Industrial equipment energy efficiency data optimization processing method based on digital twinning

Abstract

This invention relates to the field of industrial Internet of Things (IIoT) and intelligent optimization control technology, specifically disclosing a method for optimizing energy efficiency data of industrial equipment based on digital twins. The method includes: system positioning and twin construction: constructing a dynamically mapped digital twin model based on the topology and characteristic parameters of the physical equipment, and setting a dynamic energy efficiency optimization target threshold in the digital twin model; real-time acquisition of core energy efficiency parameters: collecting parameters directly related to energy efficiency in real time through sensors deployed on the physical equipment; dynamic energy efficiency evaluation and analysis: performing multi-dimensional energy efficiency evaluation in the digital twin model; and dynamic data optimization processing: initiating a dynamic optimization closed-loop control process when the energy efficiency deviation coefficient exceeds the set threshold upper limit. This invention accurately locates the root causes of energy efficiency degradation, achieves systemic energy saving through a dynamic closed-loop optimization control process, and prevents equipment failures caused by the optimization process, thereby improving energy efficiency and safety.

Description

Technical Field

[0001] This invention relates to the field of industrial Internet of Things and intelligent optimization control technology, and more specifically, to a method for optimizing the energy efficiency data of industrial equipment based on digital twins. Background Technology

[0002] High-energy-consuming industrial equipment is the core of the industrial energy system, accounting for a significant proportion of total industrial electricity consumption. For example, air compressors can account for over 25% of a company's total electricity consumption. Under the global goal of carbon neutrality, the pressure on industrial energy conservation and emission reduction has increased dramatically, making energy efficiency optimization for such equipment a key breakthrough.

[0003] Currently, existing methods for optimizing industrial equipment energy efficiency data focus on system-level modeling and data monitoring. However, these methods have some shortcomings and deficiencies in practical applications. The main shortcomings are as follows:

[0004] 1. Static system modeling, lacking dynamic mapping and adaptive capabilities: Existing systems mostly use static models or empirical formulas based on historical data, which cannot map the performance drift of physical equipment caused by environmental temperature fluctuations, equipment aging, load changes, etc. in real time. Its core defect is that the model is disconnected from the physical entity, becoming a "static snapshot" rather than a "dynamic mirror", which leads to the distortion of the benchmark for energy efficiency assessment and the optimization effect decays sharply over time.

[0005] 2. The control strategy lacks high-fidelity simulation verification, resulting in insufficient security and coordination guarantees: Traditional methods directly optimize through trial and error on the physical system or control based solely on simple rules. This "blind tuning" or "coarse tuning" mode lacks the step of conducting high-fidelity simulation verification in a virtual space before implementation, making it impossible to predict potential command conflicts, system oscillations, and potential security risks that may arise when multiple devices coordinate their actions, seriously threatening system stability and equipment lifespan;

[0006] 3. The optimization process is either open-loop or semi-closed-loop, lacking deep intelligent decision-making and fault-tolerance mechanisms: Most existing solutions lack a "digital brain" that evolves in parallel with the physical system and is continuously updated to manage the entire optimization process in a closed loop. The system cannot intelligently judge the validity of optimization commands. When encountering deep-seated faults such as pipeline blockage or permanent degradation of equipment performance, it will continue to make a large number of invalid or even harmful optimization attempts, causing the process to fail to converge and unable to autonomously diagnose the root cause and switch to maintenance mode, resulting in a low level of intelligence. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for optimizing the energy efficiency data of industrial equipment based on digital twins, which solves the problems mentioned in the background art through the following scheme.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing and processing energy efficiency data of industrial equipment based on digital twins, comprising:

[0009] S1: System localization and twin construction:

[0010] Based on the topology and characteristic parameters of physical devices, a dynamically mapped digital twin model is constructed, and a dynamic energy efficiency optimization target threshold is set in the digital twin model.

[0011] S2: Real-time collection of core energy efficiency parameters:

[0012] Sensors deployed by physical devices collect parameters directly related to energy efficiency in real time. These parameters include pressure parameters, flow parameters, power parameters, state parameters, and environmental parameters.

