Building heating ventilation air conditioner control method with dynamic energy efficiency optimization

Abstract

The invention discloses a building heating, ventilating and air conditioning control method with dynamic energy efficiency optimization, and belongs to the technical field of heating, ventilating and air conditioning control. The method comprises the following steps: step 1, acquiring temperature, humidity, carbon dioxide concentration and personnel distribution data in real time, and performing preliminary data processing through an edge computing node; step 2, realizing accurate prediction of future load change of the building; step 3, generating an optimal control strategy by using a reinforcement learning algorithm, and realizing dynamic optimization control of equipment operation; 4, the compliance of equipment operation, network communication and control instructions is monitored in real time, and a closed-loop dynamic tuning mechanism is constructed to continuously optimize control strategy parameters; step 5, dividing a control task into a group control layer, a region layer and a terminal layer, and establishing communication among the layers through a standardized interface so as to realize accurate transmission and execution of an instruction; and step 6, designing a friendly human-computer interaction interface, displaying a building operation state, energy consumption analysis and a control strategy, and supporting multi-terminal access and remote monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of heating, ventilation and air conditioning control technology, more specifically, to a dynamic energy efficiency optimized building heating, ventilation and air conditioning control method. BACKGROUND

[0002] With the increase of building energy consumption, the energy use of heating, ventilation and air conditioning (HVAC) system in buildings occupies an important proportion. How to effectively manage the energy efficiency of buildings to meet the multiple demands of comfort, energy saving and equipment life has become a key problem in building design and operation. Traditional HVAC control methods often rely on fixed setting parameters, which do not fully consider real-time changes in indoor and outdoor environmental factors such as temperature, humidity, carbon dioxide concentration and personnel activity, nor do they fully utilize modern sensing technology and data analysis methods to optimize energy efficiency. Such control methods not only fail to achieve accurate load prediction, but also easily lead to energy waste and unstable system performance.

[0003] In recent years, with the development of Internet of Things, edge computing and artificial intelligence technology, building energy efficiency management solutions based on real-time data collection and intelligent algorithms have gradually become a research hotspot. By deploying sensor networks inside buildings, real-time monitoring of environmental parameters and personnel distribution can provide real-time data support for building energy efficiency optimization. Using digital twin technology and deep learning algorithms to model the thermal dynamics of buildings, and combining reinforcement learning algorithms to optimize control strategies, it is expected to achieve dynamic adjustment and efficient energy consumption control of building HVAC systems. However, existing technologies still face the challenge of how to effectively combine multiple advanced technologies to achieve accurate prediction and dynamic control.

[0004] In summary, how to achieve dynamic energy efficiency optimization based on real-time data in complex and variable building environments, and balance energy saving, comfort and equipment life through intelligent control systems has become a technical problem that needs to be solved. SUMMARY

[0005] In order to overcome the series of defects existing in the prior art, the purpose of the present application is to provide a dynamic energy efficiency optimized building heating, ventilation and air conditioning control method, comprising the following steps: Step 1, by deploying a hierarchical distributed sensor network inside the building, real-time collection of temperature, humidity, carbon dioxide concentration and personnel distribution data, and preliminary data processing through edge computing nodes to ensure the real-time and reliability of the data; Step 2, based on the collected real-time data, combined with the digital twin model of the building, using deep learning algorithms to establish a thermal dynamics model to achieve accurate prediction of future load changes of the building; Step 3: Based on the thermodynamic characteristics model, combined with energy-saving effect, personnel comfort and equipment service life target constraints, the optimal control strategy is generated using reinforcement learning algorithm to realize dynamic optimization control of equipment operation; Step 4: Design a comprehensive safety evaluation system to monitor the compliance of equipment operation, network communication and control instructions in real time, and build a closed-loop dynamic optimization mechanism to continuously optimize control strategy parameters; Step 5: Divide the control task into group control layer, regional layer and terminal layer, and establish communication between each layer through standardized interface to realize accurate transmission and execution of instructions; Step 6: Design a friendly human-computer interaction interface to display building operation status, energy consumption analysis and control strategy, support multi-terminal access and remote monitoring.

[0006] Further, step 1 includes the following steps: According to the functional zoning, human flow density and environmental characteristics of the building, plan the sensor layout scheme, determine the installation location and quantity of various sensors, and ensure the representativeness of the monitoring points and the integrity of the coverage range; Install environmental sensor equipment and people flow monitoring equipment at the determined location, and complete the calibration and parameter setting of the equipment; Deploy edge computing nodes at each floor or key area of the building, and use wireless communication technology to establish a communication network between the equipment and the edge nodes; Deploy lightweight data processing algorithms at the edge nodes to realize data preprocessing; Establish a complete data quality control system, including sensor regular calibration mechanism, data anomaly alarm mechanism and backup mechanism, to realize remote configuration management of the equipment.

