An efficient collaborative control method for multi-subsystems in a computer room based on intelligent simulation

Abstract

The present invention relates to the technical field of air conditioning simulation optimization, specifically to an efficient computer room multi-subsystem collaborative control method based on intelligent simulation, comprising constructing a digital twin model and a virtual simulation environment for key central air conditioning equipment; performing simulation analysis on the health status and operating performance of key equipment in the digital twin model through the virtual simulation environment to establish a performance benchmark model; acquiring equipment operating parameters from an external data source; analyzing the equipment operating parameters in the virtual simulation environment based on the performance benchmark model to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model; and performing simulation design in the virtual simulation environment based on the equipment status twin model to obtain equipment compensatory collaborative control parameters. The present invention achieves collaborative control of central air conditioning subsystems through simulation analysis of the digital twin model.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning simulation optimization, and in particular to a high-efficiency computer room multi-subsystem collaborative control method based on intelligent simulation. Background Art

[0002] Due to their high density of equipment and 24 / 7 operation, university computer rooms are typically energy intensive, relying heavily on uninterrupted power supply and robust air conditioning and refrigeration systems. The optimized design and intelligent operation strategy development of efficient computer room central air conditioning control systems have become core issues in improving system energy efficiency and operational intelligence, particularly through the use of advanced computer-aided design (CAD) and computer-aided engineering (CAE) technologies, as well as advanced modeling and simulation techniques.

[0003] Most existing computer-aided engineering and simulation optimization methods still lack mature solutions that can comprehensively balance multiple, even conflicting, performance indicators such as energy efficiency, cost, operational stability, and equipment loss within a unified computing and simulation framework, and generate and verify collaborative control logic based on these indicators.

[0004] Especially when the health status of some equipment in the system is abnormal, how to use simulation models to quickly evaluate system performance and analyze the impact of faults, and intelligently design, adjust or reconfigure control strategies to dispatch other healthy equipment resources for functional compensation, thereby maintaining the reliability and continuity of the overall system service, puts higher demands on existing computer-aided design and simulation technologies.

[0005] Therefore, an efficient multi-subsystem collaborative control method for computer rooms based on intelligent simulation is proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide an efficient computer room multi-subsystem collaborative control method based on intelligent simulation, including constructing a digital twin model and a virtual simulation environment for key central air-conditioning equipment; through the virtual simulation environment, simulating and analyzing the health status and operating performance of key equipment in the digital twin model to establish a performance benchmark model; collecting external data sources to obtain equipment operating parameters; in the virtual simulation environment, analyzing based on the equipment operating parameters and the performance benchmark model to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model; based on the equipment status twin model, performing simulation design in the virtual simulation environment to obtain equipment compensatory collaborative control parameters. The present invention realizes collaborative control of central air-conditioning subsystems through simulation analysis of the digital twin model.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] An efficient computer room multi-subsystem collaborative control method based on intelligent simulation, comprising:

[0009] Build a digital twin model of key central air-conditioning equipment and construct a virtual simulation environment based on the digital twin model;

[0010] Through the virtual simulation environment, the health status and operating performance of key equipment in the digital twin model are simulated and analyzed to establish a performance benchmark model;

[0011] Collecting external data sources to obtain equipment operating parameters; analyzing the equipment operating parameters and the performance benchmark model in the virtual simulation environment to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model;

[0012] Based on the equipment state twin model, the multi-subsystem collaborative control strategy is simulated and designed in a virtual simulation environment to obtain the equipment compensatory collaborative control parameters;

[0013] The subsystems of the central air conditioner are collaboratively controlled based on the equipment compensatory collaborative control parameters.

[0014] Using computer-aided design technology, the three-dimensional geometric structure, topological connection relationship, equipment materials and physical properties of the key equipment are defined to obtain a digital twin model;

[0015] A virtual simulation environment is constructed based on the digital twin model; the virtual simulation environment includes a database for simulating the thermodynamics and fluid dynamics of the central air-conditioning system, and a physical process solver for the heat and mass transfer process.

[0016] The performance benchmark model includes a health status unit and an operating performance unit;

[0017] The health status unit obtains health status parameters based on the equipment operating parameter identification; the health status parameters include the equipment's efficiency curve, output upper limit, internal resistance, and performance degradation characteristics; the performance degradation characteristics include energy efficiency coefficient fluctuations, vibration spectrum changes, and outlet temperature anomalies;

[0018] The operation performance unit generates operation performance parameters of the central air conditioner and the air conditioning subsystem under different combinations of key equipment health states based on the health status parameters of the key equipment; the operation performance parameters include energy efficiency distribution and total energy consumption;

[0019] The air conditioning subsystem includes an electrical subsystem, a cooling water subsystem, and a chilled water subsystem.

[0020] The dynamic update process of the digital twin model includes:

[0021] Updating the data of the digital twin model according to the health status parameter and the operating performance parameter to obtain a device status twin model;

[0022] All updated digital twin model versions, the health status parameters and the operating performance parameters, as well as the feedback confirmation information from the operation and maintenance personnel, are structured and stored in the device files in the computer.