[0013] S3: Dynamic Energy Efficiency Assessment Analysis:

[0014] Performing a multi-dimensional energy efficiency assessment in a digital twin model includes the following steps:

[0015] S301: Calculate real-time energy efficiency parameters: specific power SP and isentropic efficiency. ;

[0016] S302: This combines specific power SP and isentropic efficiency. The energy efficiency deviation coefficient is generated by comparing it with the dynamic energy efficiency optimization target threshold and using a weighted algorithm. , This is a normalized weighted average of the deviation values ​​of multiple parameters;

[0017] S4: Dynamic Data Optimization Processing

[0018] When the energy efficiency deviation coefficient exceeds the set upper limit threshold, the dynamic optimization closed-loop control process is initiated, which includes the following steps:

[0019] S401: Simulate and generate an optimized instruction set in a digital twin model, the instruction set performing root cause diagnosis based on real-time pipeline differential pressure and load rate;

[0020] S402: The optimized instruction set is sent to the physical device for execution, the pressure adjustment rate does not exceed the safe rate threshold, and multiple devices are interlocked.

[0021] S403: Real-time acquisition of energy efficiency parameters after execution, and updating of the digital twin model status;

[0022] S404: Perform closed-loop verification and decision-making based on the updated energy efficiency deviation coefficient.

[0023] Preferably, the dynamic energy efficiency optimization target threshold is dynamically calibrated according to ambient temperature and equipment aging rate.

[0024] Preferably, the dynamic energy efficiency optimization target threshold includes a specific power dynamic optimization target threshold and an isentropic efficiency dynamic optimization target threshold.

[0025] Preferably, the comparison process is as follows:

[0026] The real-time specific power value is compared with the specific power dynamic optimization target threshold to generate a specific power deviation value;

[0027] The real-time isentropic efficiency is compared with the dynamic optimization target threshold of the isentropic efficiency to generate an isentropic efficiency deviation value.

[0028] Preferably, the energy efficiency deviation coefficient The calculation formula is:

[0029] ,

[0030] in, This is the actual value of the specific power. This represents the actual value of isentropic efficiency. The specific power target value, The target value for isentropic efficiency is... and Let be the weight coefficient, and satisfy... .

[0031] Preferably, the optimized instruction set includes pressure regulation instructions, unit scheduling instructions, frequency regulation instructions, pipeline maintenance instructions, and safety protection instructions.

[0032] Preferably, the closed-loop verification and decision-making in S404 includes the following steps:

[0033] (1) If the updated energy efficiency deviation coefficient is less than or equal to the lower limit of the set threshold, terminate the dynamic optimization closed-loop control process.

[0034] (2) If the exhaust pressure exceeds the preset pressure safety limit or the exhaust temperature exceeds the preset temperature safety limit, the optimization will be forcibly stopped and the alarm protocol will be triggered.

[0035] (3) If the real-time pipeline differential pressure exceeds the preset differential pressure abnormality threshold or the differential pressure change rate exceeds the preset sudden change rate threshold, the current optimization process is stopped and the pipeline maintenance subprocess is started.

[0036] (4) If the number of consecutive optimizations reaches the preset maximum number of iterations and the energy efficiency deviation coefficient decrease rate is lower than the preset minimum convergence rate, a freeze command is issued and a system fault diagnosis report is generated.

[0037] The technical effects and advantages of this invention are as follows:

[0038] 1. Dynamic Adaptation and Transparent Root Cause Location of Energy Efficiency Assessment Benchmark: This invention constructs a digital twin model that dynamically maps to physical equipment in real time, deeply integrating equipment characteristics, environmental temperature compensation, and aging degradation rate models. This allows the energy efficiency optimization target threshold to be dynamically adjusted according to actual operating conditions, completely solving the problem of static model inaccuracy. Through high-precision twins, multi-dimensional energy efficiency assessment is achieved, accurately identifying the root causes of energy efficiency degradation and providing a scientific basis for optimization and maintenance.

[0039] 2. Prospective safety verification and collaborative optimization based on digital twin simulation: This invention uses the digital twin as a "virtual test field" to perform simulation and root cause diagnosis before the instructions are issued. It verifies the effectiveness of the strategy and evaluates the safety and collaboration based on real-time data. Through constraints such as pressure adjustment rate limit and multi-device action interlock, it ensures that the optimized instruction set is efficient and reliable, completely avoids the risks of system oscillation, pressure shock and equipment overload, and achieves smooth and safe system-level optimization.