[0007] Further, step 2 includes the following steps: Integrate real-time operation data of the building with weather forecast data to build a complete data set; Build a digital twin model based on building information model to accurately restore the physical characteristics of the building and realize dynamic visual monitoring of the building state; Design a deep learning model and integrate the physical characteristics parameters of the building into it, and continuously optimize the loss function and adjust the model parameters to accurately capture the thermodynamic change law of the building; Quantitatively evaluate the model performance and continuously iterate and optimize based on the evaluation results to continuously improve the prediction accuracy.

[0008] Further, step 3 includes the following steps: Establish a multi-objective optimization framework, taking energy-saving effect, personnel comfort and equipment service life as key optimization indicators; By weight configuration, the optimization target is converted into a unified reward function as the optimization target of the reinforcement learning algorithm; A reinforcement learning environment is constructed, with a building thermal dynamics model as the environment model, defining state space and action space; A deep reinforcement learning agent is designed and trained, through continuous interaction between the agent and the environment model, to learn the optimal control strategy to find the best balance point between energy efficiency, comfort and equipment life; A hierarchical control architecture is established, dividing the control strategy into long-term planning and short-term execution: the long-term planning layer generates the overall control scheme based on weather forecasts and user demand forecasts; the short-term execution layer dynamically adjusts and optimizes control instructions based on real-time state and disturbances.

[0009] Further, step 4 includes the following steps: A multi-level safety evaluation index system is constructed, covering equipment operation safety indicators, network communication safety indicators and control instruction safety indicators; A real-time monitoring and early warning platform is built, integrating various sensor data and work logs to continuously monitor key safety indicators; Set multi-level warning thresholds and establish an automatic alarm mechanism to ensure timely warning when monitoring indicators are abnormal, and take appropriate emergency measures according to the degree of abnormality; Real-time analysis of equipment operation status, network communication quality and control instruction execution to quickly identify potential safety hazards; Establish a dynamic evaluation mechanism for control strategies, regularly collect and analyze operation data to evaluate the actual effect of control strategies, and automatically adjust control parameters based on evaluation results to continuously improve control strategies; Comprehensive analysis of safety evaluation results, anomaly detection information and control effect evaluation information to automatically identify performance improvement space and generate optimization suggestions to achieve adaptive adjustment of control strategy parameters under the premise of ensuring safety.

[0010] Further, real-time analysis of equipment operation status, network communication quality and control instruction execution to quickly identify potential safety hazards, including the following steps: Use the preset normal parameter range and historical operation data as a benchmark to compare the deviation between current equipment operation status data and standard values to promptly identify abnormal conditions in equipment operation; Comprehensive evaluation of network communication quality, verification of communication protocol integrity and data transmission security to ensure communication link reliability and security; A control instruction execution tracking mechanism is constructed to record the entire process of each control instruction from issuance to execution completion, including instruction reception confirmation, execution state feedback, and completion verification, thereby achieving precise regulation of the control process. Based on the results of multi-dimensional data analysis, potential safety hazards are predicted and risk level assessments are provided through identification and analysis of abnormal patterns.

[0011] Further, the safety assessment results, abnormal detection information, and control effect evaluation information are comprehensively analyzed to automatically identify performance improvement space and generate optimization suggestions, thereby achieving adaptive adjustment of control strategy parameters under the premise of ensuring safety, including the following steps: A multi-dimensional data integration platform is constructed to systematically archive and standardize safety assessment reports, abnormal event records, and control effect evaluation data, establish a unified data analysis framework, and ensure that various types of information can be effectively associated and cross-verified. The integrated data is deeply mined to identify key performance indicator trends and analyze the mutual influence between control parameters, thereby identifying the main factors and bottlenecks affecting system performance. The current running state is compared and analyzed with the ideal state to quantitatively evaluate the gap in efficiency, stability, and safety, and automatically identify the links and parameters with improvement space. Optimization suggestions, including control parameter adjustment schemes, operation strategy optimization measures, and safety protection enhancement suggestions, are automatically generated based on comprehensive analysis results, and each optimization suggestion is subjected to feasibility evaluation and risk analysis. Optimization suggestions are classified according to impact degree and implementation difficulty, and parameter adjustments with lower risk and significant effect are automatically implemented. The running effect after each parameter adjustment is monitored and evaluated in real time to promptly discover any abnormalities that may occur after adjustment.

