Air compressor equipment energy efficiency operation optimization method and device based on digital twinning

Abstract

The application discloses a kind of based on digital twinning air compressor equipment energy efficiency operation optimization method and device, it is related to air compressor equipment twin optimization technical field, wherein method includes: the air compressor exclusive digital twin model of integrated geometry topology, aerodynamic characteristics, motor load law and cooling heat exchange parameter is built;Sensor unit is arranged in each functional area of equipment, and multi-dimensional operating condition signal is collected;Signal standardization processing and time sequence are unified via data gateway, and compliant real-time data stream is generated;Based on real-time data driven model, carry out multi-physical field coupling simulation, synchronously solve unit gas energy consumption and key component thermal stress index, realize air compressor full-condition digital simulation research and accurate evaluation of operating state.The application builds air compressor multi-parameter coupling digital twin model, realizes equipment multi-physical field synchronous simulation deduction, solves the problem of low traditional model fitting degree, poor system linkage and large simulation deviation.

Description

Technical Field

[0001] This invention relates to the field of digital twin optimization technology for air compressor equipment, and in particular to a method and apparatus for optimizing the energy efficiency of air compressor equipment based on digital twins. Background Technology

[0002] The current energy efficiency monitoring modes for air compressor equipment are relatively simple, generally relying on various sensors deployed on-site to collect basic operating parameters of the air path, drive motor, and cooling system. Conventional processing methods are limited to on-site storage of raw parameters, simple filtering, and superficial comparison with operating conditions, resulting in insufficient depth of data utilization. Meanwhile, the application of digital twin technology in air compressor equipment scenarios is rather crude, mostly directly applying generalized mechanical equipment modeling frameworks without customizing the design to incorporate the unique operating mechanisms of air compressor compression, power matching, and heat exchange. This lacks deep integration of core, specific physical parameters such as the compressor's aerodynamic characteristics, drive motor load efficiency changes, and cooling system heat exchange patterns, leading to poor model fit and scenario adaptability.

[0003] The simultaneous acquisition of massive amounts of raw operational signals by multiple sensors of various types and in multiple areas on-site, originating from different acquisition terminals and transmission links, resulted in inconsistent formats, inconsistent encoding standards, and disordered acquisition timing. The lack of a unified standardized integration and timing correction mechanism led to fragmentation of the multi-source data. Operational data that has not undergone collaborative processing is difficult to align accurately with the unified time reference of the digital twin model, highlighting significant data integration barriers.

[0004] Building upon this, traditional digital twin architectures in application scenarios can only visualize the external structure and basic operating parameters of air compressors. They struggle to perform multi-dimensional, refined physical field coupling simulations based on the equipment's specific physical parameters, and cannot simultaneously calculate the energy consumption per unit output of compressed air and the thermal stress distribution of core mechanical components under actual operating conditions. Energy efficiency accounting analysis, motor loss monitoring, and equipment thermal status assessment are fragmented and independent, with various operating indicators unable to be linked and coupled for analysis. The resulting operational assessment is one-sided and simplistic, failing to accurately reflect the actual operating quality and long-term service risks of the air compressor.

[0005] In summary, traditional modeling methods, data processing modes, and simulation analysis systems have significant technical shortcomings, which hinder the advancement of refined energy efficiency management and integrated monitoring of equipment safety status for air compressors. There is an urgent need to build a digital twin modeling system and a multi-source sensor data collaborative processing solution that are adapted to the specific operating characteristics of air compressors. Summary of the Invention

[0006] The main objective of this invention is to provide a method for optimizing the energy efficiency of air compressor equipment based on digital twins.

[0007] Another objective of this invention is to propose an energy efficiency optimization device for air compressor equipment based on digital twins.

[0008] The third objective of this invention is to provide an electronic device.

[0009] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0010] To achieve the above objectives, a first aspect of the present invention proposes a method for optimizing the energy efficiency of air compressor equipment based on digital twins, comprising:

[0011] Construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model shall at least cover the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system. Based on the equipment pipelines, power components and cooling structure covered by the digital twin model, several sensor acquisition units are deployed in the corresponding functional areas of the air compressor to acquire multi-dimensional original operating condition signals during equipment operation. The original operating condition signals are standardized and time-series coordinated and corrected by the data gateway to achieve data format normalization and time reference unification, forming a standardized operating data stream that is compatible with digital twin access applications. The real-time operating status data stream is imported into the digital twin model. Based on the real-time data, the digital twin model is driven to carry out multi-physics field coupled simulation calculations. The comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of the core components are output, and the digital simulation and operating status assessment of the air compressor under all working conditions are completed.

[0012] Optionally, construct a digital twin model of the physical air compressor equipment, including: The physical structure of the physical air compressor is comprehensively surveyed and digitally modeled to fully restore the layout of the equipment and the connection of components. The geometric topology corresponding to the air compressor body is constructed and solidified, and the assembly association and spatial constraint relationship of each mechanical component are determined. Combining the equipment's factory calibration data, on-site test results, and component performance files, we determined the aerodynamic characteristic mapping table corresponding to the compressor host's full range of operating conditions, the efficiency-load characteristic curve corresponding to the drive motor's full load range, and the various heat exchange parameters of the cooling system to match the operating conditions. Based on the internal medium flow path, mechanical transmission relationship and energy transfer logic of the physical air compressor, the geometric topology, main unit aerodynamic characteristics, motor efficiency load characteristics and cooling system heat exchange parameters are integrated and coupled across dimensions to form an integrated digital twin model with the ability to receive real-time parameters, respond to multi-physics fields and output operating performance.

[0013] Optionally, based on the equipment piping, power components, and cooling structure covered by the digital twin model, several sensor acquisition units are deployed in the corresponding functional areas of the air compressor to acquire multi-dimensional raw operating condition signals during equipment operation, including: Based on the air pipeline structure, drive motor power components and cooling system circulation structure covered by the digital twin model, the target operation monitoring points of the air compressor are delineated, and the corresponding operation parameter collection requirements for each monitoring point are determined. For different functional areas including intake branch, exhaust branch, motor power input terminal and cooling water circulation circuit, sensor acquisition units adapted to the on-site working environment are deployed in different zones. By utilizing sensor acquisition units deployed throughout the entire area, basic signals such as pressure, temperature, power, and fluid are continuously collected during the operation of the equipment, and multi-dimensional original operating condition signals that fully reflect the real-time operating status of the equipment are continuously accumulated and output.

[0014] Optionally, the original operating condition signals are standardized and time-series coordinated and corrected through a data gateway to achieve data format normalization and time base unification, forming a standardized operating data stream adapted for digital twin access applications, including: By centrally accessing multi-source heterogeneous raw operating condition signals uploaded by sensor acquisition units through data gateways deployed across the entire domain, the system can uniformly collect and transmit dispersed data and integrate multi-link acquired data. To address the differences in format, encoding, and transmission protocols among different types of sensor signals, we will carry out unified format standardization and data cleaning to unify the output specifications of multi-source signals. Using the unified time reference preset by the digital twin model as the calibration basis, all acquired signals are subjected to time-series collaborative correction and global timestamp synchronization processing to eliminate the time-series misalignment problem caused by multi-node acquisition and generate a time-consistent, format-unified, and standardized running data stream that is compatible with the digital twin model call.

[0015] Optionally, the real-time runtime data stream can be imported into the digital twin model, and the digital twin model can be driven to perform multiphysics coupling simulation calculations based on the real-time data, including: The real-time operating parameters in the standardized operating data stream are used as the input boundary conditions of the digital twin model to drive the digital twin model to synchronously match the current operating status of the physical air compressor equipment. Based on the aerodynamic characteristic mapping table of the compressor host in the digital twin model, aerodynamic and thermodynamic simulation calculations are performed on the compression process to generate compression performance characterization results that match the current operating conditions. Based on the drive motor efficiency-load characteristic curve built into the digital twin model, the energy conversion process of the drive motor in the current load range is analyzed, and the corresponding characterization results of power input conditions and energy efficiency changes are generated. Based on the heat exchange parameters of the cooling system built into the digital twin model, the heat exchange process of the air compressor cooling structure under the current operating conditions is calculated, and the effect of the cooling process on the overall thermal field of the equipment is quantified. By combining the inherent physical connection relationship of the equipment with the global energy transfer logic, the simulation results of the compression process, the power input simulation results, and the cooling process simulation results are jointly solved to form a global multi-physics field coupled simulation result.

