there are changes to the equipment during operation and maintenance, the relevant files of the entire equipment can also be updated. All data changes will be synchronized into the database module, and then the entire system will be updated to ensure data consistency;

[0013] The simulation data display unit is used to obtain simulation optimization data from the database, display the load forecast results of the prediction model and the outdoor meteorological condition forecast results, the system simulation model, the hyperparameter setting values ​​of the AI ​​agent model, the optimization algorithm, the optimization objective function, the parameter constraint range, and interact with the BIM model dynamic display unit to display the historical control parameters of the corresponding equipment. When there is an equipment change, the property management personnel can update the simulation model parameters according to the actual changes and correct the training of the AI ​​agent model.

[0014] The simulation optimization module is used to establish a system simulation model and obtain the system's equipment control optimization parameters based on the system's real-time operation data;

[0015] The simulation optimization module includes a prediction model unit, a system simulation model unit, an AI agent model unit and an optimization control model unit;

[0016] A prediction model unit, for establishing a sequence-to-sequence (Seq2Seq) prediction model for predicting building load using a long short-term memory network (LSTM);

[0017] The prediction model includes an encoder and a decoder;

[0018] The prediction model uses an encoder to encode input data into an intermediate vector, and then uses a decoder to decode the intermediate vector and output predicted data. The prediction model uses the historical 24-hour collected data to predict the data for the next 4 hours, which is used to predict meteorological parameters such as building load and outdoor temperature. The input data of the prediction model comes from the measured historical data in the database, and the prediction results serve as the input of the system simulation model unit and the AI ​​agent model unit;

[0019] System simulation model unit, the system simulation model is used to provide a system simulation model for simulating the actual system operation status; it uses the Modelica language and is directly mapped and generated by the extended attributes and connection logic of the BIM model to ensure the consistency of the system simulation model parameters with the actual equipment parameters. At the same time, it will generate mathematical models of the corresponding equipment to form a mathematical model equipment library, reflecting the two-dimensional and three-dimensional linkage of the PLM system based on the same data source.

[0020] The AI ​​agent model unit is used to establish an AI agent model using a neural network and train the AI ​​agent model using the system's real-time operation data until the error between the output measured results of the system simulation model unit and the predicted results of the prediction model reaches an allowable range. In the initial operation period, there is no measured data, and the model training is performed using local typical meteorological data and equipment rated parameters;

[0021] The optimization control model unit is used to establish an optimization control model for the AI ​​agent model. The optimization variable is the control signal of the equipment in the system. The optimization target can be indoor thermal comfort, total system energy consumption, system stability, etc. The constraint range is determined according to the actual needs of the equipment. The optimization algorithm can be a genetic algorithm, a non-dominated sorting genetic algorithm, or a fast non-dominated sorting genetic algorithm based on a reference point. A set of optimal control solutions is calculated at each moment and then brought into the system simulation model for verification;

[0022] The system equipment module includes a building air-conditioning system equipment unit, a sensor unit arranged on the equipment, and an equipment control system unit.

[0023] Building air conditioning system equipment unit: all physical equipment of the air conditioning system, including air handling units, cold and hot sources, fan coil units, air ducts, water pipes and water pumps;

[0024] Sensor unit: used to detect equipment and indoor environment and send data to the database module, including temperature and humidity sensors, wind speed sensors, pressure sensors, flow meters and energy meters;

[0025] Equipment control system unit: receives control data from the simulation optimization module and sends specified control signals to the equipment remotely.

[0026] Database module, including building information database, equipment operation database, and simulation optimization database;

[0027] Building information database, used to store basic information of building space and equipment attribute information, is the source of BIM model data in the visualization module;

[0028] Equipment operation database unit, used to store the actual equipment operation data monitored by sensors;

[0029] The simulation optimization database unit is used to store all parameter data of the optimization simulation module.