[0012] Further, step 5 includes the following steps: A three-layer control architecture is planned and established to achieve precise control and collaborative optimization of building group energy management: the group control layer is set as the highest decision-making layer, responsible for overall energy distribution and collaborative optimization; the regional layer serves as the intermediate layer, responsible for independent control and management of each functional area; and the terminal layer serves as the execution layer, directly controlling the running state of specific heating, ventilation, and air conditioning systems. Special data conversion modules are developed to meet the data interaction needs between different levels, ensuring the accuracy and real-time performance of information transmission between levels. A collaborative optimization mechanism is constructed for the group control layer to achieve optimal allocation of resources by integrating building group energy demand and supply information. Develop a task management mechanism at the regional level to receive control instructions from the group control level, and further refine the control tasks based on regional characteristics and real-time status; After receiving specific instructions from the regional level, the terminal controller directly adjusts the HVAC operating parameters; By monitoring the device status in real time, quickly responding to control instructions, and feeding back the execution results to the upper layer, a complete control closed loop is formed.

[0013] Compared with the prior art, this application has the following beneficial effects: This application collects data by deploying a layered distributed sensor network in the building, combines the building's digital twin model with a deep learning algorithm to predict thermodynamic characteristics, uses a reinforcement learning algorithm to generate an optimal control strategy, and implements a dynamic optimization control and safety assessment system for equipment through a three-layer control architecture to improve building energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a building HVAC control method for dynamic energy efficiency optimization disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0015] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Throughout the drawings, identical or similar reference numerals represent identical or similar elements or elements having identical or similar functions. The described embodiments are only some, not all, of the embodiments of the present invention.

[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0017] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0018] like Figure 1 As shown, a building HVAC control method with dynamic energy efficiency optimization includes the following steps: Step 1: Deploy a hierarchical distributed sensor network within the building to collect real-time data on temperature, humidity, carbon dioxide concentration, and occupant distribution. Initial data processing is performed through edge computing nodes to ensure real-time and reliable data. Step 2: Based on the collected real-time data and the building's digital twin model, a deep learning algorithm is used to establish a thermodynamic characteristic model to accurately predict future load changes in the building. Step 3, based on the thermodynamic characteristics model, combined with energy saving effect, personnel comfort and equipment service life target constraints, the optimal control strategy is generated by using reinforcement learning algorithm to realize dynamic optimization control of equipment operation; Step 4, design a comprehensive safety evaluation system to monitor the compliance of equipment operation, network communication and control instructions in real time, and build a closed-loop dynamic optimization mechanism to continuously optimize control strategy parameters; Step 5, divide the control task into group control layer, regional layer and terminal layer, and establish communication between each layer through standardized interface to realize accurate transmission and execution of instructions; Step 6, design a friendly human-computer interaction interface to show the building operation state, energy consumption analysis and control strategy, support multi-terminal access and remote monitoring.

[0019] The dynamic energy efficiency optimization building HVAC control method disclosed in this embodiment realizes efficient management of temperature, humidity, air quality and energy consumption in the building through a multi-step process, ultimately improving energy utilization efficiency while improving building comfort and equipment life.

[0020] First, in step 1, a hierarchical distributed sensor network is adopted to collect environmental data in real time, and preliminary processing is carried out on the edge computing nodes. This network structure ensures the accuracy and real-time nature of the data, so that it can quickly respond to changes in indoor and outdoor environments. Distributed sensors can comprehensively cover multiple areas within the building, collecting temperature, humidity, carbon dioxide concentration and personnel distribution data in real time, ensuring that dynamic information within the building can be obtained. Edge computing processes data on computing nodes closer to sensors, effectively reducing transmission delay and computational burden, improving response speed and data reliability, providing accurate and real-time data support for subsequent control decisions.

[0021] In step 2, based on the collected real-time data and the digital twin model of the building, a thermodynamic characteristics model of the building is constructed through a deep learning algorithm. The digital twin model digitally maps the structure, equipment layout and thermal physical properties of the building, and combined with real-time data, an accurate thermodynamic model can be generated, which can accurately simulate and predict energy efficiency under different load changes. The deep learning algorithm trains a prediction model for future load changes through large-scale historical data, which can identify the rules of temperature changes in the building and the influence of external environment, achieving accurate load prediction. This step combines the thermal characteristics of the building with real-time environmental data to support early control strategies.