[0023] The multi-subsystem collaborative control strategy includes: defining an objective function, setting variable constraints and iterative optimization strategy;

[0024] Define the objective function: The objective function integrates the operating energy efficiency, operating cost, cooling stability and equipment loss of the central air conditioning and air conditioning subsystems;

[0025] Set variable constraints: Set design variables and constraints. The design variables include the start-stop combinations of key equipment under different health states, the load percentage of each operating equipment, the operating frequency of variable-frequency equipment, the opening of electric valves, and system-level temperature and pressure differential set points. The constraints include the safe operating range of key equipment.

[0026] Iterative optimization strategy: Through the optimization engine, the combination of design variables is repeatedly adjusted in the virtual simulation environment, the equipment state twin model is called to simulate and predict the system performance, and each simulation result is evaluated according to the objective function and constraints. The system gradually searches and converges to a set of design variable values ​​that can optimize the objective function, forming the equipment compensatory collaborative control parameters.

[0027] The optimization engine is obtained by integrating the global optimization algorithm;

[0028] When abnormalities occur in the health status parameters and operating performance parameters of key equipment in the equipment status twin model, the optimization engine automatically adjusts the penalty weight of the key equipment in the objective function during the iteration process, dynamically tightens the safe operating range constraints of the key equipment, and coordinates the operating parameters of related subsystems to compensate for the overall supply demand.

[0029] The equipment compensatory collaborative control parameters include logical control instructions designed for each central air-conditioning key equipment and central air-conditioning subsystem, trigger conditions for the effectiveness of the strategy, adjustment data of equipment parameters, expected energy efficiency improvement and cost saving estimates, and design version number and timestamp for generating equipment compensatory collaborative control parameters.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. Significantly improve operational energy efficiency: By building a high-precision digital twin model and virtual simulation environment, multiple subsystems are collaboratively optimized, breaking down the barriers to independent operation of each subsystem and finding the global optimal operating point, thereby minimizing the total energy consumption of the central air-conditioning system.

[0032] 2. Enhance system reliability and equipment life: By real-time monitoring and evaluation of the health status of key equipment and considering equipment performance degradation and potential risks in the control strategy, compensatory control can be achieved to avoid long-term operation of equipment under adverse conditions, thereby improving the reliability of the overall system operation and extending the service life of the equipment.

[0033] 3. Improved control adaptability and intelligence: Digital twin models can be dynamically updated based on actual operating data, ensuring that control strategies are always optimized based on the latest equipment status. Combined with the application of algorithms, the system also has self-learning and self-optimization capabilities, able to adapt to various dynamic factors such as environmental changes, load fluctuations, and equipment aging.

[0034] 4. Reduced Operation and Maintenance Costs and Complexity: Intelligent simulation predictions and health status assessments provide early warning of potential failures, assisting in developing more scientific maintenance plans. Automated collaborative optimization control reduces the frequency and complexity of manual intervention and reduces the reliance on the experience of operation and maintenance personnel.

[0035] 5. Provide decision support and transparency: Structured control parameter output includes expected benefit evaluation and strategy version information, making the control decision process more transparent and easier to manage and verify the effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of a flow chart of an efficient computer room multi-subsystem collaborative control method based on intelligent simulation of the present invention;

[0037] Figure 2 is a schematic structural diagram of the performance benchmark model of the present invention;

[0038] Figure 3 Schematic diagram of the process of obtaining the compensatory collaborative control parameters of the equipment according to the present invention. DETAILED DESCRIPTION

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0040] Example 1

[0041] The present invention provides an efficient computer room multi-subsystem collaborative control method based on intelligent simulation, the process is as follows Figure 1 As shown, including:

[0042] Build a digital twin model of key central air-conditioning equipment and construct a virtual simulation environment based on the digital twin model;

[0043] Through the virtual simulation environment, the health status and operating performance of key equipment in the digital twin model are simulated and analyzed to establish a performance benchmark model;

[0044] Collecting external data sources to obtain equipment operating parameters; analyzing the equipment operating parameters and the performance benchmark model in the virtual simulation environment to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model;

[0045] Based on the equipment state twin model, the multi-subsystem collaborative control strategy is simulated and designed in a virtual simulation environment to obtain the equipment compensatory collaborative control parameters;

[0046] The subsystems of the central air conditioner are collaboratively controlled based on the equipment compensatory collaborative control parameters.

[0047] Furthermore, by using computer-aided design technology, the three-dimensional geometric structure, topological connection relationship, equipment materials and physical properties of the key equipment are defined to obtain a digital twin model;

[0048] A virtual simulation environment is constructed based on the digital twin model; the virtual simulation environment includes a database for simulating the thermodynamics and fluid dynamics of the central air-conditioning system, and a physical process solver for the heat and mass transfer process.

[0049] The key energy-consuming equipment in the high-efficiency computer room, including but not limited to chillers, chilled water pumps, cooling water pumps, cooling tower fans, main heat exchangers, etc., are selected as modeling objects.

[0050] Using computer-aided design (CAD) software (such as AutoCAD and Revit) or building information modeling (BIM) data, obtain the 3D geometry of the equipment. Define the topological connections between the equipment, such as how the chiller's evaporator connects to the chilled water circulation piping, and how the condenser connects to the cooling water circulation piping. Based on the equipment nameplate, design manual, or material library, assign accurate material and physical properties to each component. Material properties include, but are not limited to, metal thermal conductivity, fluid density, and specific heat capacity. Physical properties include, but are not limited to, pump and fan characteristic curves and heat exchanger heat transfer coefficient models. This information together constitutes the initial version of the digital twin model.