[0016] Optionally, the comprehensive assessment results of the unit gas production energy consumption and the thermal stress state of the core components of the output equipment are used to complete the digital simulation and operational status evaluation of the air compressor under full operating conditions, including: Based on the real-time operating status data stream and multi-physics field coupled simulation results, the total energy input and total compressed air output of the equipment per unit time are statistically analyzed, and the unit air production energy consumption index of the equipment is calculated. Based on the structural connection relationships, component positional relationships, and dimensional constraint relationships represented by the geometric topology, the temperature distribution, pressure distribution, and energy transfer changes during the equipment operation phase are analyzed to determine the force evolution law of the target component. By combining the dynamic response of the compressor host, drive motor and cooling structure in multi-physics field coupled simulation, the key component areas where thermal loads are concentrated are located, and the thermal stress state characterization results of the whole equipment components are formed. The unit gas production energy consumption of the equipment is correlated with the thermal stress state characterization results of key components to form a comprehensive judgment result that includes operating efficiency indicators and operating safety indicators, thus completing the digital simulation and operating status assessment of the air compressor under the current working conditions.

[0017] To achieve the above objectives, a second aspect of the present invention provides an energy efficiency optimization device for air compressor equipment based on digital twins, comprising: The twin modeling module is used to construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model at least covers the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system. The signal acquisition module is used to deploy several sensor acquisition units in the corresponding functional area of ​​the air compressor based on the equipment pipelines, power components and cooling structure covered by the digital twin model, to acquire multi-dimensional original operating condition signals during equipment operation. The data synchronization module is used to standardize and correct the original operating condition signals through the data gateway, complete the data format normalization and time reference unification, and form a standardized operating data stream that is compatible with the digital twin access application. The simulation evaluation module is used to import real-time operating status data streams into the digital twin model, drive the digital twin model to carry out multi-physics field coupled simulation calculations based on real-time data, and output the comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of core components, thus completing the digital simulation and operating status evaluation of the air compressor under all operating conditions.

[0018] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0019] To achieve the above objectives, a third aspect of this application provides an electronic device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a digital twin-based energy efficiency operation optimization method for air compressor equipment as described in the first aspect embodiment.

[0020] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a digital twin-based method for optimizing the energy efficiency of air compressor equipment as described in the first aspect embodiment.

[0021] The embodiments of the present invention have the following beneficial effects: 1. This solution builds a dedicated digital twin model that integrates the air compressor's geometry, main unit aerodynamic characteristics, motor load characteristics, and cooling heat exchange parameters. It fully matches the physical structure of the physical equipment with the operating mechanisms of each system, ensuring that the simulation's basic parameters are highly consistent with the actual operating attributes of the equipment. It strengthens the linkage logic of parameters of each functional system, reduces the deviation between the model's deduction process and the actual operating conditions of the physical equipment, and improves the overall fit of the simulation.

[0022] 2. Based on the data gateway, the format normalization and timing synchronization correction of multiple parts and types of original operating signals are completed, the time reference of the data collected across the entire domain is unified, the timing misalignment problem in the transmission and acquisition of multi-source heterogeneous data is eliminated, and reliable data support is provided for the stable and accurate access of the digital twin model to real-time operating data.

[0023] 3. Drive the digital twin model to carry out multi-physics field coupled simulation with standardized real-time operation data stream, realize the synchronous calculation of equipment gas generation energy consumption and component thermal stress state, break the limitation of the separation between energy efficiency analysis and thermal state monitoring under the traditional monitoring mode, and enrich the coverage dimensions of equipment operation judgment.

[0024] 4. By leveraging the synergistic coupling calculation of aerodynamic characteristics, motor efficiency, and heat exchange parameters, the dynamic correlation between air compressor compression work, power loss, and cooling heat exchange is realistically reproduced. This effectively reduces the bias in judgment caused by single parameter analysis and ensures that the simulation output results can objectively represent the overall operating level of the equipment under all working conditions.

[0025] 5. By combining precise timing calibration with customized model construction, simulation errors caused by data anomalies and insufficient model adaptation are reduced, providing accurate and comprehensive data and simulation support for subsequent energy efficiency regulation, condition assessment and stable operation management of air compressors. Attached Figure Description

[0026] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating an energy efficiency optimization method for air compressor equipment based on digital twins, provided as an embodiment of the present invention; Figure 2 This is a structural diagram of an air compressor equipment energy efficiency operation optimization device based on digital twin, provided in an embodiment of the present invention. Detailed Implementation

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] The following description, with reference to the accompanying drawings, describes a method and apparatus for optimizing the energy efficiency of air compressor equipment based on digital twins, according to an embodiment of the present invention.

[0030] Example 1 This invention provides a method for optimizing the energy efficiency of air compressor equipment based on digital twins. Figure 1 This is a schematic flowchart illustrating a method for optimizing the energy efficiency of air compressor equipment based on digital twins, as provided in an embodiment of the present invention. Figure 1 As shown, the method includes the following steps: Step S1: Construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model shall at least cover the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system.

[0031] A digital twin model simulation platform adapted for the entire process of air compressor simulation was built. This platform is a digital twin development and operation environment that supports multi-physics coupling and real-time data-driven operation. The overall underlying architecture is built using a component-based modeling approach, ensuring that the model has modular expansion and independent computing capabilities. The physical structure of the physical air compressor is comprehensively surveyed, scanned, and digitally modeled to fully restore the spatial layout, pipeline routing, and mechanical connection forms of each component. The geometric topology corresponding to the air compressor body is constructed and solidified, clarifying the spatial positions, dimensions, and assembly constraints between the main casing, oil-gas cylinder, various stages of coolers, pipeline flanges, and valve connection points.

[0032] Based on the equipment's factory calibration file, on-site bench test results, and core component factory performance test reports, we systematically reviewed and determined the aerodynamic characteristic mapping table corresponding to the compressor host's full range of operating conditions. This mapping table fully associates key parameters such as volumetric efficiency, isentropic efficiency, and internal leakage rate under different intake pressure, exhaust pressure, and operating speed conditions.

[0033] Based on the rated parameters on the drive motor nameplate and factory type test data, efficiency-load characteristic curves corresponding to the full load range of the drive motor are fitted and generated, clearly representing the dynamic correspondence between motor operating efficiency and power factor under different load rates during equipment operation. Valid data are systematically extracted from engineering design drawings and heat exchanger parameter manuals, integrating and determining various heat exchange parameters of the cooling system to match various operating conditions, specifically including the effective heat exchange area of ​​each stage of heat exchanger, the comprehensive heat transfer coefficient, the rated design flow rate of cooling water, and the standard design temperature rise. Based on the internal flow path of the compressed medium, the mechanical power transmission relationship, and the overall energy transfer logic of the air compressor, the air compressor body geometry topology, the compressor host aerodynamic characteristics, the drive motor efficiency-load characteristics, and the cooling system heat exchange parameters are deeply integrated and coupled across dimensions.

[0034] Based on the preset energy flow and signal flow transmission ports of the simulation platform, and strictly following the step-by-step energy transfer path of electrical energy-mechanical energy-internal energy, the topological assembly and linkage of each functional module are completed. This ultimately forms an integrated digital twin model capable of receiving real-time operating parameters, dynamic response to multi-physics fields, and quantitative output of equipment operating performance. The model incorporates a high-precision multi-physics joint solver and a global time synchronization engine, which can simultaneously support collaborative calculations of various physical field processes such as aerodynamics and thermodynamics, electromagnetic-thermal coupling of motors, and fluid flow and heat transfer. It also features a dedicated standardized data interaction interface, effectively ensuring the time consistency and bidirectional data exchange between the digital twin virtual entity and the physical entity, providing a stable and reliable underlying operating platform for subsequent real-time data-driven simulation calculations, energy efficiency benchmark construction, and parameter optimization and control.