[0022] In step 3, the optimal control strategy is generated by combining the thermodynamic property model with energy-saving, comfort, and equipment life target constraints using a reinforcement learning algorithm. Reinforcement learning can adapt to different operating environments and load conditions, dynamically adjusting the operating parameters of the equipment. This algorithm takes energy-saving effect, personnel comfort, and equipment life as multiple targets, continuously optimizing the control strategy through trial and error and feedback, and maintaining the best operating state in a changing environment. Through this learning method, the maximum energy utilization efficiency can be achieved, unnecessary energy consumption is reduced, and the working life of the equipment is extended.

[0023] Step 4 designs a comprehensive safety assessment system to monitor the compliance of equipment operation, network communication, and control instructions in real time, and builds a closed-loop dynamic optimization mechanism to continuously optimize control strategy parameters. This ensures the safety and compliance of each control strategy, and through the closed-loop feedback mechanism, real-time collection of equipment operation data, identification of possible abnormalities or potential failures, and adjustment of control strategy parameters, the reliability and stability are maximized. The security monitoring of network communication has the ability to defend against possible network attacks and data hijacking, ensuring the safe operation of the entire HVAC system.

[0024] In step 5, by dividing the control task into group control layer, regional layer, and terminal layer, a standardized interface between different levels is established. The group control layer is responsible for global strategy formulation, the regional layer controls specific regions, and the terminal layer executes specific control instructions. This hierarchical architecture enhances the flexibility and accuracy of the control mechanism, allowing different regions to adaptively regulate according to specific environmental needs, ensuring individual region comfort and optimizing overall energy efficiency. The application of standardized interfaces improves the accuracy of data and instruction transmission between different control layers, making communication and cooperation between layers smoother.

[0025] Finally, step 6 designs a friendly human-machine interface to display building operation status, energy consumption analysis, and control strategy, allowing managers to intuitively understand the energy efficiency of the building. The interface supports multi-terminal access and remote monitoring, allowing managers to control and monitor at any time and anywhere. The multi-terminal design improves usability, ensuring real-time information acquisition and rapid response. This function not only facilitates the operation of managers, but also provides more flexible control means for building energy efficiency management.

[0026] Overall, this dynamic energy efficiency optimization building HVAC control method builds an efficient, stable, and safe building energy efficiency management system through multi-dimensional data collection, model prediction, intelligent control, and hierarchical architecture. The entire system combines real-time data with machine learning technology to maximize energy savings while ensuring comfort, while also having strong adaptability and intelligence levels to effectively respond to different building environments and energy consumption needs, providing an innovative solution for intelligent building HVAC systems.

[0027] Further, step 1 includes the following steps: According to the functional zoning of the building, the density of human flow and the environmental characteristics, plan the sensor layout scheme, determine the installation location and quantity of various sensors, and ensure the representativeness of the monitoring points and the integrity of the coverage range; Install environmental sensor devices and people flow monitoring devices at the determined locations, and complete the calibration and parameter setting of the devices; Deploy edge computing nodes at each floor or key area of the building, and use wireless communication technology to establish a communication network between the devices and the edge nodes; Deploy lightweight data processing algorithms at the edge nodes to achieve data preprocessing; Establish a complete data quality control system, including regular calibration mechanism, data anomaly alarm mechanism and backup mechanism for sensors, to realize remote configuration management of devices.

[0028] In summary, the decomposition and implementation of step 1 lay a solid foundation for efficient control. From the scientific layout of sensors, to calibration and installation, deployment of edge computing nodes, lightweight data processing, and construction of data quality control system, each step improves the accuracy of data collection and the response speed of the system. These measures ensure the timeliness, accuracy and reliability of the data, and can flexibly and efficiently operate under different environmental conditions, thereby realizing the intelligentization and energy optimization of building HVAC systems.

[0029] Further, step 2 includes the following steps: Integrate building real-time operation data and weather forecast data to build a complete data set; Build a digital twin model based on building information model to accurately restore the physical characteristics of the building and realize dynamic visual monitoring of the building state; Design a deep learning model and integrate the physical characteristics parameters of the building into it, and continuously optimize the loss function and adjust the model parameters to accurately capture the thermodynamic change law of the building; Quantitatively evaluate the model performance and continuously iterate and optimize based on the evaluation results to continuously improve the prediction accuracy.