[0030] According to the above solution, the BIM model in the visualization module is obtained by lightweighting the building air-conditioning system in the electromechanical detailed design model in the PLM system.

[0031] The present invention also provides a PLM system-based semi-physical simulation control method for a building air-conditioning system, comprising the following steps:

[0032] 1) Arrange sensors and equipment control systems on all physical equipment in the building air conditioning system;

[0033] 2) Forecast building load and outdoor weather conditions based on historical data including building load and outdoor weather conditions;

[0034] 3) Establish a system simulation model of the operating status of the building air-conditioning system, and keep the system simulation model parameters consistent with the actual equipment parameters;

[0035] 4) Use a neural network to build an AI agent model, and use the real-time operating data of the building air conditioning system to train the AI ​​agent model until the error between the predicted results and the measured results of the simulation model system reaches the allowable range, thereby obtaining the optimal control parameters for each device in the system;

[0036] 5) Establish an optimization control model for the AI ​​agent model. The optimization variables are the control signals of the equipment in the system. The optimization objectives include indoor thermal comfort, total system energy consumption, and system stability. The constraints are determined based on the actual needs of the equipment. The optimization algorithm is selected from one or a combination of genetic algorithm, non-dominated sorting genetic algorithm, and reference point-based fast non-dominated sorting genetic algorithm. At each moment, a set of optimal control solutions is calculated and then brought into the system simulation model for verification until the error between the predicted results and the measured results of the simulation model system reaches the allowable range.

[0037] The solution process is rolling forward, that is, at each moment the optimal control parameters for the next moment are optimized and solved based on real-time data.

[0038] 6) The device control system receives the optimal control parameters of each device in the system from the AI ​​agent model and sends control signals to the devices;

[0039] 7) Use the database to store basic information of the building space and equipment attribute information, the actual operation data of the equipment monitored by sensors, and all parameter data of the optimization simulation of each device in the system;

[0040] 8) Visual display;

[0041] Based on the data in the database, the building BIM model including the three-dimensional spatial layout and equipment properties of the building air-conditioning system, the measured data and the simulation optimization data are dynamically displayed.

[0042] According to the above scheme, the physical equipment in step 1) includes an air handling unit, a cold and hot source, a fan coil, an air duct, a water pipe and a water pump; the sensors include a temperature and humidity sensor, a wind speed sensor, a pressure sensor, a flow meter and an energy meter.

[0043] According to the above solution, the BIM model in step 7) is obtained by lightweighting the building air-conditioning system in the electromechanical detailed design model in the PLM system.

[0044] The beneficial effects produced by the present invention are:

[0045] The present invention proposes a semi-physical simulation control management platform for building air-conditioning systems based on a full life cycle management PLM system. The platform realizes the 5D dynamic BIM model display of the three-dimensional spatial layout, equipment properties and real-time operating status of the building air-conditioning system. The BIM model is used to generate a simulation model to realize two-dimensional and three-dimensional linkage to make it have logical significance. The AI ​​agent model is used to quickly solve the optimal control parameters of the building air-conditioning system, ensuring that the entire system is always operating in the optimal state and realizing optimal control of the building air-conditioning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0047] Figure 1 is a schematic diagram of the management platform structure of an embodiment of the present invention;

[0048] Figure 2 is an optimization control flow chart of the simulation optimization module according to an embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram of the AI ​​agent model according to an embodiment of the present invention;

[0050] Figure 4 is a schematic diagram of a prediction model according to an embodiment of the present invention;

[0051] Figure 5 This is a schematic diagram of a system simulation model generation principle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] like Figure 1 As shown in FIG, a PLM-based semi-physical simulation control and management platform for building air conditioning systems includes:

[0054] Visualization module, used for dynamic display of building BIM models, measured data, and simulation optimization data in building air-conditioning systems;

[0055] The visualization module includes a building BIM model dynamic display unit, a measured data display unit, and a simulation optimization data display unit;

[0056] The BIM model dynamic display unit is used to display the building information model and equipment attribute information obtained from the database module, as well as real-time operating data obtained from sensors. This real-time operating data includes valve opening, fan speed, and pressure gauge data. The relevant equipment components in the BIM model dynamically change in response to changes in real-time operating data, such as valve opening, fan speed, and pressure gauge indications. The BIM model is derived from the lightweight design of the building air conditioning system in the PLM system's detailed electromechanical design model.