[0035] Step S2: Based on the equipment pipelines, power components and cooling structures covered by the digital twin model, several sensor acquisition units are deployed in the corresponding functional areas of the air compressor to acquire multi-dimensional original operating condition signals during equipment operation.

[0036] By combining a digital twin model to fully cover the air delivery pipeline system, the core components of the drive motor power transmission, and the overall structure of the cooling system circulation loop, the dynamic monitoring requirements for the entire continuous operation of the air compressor are comprehensively analyzed. Target monitoring points are delineated across the entire equipment area, and the type of operating parameters to be collected, the accuracy standard, and the continuous monitoring range for each monitoring point are clearly defined. Strictly following the structural division of the three core functional units—the air compression system, the power drive system, and the circulating cooling system—sensor acquisition units are deployed in a zoned, differentiated, and comprehensive manner. High-precision pressure and temperature sensors are stably deployed at key nodes and diameter changes in the air compressor's inlet and outlet pipelines to monitor the gas medium's operating status in real time. An intelligent power acquisition and monitoring module is installed at the high-voltage power input end of the drive motor to continuously capture real-time power parameters such as motor load and input power. Temperature sensors and fluid monitoring units are deployed at the pipe sections of the cooling water inlet and outlet loops of the cooling system to achieve full-process tracking of cooling medium operating parameters.

[0037] Based on a comprehensive, layered, zoned, and categorized deployment of various sensor acquisition units, continuous on-site data acquisition is conducted, accumulating fundamental operational signals such as pressure, temperature, electrical, and fluid parameters during the steady-state and variable-condition operation of the air compressor. The entire sensor acquisition system can comprehensively capture the equipment's intake pressure. Exhaust pressure Intake pipe temperature Exhaust temperature Motor input power Motor stator temperature Cooling water inlet temperature Cooling water outlet temperature Ambient temperature and gas load These key physical operating quantities are used to form a multi-dimensional original operating condition signal that can comprehensively, three-dimensionally, and in all time periods reflect the real-time load conditions, thermal operating status, power output level, and cooling cycle status of the physical air compressor. This provides real, effective, dimensional, and time-series continuous original data support for backend multi-source data integration and processing, full-domain time-series collaborative correction, high-precision simulation calculation of digital twin models, and the construction of a long-term energy efficiency benchmark library.

[0038] Step S3: The original operating condition signal is standardized and time-series coordinated and corrected through the data gateway to complete the data format normalization and time reference unification, forming a standardized operating data stream that is compatible with digital twin access applications.

[0039] Based on high-performance data gateway devices deployed across the entire industrial site, the system centrally accesses multi-source heterogeneous raw operating condition signals independently uploaded by each sensor acquisition unit. This effectively breaks down the isolation barriers of traditional distributed data transmission, enabling centralized collection, unified transmission, and integrated management of data collected from multiple monitoring points and various types of sensor devices.

[0040] To address the differences in encoding formats, communication transmission protocols, physical quantity data units, and signal delays caused by long-distance transmission between different sensor models and operating principles in the field, a systematic and standardized data standardization and processing was implemented. This process simultaneously removed invalid data from the original signals, cleaned abnormal fluctuation data, and intelligently completed short-term missing data. It unified the output specifications, data encoding formats, and physical quantity measurement standards of all acquired signals, completely eliminating compatibility conflicts during heterogeneous data integration.

[0041] Using the pre-defined global unified time reference of the completed digital twin model as the core calibration basis, the monitoring signals uploaded by all sensor acquisition units undergo full-domain, large-scale time-series collaborative correction and high-precision timestamp synchronization matching processing. This effectively corrects common data defects caused by asynchronous acquisition from multiple nodes, such as time-series misalignment, data asynchrony, and time axis offset. After centralized integration of multi-source data, unified cleaning of heterogeneous formats, full-domain time-series synchronization correction, and filtering of abnormal data, all operational data is systematically integrated and summarized according to the model access specifications. This results in a standardized operational data stream with highly consistent time series, complete operational parameter dimensions, standardized data format, and seamless compatibility with digital twin model interface calls. This data stream can stably adapt to the model's real-time access requirements, ensuring continuous stability of the equipment data transmission link and accurate and effective input simulation parameters. It avoids error interference from the data source, providing compliant, standardized, and reliable full-process data support for the digital twin model to conduct high-precision multi-physics coupling simulation calculations, historical operating condition data analysis, and real-time intelligent parameter optimization.

[0042] Step S4: Import the real-time operating status data stream into the digital twin model, drive the digital twin model to carry out multi-physics field coupling simulation calculation based on the real-time data, output the comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of the core components, and complete the digital simulation and operating status assessment of the air compressor under all working conditions.

[0043] The generated standardized real-time operating status data stream is completely imported into the digital twin model. The various real-time operating parameters contained in the data stream are accurately analyzed. The key operating parameters that are selected and extracted are uniformly set as the boundary conditions and dynamic excitation inputs for model simulation. This continuously drives the digital twin model to dynamically iterate and change in real time with the actual operating conditions of the physical air compressor, truly realizing the synchronous linkage of the operating conditions and real-time mapping of the digital virtual body and the physical physical equipment.

[0044] Based on its built-in parallel high-speed computing mechanism, the digital twin model simultaneously launches two independent simulation threads: aerodynamic and thermodynamic calculations and motor loss calculations. Combined with the compressor host aerodynamic characteristic mapping table, drive motor efficiency-load characteristic curve, and a complete set of heat exchange parameters of the cooling system pre-embedded in the model, it performs multi-physics field coupled joint solution calculations.

[0045] Based on the collected intake pressure Intake pipe temperature Exhaust pressure The core operating parameters are matched and retrieved from the aerodynamic characteristic mapping table using a three-dimensional interpolation algorithm to calculate the theoretical exhaust flow rate of the compressor host under real-time operating conditions. Equipment power indication Theoretical temperature rise with adiabatic compression Based on the inlet and outlet temperature parameters of the cooling water, the following calculation formula is used:

[0046] in, This represents the actual heat exchange temperature difference in the cooling water circulation loop. The measured temperature at the cooling water outlet; This is the measured temperature at the cooling water inlet.

[0047] Based on the actual heat exchange efficiency of the cooling system By reasonably correcting the theoretical temperature rise of adiabatic compression, the temperature deviation caused by heat exchange loss can be effectively offset, and the actual exhaust temperature change law of the compression process can be restored.

[0048] Synchronous acquisition of motor input power With motor stator temperature Real-time parameters are used to perform interpolation calculations based on the efficiency-load characteristic curve of the drive motor to obtain the motor operating efficiency under the current load range. Then, the theoretical power loss of the motor is calculated using the motor loss calculation formula:

[0049] in, This represents the theoretical power loss of the motor. This refers to the actual electrical power input to the motor. The operating efficiency of the motor under the current load is used to quantify the theoretical power loss of the motor, clearly characterizing the energy loss composition and heat generation characteristics of the power end.

[0050] An exhaust temperature residual determination mechanism is introduced, using the following residual calculation formula:

[0051] in, This refers to the exhaust temperature residual. Simulated exhaust temperature values ​​for digital twin models; The exhaust temperature was measured on-site.

[0052] Using the law of conservation of energy as the core constraint, multi-dimensional coupled iterative calculations of compression thermodynamics, motor losses, and fluid heat dissipation are conducted to dynamically correct the aerodynamic loss coefficient and overall heat transfer coefficient within the model, continuously reducing the deviation between simulated and measured values. When the absolute value of the residual... When the temperature is below the system's preset threshold, the digital twin model and the physical air compressor are determined to have reached a highly matched thermodynamic equilibrium state, and the cycle iteration is terminated to ensure that the simulation conditions are highly consistent with the actual on-site conditions.

[0053] Under steady-state equilibrium conditions, the energy consumption per unit output is calculated using the following formula:

[0054] in, Energy consumption per unit output of compressed air; The input power to the motor is in thermal equilibrium. The core energy efficiency indicators of the air compressor are calculated based on the theoretical discharge flow rate of the compressor.