[0030] Overall, step 2 achieves efficient prediction of building thermodynamic characteristics through multi-level data integration, dynamic modeling, and model optimization, laying a solid foundation for intelligent control of HVAC systems. Each sub-step complements each other, from data collection, model construction to prediction optimization, forming a closed-loop prediction system. Complete data set, accurate digital twin and application of deep learning algorithm, thus have advanced prediction ability and adaptability. With this prediction model, building HVAC systems can maximize energy efficiency while ensuring indoor comfort, meeting the dual needs of energy saving and efficient operation of modern buildings.

[0031] Further, the optimization objective of the loss function is represented as: where, and are weight coefficients used to balance the influence between data fitting error, physical constraint term and regularization term; is the true observation value; is the model prediction value; represents the weight of different thermodynamic indicators, which is used to strengthen the attention to energy consumption; represents the temperature at a certain point; is the thermal diffusivity; represents the parameter set of the model; is the number of data samples used to train the model; represents the number of spatial points used to calculate the thermodynamic law constraints; is the gradient operator used to calculate the partial derivative; represents the Laplacian operator of the temperature at position .

[0032] In summary, the weight of different thermodynamic indicators in this optimization objective is particularly used to enhance the attention to energy consumption. This part of the setting makes the model pay more attention to energy-related thermodynamic variables in the optimization process, thereby better serving the dynamic energy-saving needs of HVAC systems. This comprehensive multi-constraint and objective loss function design enables the model to achieve high-precision prediction while maintaining good robustness and energy efficiency.

[0033] Further, step 3 includes the following steps: Establish a multi-objective optimization framework, taking energy saving effect, personnel comfort and equipment service life as key optimization indicators; Convert the optimization objective into a unified reward function through weight configuration, which serves as the optimization objective of the reinforcement learning algorithm; Build a reinforcement learning environment, taking the building thermodynamic characteristics model as the environment model, and define the state space and action space; Design and train a deep reinforcement learning agent that continuously interacts with an environment model to learn the optimal control strategy to find the best balance between energy efficiency, comfort, and equipment lifespan. A hierarchical control architecture is established, dividing the control strategy into two levels: long-term planning and short-term execution. The long-term planning layer generates an overall control plan based on weather forecasts and user demand predictions; the short-term execution layer dynamically adjusts and optimizes control instructions based on real-time status and disturbances.

[0034] In summary, Step 3 achieves intelligent control of the building's HVAC system through the combined application of a multi-objective optimization framework, a reinforcement learning algorithm, and a hierarchical control architecture. The multi-objective optimization framework balances energy conservation, comfort, and equipment lifespan, while the reinforcement learning algorithm enhances adaptability and flexibility. The hierarchical control architecture provides coordinated long- and short-term control, meeting user needs while better adapting to the dynamic changes in complex environments. This innovative optimization approach significantly improves the energy efficiency, comfort, and equipment stability of the building's HVAC system, meeting the requirements of modern buildings for high efficiency, energy conservation, and intelligence.

[0035] Furthermore, the unified reward function is expressed as follows: ,in, In state Take action The reward value reflects the effect of the control strategy in the current state; is the weight of the energy efficiency target, indicating the importance of energy efficiency to the reward in the overall optimization; is the weight of the comfort objective, indicating the importance of comfort to the reward in the overall optimization; is the weight of the equipment life target, which indicates the importance of equipment life to the reward in the overall optimization; Indicates the target energy efficiency value, which is the ideal energy consumption level that the building is expected to achieve during operation; Indicates that the status Take action Actual energy efficiency indicators at the time of implementation; Indicates the target comfort value, which is the ideal comfort standard to be achieved; Indicates that the status Take action Actual comfort index when Indicates that the status Take action Equipment service life indicator when

[0036] In this formula, the combination of the three terms weighs energy saving, comfort, and equipment life. The agent optimizes its control strategy through this reward function to meet energy saving requirements while ensuring comfort and extending equipment life. In practical applications, the proportion of , and can be adjusted based on the building's needs to form a flexible optimization control strategy. For example, when the building operates under an energy-saving priority strategy, the weight of can be increased, and when the comfort needs of personnel are high, the proportion of can be increased. In addition, the design of this reward function can also optimize the weight settings through multiple training and fine-tuning to more accurately reflect the building control needs in actual use.