[0040] 3. Deep self-learning and fault-tolerant optimization closed loop with digital twin as the intelligent hub: The system continuously updates the model through real-time data feedback and performs intelligent verification based on multi-objective decision-making. It can not only dynamically track the optimal energy efficiency point, but also autonomously stop ineffective optimization when encountering deep faults, accurately locate the root cause of the fault, and intelligently switch to the maintenance process or trigger alarms, forming a highly adaptive and fault-tolerant intelligent closed loop, which significantly improves the system's reliability, economy, and intelligence level. Attached Figure Description

[0041] Figure 1 This is a flowchart of the energy efficiency data optimization processing method for the device of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0043] refer to Figure 1 The method for optimizing energy efficiency data of industrial equipment based on digital twins, as shown, includes:

[0044] S1: System localization and twin construction:

[0045] Based on the topology and characteristic parameters of physical devices, a dynamically mapped digital twin model is constructed, and a dynamic energy efficiency optimization target threshold is set in the digital twin model.

[0046] S2: Real-time collection of core energy efficiency parameters:

[0047] Sensors deployed by physical devices collect parameters directly related to energy efficiency in real time. These parameters include pressure parameters, flow parameters, power parameters, state parameters, and environmental parameters.

[0048] S3: Dynamic Energy Efficiency Assessment Analysis:

[0049] Performing a multi-dimensional energy efficiency assessment in a digital twin model includes the following steps:

[0050] S301: Calculate real-time energy efficiency parameters: specific power SP and isentropic efficiency. ;

[0051] S302: This combines specific power SP and isentropic efficiency. The energy efficiency deviation coefficient is generated by comparing it with the dynamic energy efficiency optimization target threshold and using a weighted algorithm. , This is a normalized weighted average of the deviation values ​​of multiple parameters;

[0052] S4: Dynamic Data Optimization Processing

[0053] When the energy efficiency deviation coefficient exceeds the set upper limit threshold, the dynamic optimization closed-loop control process is initiated, which includes the following steps:

[0054] S401: Simulate and generate an optimized instruction set in a digital twin model, the instruction set performing root cause diagnosis based on real-time pipeline differential pressure and load rate;

[0055] S402: The optimized instruction set is sent to the physical device for execution, the pressure adjustment rate does not exceed the safe rate threshold, and multiple devices are interlocked.

[0056] S403: Real-time acquisition of energy efficiency parameters after execution, and updating of the digital twin model status;

[0057] S404: Perform closed-loop verification and decision-making based on the updated energy efficiency deviation coefficient.

[0058] In a preferred embodiment of the present invention, taking a compressor system as an example, the present invention will be further described in detail. S1 specifically includes the following steps:

[0059] S101: Physical system topology mapping:

[0060] Identify physical devices and label them with unique identifiers; construct a topological directed graph by expressing the flow of energy or working fluid through directed edges; divide the system into subsystems according to function and define the coupling interfaces between subsystems.

[0061] S102: Integration of Equipment Characteristic Parameters:

[0062] The static parameters of the physical devices are injected into the twin model, and the isentropic efficiency and specific power calculation models and their respective environmental temperature compensation coefficients are embedded, and their respective device aging models are preset.

[0063] In this embodiment, it should be noted that the static parameters include rated power, rated flow rate, rated pressure, maximum allowable operating pressure, and volume, which are directly imported from the equipment database. The isentropic efficiency calculation model is as follows: The specific power calculation model is as follows: The equipment aging model is defined as aging decay rate = f(operating hours, load rate (%)).

[0064] The formula for calculating the specific power environmental compensation coefficient is as follows:

[0065] ,

[0066] in, It is the specific power environmental compensation coefficient, and k1 is the temperature sensitivity coefficient. k1 is a positive number, which is derived from the energy consumption-temperature characteristics. For example, a typical value is 0.0015 / ℃. It is the actual ambient temperature, i.e., the intake air temperature, which is obtained through a temperature sensor; The standard reference temperature is set at 20°C, as specified by international standards for dry air intake conditions.

[0067] The formula for calculating the isentropic efficiency environmental compensation coefficient is as follows:

[0068] ,

[0069] in, K is the isentropic efficiency environmental compensation coefficient, and k2 is the temperature sensitivity coefficient. K2 is a negative number, which is derived from thermodynamic properties. For example, a typical value is -0.0018 / ℃. It is the actual ambient temperature, i.e., the intake air temperature, which is obtained through a temperature sensor; The standard reference temperature is set at 20°C, as specified by international standards for dry air intake conditions.