[0057] The BIM model dynamic display unit can interact with the measured data display unit and the simulation data display unit. By clicking on the device node, the measured data display unit and the simulation data display unit will display the data of the corresponding device respectively. By clicking on the device in the measured data display unit and the simulation data display unit, you can locate and select the specified device in the BIM model dynamic display unit;

[0058] The measured data display unit is used to display the historical operation data of the equipment and the real-time collected data, and to evaluate the equipment operation status based on the real-time operation data;

[0059] If there are changes to the equipment during operation and maintenance, the relevant files of the entire equipment can also be updated. All data changes will be synchronized into the database module, and then the entire system will be updated to ensure data consistency;

[0060] The simulation data display unit is used to obtain simulation optimization data from the database, display the load forecast results of the prediction model and the outdoor meteorological condition forecast results, the system simulation model, the hyperparameter setting values ​​of the AI ​​agent model, the optimization algorithm, the optimization objective function, the parameter constraint range, and interact with the BIM model dynamic display unit to display the historical control parameters of the corresponding equipment. When there is an equipment change, the property management personnel can update the simulation model parameters according to the actual changes and correct the training of the AI ​​agent model.

[0061] The simulation optimization module is used to establish a system simulation model and obtain the system's equipment control optimization parameters based on the system's real-time operation data;

[0062] The simulation optimization module includes a prediction model unit, a system simulation model unit, an AI agent model unit and an optimization control model unit;

[0063] A prediction model unit, for establishing a sequence-to-sequence (Seq2Seq) prediction model for predicting building load using a long short-term memory network (LSTM);

[0064] The prediction model includes an encoder and a decoder;

[0065] The prediction model uses an encoder to encode input data into an intermediate vector, and then uses a decoder to decode the intermediate vector and output predicted data. The prediction model uses the historical 24-hour collected data to predict the data for the next 4 hours, which is used to predict meteorological parameters such as building load and outdoor temperature. The input data of the prediction model comes from the measured historical data in the database, and the prediction results serve as the input of the system simulation model unit and the AI ​​agent model unit;

[0066] The principle of the prediction model is as follows Figure 3 As shown, a long short-term memory (LSTM) network is used to build a sequence-to-sequence (seq2seq) prediction model. The model consists of an encoder, an intermediate vector, and a decoder. The encoder encodes the input sequence into a fixed-length intermediate vector, which is then decomposed by the decoder to output a set of vectors for data prediction. The parameters in the prediction model are all determined using hyperparameter optimization.

[0067] System simulation model unit, the system simulation model is used to provide a system simulation model for simulating the actual system operation status; it uses the Modelica language and is directly mapped and generated by the extended attributes and connection logic of the BIM model to ensure the consistency of the system simulation model parameters with the actual equipment parameters. At the same time, it will generate mathematical models of the corresponding equipment to form a mathematical model equipment library, reflecting the two-dimensional and three-dimensional linkage of the PLM system based on the same data source.

[0068] The property expansion of the three-dimensional model of the building air conditioning system is completed by generating a simulation model from the three-dimensional model. The principle is as follows Figure 5 As shown in the figure, the three-dimensional model of the building air-conditioning system is expanded and mapped with the dymola model to generate a system simulation model.