[0051] Construct a virtual simulation environment, which includes professional databases (e.g., refrigerant property libraries, water and air property libraries, equipment performance curve libraries) for simulating the thermodynamic characteristics (e.g., refrigeration cycle, heat exchange process) and fluid dynamic characteristics (e.g., water flow resistance, pressure distribution) of central air-conditioning systems, as well as physical process solvers capable of solving complex heat and mass transfer processes, such as solution engines based on finite element method, finite volume method, or empirical models.

[0052] By constructing a digital twin model and a virtual simulation environment, the present invention ensures the physical authenticity of the digital twin model and the professionalism of the simulation environment, laying the foundation for subsequent precise simulation analysis and optimization, enabling the model to accurately reflect the complex physical behavior of the real system.

[0053] The structure of the performance benchmark model is as follows Figure 2 As shown, it includes a health status unit and an operation performance unit;

[0054] The health status unit obtains health status parameters based on the equipment operating parameter identification; the health status parameters include the equipment's efficiency curve, output upper limit, internal resistance, and performance degradation characteristics; the performance degradation characteristics include energy efficiency coefficient fluctuations, vibration spectrum changes, and outlet temperature anomalies;

[0055] The operation performance unit generates operation performance parameters of the central air conditioner and the air conditioning subsystem under different combinations of key equipment health states based on the health status parameters of the key equipment; the operation performance parameters include energy efficiency distribution and total energy consumption;

[0056] The air conditioning subsystem includes an electrical subsystem, a cooling water subsystem, and a chilled water subsystem.

[0057] Furthermore, the equipment operating parameters are operating parameters obtained by monitoring key equipment and subsystems during the operation of the central air conditioner through building automation systems (BAS), Internet of Things (IoT) sensors or manual inspections.

[0058] The health status unit identifies equipment status parameters based on equipment operating parameters through statistical analysis, mechanism modeling, or machine learning methods. These equipment status parameters include, but are not limited to, equipment efficiency curves (e.g., a chiller's COP curve showing changes with load factor, cooling water inlet temperature, etc.), output upper limit (maximum output upper limit under rated operating conditions), internal resistance (equipment internal fluid resistance characteristics, e.g., pump head curve, heat exchanger water resistance characteristics), and performance degradation characteristics (e.g., the time-varying pattern of heat exchanger fouling thermal resistance, compressor efficiency degradation model). Machine learning models include support vector machines (SVMs) and neural networks. The COP refers to the energy efficiency ratio of key equipment.

[0059] Based on the current health status parameters of key equipment, the operational performance unit generates operational performance parameters for the central air conditioning system and its various air conditioning subsystems (such as the electrical subsystem, cooling water subsystem, and chilled water subsystem) under different combinations of key equipment health statuses by invoking a physical process solver and database within a virtual simulation environment. These operational performance parameters include, but are not limited to, energy efficiency distribution and total energy consumption; specifically, they include real-time energy efficiency distribution diagrams for the system or subsystem, predicted total energy consumption under specific operating conditions, and the energy consumption contribution of each device.

[0060] This invention constructs a performance baseline model, a dynamic model that reflects equipment health and performance degradation. The introduction of a health status unit enables the system to quantitatively assess equipment health, providing a basis for predictive maintenance and compensatory control. Based on this health status, the operational performance unit accurately predicts the system's performance under current conditions, improving the accuracy of subsequent control decisions.

[0061] The dynamic update process of the digital twin model includes:

[0062] Updating the data of the digital twin model according to the health status parameter and the operating performance parameter to obtain a device status twin model;

[0063] All updated digital twin model versions, the health status parameters and the operating performance parameters, as well as the feedback confirmation information from the operation and maintenance personnel, are structured and stored in the device files in the computer.

[0064] Furthermore, the dynamic update process of the digital twin model includes updating and calibrating the relevant parameters of the constructed digital twin model (such as the efficiency curve correction coefficient, internal resistance correction coefficient, output upper limit adjustment value, etc.) based on the obtained operating performance parameters and the current health status parameters obtained by the health status unit analysis, so as to obtain an equipment status twin model that can more accurately reflect the current actual condition of the equipment.

[0065] All updated digital twin model versions, corresponding health status parameters, operating performance parameters, and feedback confirmation information from operation and maintenance personnel on control effects or equipment status, such as records of actual energy-saving effects after a control parameter adjustment, or status confirmation after equipment maintenance, are structured and stored in the equipment archive database in the computer.

[0066] This invention empowers the digital twin model with the ability to self-evolve and continuously optimize through model modification. Continuous comparison and calibration with actual operating data ensures the fidelity of the twin model. The establishment of equipment archives provides data accumulation for historical tracing, fault diagnosis, and further optimization of models and control strategies using machine learning methods.