[0055] Combining the equipment's structural geometric parameters with the inherent physical properties of the metallic material, the thermal stress solution formula is used:

[0056] in, For component thermal stress; The coefficient of thermal expansion of the material; The elastic modulus of the material; This represents the actual temperature rise of the component. To determine the Poisson's ratio of the material, the thermal stress level of key components such as the main bearing and high-pressure housing is calculated to fully understand the structural safety status of the equipment under high-temperature operation.

[0057] This process involves compiling all data results from pneumatic compression simulation, motor loss calculation, cooling heat transfer analysis, overall machine energy efficiency calculation, and component thermal stress solution. These results are then integrated to form a comprehensive assessment that considers both energy efficiency and safety indicators, enabling a full-condition digital simulation and comprehensive evaluation of the air compressor's operating status. Simultaneously, leveraging long-term, continuously accumulated standardized operating data streams, effective operating segments with stable and sustained pressure, temperature, and air load fluctuations are selected. Each stable operating condition combination is then linked to the model simulation energy consumption results, constructing a historical operating condition-energy efficiency data set.

[0058] The K-means clustering algorithm is used to intelligently divide the working condition intervals, and the Euclidean distance between the working conditions is calculated using the following formula:

[0059] in, The Euclidean distance in the characteristic space of the working condition; This is the intake pressure difference. This is the exhaust pressure difference. This represents the difference in ambient temperature. This represents the gas load difference; the smaller the distance value, the higher the operating condition matching degree.

[0060] In accordance with clustering assignment rules:

[0061] in, This represents the set of samples contained in the j-th cluster; It is a feature vector of a single historical operating condition sample, which is composed of a combination of intake pressure, exhaust pressure, ambient temperature, and gas load; Let be the cluster center vector of the j-th class; Let p be the cluster center vector of any other p-th cluster. It is the L2 norm, used to characterize the spatial distance between working condition vectors; This is the preset total number of cluster categories.

[0062] and the cluster center iterative update formula ,in, The new cluster center of the j-th class after iterative update; Let be the total number of valid samples in the j-th cluster set. The cluster centers are iteratively adjusted using the mean of samples within the set to achieve continuous convergence optimization of the working condition partitioning.

[0063] This enables the automatic grouping of massive historical data, selects the energy-efficient sample within a single operating condition range as the benchmark operating state, and integrates benchmark data from across the entire domain to build a structured energy efficiency benchmark library, providing a standard reference for subsequent intelligent optimization.

[0064] Real-time capture of on-site operating parameters is performed and matched with the energy efficiency benchmark database to retrieve the optimal energy consumption target value for the corresponding operating range. Using the energy consumption deviation rate formula:

[0065] in, Energy consumption deviation rate; Real-time energy consumption; To determine the optimal target energy consumption under operating conditions, the difference between real-time energy consumption and optimal energy consumption is quantitatively compared to clarify the space for energy efficiency optimization.

[0066] Set the loading pressure setpoint, unloading pressure setpoint, cooling water circulation valve opening, and motor inverter output frequency as global optimization variables, and construct a variable vector:

[0067] in, For the load pressure setpoint, Set the unloading pressure setpoint. For the opening degree of the cooling water circulation valve, Set the output frequency value for the motor inverter.

[0068] Constructing a constrained objective optimization model and writing the objective function. ,in, To optimize the objective function; The simulated energy consumption after parameter adjustment; This represents the optimal energy consumption target value. These are the weighting coefficients; To adjust the parameters, under the premise of satisfying the lower limit constraint of gas pressure and the boundary constraints of equipment parameters, the operating parameters are iteratively adjusted to search for the optimal energy efficiency control combination.

[0069] A safety operation boundary rule library for air compressors was simultaneously established, clearly defining the permissible limits for key parameters such as temperature, current, oil pressure, and flow rate, and employing safety verification formulas. ,in, This is a security check coefficient; These are simulation parameter values; To establish safety boundary limits, the optimized simulation parameters are compared for compliance, and control schemes that exceed the safety threshold are strictly eliminated to ensure that the optimization operation is safe and controllable.

[0070] The compliant optimization parameters are converted into a sequence of control commands that can be recognized by the field controller. Command pre-simulation is carried out in the digital twin model. The smooth correction algorithm suppresses abnormal transition processes such as pressure change, temperature overshoot, and power oscillation, ensuring that parameter switching is smooth and controllable.

[0071] After the stability of the control commands is verified, they are sent down to the physical air compressor field controller for execution, continuously collecting the optimized new operating data stream, and using the real-time residual calculation formula:

[0072] in, The parameter residual at time t; These are the actual measured parameters of the equipment; The parameters for model prediction are used to compare the deviation between the actual field measurement data and the model prediction data.

[0073] Model drift judgment rules are established based on the statistical characteristics of residual mean and standard deviation, and mean drift judgment formula is used. Discreteness drift criterion Monitoring will be conducted to identify systematic shifts in the model in real time. Among these, The mean of the current residual sequence. The historical baseline residual mean The standard deviation of the historical baseline residuals. , A preset threshold coefficient (dimensionless) is used. Once model drift is detected, parameter calibration is automatically triggered, updating the compressor host aerodynamic characteristic mapping table and the motor efficiency-load characteristic curve. Finally, the optimized operating data, control parameters, and incremental energy efficiency results are entered into the energy efficiency benchmark library, realizing adaptive iteration of the digital twin model and closed-loop operation of air compressor energy efficiency optimization.

[0074] In the application of one embodiment of the present invention, the implementation process is as follows: In an embodiment of the present invention, the construction of the digital twin model begins with three-dimensional laser scanning of a screw air compressor physical device of model SA-250A. The scanning operation covers the entire structure of the device body, and high-density point cloud data on the surface of the device is completely collected. The obtained original point cloud data is imported into reverse engineering software to complete denoising, simplification, splicing, and surface reconstruction processing, generating a geometric structure topology including the spatial position coordinates, external dimensions, and assembly spacing of the main engine housing, oil-gas cylinder, two-stage aftercooler, air-water separator, and all connecting pipeline flanges. This geometric structure topology accurately defines the mechanical assembly limit relationship between each functional component, the pipeline docking position, and the complete flow directions of the internal air flow channel and cooling water flow channel of the device, providing an accurate geometric basis for subsequent multi-physics field simulations.

[0075] Simultaneously, the collection of equipment performance parameters and basic characteristic data is carried out. The compressor main engine factory performance test report is retrieved from the standardized technical documents issued by the air compressor factory. The report contains a multi-dimensional working condition data table, and the data covers the intake pressure range of 0.1 MPa to 0.1013 MPa, the exhaust pressure range of 0.7 MPa to 1.2 MPa, and the main shaft speed range of 2000 rpm to 5000 rpm. The data table completely records the measured values of three core performance indicators, namely the volumetric efficiency, isentropic efficiency, and internal leakage rate caused by rotor clearance under different working condition combinations. The original data table is standardized, dimensionally classified, and format unified to construct a compressor pneumatic characteristic mapping table that can be quickly retrieved and interpolated for calling, enabling quick query and calculation of any intermediate working condition parameters.

[0076] From the factory type test report of the supporting drive motor of the air compressor, the electrical performance parameters corresponding to the gradient change of the load rate are extracted. The load rate covers the full range of 20% to 100%, and the key parameters include the real-time input electric power, mechanical output power, and operating power factor. Based on the least squares nonlinear fitting algorithm, the discrete load rate sample data is curve-fitted to accurately construct two characteristic models, namely the motor efficiency function and the power factor function, forming an efficiency-load characteristic curve function that can be quantitatively calculated, enabling accurate solution of the motor energy efficiency parameters at any continuous load rate.