[0037] Further, step 4 includes the following steps: Construct a multi-level safety evaluation index system covering device operation safety indicators, network communication safety indicators, and control instruction safety indicators; Build a real-time monitoring and early warning platform to integrate various sensor data and work logs to continuously monitor key safety indicators; Set multiple warning thresholds and establish an automatic alarm mechanism to ensure that when monitoring indicators are abnormal, timely warning information is issued, and appropriate emergency measures are taken according to the severity of the anomaly; Real-time analysis of device operation status, network communication quality, and control instruction execution to quickly identify potential safety hazards; Establish a dynamic evaluation mechanism for control strategies, regularly collect and analyze operation data to evaluate the actual effect of control strategies, and automatically adjust control parameters based on evaluation results to continuously improve control strategies; Comprehensive analysis of safety evaluation results, anomaly detection information, and control effect evaluation information to automatically identify performance improvement space and generate optimization suggestions to achieve adaptive adjustment of control strategy parameters under the premise of ensuring safety.

[0038] Through the cooperation of the above steps, not only is multi-objective optimization of building HVAC control achieved, but also safety and stability are ensured. The multi-level safety evaluation system combined with real-time monitoring, hierarchical warning, dynamic evaluation, and adaptive optimization mechanisms provides comprehensive safety protection for buildings, ensuring that control strategies meet energy saving, comfort, and equipment life requirements while maintaining high levels of operational safety.

[0039] Further, real-time analysis of device operation status, network communication quality, and control instruction execution to quickly identify potential safety hazards, including the following steps: Utilizing the preset normal parameter range and historical operation data as the benchmark, the deviation between the current equipment operation state data and the standard value is compared to timely discover abnormal situations in equipment operation; The network communication quality is comprehensively evaluated, and the integrity of the communication protocol and the security of data transmission are verified to ensure the reliability and security of the communication link; A control instruction execution tracking mechanism is constructed to record the entire process of each control instruction from issuance to execution completion, including instruction reception confirmation, execution state feedback, and completion verification, achieving precise regulation of the control process; Based on multi-dimensional data analysis results, potential safety hazards are predicted through identification and analysis of abnormal patterns, and risk level assessment is given.

[0040] In summary, real-time analysis of equipment operation state, network communication quality, and control instruction execution is the core link in the HVAC control safety guarantee system. Through multi-dimensional data evaluation of this system, potential safety hazards can be quickly identified and predicted on multiple levels, achieving dynamic and intelligent safety management. Combined with safety assessment and early warning mechanisms, not only is the stability of equipment, the reliability of communication, and the accuracy of control guaranteed, but also continuous improvement and self-optimization are achieved, providing strong support for the safety and energy-saving goals of buildings.

[0041] Further, the safety evaluation results, abnormal detection information, and control effect evaluation information are comprehensively analyzed to automatically identify performance improvement space and generate optimization suggestions, realizing adaptive adjustment of control strategy parameters under the premise of ensuring safety, including the following steps: A multi-dimensional data integration platform is constructed to systematically archive and standardize safety evaluation reports, abnormal event records, and control effect evaluation data, establishing a unified data analysis framework to ensure that various types of information can be effectively associated and cross-verified; The integrated data is deeply mined to identify key performance indicator trends and analyze the mutual influence between control parameters, identifying the main factors and bottlenecks affecting system performance; The current operating state is compared and analyzed with the ideal state to quantitatively evaluate the gap in efficiency, stability, and safety, automatically identifying areas and parameters with improvement space; Optimization suggestions are automatically generated based on comprehensive analysis results, including control parameter adjustment schemes, operation strategy optimization measures, and safety protection enhancement suggestions, and each optimization suggestion is subjected to feasibility evaluation and risk analysis; Optimization suggestions are classified according to impact degree and implementation difficulty, and parameter adjustments with lower risk and significant effect are automatically implemented; Real-time monitoring and evaluation of the operation effect after each parameter adjustment is performed to timely find abnormal situations that may occur after adjustment.

[0042] In summary, through multi-dimensional data integration, deep mining and automatic generation of optimization suggestions, adaptive parameter adjustment can be realized, and the safety and effectiveness of adjustment can be ensured under the closed-loop feedback mechanism. This adaptive optimization process significantly improves the intelligent level of building control and provides a strong guarantee for efficient and stable operation. Under the premise of safety, control strategies and operating parameters can be continuously iterated and optimized to ultimately achieve the goal of energy-efficient, efficient and intelligent building management.