[0070] The formula for calculating the specific power aging degradation rate is as follows:

[0071] ,

[0072] in, It is the power aging attenuation rate. This is the power-to-time decay factor, with a default value of 0.15; H o This is the cumulative running hours, derived from system records; H l This is the design life, usually 60,000 hours, derived from the equipment technical manual; This is the power load attenuation factor, with a default value of 0.05; It is the average load factor (%), which is derived from the average of the operating data over the past 30 days; This is the maximum permissible load rate (%), typically 110%, derived from equipment safety limits; specific power attenuation factor. and This is based on statistical data on increased power consumption and mechanical wear models.

[0073] The formula for calculating the isentropic efficiency aging degradation rate is as follows:

[0074] ,

[0075] in, It is the power aging attenuation rate. This is the power-to-time decay factor, with a default value of 0.18; H o This is the cumulative running hours, derived from system records; H l This is the design life, usually 60,000 hours, derived from the equipment technical manual; It is the power load attenuation factor, with a default value of 0.06; It is the average load factor (%), which is derived from the average of the operating data over the past 30 days; This is the maximum permissible load rate (%), typically 110%, derived from equipment safety limits; isentropic efficiency decay factor. and It is based on statistical data on performance degradation and aerodynamic thermodynamic models.

[0076] S103: Setting the target threshold for dynamic energy efficiency optimization:

[0077] In the twin model, two dynamic optimization target thresholds for specific power and isentropic efficiency are preset: the dynamic energy efficiency optimization target thresholds are dynamically calibrated with the ambient temperature and equipment aging rate, and are adaptively adjusted as the twin model state is updated.

[0078] In this embodiment, it should be noted that the dynamic energy efficiency optimization target threshold = design value × (1 + environmental compensation coefficient - aging attenuation rate), the specific power design value is calculated using the rated input power and rated gas production provided by static parameters, and the isentropic efficiency design value is provided by an international standard test report.

[0079] In a preferred embodiment of the present invention, step S2 specifically includes the following steps:

[0080] S201: Sensor Network Deployment and Data Acquisition

[0081] The identification code and installation location of each sensor are determined from the topological directed graph, and the raw parameter data of each sensor that are directly related to energy efficiency are collected at a preset frequency.

[0082] In this embodiment, it should be noted that the sensors deployed in the physical equipment include pressure sensors, gas flow meters, power sensors, status sensors, and temperature and humidity sensors. The parameters directly related to energy efficiency include pressure parameters, flow parameters, power parameters, status parameters, and environmental parameters. Pressure parameters include exhaust pressure, intake pressure, and pipeline pressure differential. Flow parameters refer to the actual gas production. Environmental parameters include temperature and humidity, with temperature referring to intake temperature and exhaust temperature. Power parameters refer to operating power. Status parameters refer to operating status, total operating hours, loaded operating hours, and load rate. Operating status includes loading and unloading status, running and stopping status, automatic and manual status, and fault alarm status.

[0083] S202: Data Quality Assurance and Outlier Removal

[0084] By setting reasonable intervals for the range, rate of change, correlation, and equipment status of the data, calculating the slope of adjacent sampling points, calculating the goodness of fit of two related data, and checking whether the shutdown parameters are zero, real-time data verification is performed, and abnormal data is identified and removed using data cleaning algorithms.

[0085] In this embodiment, it should be noted that the trigger actions for verifying the range, rate of change, correlation, and device status of the data are as follows: marking out-of-limit data as invalid, temporarily storing mutated data for manual review, triggering sensor diagnosis when correlation fails, and automatically discarding non-zero data.

[0086] S203: Real-time Data Stream Construction and Storage

[0087] The three types of data—raw data, cleaned data, and statistical data—are stored in different storage media with different retention strategies, and these three types of data are then input into the digital twin model.

[0088] In this embodiment, it should be noted that the storage media are edge SD card, InfluxDB and PostgreSQL, respectively, and the retention policies are rolling storage for 7 days, retention for 1 year and permanent retention. Rolling storage for 7 days is used for fault tracing, retention for 1 year is used for trend analysis, and permanent retention is used for report generation.

[0089] In a preferred embodiment of the present invention, step S3 specifically includes the following steps:

[0090] Perform multi-dimensional energy efficiency assessments in digital twin models:

[0091] S301: Calculate real-time energy efficiency parameters: specific power SP and isentropic efficiency. .