[0077] Based on the overall machine design drawings of the air compressor and the assembly drawings of the cooling system, the core structures and design parameters of two shell-and-tube heat exchangers, namely the first-stage aftercooler and the second-stage aftercooler, are accurately extracted, specifically including the total number of heat exchange tubes, the effective length of a single heat exchange tube, the inner diameter and outer diameter of the heat exchange tube, the total effective heat exchange area of the device, and the rated design heat transfer coefficient. Simultaneously, the rated delivery flow rate of the circulating water pump supporting the cooling system and the standard temperature difference at the inlet and outlet under the design working condition are collected. All heat exchange structure parameters and fluid operation parameters are unified and integrated to form a complete set of heat exchange parameters for the cooling system.

[0078] The aforementioned collected and processed geometric topology, compressor aerodynamic characteristic mapping table, motor efficiency-load characteristic curve function, and cooling system heat exchange parameter set are comprehensively integrated. The physical connection logic of the equipment is strictly followed: the compressor main unit exhaust port is fixedly connected to the oil-gas cylinder medium inlet; the oil-gas cylinder gas outlet is connected to the first-stage aftercooler medium inlet; and the drive motor output shaft is rigidly connected to the main unit rotor via a coupling. Simultaneously, the energy transfer path across the entire equipment is matched, sequentially: electrical energy is converted into motor mechanical energy; mechanical energy drives the rotor to perform work, which is then converted into compressed gas internal energy; and the high-temperature gas internal energy is dissipated through forced convection heat transfer of cooling water in the cooling system. All data and models are coupled and integrated in a multi-physics simulation development platform, ultimately building a dedicated digital twin model for the air compressor that can receive signals from field sensors in real time, synchronously calculate and output compressed air unit output energy consumption, and thermal stress values ​​of key structural components.

[0079] The digital twin model simulation platform is an integrated, dedicated development and runtime environment that natively supports multi-physics field coupling calculations involving fluid mechanics, thermodynamics, electromagnetics, and structural mechanics. It features real-time data access, dynamic driving simulation, and online interactive computation capabilities. The platform adopts a modular, component-based modeling architecture, independently encapsulating each subsystem of the air compressor into callable simulation components with standardized input / output interfaces. These components include a compressor main unit module, a drive motor module, and a two-stage cooler integration module.

[0080] Each simulation component is digitally assembled and laid out strictly according to the mechanical topology of the physical equipment. Data interaction links are established through the platform's pre-set standardized energy flow ports and signal flow ports, perfectly matching the complete energy transfer path from electrical energy to mechanical energy to gas internal energy to heat loss. This enables the linkage and coupling between sub-modules, combining them to construct a full-domain system-level simulation model. The platform has a built-in high-precision multiphysics joint solver that can simultaneously and in parallel perform aerodynamic and thermodynamic calculations, motor electromagnetic and thermal loss calculations, and fluid heat transfer calculations for shell-and-tube heat exchangers, ensuring the synchronization and accuracy of multi-field coupled calculations.

[0081] The platform is equipped with a dedicated data interface module that supports industrial Ethernet and edge gateway protocol adaptation. It can receive and parse standardized real-time operating status data streams of air compressors uploaded from field data gateways. The parsed operating data is uniformly converted into boundary conditions and dynamic excitation inputs recognizable by the simulation model, driving dynamic iterative calculations. After completing single-step calculations, the model publishes simulation results such as energy consumption, temperature, stress, and flow rate in real time through the platform's standardized output interface. The platform features a high-precision time synchronization engine that uniformly calibrates the time reference between the simulation model's calculation step size and the real-time data stream, eliminating timing deviations and ensuring real-time synchronous mapping and bidirectional interactive linkage between the digital twin virtual model and the physical equipment's operating status.

[0082] Within the digital twin model, the physics simulation task is divided into two parallel concurrent computation threads: aerodynamic thermodynamic calculation and motor loss calculation, which are uniformly scheduled and started by the simulation engine. The simulation engine continuously connects to the data gateway data stream, analyzes real-time operating parameters cycle by cycle, and extracts all boundary condition parameters under a single simulation step: intake pressure. Intake pipe ambient temperature Equipment exhaust pressure On-site measured exhaust temperature Actual input power of the motor motor stator operating temperature Cooling water inlet temperature Cooling water outlet temperature .

[0083] The three core operating parameters—intake pressure, intake temperature, and exhaust pressure—are imported into an aerodynamic characteristic mapping table. A three-dimensional linear interpolation algorithm is then used for precise matching and querying to obtain the core performance parameters of the compressor unit under the current steady-state operating conditions: theoretical volumetric exhaust flow rate. Compressor power indication Theoretical temperature rise during ideal adiabatic compression process .

[0084] Based on the complete set of heat exchange parameters of the cooling system, the actual operating temperature difference of the cooling water is first calculated. Substituting the values, we get: .

[0085] By combining the rated flow rate of the cooling water pump and the parameters of the heat exchange structure to calculate the actual heat exchange capacity of the equipment, and comparing it with the theoretical maximum heat exchange capacity to complete the ratio calculation, the actual heat exchange efficiency of the cooling system under the current operating conditions is obtained. Based on this heat transfer efficiency, the theoretical temperature rise of adiabatic compression is corrected, and the influence of cooling loss is introduced to calculate the simulated value of the uncalibrated exhaust temperature. Substituting all the parameters, the value is: .

[0086] Substituting the real-time input power of the motor and the stator operating temperature into the pre-fitted efficiency-load characteristic curve function, the interpolation solution is used to obtain the equipment operating parameters corresponding to the current load rate: motor operating efficiency. Power factor on the grid side Based on the energy balance relationship, the total power loss during motor operation is calculated:

[0087] Substitute the numerical values ​​into the calculation:

[0088] In some preferred embodiments, a multi-parameter coupled iterative calculation logic is constructed based on the law of conservation of energy. This calculation logic is the core key to achieving virtual-real synchronization and high-precision simulation of the digital twin model. The iterative calculation establishes a heat balance equation based on three main factors: the increment of compressed air enthalpy change, the heat generated by motor losses, and the real-time heat dissipation of the cooling system. A residual model is constructed by quantitatively comparing the difference between the simulated exhaust temperature and the actual measured exhaust temperature. The exhaust temperature residual is defined by the following formula:

[0089] in, The simulated exhaust temperature value output by the iterative calculation of the digital twin model; The actual measured value of exhaust temperature is obtained from the real-time data stream of the field sensors.

[0090] During the iterative calculation, the system dynamically fine-tunes two key correction parameters within the model: the aerodynamic drag loss coefficient and the overall heat transfer coefficient of the heat exchanger. This continuously reduces the residual value of the exhaust temperature, causing the simulated exhaust temperature to gradually approach the measured temperature of the actual equipment. The preset synchronization judgment threshold is... When satisfied At that time, it is determined that the digital twin model has completed thermodynamic balance calibration, realizing synchronous matching of the working conditions between the virtual model and the physical entity.

[0091] After the model reaches a stable thermal equilibrium state, the simulation engine simultaneously calculates core energy efficiency and structural mechanics indicators. The formula for calculating the energy consumption per unit output of compressed air is as follows: ,in, Energy consumption per unit output of compressed air; This refers to the input electrical power of the motor under thermal equilibrium steady state. The theoretical discharge flow rate of the compressor under steady-state conditions; taking into account the time unit conversion relationship 1h=60min, the measured stable value under equilibrium conditions is substituted. , The calculations ultimately yield the energy consumption per unit output under equilibrium conditions. .

[0092] Simultaneously, based on the global distribution of the steady-state temperature field of the equipment, the thermal stress values ​​at key locations of the air compressor main bearing are calculated. The formula for the thermal stress calculation model is as follows: ,in, For component thermal stress; The linear thermal expansion coefficient of the bearing alloy material; The elastic modulus of the bearing material; This represents the actual temperature rise of the bearing operating temperature relative to the installation reference room temperature. Let be the Poisson's ratio of the bearing material. Combining the equipment material parameters and steady-state temperature rise and fall data, the steady-state thermal stress at the location of the main bearing is calculated as follows: .