[0067] The multi-subsystem collaborative control strategy includes: defining an objective function, setting variable constraints and iterative optimization strategy;

[0068] Define the objective function: The objective function integrates the operating energy efficiency, operating cost, cooling stability and equipment loss of the central air conditioning and air conditioning subsystems;

[0069] Set variable constraints: Set design variables and constraints. The design variables include the start-stop combinations of key equipment under different health states, the load percentage of each operating equipment, the operating frequency of variable-frequency equipment, the opening of electric valves, and system-level temperature and pressure differential set points. The constraints include the safe operating range of key equipment.

[0070] Iterative optimization strategy: Through the optimization engine, the combination of design variables is repeatedly adjusted in the virtual simulation environment, the equipment state twin model is called to simulate and predict the system performance, and each simulation result is evaluated according to the objective function and constraints. The system gradually searches and converges to a set of design variable values ​​that can optimize the objective function, forming the equipment compensatory collaborative control parameters.

[0071] The process of obtaining the device compensatory collaborative control parameters is as follows: Figure 3 shown.

[0072] The objective function is a comprehensive evaluation metric designed to minimize (or maximize) one or more objectives. Examples include the total energy consumption of the central air conditioning system, operating costs (which may factor in electricity prices), cooling stability indicators (such as the average deviation of terminal temperature or the probability of exceeding the standard), and equipment loss assessment values ​​(which can be quantified based on equipment operating load, number of starts and stops, and the degree of deviation from recommended operating conditions).

[0073] The design variables include, but are not limited to, start-stop combination decisions for key equipment (such as chillers, water pumps, and cooling tower fans) in different health states; the load percentage of each operating equipment (such as chiller load rate); the operating frequency of variable-frequency equipment (such as water pump frequency and fan frequency); the opening degree of electric valves (such as cooling water valves, chilled water bypass valves, and terminal regulating valves); and key system-level set points, such as the temperature difference between the chilled water supply and return mains, the temperature difference between the cooling water supply and return mains, and the chilled water or cooling water system pressure set point.

[0074] The constraints include the safe operating range of each device; for example, the lower limit of the chiller evaporation temperature, the upper limit of the condensing pressure, the upper limit of the motor current, the water pump NPSH requirement, the process parameter range (such as the chilled water outlet temperature range, the lower limit of the terminal cooling demand satisfaction), and other operating specifications.

[0075] The iterative optimization strategy uses an optimization engine in a virtual simulation environment, using the aforementioned design variables as adjustable parameters and the device state twin model as the simulation object. The optimization engine repeatedly adjusts the combination of design variables, calling the device state twin model after each adjustment to simulate and predict system performance and obtain the corresponding objective function value. Each simulation result is evaluated based on the objective function value and the satisfaction of the constraints, gradually searching for and converging on a set of design variable values ​​that optimize (or achieve a satisfactory level of) the objective function. These values ​​constitute the device compensatory collaborative control parameters.

[0076] By defining comprehensive objective functions and design variables and imposing actual operational constraints, the present invention utilizes an iterative optimization algorithm to systematically find the control parameter combination that optimizes the overall system performance while satisfying all constraints. This avoids the limitations of manual trial and error or local optimization and provides decision-making support for achieving collaborative control of multiple subsystems.

[0077] The optimization engine is obtained by integrating the global optimization algorithm;

[0078] When abnormalities occur in the health status parameters and operating performance parameters of key equipment in the equipment status twin model, the optimization engine automatically adjusts the penalty weight of the key equipment in the objective function during the iteration process, dynamically tightens the safe operating range constraints of the key equipment, and coordinates the operating parameters of related subsystems to compensate for the overall supply demand.

[0079] Furthermore, the optimization engine is obtained by integrating one or more global optimization algorithms (such as multi-island genetic algorithm, differential evolution algorithm, Bayesian optimization, particle swarm optimization algorithm, etc.) to enhance the ability to search for the global optimal solution.

[0080] The integrated approach includes parallel independent operation, combinatorial optimization and weighted combination;

[0081] The parallel independent operation: multiple different optimization algorithms or different parameter configurations of the same algorithm run independently at the same time, and the global optimal solution is selected from the optimal solutions obtained by all algorithms; the combined optimization: based on the characteristics of the problem or the current search status, the most appropriate algorithm is dynamically selected to perform the next step of optimization; the weighted combination: multiple candidate solutions can be output through multiple algorithms, and different solutions are weightedly combined.

[0082] When abnormalities occur in the health status parameters and operating performance parameters of key equipment in the equipment status twin model, for example, the efficiency obtained by the health status unit evaluation decreases significantly, the internal resistance increases, and the performance predicted by the simulation based on the operating performance parameters cannot meet the requirements; the optimization engine will automatically adjust the penalty weight of the key equipment in the objective function during the iteration process; for example, if the health of a chiller decreases, its weight in the energy-saving target can be appropriately reduced, and more emphasis can be placed on ensuring its stable operation or reducing its losses, or dynamically tightening the safe operating range constraints of the key equipment; for example, stricter temporary constraints can be imposed on its output upper limit, operating current upper limit, etc., to prioritize the reliability and operational safety of the key equipment.

[0083] At the same time, the optimization engine will coordinate operating parameters of other related subsystems, such as increasing the output of other healthy chillers, adjusting the water pump frequency to adapt to the changed system resistance, etc., to compensate for the overall supply capacity shortage caused by the performance degradation of abnormal equipment, and ensure that the overall cooling or other service needs of the system are met.