[0043] Further, step 5 includes the following steps: A three-layer control architecture is planned and established to achieve precise control and collaborative optimization of building group energy management: the group control layer is set as the highest decision-making layer, responsible for overall energy distribution and collaborative optimization; the regional layer is the intermediate layer, responsible for independent control and management of each functional area; and the terminal layer is the execution layer, directly controlling the operation state of specific HVAC; A dedicated data conversion module is developed to meet the data interaction needs between different levels, ensuring the accuracy and real-time performance of information transmission between levels; A collaborative optimization mechanism is established for the group control layer to achieve optimal allocation of resources by integrating building group energy demand and supply information; A task management mechanism is developed for the regional layer to receive control instructions from the group control layer, further refine control tasks based on regional characteristics and real-time state; After receiving specific instructions from the regional layer, the terminal controller directly adjusts the operating parameters of the HVAC; Through real-time monitoring of equipment status, the control instructions are quickly responded to, and the execution results are fed back to the upper layer to form a complete control loop.

[0044] Overall, the three-layer control architecture realizes efficient operation of building group energy management through the collaborative operation of decision-making, refinement, execution and feedback. The data interaction and collaborative optimization mechanism between levels ensures timely response and flexible adjustment of control strategies, forming a complete closed-loop control system. The design of this architecture not only improves energy efficiency and operational stability, but also allows for flexible adjustment under different demand and environmental conditions, thereby achieving efficient and intelligent energy management.

[0045] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the same. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A building HVAC control method with dynamic energy efficiency optimization, characterized in that: The following steps are involved: Step 1: Deploy a hierarchical distributed sensor network within the building to collect real-time data on temperature, humidity, carbon dioxide concentration, and occupant distribution. Initial data processing is performed through edge computing nodes to ensure real-time and reliable data. Step 2: Based on the collected real-time data and the building's digital twin model, a deep learning algorithm is used to establish a thermodynamic characteristic model to accurately predict future load changes in the building. Step 3: Based on the thermodynamic characteristic model, combined with the energy-saving effect, personnel comfort and equipment service life target constraints, the reinforcement learning algorithm is used to generate the optimal control strategy to achieve dynamic optimization control of equipment operation; Step 4: Design a comprehensive safety assessment system to monitor the compliance of equipment operation, network communication, and control instructions in real time, and establish a closed-loop dynamic tuning mechanism to continuously optimize control strategy parameters; Step 5: Divide the control tasks into the group control layer, regional layer, and terminal layer, and establish communication between each layer through standardized interfaces to achieve accurate transmission and execution of instructions; Step 6: Design a user-friendly human-computer interaction interface to display the building operation status, energy consumption analysis and control strategy, and support multi-terminal access and remote monitoring.

2. A building HVAC control method with dynamic energy efficiency optimization according to claim 1, characterized in that: Step 1 includes the following steps: Plan the sensor layout based on the building's functional zoning, pedestrian density, and environmental characteristics, determine the installation locations and quantity of various sensors, and ensure representativeness of monitoring points and completeness of coverage; Install environmental sensor equipment and pedestrian flow monitoring equipment at the determined locations, and complete the equipment calibration and parameter setting; Deploy edge computing nodes on each floor or in key areas of a building, and use wireless communication technology to establish a communication network between devices and edge nodes; Deploy lightweight data processing algorithms on edge nodes to implement data preprocessing; Establish a complete data quality control system, including regular sensor calibration mechanism, data anomaly alarm mechanism and backup mechanism, to realize remote configuration management of equipment.

3. The building HVAC control method with dynamic energy efficiency optimization according to claim 1, characterized in that: Step 2 includes the following steps: By integrating real-time building operation data with weather forecast data, a complete data set is constructed; Build a digital twin model based on the building information model to accurately restore the physical characteristics of the building and realize dynamic visual monitoring of the building status; Design a deep learning model and incorporate the building's physical characteristics into it, and accurately capture the building's thermodynamic changes by continuously optimizing the loss function and adjusting the model parameters; Conduct quantitative evaluation of model performance, and continuously iterate and optimize based on the evaluation results to continuously improve prediction accuracy.

4. The building HVAC control method for dynamic energy efficiency optimization according to claim 1, characterized in that: Step 3 includes the following steps: Establish a multi-objective optimization framework, taking energy saving, personnel comfort and equipment service life as key optimization indicators; Through weight configuration, the optimization objective is converted into a unified reward function as the optimization objective of the reinforcement learning algorithm; Construct a reinforcement learning environment, use the building's thermodynamic characteristics model as the environment model, and define the state space and action space; Design and train a deep reinforcement learning agent that continuously interacts with an environment model to learn the optimal control strategy to find the best balance between energy efficiency, comfort, and equipment lifespan. A hierarchical control architecture is established, dividing the control strategy into two levels: long-term planning and short-term execution. The long-term planning layer generates an overall control plan based on weather forecasts and user demand predictions; the short-term execution layer dynamically adjusts and optimizes control instructions based on real-time status and disturbances.