[0093] In one embodiment of the present invention, a comprehensive energy efficiency benchmark library for air compressors is constructed. The energy efficiency benchmark library is built based on a massive real-time status data stream accumulated from the long-term continuous operation of the equipment. The system connects to an industrial time-series database to read batches of continuous operation data from the air compressor throughout its entire lifecycle. Stable operation segment filtering rules are set to extract effective steady-state data segments. Specific filtering constraints include: the intake pressure fluctuation range is less than... The exhaust pressure fluctuation range is less than Ambient temperature fluctuation range is less than The fluctuation range of on-site gas load is less than Furthermore, the duration for which the above multiple constraints are simultaneously and continuously satisfied is no less than 300 seconds.

[0094] All continuous data segments meeting the screening criteria are uniformly defined as a single stable operating segment. Each stable operating segment uniquely corresponds to a set of representative steady-state operating condition parameters. An example of a typical steady-state operating condition is selected: inlet pressure 0.100 MPa, exhaust pressure 0.80 MPa, ambient temperature 20℃, and stable gas load 38.5. The system automatically adds timestamp tags and operating condition feature vector tags to all identified and filtered stable operating segments, completes classification index storage, and generates a structured segment list for easy batch processing later.

[0095] In some embodiments, for each stable running segment that has completed its indexing and marking, the standardized simulation service interface provided by the digital twin model is invoked to batch input the steady-state operating condition parameter vector corresponding to the segment into the simulation model. This drives the digital twin model to complete the full-process steady-state coupled simulation calculation, and each segment simulation independently outputs the simulation result of the unit output energy consumption of compressed air under the corresponding operating condition. Taking the above typical operating condition combination as an example, the simulation calculation yields the simulated value of the unit output energy consumption under this operating condition. .

[0096] All operating parameters of a single stable operating segment and the energy consumption simulation results output by the digital twin model are structurally bound and associated to generate standardized historical operating condition-energy efficiency data pairs. Each data pair contains five core fields: intake pressure, exhaust pressure, ambient temperature, gas load, and unit output energy consumption. All historical operating condition-energy efficiency data pairs generated in batches are uniformly stored in a dedicated industrial analysis database, completing the accumulation of original benchmark data.

[0097] Based on tens of thousands of historical operating condition-energy efficiency data samples accumulated in the analysis database, a comprehensive operating condition clustering analysis was conducted. The clustering algorithm selected was the K-means clustering algorithm based on Euclidean distance. A four-dimensional feature space was constructed using four types of parameters: intake pressure, exhaust pressure, ambient temperature, and gas load. The massive data was automatically divided and classified based on the distance in the feature space, realizing the aggregation of data samples with highly similar operating conditions into the same operating condition range, and completing the fine-grained partitioning of the entire operating condition range.

[0098] The core objective of the K-means clustering algorithm is to iteratively find the optimal cluster centers and minimize the sum of squared distances from all data samples to their respective cluster centers. The algorithm consists of two main iterative steps: allocation and update.

[0099] First, the allocation steps: Based on the Euclidean distance calculation formula in four-dimensional feature space, each historical working condition data vector is... Assign to the nearest cluster center The allocation rules for the set to which it belongs are as follows:

[0100] In the formula, This is a four-dimensional working condition parameter vector for a single data point. These are the cluster center vectors for different groups. This is the preset total number of clusters.

[0101] Second, the update step: Based on the grouping results of the allocation step, statistically analyze the individual cluster sets. The arithmetic mean of all internal data samples is used to iteratively update and generate new cluster centers. The update calculation formula is as follows:

[0102] The allocation and update steps are executed alternately in a loop, and the change in the coordinates of the cluster centers is monitored in real time. When the position of the cluster centers has no significant shift or the algorithm reaches the preset maximum number of iterations, the calculation is automatically terminated, and the interval division of all historical working conditions is completed.

[0103] After clustering and partitioning, all historical operating condition-energy efficiency data pairs within a single operating condition interval are traversed, and the samples with the lowest unit output energy consumption within the interval are selected as benchmark data. An example of selecting a limited operating condition interval: intake pressure. Exhaust pressure Ambient temperature Gas load Three optimal benchmark data points were obtained within this interval, with corresponding unit output energy consumption of [missing data]. , , These low-energy-consumption samples are uniformly defined as the benchmark operating conditions for the corresponding operating range.

[0104] The benchmark operating status records obtained from filtering all operating condition ranges are integrated and standardized to construct a comprehensive air compressor energy efficiency benchmark library. A key-value pair database is preferred for the storage architecture, using a normalized four-dimensional operating condition parameter vector as the retrieval key and the optimal unit output energy consumption target value for the corresponding operating condition range as the binding storage value, enabling rapid matching of operating conditions and second-level retrieval of optimal energy consumption parameters.

[0105] Throughout the construction of the energy efficiency benchmark library, the similarity between two sets of operating conditions is quantitatively assessed using Euclidean distance in the operating condition feature space. The distance calculation formula is as follows:

[0106] in, The four-dimensional feature space Euclidean distance between the two sets of working condition data; This represents the difference in intake pressure parameters between the two data points. The difference between the two data points is the exhaust pressure parameter. This represents the difference in ambient temperature parameters between the two data points. This represents the difference in gas load parameters between the two data points.

[0107] The smaller the distance calculation value, the higher the overlap of the two sets of operating conditions and the closer their operating characteristics are. Cluster analysis relies on this distance formula to complete data grouping, and the optimal energy consumption sample is selected within each group to jointly constitute the core benchmark data source of the energy efficiency benchmark library.

[0108] In one embodiment of the present invention, real-time intelligent optimization control of air compressor operating parameters is carried out based on the established energy efficiency benchmark library and the real-time linked digital twin model. The real-time optimization process first parses the real-time operating status data stream sent by the field edge gateway to extract the current actual operating parameters of the equipment: real-time intake pressure 0.102MPa, real-time exhaust pressure 0.82MPa, real-time ambient temperature 22℃, and real-time air flow rate. .

[0109] Using the current four-dimensional operating condition parameter vector as the search criteria, a full-domain matching search is performed within the energy efficiency benchmark database. The search logic is based on the Euclidean distance calculation between the real-time operating condition vector and the benchmark operating condition vectors within the database, selecting the target operating condition interval with the closest distance within the allowable deviation range. Example matching target intervals: Intake pressure 0.100–0.103 MPa, exhaust pressure 0.81–0.83 MPa, ambient temperature 20–24℃, air flow rate 40–42 liters per minute. ; Retrieve the optimal energy consumption target value bound to this operating condition range. .

[0110] Simultaneously, real-time simulation calculations using a digital twin model are performed to determine the actual energy consumption of the equipment under the current unoptimized state, obtaining the real-time energy consumption per unit output. The deviation between real-time energy consumption and the optimal benchmark energy consumption is quantitatively calculated. The formula for calculating the energy consumption deviation rate is as follows: The numerical values ​​are substituted to complete the calculation, quantifying the degree of energy waste in the equipment and providing a quantitative basis for subsequent parameter optimization.

[0111] In some embodiments, four sets of dynamically adjustable equipment operation control parameters are preset within the digital twin model as core decision variables for optimization iteration. These parameters specifically include: loading pressure setpoint, unloading pressure setpoint, cooling water circulation electric valve opening percentage, and motor inverter output frequency setpoint. The initial default operating parameters of the equipment are configured as follows: loading pressure setpoint 0.80MPa, unloading pressure setpoint 0.85MPa, cooling water circulation valve opening 60%, and motor inverter output frequency 49.5Hz.

[0112] Under the premise of fixed, non-adjustable boundary parameters at the site, the real-time intake pressure is kept constant at 0.102 MPa, the ambient temperature at 22℃, and the downstream gas flow rate at 41.2 liters per cubic meter. The process remains unchanged; with the core optimization objective being to reduce the difference between real-time energy consumption and the optimal benchmark energy consumption, a parameter optimization mathematical model with multiple constraints is constructed within the digital twin simulation environment, and closed-loop iterative simulation optimization of adjustable operating parameters is initiated.

[0113] The complete mathematical model for the optimization problem is as follows: 1. Decision variable vector:

[0114] in, For the load pressure setpoint, Set the unloading pressure setpoint. For the opening degree of the cooling valve, This refers to the output frequency of the frequency converter.