[0084] The present invention applies a global optimization algorithm to improve the probability of finding the truly optimal solution. When the performance of some equipment deteriorates, the system can not only perceive it, but also actively adjust the optimization strategy, ensuring the safety of the faulty equipment by sacrificing some economic efficiency, and compensating for the lost performance by mobilizing the potential of other healthy equipment. In this way, the service capability of the overall system is maintained while ensuring the reliability of key equipment, reflecting a high degree of intelligence and adaptability.

[0085] The equipment compensatory collaborative control parameters include logical control instructions designed for each central air-conditioning key equipment and central air-conditioning subsystem, trigger conditions for the effectiveness of the strategy, adjustment data of equipment parameters, expected energy efficiency improvement and cost saving estimates, and design version number and timestamp for generating equipment compensatory collaborative control parameters.

[0086] Furthermore, the output device compensatory collaborative control parameters are in a standardized, computer-parseable structured data format (such as JSON, XML, or a specific protocol format).

[0087] The equipment compensatory collaborative control parameters specifically include, but are not limited to: logical control instructions (such as start, stop, and mode switching) designed for each key central air-conditioning equipment (such as chillers, refrigeration pumps) and central air-conditioning subsystems (such as the pressure difference setting of the chilled water system main pipe), preset trigger conditions for the policy to take effect (for example, when the outdoor temperature is greater than X and the total cooling load is greater than Y), adjustment data for specific equipment parameters (for example, the chiller load percentage is set to 80%, the water pump frequency is set to 45Hz, and the chilled water supply temperature is set to 7°C), the estimated energy efficiency improvement and cost savings expected after adoption, and the design version number and timestamp for generating the equipment compensatory collaborative control parameters.

[0088] The present invention defines in detail the specific contents of the equipment compensatory collaborative control parameters, such as logical instructions, trigger conditions, adjustment data, expected benefits, version numbers, etc., so that the optimization results can be output in a clear, executable and traceable form; it facilitates the deployment, execution, effect evaluation and version management of control strategies, and ensures the effective transformation from simulation optimization to actual control.

[0089] Example 2:

[0090] The present invention proposes an efficient computer room multi-subsystem collaborative control method based on intelligent simulation, comprising:

[0091] Build a digital twin model of key central air-conditioning equipment and construct a virtual simulation environment based on the digital twin model;

[0092] Through the virtual simulation environment, the health status and operating performance of key equipment in the digital twin model are simulated and analyzed to establish a performance benchmark model;

[0093] Collecting external data sources to obtain equipment operating parameters; analyzing the equipment operating parameters and the performance benchmark model in the virtual simulation environment to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model;

[0094] Based on the equipment state twin model, the multi-subsystem collaborative control strategy is simulated and designed in a virtual simulation environment to obtain the equipment compensatory collaborative control parameters;

[0095] The subsystems of the central air conditioner are collaboratively controlled based on the equipment compensatory collaborative control parameters.

[0096] Using computer-aided design technology, the three-dimensional geometric structure, topological connection relationship, equipment materials and physical properties of the key equipment are defined to obtain a digital twin model;

[0097] A virtual simulation environment is constructed based on the digital twin model; the virtual simulation environment includes a database for simulating the thermodynamics and fluid dynamics of the central air-conditioning system, and a physical process solver for the heat and mass transfer process.

[0098] By constructing a digital twin model and a virtual simulation environment, the present invention ensures the physical authenticity of the digital twin model and the professionalism of the simulation environment, laying the foundation for subsequent precise simulation analysis and optimization, enabling the model to accurately reflect the complex physical behavior of the real system.

[0099] The performance benchmark model includes a health status unit and an operating performance unit;

[0100] The health status unit obtains health status parameters based on the equipment operating parameter identification; the health status parameters include the equipment's efficiency curve, output upper limit, internal resistance, and performance degradation characteristics; the performance degradation characteristics include energy efficiency coefficient fluctuations, vibration spectrum changes, and outlet temperature anomalies;

[0101] The operation performance unit generates operation performance parameters of the central air conditioner and the air conditioning subsystem under different combinations of key equipment health states based on the health status parameters of the key equipment; the operation performance parameters include energy efficiency distribution and total energy consumption;

[0102] The air conditioning subsystem includes an electrical subsystem, a cooling water subsystem, and a chilled water subsystem.

[0103] The initial digital twin model is simulated and analyzed through a virtual simulation environment.

[0104] Data collection and preprocessing:

[0105] For new equipment, its initial health state can be considered ideal, and performance parameters can be directly taken from design values ​​or factory test data. For example, the COP curve of a chiller, the efficiency curve of a water pump, and the output limit.

[0106] For operating equipment, collect historical operating data and maintenance records. Analyze this data to identify equipment performance at different stages of operation. For example, by analyzing the actual COP of a chiller at different load rates and cooling water temperatures and comparing it with the designed COP, you can determine initial performance degradation characteristics.