[0069] The AI ​​agent model unit is used to establish an AI agent model using a neural network and train the AI ​​agent model using the system's real-time operation data until the error between the output measured results of the system simulation model unit and the predicted results of the prediction model reaches an allowable range. In the initial operation period, there is no measured data, and the model training is performed using local typical meteorological data and equipment rated parameters;

[0070] The optimization control model unit is used to establish an optimization control model for the AI ​​agent model. The optimization variable is the control signal of the equipment in the system. The optimization target can be indoor thermal comfort, total system energy consumption, system stability, etc. The constraint range is determined according to the actual needs of the equipment. The optimization algorithm can be a genetic algorithm, a non-dominated sorting genetic algorithm, or a fast non-dominated sorting genetic algorithm based on a reference point. A set of optimal control solutions is calculated at each moment and then brought into the system simulation model for verification;

[0071] The optimized control process of the semi-physical simulation control management platform of the building air-conditioning system based on the PLM system in this embodiment is as follows: Figure 2 As shown in the figure, the sensor saves the collected measured data in the database, where the measured historical data is input into the prediction model to predict future loads. The predicted load is then input into the AI ​​agent model together with the measured real-time data. The optimization algorithm generates a set of control strategy solutions through iterative optimization, which are then brought into the simulation model for verification. Finally, the TOPSIS comprehensive analysis method is used to analyze and select the final control strategy and send it to the equipment control system. Finally, the equipment control system sends the final control signal to the building air-conditioning equipment terminal. The above cycle is repeated to achieve optimal control of the building air-conditioning equipment.

[0072] The principle of AI agent model is as follows Figure 4 As shown in the figure, it consists of an input layer, a hidden layer, and an output layer. The input data of the input layer are weather forecast data, load forecast data, and the current operating status of the equipment. The weather forecast data includes meteorological parameters such as outdoor dry-bulb temperature, outdoor wet-bulb temperature, wind speed, and solar radiation. The hidden layer is a fully connected neuron. The specific number of hidden layers is obtained through parameter optimization. The output is parameters such as indoor thermal comfort, total system energy consumption, and system stability. Different input and output data can also be defined according to actual needs.

[0073] The system equipment module includes a building air-conditioning system equipment unit, a sensor unit arranged on the equipment, and an equipment control system unit.

[0074] Building air conditioning system equipment unit: all physical equipment of the air conditioning system, including air handling units, cold and hot sources, fan coil units, air ducts, water pipes and water pumps;

[0075] Sensor unit: used to detect equipment and indoor environment and send data to the database module, including temperature and humidity sensors, wind speed sensors, pressure sensors, flow meters and energy meters;

[0076] Equipment control system unit: receives control data from the simulation optimization module and sends specified control signals to the equipment remotely.

[0077] Database module, including building information database, equipment operation database, and simulation optimization database;

[0078] Building information database, used to store basic information of building space and equipment attribute information, is the source of BIM model data in the visualization module;

[0079] Equipment operation database unit, used to store the actual equipment operation data monitored by sensors;

[0080] The simulation optimization database unit is used to store all parameter data of the optimization simulation module.

[0081] The present invention also provides a PLM system-based semi-physical simulation control method for a building air-conditioning system, comprising the following steps:

[0082] 1) Arrange sensors and equipment control systems on all physical equipment in the building air conditioning system;

[0083] Physical equipment includes air handling units, cold and heat sources, fan coil units, air ducts, water pipes and water pumps; sensors include temperature and humidity sensors, wind speed sensors, pressure sensors, flow meters and energy meters.

[0084] 2) Forecast building load and outdoor weather conditions based on historical data including building load and outdoor weather conditions;

[0085] 3) Establish a system simulation model of the operating status of the building air-conditioning system, and keep the system simulation model parameters consistent with the actual equipment parameters;

[0086] 4) Use a neural network to build an AI agent model, and use the real-time operating data of the building air conditioning system to train the AI ​​agent model until the error between the predicted results and the measured results of the simulation model system reaches the allowable range, thereby obtaining the optimal control parameters for each device in the system;

[0087] 5) Establish an optimization control model for the AI ​​agent model. The optimization variables are the control signals of the equipment in the system. The optimization objectives include indoor thermal comfort, total system energy consumption, and system stability. The constraints are determined based on the actual needs of the equipment. The optimization algorithm is selected from one or a combination of genetic algorithm, non-dominated sorting genetic algorithm, and reference point-based fast non-dominated sorting genetic algorithm. At each moment, a set of optimal control solutions is calculated and then brought into the system simulation model for verification until the error between the predicted results and the measured results of the simulation model system reaches the allowable range.