5. The building HVAC control method with dynamic energy efficiency optimization according to claim 1, characterized in that: Step 4 includes the following steps: Build a multi-level security assessment indicator system, covering equipment operation safety indicators, network communication security indicators, and control instruction security indicators; Build a real-time monitoring and early warning platform, integrate various sensor data and work logs, and achieve continuous monitoring of key safety indicators; Set up multi-level warning thresholds and establish an automatic alarm mechanism to ensure that when monitoring indicators show abnormalities, early warning information is issued in a timely manner and appropriate emergency response measures are taken according to the degree of abnormality; Conduct real-time analysis of equipment operating status, network communication quality, and control instruction execution to quickly identify potential safety hazards; Establish a dynamic evaluation mechanism for control strategies, regularly collect and analyze operating data, evaluate the actual effects of control strategies, and automatically adjust control parameters based on the evaluation results to achieve continuous improvement of control strategies; Comprehensively analyze safety assessment results, anomaly detection information, and control effect evaluation information to automatically identify performance improvement space, generate optimization suggestions, and achieve adaptive adjustment of control strategy parameters while ensuring safety.

6. A building HVAC control method with dynamic energy efficiency optimization according to claim 5, characterized in that: Real-time analysis of device operating status, network communication quality, and control instruction execution to quickly identify potential safety hazards includes the following steps: Using the preset normal parameter range and historical operating data as a benchmark, the deviation between the current equipment operating status data and the standard value is compared to promptly detect abnormal conditions in equipment operation; Conduct a comprehensive assessment of network communication quality, verify the integrity of communication protocols and the security of data transmission, and ensure the reliability and security of communication links; Build a control instruction execution tracking mechanism to record the entire process of each control instruction from issuance to completion, including instruction receipt confirmation, execution status feedback and completion verification, to achieve precise supervision of the control process; Based on the results of multi-dimensional data analysis, by identifying and analyzing abnormal patterns, potential safety hazards are predicted and a risk level assessment is given.

7. A building HVAC control method with dynamic energy efficiency optimization according to claim 5, characterized in that: Comprehensively analyze safety assessment results, anomaly detection information, and control effect evaluation information to automatically identify performance improvement areas, generate optimization suggestions, and implement adaptive adjustment of control strategy parameters while ensuring safety. This includes the following steps: Build a multi-dimensional data integration platform to systematically archive and standardize safety assessment reports, abnormal event records, and control effect evaluation data, establish a unified data analysis framework, and ensure effective correlation analysis and cross-validation of various types of information; Conduct in-depth mining of the integrated data to identify the changing trends of key performance indicators, analyze the mutual influence between various control parameters, and find out the main factors and bottlenecks affecting system performance; Compare and analyze the current operating status with the ideal state, quantitatively assess the gaps in efficiency, stability, and safety, and automatically identify links and parameters with room for improvement; Automatically generate optimization suggestions based on comprehensive analysis results, including control parameter adjustment plans, operation strategy optimization measures, and safety protection enhancement suggestions, and conduct feasibility assessment and risk analysis for each optimization suggestion; Categorize optimization suggestions by impact and implementation difficulty, and automate the implementation of parameter adjustments with low risk and significant effects. Monitor and evaluate the operating results after each parameter adjustment in real time to promptly detect any abnormal situations that may occur after the adjustment.

8. The building HVAC control method with dynamic energy efficiency optimization according to claim 1, characterized in that: Step 5 includes the following steps: Plan and establish a three-tier control architecture to achieve precise control and coordinated optimization of building complex energy management: the group control layer is set as the highest decision-making layer, responsible for overall energy allocation and coordinated optimization; the regional layer serves as the middle layer, responsible for the independent control and management of each functional area; and the terminal layer serves as the execution layer, directly controlling the operating status of specific HVAC systems. Develop dedicated data conversion modules to meet the data interaction needs between different levels, ensuring the accuracy and real-time nature of information transmission between levels; Build a coordinated optimization mechanism at the group control layer to achieve optimal resource allocation by integrating the energy demand and supply information of the building complex; Develop a task management mechanism at the regional level to receive control instructions from the group control level, and further refine the control tasks based on regional characteristics and real-time status; After receiving specific instructions from the regional level, the terminal controller directly adjusts the HVAC operating parameters; By monitoring the device status in real time, quickly responding to control instructions, and feeding back the execution results to the upper layer, a complete control closed loop is formed.

Images