[0107] Virtual simulation analysis:

[0108] Using the constructed digital twin model, simulation experiments are conducted in a virtual simulation environment. The equipment's performance in an "ideal healthy state" is simulated under different operating conditions (such as varying outdoor temperatures and load demands), and its efficiency curve, output, and energy consumption are recorded. These simulation results form the baseline efficiency curve and output limit for the healthy state unit, as well as the baseline energy efficiency distribution and total energy consumption for the operating performance unit.

[0109] Calibration with real data:

[0110] By collecting data from external data sources, the actual operating parameters of each key equipment are obtained from the BAS / SCADA system of the high-efficiency computer room in real time or periodically as equipment operating parameters, such as the inlet and outlet water temperature, flow rate, and power of the chiller, the operating frequency, current, inlet and outlet pressure of the water pump, the frequency and power of the cooling tower fan, and the outdoor dry-bulb and wet-bulb temperatures.

[0111] Based on the equipment operating parameters, the equipment status parameters are identified through statistical analysis, mechanism modeling or machine learning methods; the equipment status parameters include but are not limited to the equipment's efficiency curve (for example, a curve showing the COP of a chiller changing with load rate, cooling water inlet temperature, etc.), output upper limit (maximum output upper limit under rated operating conditions), internal resistance (fluid resistance characteristics within the equipment, for example, water pump head curve, heat exchanger water resistance characteristics), and performance degradation characteristics (for example, the variation pattern of heat exchanger fouling thermal resistance over time, compressor efficiency degradation model).

[0112] Based on the current health status parameters of key equipment, the system generates operating performance parameters for the central air conditioning system and its various subsystems (such as the electrical subsystem, cooling water subsystem, and chilled water subsystem) under different combinations of key equipment health statuses by invoking a physical process solver and database within a virtual simulation environment. These operating performance parameters include, but are not limited to, energy efficiency distribution and total energy consumption; specifically, they include real-time energy efficiency distribution diagrams for the system or subsystem, predicted total energy consumption under specific operating conditions, and the energy consumption contribution of each device.

[0113] The identified actual health parameters and performance degradation characteristics are fed back into the performance benchmark model to calibrate and modify the parameters. For example, if the COP degradation rate of a particular chiller model is found to be generally faster than the initial benchmark setting during actual operation, the degradation model parameters of that chiller model in the benchmark model will be adjusted.

[0114] The core purpose of the performance benchmark model is to establish a reference system for evaluating the ideal or expected operating performance of key central air-conditioning equipment under different health conditions, and serve as the basis for subsequent dynamic updates of digital twin models, health status assessments, and control strategy optimization. It can also evolve as the understanding of the system deepens and data accumulates.

[0115] The performance benchmark model can be regarded as a combination of a structured model library and a database, which is used to identify and dynamically optimize the performance of key equipment, subsystems and the entire central air-conditioning system.

[0116] This invention constructs a performance baseline model, a dynamic model that reflects equipment health and performance degradation. The introduction of a health status unit enables the system to quantitatively assess equipment health, providing a basis for predictive maintenance and compensatory control. Based on this health status, the operational performance unit accurately predicts the system's performance under current conditions, improving the accuracy of subsequent control decisions.

[0117] The dynamic update process of the digital twin model includes:

[0118] Updating the data of the digital twin model according to the health status parameter and the operating performance parameter to obtain a device status twin model;

[0119] During the dynamic update process of the digital twin model, the health status parameters identified by the health status unit (for example, the energy efficiency coefficient fluctuation percentage, the amplitude change at a specific vibration frequency, and the deviation of the outlet temperature from the baseline model) will be quantified and mapped into adjustment values ​​for the specific equipment characteristic parameters in the digital twin model through a conversion rule library trained based on historical data.

[0120] For example, if a chiller's actual COP is detected to be 15% lower than the baseline model's expected COP under current operating conditions, and its internal sensors detect vibrations exceeding a warning threshold at a specific frequency, the health status unit might comprehensively determine that a compressor component is worn or a small refrigerant leak is occurring. This determination triggers a corresponding downward adjustment to the device's isentropic efficiency curve in the digital twin model, or fine-tuning the virtual charge level of its refrigerant cycle module. This ensures that the resulting device status twin model more accurately reflects the device's current health and performance boundaries.

[0121] All updated digital twin model versions, the health status parameters and the operating performance parameters, as well as the feedback confirmation information from the operation and maintenance personnel, are structured and stored in the device files in the computer.

[0122] This invention empowers the digital twin model with the ability to self-evolve and continuously optimize through model modification. Continuous comparison and calibration with actual operating data ensures the fidelity of the twin model. The establishment of equipment archives provides valuable data for historical tracing, fault diagnosis, and further optimization of models and control strategies using machine learning methods.

[0123] The multi-subsystem collaborative control strategy includes: defining an objective function, setting variable constraints and iterative optimization strategy;

[0124] Define the objective function: The objective function integrates the operating energy efficiency, operating cost, cooling stability and equipment loss of the central air conditioning and air conditioning subsystems;

[0125] Set variable constraints: Set design variables and constraints. The design variables include the start-stop combinations of key equipment under different health states, the load percentage of each operating equipment, the operating frequency of variable-frequency equipment, the opening of electric valves, and system-level temperature and pressure differential set points. The constraints include the safe operating range of key equipment.