[0115] 2. Objective function:

[0116] in, The simulation energy consumption value corresponding to the combination of decision variables. To achieve the optimal energy consumption target value, For parameter adjustment range, These are the balancing weighting coefficients.

[0117] 3. Constraints: Fixed operating condition constraints: , , ; Gas supply guarantee constraints: ; Variable boundary constraints:

[0118]

[0119] .

[0120] During the iterative optimization process, the updated decision variable set is input into the simulation engine, and the model control logic is switched synchronously: the pressure regulation logic changes from following the real-time pressure on site to being directly controlled by the loading and unloading pressure setpoints; the cooling system flow rate and heat exchange efficiency are corrected based on the valve opening parameters, and the heat exchange boundary conditions are updated synchronously; the motor speed and operating efficiency are corrected based on the inverter frequency parameters, and the power input boundary is updated. After completing the multi-boundary condition update, the multiphysics coupling simulation is restarted, and new exhaust pressure and energy consumption simulation results are output.

[0121] After each iteration, a dual-judgment constraint is applied: first, the simulated exhaust pressure meets the minimum downstream gas pressure requirement; second, the simulated energy consumption continuously approaches the optimal target value. If the parameter combination does not meet the constraints or the optimization effect decreases, the decision variables are fine-tuned according to a fixed step size, and the iteration continues in a loop until the optimal parameter combination is finally obtained: loading pressure 0.81MPa, unloading pressure 0.84MPa, cooling valve opening 65%, and inverter frequency 48.8Hz.

[0122] In one embodiment of the present invention, after the parameter optimization combination is determined, the equipment safety boundary verification process is initiated to eliminate the risks of over-temperature, over-pressure, and overload operation caused by parameter adjustments. A dedicated safety operation boundary rule library for the air compressor is pre-built. The rule library is stored using structured data tables, and the data sources are integrated from the original equipment manufacturer's technical manual, national industrial safety standards, and on-site equipment operation and maintenance safety procedures. The safety limits of each core operating parameter are clearly defined, as shown in Table 1 below.

[0123] Table 1 Parameter Table of Safe Operation Boundary Rule Base

[0124] The optimal combination of operating parameters was imported into the digital twin model to conduct full-condition simulation. The complete set of operating state parameters corresponding to the optimized scheme were extracted: simulated exhaust temperature 102℃, simulated motor operating current 305A, simulated lubricating oil pressure 0.28MPa, simulated lubricating oil temperature 82℃, and simulated cooling water flow rate 115L / min.

[0125] All parameters are verified using a unified verification formula. The single-parameter verification coefficient calculation model is as follows:

[0126] in, For security verification coefficient, For simulation calculation parameter values, These are safety boundary limits.

[0127] The judgment rules are uniformly set: the verification coefficient of the upper limit type parameter must not be greater than 0, the verification coefficient of the lower limit type parameter must not be less than 0, and the verification is qualified if all parameters meet the judgment requirements simultaneously.

[0128] Numerical verification is completed using key parameters as an example: Exhaust temperature verification coefficient:

[0129] Motor current verification coefficient:

[0130] If all simulation parameter verification coefficients are within the safe range, the optimized parameter combination is deemed to have passed the safety boundary verification. If any parameter exceeds the limit, the system automatically feeds back the exceeding parameter and its corresponding safety limit to the optimization model for iterative optimization until a qualified parameter scheme that balances energy saving and equipment safety is output.

[0131] The optimized parameter combination, verified by safety checks, is converted into standardized control instructions recognizable by the field PLC controller, electric regulating valve, and industrial frequency converter. Pressure parameters are converted into register write instructions: "WRITE_PS_SETPOINT_HIGH0.84; WRITE_PS_SETPOINT_LOW 0.81"; cooling valve opening is converted into analog control instructions: "SET_AV01_OPENING 65.0"; and frequency converter frequency is converted into frequency setpoint instructions: "SET_FREQ_SETPOINT 48.8".

[0132] The execution order of instructions is arranged according to the control logic of equipment operation stability. Priority is given to issuing instructions for valve opening and frequency converter frequency adjustment. After the fluid system and power system are running stably, the pressure control setpoint is updated, and a complete optimized control instruction sequence is formed.

[0133] Before the field equipment commands are issued and executed, a control command pre-simulation is conducted based on a digital twin model. The simulation time acceleration is set to 10 times to simulate the complete process of the field controller executing commands step by step according to time nodes. The continuous dynamic change curves of exhaust pressure, exhaust temperature, and motor input power during the pre-simulation are fully recorded, and abnormal fluctuation thresholds are set: exhaust pressure change rate ≤ 0.05MPa / s, exhaust temperature ≤ 105℃, and motor power fluctuation amplitude ≤ 10% of steady-state value.

[0134] To address abnormal operating conditions such as sudden pressure changes, temperature overshoot, and power oscillations that occur during the pre-simulation process, the original commands are smoothly corrected using ramp transition functions, S-shaped transition curves, and first-order low-pass filter functions. Step parameters are split and adjusted into multi-period linear transitions, the execution interval of key commands is delayed, the control command sequence is reconstructed and optimized, and the pre-simulation is repeated until the equipment transition process is stable and without abnormalities.

[0135] In one embodiment of the present invention, the final control command sequence that passes the stability verification is sent to the physical air compressor field controller for execution. During equipment operation, the global sensors continuously collect the optimized and adjusted new real-time operating data stream, and fully record the timing data such as intake pressure, exhaust temperature, electrical power, and cooling water parameters.

[0136] By comparing the actual operating parameters of physical equipment with the predicted parameters of digital twin simulation in real time, a dynamic residual monitoring model is constructed, and the formula for calculating the real-time residual of a single parameter is as follows:

[0137] in, Let be the parameter residual value at time t. These are actual measured values. These are the predicted values ​​from the model simulation.

[0138] Long-term monitoring of the mean of the residual sequence within a fixed time window with standard deviation Combined with the historical benchmark residual mean during the steady-state operation phase of the equipment Benchmark standard deviation Set the quantization drift detection criteria: Mean drift determination: ; Discreteness drift determination: or .

[0139] When residual statistics continuously trigger drift criteria, a systematic calculation bias is identified in the digital twin model, and the model parameter self-calibration process is automatically initiated. Using recent stable operating data of the equipment, core model parameters such as the compressor aerodynamic characteristic mapping table and the motor efficiency fitting curve are corrected to restore simulation calculation accuracy.

[0140] After model calibration, the operating parameters, control commands, and steady-state energy efficiency results of this optimization event are uniformly packaged into a new data sample and incrementally written into the energy efficiency benchmark library. If the energy consumption index of the new sample is better than the original benchmark data in the same range, the benchmark record is automatically replaced and updated, realizing continuous iteration, incremental learning, and dynamic improvement of the energy efficiency benchmark library, and continuously improving the energy-saving optimization accuracy of the air compressor under all operating conditions.

[0141] Example 2 This invention provides an energy efficiency optimization device for air compressor equipment based on digital twins. Figure 2 This is a schematic diagram of a digital twin-based energy efficiency optimization device for air compressor equipment, provided as an embodiment of the present invention. Figure 2 As shown, the device includes: The twin modeling module 100 is used to construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model at least covers the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system. The signal acquisition module 200 is used to deploy several sensor acquisition units in the corresponding functional area of ​​the air compressor based on the equipment pipelines, power components and cooling structure covered by the digital twin model, to acquire multi-dimensional original operating condition signals during equipment operation. The data synchronization module 300 is used to standardize and correct the original operating condition signal through the data gateway, complete the data format normalization and time reference unification, and form a standardized operating data stream that is compatible with the digital twin access application. The simulation evaluation module 400 is used to import real-time operating status data streams into the digital twin model, drive the digital twin model to carry out multi-physics field coupled simulation calculations based on real-time data, and output the comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of core components, thus completing the digital simulation and operating status evaluation of the air compressor under all operating conditions.

[0142] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0143] Example 3 To implement the methods of the above embodiments, the present invention also provides an electronic device, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the various steps of the methods described above.