[0088] The solution process is rolling forward, that is, at each moment the optimal control parameters for the next moment are optimized and solved based on real-time data.

[0089] 6) The device control system receives the optimal control parameters of each device in the system from the AI ​​agent model and sends control signals to the devices;

[0090] 7) Use the database to store basic information of the building space and equipment attribute information, the actual operation data of the equipment monitored by sensors, and all parameter data of the optimization simulation of each device in the system;

[0091] 8) Visual display;

[0092] Based on the data in the database, the building BIM model including the three-dimensional spatial layout and equipment properties of the building air-conditioning system, the measured data and the simulation optimization data are dynamically displayed.

[0093] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A PLM-based semi-physical simulation control and management platform for building air-conditioning systems, characterized by: include: The simulation optimization module is used to establish a system simulation model and obtain the system's equipment control optimization parameters based on the system's real-time operation data; The simulation optimization module includes a prediction model unit, a system simulation model unit, an AI agent model unit and an optimization control model unit; A prediction model unit, used to establish a sequence-to-sequence prediction model for predicting building load using a long short-term memory network (LSTM); The prediction model includes an encoder and a decoder; The prediction model uses an encoder to encode input data into an intermediate vector, and then uses a decoder to decode the intermediate vector and output predicted data. The prediction model uses the historical 24-hour collected data to predict the data for the next 4 hours, which is used to predict meteorological parameters such as building load and outdoor temperature. The input data of the prediction model comes from the measured historical data in the database, and the prediction results serve as the input of the system simulation model unit and the AI ​​agent model unit; A system simulation model unit, used to provide a system simulation model for simulating the actual system operation state; The AI ​​agent model unit is used to establish an AI agent model using a neural network and train the AI ​​agent model using the system's real-time operation data until the error between the output of the system simulation model unit's actual measurement results and the prediction results of the prediction model reaches an allowable range; The optimization control model unit is used to establish an optimization control model for the AI ​​agent model. The optimization variables are the control signals of the equipment in the system, and the optimization objectives are indoor thermal comfort, total system energy consumption, and system stability. The constraints are determined according to the actual needs of the equipment. The optimization algorithm is one of the following: genetic algorithm, non-dominated sorting genetic algorithm, and reference point-based fast non-dominated sorting genetic algorithm, or a combination thereof. At each moment, the optimal solution for the system equipment control parameters is calculated and then brought into the system simulation model for verification. System equipment module, used to obtain equipment properties and operating parameters of building air-conditioning system equipment and perform equipment control; Database module, including building information database, equipment operation database, and simulation optimization database; The visualization module is used to dynamically display the building BIM model, measured data and simulation optimization data in the building air-conditioning system based on the data in the database module.

2. The PLM-based building air conditioning system semi-physical simulation control management platform according to claim 1 is characterized in that: The visualization module includes a building BIM model dynamic display unit, a measured data display unit and a simulation optimization data display unit; The details are as follows: The BIM model dynamic display unit obtains the building information model and equipment attribute information from the database module, as well as real-time operating data obtained from sensors. The sensor data includes valve opening, fan speed, and pressure gauge data. The relevant equipment components in the BIM model are in dynamic change in response to changes in the real-time operating data. The measured data display unit is used to display the historical operation data of the equipment and the real-time collected data, and to evaluate the equipment operation status based on the real-time operation data; The simulation data display unit is used to obtain simulation optimization data from the database, display the load forecast results of the prediction model and the outdoor meteorological condition forecast results, as well as the system simulation model, the hyperparameter setting values ​​of the AI ​​agent model, the optimization algorithm adopted, the optimization objective function, the parameter constraint range, and interact with the BIM model dynamic display unit to display the historical control parameters of the corresponding equipment.