[0126] Iterative optimization strategy: Through the optimization engine, the combination of design variables is repeatedly adjusted in the virtual simulation environment, the equipment state twin model is called to simulate and predict the system performance, and each simulation result is evaluated according to the objective function and constraints. The system gradually searches and converges to a set of design variable values ​​that can optimize the objective function, forming the equipment compensatory collaborative control parameters.

[0127] Define the objective function: For example, set the objective function to minimize total energy consumption while considering the cooling stability penalty.

[0128] Set the constraint variables:

[0129] Design variables: start / stop status of chillers A and B (0 or 1), chiller A load factor (30%-100%), chiller B load factor (30%-100%), chilled water pump P1 frequency (30Hz-50Hz), cooling water pump P2 frequency (30Hz-50Hz), cooling tower fan F1 frequency (25Hz-50Hz), chilled water main supply water temperature set point (6℃-8℃).

[0130] Constraints: Chiller evaporation temperature > 3°C, condensing pressure < 1.8 MPa, total cooling capacity ≥ current total cooling load demand, power of each device ≤ rated power, etc.

[0131] Iterative optimization strategy: Particle swarm optimization (PSO) algorithm is selected as the optimization engine.

[0132] Initialize a group of particles, each particle represents a combination of a set of design variables.

[0133] For each particle (i.e., each set of control parameter combinations), the device state twin model is called to perform system performance simulation prediction in a virtual simulation environment to calculate the value of the objective function.

[0134] According to the rules of the PSO algorithm, the speed and position of the particles are updated to search for a better solution space.

[0135] Exception Handling and Compensation (Key Point): Suppose during the iteration process, the health parameters of chiller A indicate that its COP has significantly decreased by 20%. At this point, the optimization engine automatically triggers compensation logic: increasing the weight of chiller A's losses in the objective function or temporarily lowering its maximum load factor from 100% to 70% in the constraints. Simultaneously, the optimization process may favor increasing the load factor of chiller B, activating a backup chiller, and adjusting the pump and fan frequencies to compensate for the cooling capacity shortfall caused by chiller A's performance degradation, ensuring that the total cooling load demand is met.

[0136] Iterations are repeated until a preset convergence condition is reached (e.g., the objective function value change is less than a threshold or the maximum number of iterations is reached). The optimal (or satisfactory) set of design variable values ​​that are finally converged to constitute the equipment compensatory collaborative control parameters.

[0137] By defining comprehensive objective functions and design variables and imposing actual operational constraints, the present invention utilizes an iterative optimization algorithm to systematically find the control parameter combination that optimizes the overall system performance while satisfying all constraints. This avoids the limitations of manual trial and error or local optimization and provides decision-making support for achieving collaborative control of multiple subsystems.

[0138] The optimization engine is obtained by integrating the global optimization algorithm;

[0139] When abnormalities occur in the health status parameters and operating performance parameters of key equipment in the equipment status twin model, the optimization engine automatically adjusts the penalty weight of the key equipment in the objective function during the iteration process, dynamically tightens the safe operating range constraints of the key equipment, and coordinates the operating parameters of related subsystems to compensate for the overall supply demand.

[0140] Automatic adjustment of penalty weights: For example, if the health status and operating performance of a chiller fall below a preset threshold, the weight of the corresponding equipment loss penalty term in the overall objective function will be automatically adjusted from the default value. The extent of the adjustment is determined by the degree of deviation of the key equipment's own performance from the preset threshold, thereby placing greater emphasis on reducing the operating intensity of the unhealthy equipment during optimization.

[0141] Dynamic tightening of safe operating range constraints: For example, if a pump's vibration intensity is assessed to be exceeding the specified limit, indicating a risk of cavitation, the optimization engine not only adjusts its weight in the objective function but also queries the device state twin associated with the pump to obtain a recommended temporary upper limit for its current state, which it then uses as an immediate constraint for this optimization iteration. These adjustment strategies and parameter thresholds can be pre-defined by domain experts and validated and optimized through analysis of historical operating data and simulation experiments.

[0142] The equipment compensatory collaborative control parameters include logical control instructions designed for each central air-conditioning key equipment and central air-conditioning subsystem, trigger conditions for the effectiveness of the strategy, adjustment data of equipment parameters, expected energy efficiency improvement and cost saving estimates, and design version number and timestamp for generating equipment compensatory collaborative control parameters.

[0143] The obtained equipment compensatory collaborative control parameters are output in a predefined structured data format, and actual regulation and control are performed on each subsystem equipment of the central air-conditioning (chillers, water pumps, valves, etc.).

[0144] Self-learning and optimization:

[0145] By utilizing the historical digital twin model version sequences stored in the equipment archive, the corresponding health status parameters, operating performance parameters, and performance feedback data of actual good control parameters (i.e., those control strategies that have been verified to be effective), ensemble learning (such as RandomForest and GradientBoosting) or reinforcement learning (such as Q-learning and DeepQ-Network) algorithms can be used to self-learn and optimize the following two aspects:

[0146] Parameter correction logic for digital twin models: For example, learning the deviation patterns between actual performance parameters in historical data and model-predicted parameters, and optimizing the parameter correction algorithm to enable it to approach the actual device status faster and more accurately.