[0144] Example 4 To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing embodiments.

[0145] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0146] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0147] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for optimizing the energy efficiency of air compressor equipment based on digital twins, characterized in that, include: Construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model shall at least cover the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system. Based on the equipment pipelines, power components and cooling structure covered by the digital twin model, several sensor acquisition units are deployed in the corresponding functional areas of the air compressor to acquire multi-dimensional original operating condition signals during equipment operation. The original operating condition signals are standardized and time-series coordinated and corrected by the data gateway to achieve data format normalization and time reference unification, forming a standardized operating data stream that is compatible with digital twin access applications. The real-time operating status data stream is imported into the digital twin model. Based on the real-time data, the digital twin model is driven to carry out multi-physics field coupled simulation calculations. The comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of the core components are output, and the digital simulation and operating status assessment of the air compressor under all working conditions are completed.

2. The method according to claim 1, characterized in that, Constructing a digital twin model of the physical air compressor equipment, including: The physical structure of the physical air compressor is comprehensively surveyed and digitally modeled to fully restore the layout of the equipment and the connection of components. The geometric topology corresponding to the air compressor body is constructed and solidified, and the assembly association and spatial constraint relationship of each mechanical component are determined. Combining the equipment's factory calibration data, on-site test results, and component performance files, we determined the aerodynamic characteristic mapping table corresponding to the compressor host's full range of operating conditions, the efficiency-load characteristic curve corresponding to the drive motor's full load range, and the various heat exchange parameters of the cooling system to match the operating conditions. Based on the internal medium flow path, mechanical transmission relationship and energy transfer logic of the physical air compressor, the geometric topology, main unit aerodynamic characteristics, motor efficiency load characteristics and cooling system heat exchange parameters are integrated and coupled across dimensions to form an integrated digital twin model with the ability to receive real-time parameters, respond to multi-physics fields and output operating performance.

3. The method according to claim 2, characterized in that, Based on the equipment piping, power components, and cooling structure covered by the digital twin model, several sensor acquisition units are deployed in the corresponding functional areas of the air compressor to acquire multi-dimensional raw operating condition signals during equipment operation, including: Based on the air pipeline structure, drive motor power components and cooling system circulation structure covered by the digital twin model, the target operation monitoring points of the air compressor are delineated, and the corresponding operation parameter collection requirements for each monitoring point are determined. For different functional areas including intake branch, exhaust branch, motor power input terminal and cooling water circulation circuit, sensor acquisition units adapted to the on-site working environment are deployed in different zones. By utilizing sensor acquisition units deployed throughout the entire area, basic signals such as pressure, temperature, power, and fluid are continuously collected during the operation of the equipment, and multi-dimensional original operating condition signals that fully reflect the real-time operating status of the equipment are continuously accumulated and output.

4. The method according to claim 3, characterized in that, The original operating condition signals are standardized and time-series coordinated and corrected through a data gateway to achieve data format normalization and time base unification, forming a standardized operating data stream adapted for digital twin access applications, including: By centrally accessing multi-source heterogeneous raw operating condition signals uploaded by sensor acquisition units through data gateways deployed across the entire domain, the system can uniformly collect and transmit dispersed data and integrate multi-link acquired data. To address the differences in format, encoding, and transmission protocols among different types of sensor signals, we will carry out unified format standardization and data cleaning to unify the output specifications of multi-source signals. Using the unified time reference preset by the digital twin model as the calibration basis, all acquired signals are subjected to time-series collaborative correction and global timestamp synchronization processing to eliminate the time-series misalignment problem caused by multi-node acquisition and generate a time-consistent, format-unified, and standardized running data stream that is compatible with the digital twin model call.

5. The method according to claim 4, characterized in that, Import real-time operational status data streams into the digital twin model, and drive multiphysics coupling simulation calculations based on the real-time data, including: The real-time operating parameters in the standardized operating data stream are used as the input boundary conditions of the digital twin model to drive the digital twin model to synchronously match the current operating status of the physical air compressor equipment. Based on the aerodynamic characteristic mapping table of the compressor host in the digital twin model, aerodynamic and thermodynamic simulation calculations are performed on the compression process to generate compression performance characterization results that match the current operating conditions. Based on the drive motor efficiency-load characteristic curve built into the digital twin model, the energy conversion process of the drive motor in the current load range is analyzed, and the corresponding characterization results of power input conditions and energy efficiency changes are generated. Based on the heat exchange parameters of the cooling system built into the digital twin model, the heat exchange process of the air compressor cooling structure under the current operating conditions is calculated, and the effect of the cooling process on the overall thermal field of the equipment is quantified. By combining the inherent physical connection relationship of the equipment with the global energy transfer logic, the simulation results of the compression process, the power input simulation results, and the cooling process simulation results are jointly solved to form a global multi-physics field coupled simulation result.

6. The method according to claim 5, characterized in that, The comprehensive analysis results of the unit air production energy consumption of the output equipment and the thermal stress state of the core components complete the digital simulation and operation status assessment of the air compressor under all operating conditions, including: Based on the real-time operating status data stream and multi-physics field coupled simulation results, the total energy input and total compressed air output of the equipment per unit time are statistically analyzed, and the unit air production energy consumption index of the equipment is calculated. Based on the structural connection relationships, component positional relationships, and dimensional constraint relationships represented by the geometric topology, the temperature distribution, pressure distribution, and energy transfer changes during the equipment operation phase are analyzed to determine the force evolution law of the target component. By combining the dynamic response of the compressor host, drive motor and cooling structure in multi-physics field coupled simulation, the key component areas where thermal loads are concentrated are located, and the thermal stress state characterization results of the whole equipment components are formed. The unit gas production energy consumption of the equipment is correlated with the thermal stress state characterization results of key components to form a comprehensive judgment result that includes operating efficiency indicators and operating safety indicators, thus completing the digital simulation and operating status assessment of the air compressor under the current working conditions.

7. A digital twin-based energy efficiency optimization device for air compressor equipment, characterized in that, include: The twin modeling module is used to construct a digital twin model corresponding to the physical air compressor equipment. The digital twin model at least covers the geometric topology of the air compressor body, the aerodynamic characteristic mapping table of the compressor host, the efficiency-load characteristic curve of the drive motor, and the heat exchange parameters of the cooling system. The signal acquisition module is used to deploy several sensor acquisition units in the corresponding functional area of ​​the air compressor based on the equipment pipelines, power components and cooling structure covered by the digital twin model, to acquire multi-dimensional original operating condition signals during equipment operation. The data synchronization module is used to standardize and correct the original operating condition signals through the data gateway, complete the data format normalization and time reference unification, and form a standardized operating data stream that is compatible with the digital twin access application. The simulation evaluation module is used to import real-time operating status data streams into the digital twin model, drive the digital twin model to carry out multi-physics field coupled simulation calculations based on real-time data, and output the comprehensive judgment results of the unit gas production energy consumption and the thermal stress state of core components, thus completing the digital simulation and operating status evaluation of the air compressor under all operating conditions.

8. The apparatus according to claim 7, characterized in that, The twin modeling module is also used for: The physical structure of the physical air compressor is comprehensively surveyed and digitally modeled to fully restore the layout of the equipment and the connection of components. The geometric topology corresponding to the air compressor body is constructed and solidified, and the assembly association and spatial constraint relationship of each mechanical component are determined. Combining the equipment's factory calibration data, on-site test results, and component performance files, we determined the aerodynamic characteristic mapping table corresponding to the compressor host's full range of operating conditions, the efficiency-load characteristic curve corresponding to the drive motor's full load range, and the various heat exchange parameters of the cooling system to match the operating conditions. Based on the internal medium flow path, mechanical transmission relationship and energy transfer logic of the physical air compressor, the geometric topology, main unit aerodynamic characteristics, motor efficiency load characteristics and cooling system heat exchange parameters are integrated and coupled across dimensions to form an integrated digital twin model with the ability to receive real-time parameters, respond to multi-physics fields and output operating performance.

9. An electronic device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in any one of claims 1-6.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images