3. The PLM-based building air conditioning system semi-physical simulation control management platform according to claim 1 is characterized in that: The BIM model displayed in the visualization module is obtained by lightweighting the building air-conditioning system in the electromechanical detailed design model in the PLM system.

4. The PLM-based building air conditioning system semi-physical simulation control management platform according to claim 1 is characterized in that: The system equipment module includes a building air conditioning system equipment unit, a sensor unit arranged on the equipment, and an equipment control system unit; Building air conditioning system equipment unit: all physical equipment of the air conditioning system, including air handling units, cold and hot sources, fan coil units, air ducts, water pipes and water pumps; Sensor unit: used to detect equipment and indoor environment and send data to the database module, including temperature and humidity sensors, wind speed sensors, pressure sensors, flow meters and energy meters; Equipment control system unit: receives control data from the simulation optimization module and sends specified control signals to the equipment remotely.

5. The PLM-based building air conditioning system semi-physical simulation control management platform according to claim 1 is characterized in that: The database includes: Building information database, used to store basic information of building space and equipment attributes, is the source of BIM model data in the visualization module; Equipment operation database unit, used to store the actual equipment operation data monitored by sensors; The simulation optimization database unit is used to store all parameter data of the optimization simulation module.

6. A semi-physical simulation control method for a building air conditioning system based on a PLM system, characterized in that: The following steps are involved: 1) Arrange sensors and equipment control systems on all physical equipment in the building air conditioning system; 2) Forecast building loads and outdoor weather conditions based on historical data including building loads and outdoor weather conditions; 3) Establish a system simulation model of the building air conditioning system operating status, and keep the system simulation model parameters consistent with the actual equipment parameters; 4) Establish an AI agent model and train it using real-time operating data from the building air conditioning system until the error between the predicted results and the measured results of the simulation model system reaches the allowable range, thereby obtaining the optimal control parameters for each device in the system; 5) Establish an optimization control model for the AI ​​agent model. The optimization variables are the control signals of the equipment in the system. The optimization objectives include indoor thermal comfort, total system energy consumption, and system stability. The constraints are determined based on the actual needs of the equipment. The optimization algorithm is selected from one or a combination of genetic algorithm, non-dominated sorting genetic algorithm, and reference point-based fast non-dominated sorting genetic algorithm. At each moment, a set of optimal control solutions is optimized and calculated. These solutions are then brought into the system simulation model for verification until the error between the predicted results and the measured results of the simulation model system reaches the allowable range. 6) The device control system receives the optimal control parameters of each device in the system from the AI ​​agent model and sends control signals to the devices; 7) Use the database to store basic information of the building space and equipment attributes, the actual operation data of the equipment monitored by sensors, and all parameter data of the optimization simulation of each device in the system; 8) Visual display; Based on the data in the database, the building BIM model including the three-dimensional spatial layout and equipment properties of the building air-conditioning system, the measured data and the simulation optimization data are dynamically displayed.

7. The PLM-based semi-physical simulation control method for building air-conditioning systems according to claim 6 is characterized in that: The physical equipment in step 1) includes an air handling unit, a cold and hot source, a fan coil unit, an air duct, a water pipe, and a water pump; the sensors include a temperature and humidity sensor, a wind speed sensor, a pressure sensor, a flow meter, and an energy meter.

8. The PLM-based semi-physical simulation control method for building air-conditioning systems according to claim 6 is characterized in that: The BIM model in step 7) is obtained by lightweighting the building air-conditioning system in the electromechanical detailed design model in the PLM system.

Images