[0147] Performance benchmark model degradation patterns: For example, by analyzing large amounts of long-term equipment operating data and health status evolution, the accuracy of health status units in predicting equipment performance degradation characteristics (such as heat exchanger fouling accumulation rate models and compressor wear models) can be learned and optimized. This self-learning and optimization process can be performed periodically offline or incrementally online, continuously improving the accuracy of subsequent digital twin model updates and the long-term effectiveness and predictability of recommended collaborative control strategies.

[0148] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An efficient computer room multi-subsystem collaborative control method based on intelligent simulation, characterized in that: include: Build digital twin models of key central air conditioning equipment, including chillers, chilled water pumps, cooling water pumps, cooling tower fans, and heat exchangers; Building a virtual simulation environment based on the digital twin model; Through the virtual simulation environment, the health status and operating performance of key equipment in the digital twin model are simulated and analyzed to establish a performance benchmark model; Collecting external data sources to obtain equipment operating parameters; analyzing the equipment operating parameters and the performance benchmark model in the virtual simulation environment to obtain health status parameters and operating performance parameters; dynamically updating the digital twin model based on the health status parameters and operating performance parameters to obtain an equipment status twin model; Based on the equipment state twin model, the optimization engine is used in a virtual simulation environment to simulate and design the multi-subsystem collaborative control strategy, and the equipment compensatory collaborative control parameters are obtained; The optimization engine is obtained by integrating the global optimization algorithm; When abnormalities occur in the health status parameters and operating performance parameters of key equipment in the equipment status twin model, the optimization engine automatically adjusts the penalty weight of the key equipment in the objective function during the iteration process, dynamically tightens the safe operating range constraints of the key equipment, and coordinates the operating parameters of related subsystems to compensate for the overall supply demand; The automatic adjustment of the penalty weight includes: the adjustment range is determined according to the degree of deviation between the state performance of the key equipment itself and the preset threshold; The subsystems of the central air conditioner are collaboratively controlled based on the equipment compensatory collaborative control parameters.

2. The method for efficient multi-subsystem coordinated control of a computer room based on intelligent simulation according to claim 1 is characterized in that: Using computer-aided design technology, the three-dimensional geometric structure, topological connection relationship, equipment materials and physical properties of the key equipment are defined to obtain a digital twin model; A virtual simulation environment is constructed based on the digital twin model; the virtual simulation environment includes a database for simulating the thermodynamics and fluid dynamics of the central air-conditioning system, and a physical process solver for the heat and mass transfer process.

3. The method for efficient computer room multi-subsystem collaborative control based on intelligent simulation according to claim 1 is characterized by: The performance benchmark model includes a health status unit and an operating performance unit; The health status unit obtains health status parameters based on the equipment operation parameter identification; the health status parameters include the efficiency curve, output upper limit, internal resistance and performance degradation characteristics of the equipment; The performance degradation characteristics include fluctuations in energy efficiency coefficient, changes in vibration spectrum, and abnormal outlet temperature; The operation performance unit generates operation performance parameters of the central air conditioner and the air conditioning subsystem under different combinations of key equipment health states based on the health status parameters of the key equipment; the operation performance parameters include energy efficiency distribution and total energy consumption; The air conditioning subsystem includes an electrical subsystem, a cooling water subsystem, and a chilled water subsystem.

4. The method for efficient computer room multi-subsystem collaborative control based on intelligent simulation according to claim 1 is characterized in that: The dynamic update process of the digital twin model includes: Updating the data of the digital twin model according to the health status parameter and the operating performance parameter to obtain a device status twin model; All updated digital twin model versions, the health status parameters and the operating performance parameters, as well as the feedback confirmation information from the operation and maintenance personnel, are structured and stored in the device files in the computer.

5. The method for efficient multi-subsystem coordinated control of a computer room based on intelligent simulation according to claim 1 is characterized in that: The multi-subsystem collaborative control strategy includes: defining an objective function, setting variable constraints and iterative optimization strategy; Define the objective function: The objective function integrates the operating energy efficiency, operating cost, cooling stability and equipment loss of the central air conditioning and air conditioning subsystems; Set variable constraints: Set design variables and constraints. The design variables include the start-stop combinations of key equipment under different health states, the load percentage of each operating equipment, the operating frequency of variable-frequency equipment, the opening of electric valves, and system-level temperature and pressure differential set points. The constraints include the safe operating range of key equipment. Iterative optimization strategy: Through the optimization engine, the combination of design variables is repeatedly adjusted in the virtual simulation environment, the equipment state twin model is called to simulate and predict the system performance, and each simulation result is evaluated according to the objective function and constraints. The system gradually searches and converges to a set of design variable values ​​that can optimize the objective function, forming the equipment compensatory collaborative control parameters.

6. The method for efficient multi-subsystem coordinated control of a computer room based on intelligent simulation according to claim 1, characterized in that: The equipment compensatory collaborative control parameters include logical control instructions designed for each central air-conditioning key equipment and central air-conditioning subsystem, trigger conditions for the effectiveness of the strategy, adjustment data of equipment parameters, expected energy efficiency improvement and cost saving estimates, and design version number and timestamp for generating equipment compensatory collaborative control parameters.

Images