[0092] In this embodiment, it should be noted that the specific power represents the power consumed by the compressor per unit displacement, and the calculation formula is as follows:

[0093] ,

[0094] Where SP represents the specific power of the compressor, the unit is kW / (m²). 3. (min), P represents the compressor's input power, which is the total power actually consumed to drive the compressor, in kW; Q represents the compressor's actual gas output, which is the volume of gas delivered to the exhaust pipe per unit time under actual operating conditions, and is the actual output gas volume measured or calculated at the compressor's exhaust port, in m³ / s. 3 / min.

[0095] isentropic efficiency It measures how close the actual compression process of a compressor is to an isentropic process, and the calculation formula is:

[0096] ,

[0097] in, Indicates the intake enthalpy, which refers to the total energy contained in a unit mass of gas at the compressor inlet; The actual exhaust enthalpy indicates the actual total energy per unit mass of gas at the compressor outlet. Specific enthalpy, representing isentropic exhaust enthalpy, refers to the theoretical specific enthalpy value corresponding to the same exhaust pressure when the compression process is assumed to be isentropic. The unit of specific enthalpy is kJ / kg.

[0098] Taking air as an ideal gas as an example, the formula for calculating specific enthalpy is:

[0099] ，

[0100] ，

[0101] ，

[0102] in, It is the isobaric specific heat capacity, which represents the amount of heat that a unit mass of air needs to absorb or release to increase or decrease its temperature by 1 Kelvin under constant pressure. The unit is kJ / (kg·K). , These represent the actual intake air temperature and the actual exhaust air temperature, respectively. It represents the theoretically achievable exhaust temperature when the gas undergoes an isentropic process from the intake state to the exhaust state, and is expressed in K.

[0103] The specific enthalpy and specific heat capacity of a gas can be obtained by looking up a table, which can be integrated into a digital twin model for dynamic access.

[0104] S302: This combines specific power SP and isentropic efficiency. The energy efficiency deviation coefficient is generated by comparing it with the dynamic energy efficiency optimization target threshold and using a weighted algorithm. , It is the normalized weighted value of the multi-parameter deviation values.

[0105] The comparison process is as follows:

[0106] The real-time specific power value is compared with the specific power dynamic optimization target threshold to generate a specific power deviation value;

[0107] The real-time isentropic efficiency is compared with the dynamic optimization target threshold of the isentropic efficiency to generate an isentropic efficiency deviation value.

[0108] The energy efficiency deviation coefficient The calculation formula is:

[0109] ,

[0110] in, This is the actual value of the specific power. This represents the actual value of isentropic efficiency. The specific power target value, The target value for isentropic efficiency is... and Let be the weight coefficient, and satisfy... .

[0111] It should be added to this embodiment that, The value range is from 0.60 to 0.70. The value range is from 0.30 to 0.40. and The value can be 0.6 or 0.4.

[0112] The above weighting coefficients are based on energy consumption contribution, optimization cost, and management priority in engineering practice, for the following reasons:

[0113] 1. The highest weighting for specific power is based on cost sensitivity. Electricity costs account for up to 80% of the life cycle cost of an air compressor system, and specific power is a direct determinant of electricity costs. Specific power can be monitored online, the data is easy to obtain and has strong real-time performance. The 0.6 weighting focuses on specific power, responding to the core demand of enterprises to reduce costs.

[0114] 2. The next weight for isentropic efficiency is local technical health. A decrease in efficiency indicates mechanical failure, and the maintenance cost is 5-10 times the energy consumption. Equipment health warnings are issued, and when the efficiency of some types of compressors is lower than the standard value, they need to be shut down for repair. The weight of 0.4 is used to monitor equipment health and avoid unnecessary maintenance costs and downtime losses.

[0115] In a preferred embodiment of the present invention, step S4 specifically includes the following steps:

[0116] When the energy efficiency deviation coefficient δ is greater than 0.1, the following dynamic optimization closed-loop control process is executed:

[0117] S401: Simulate and generate an optimized instruction set in a digital twin model, the instruction set being used for root cause diagnosis based on real-time pipeline differential pressure and load rate.

[0118] In this embodiment, it should be noted that the optimized instruction set includes pressure regulation instructions, unit scheduling instructions, frequency regulation instructions, pipeline maintenance instructions, and safety protection instructions. The energy efficiency parameters after execution include power parameters, flow parameters, pressure parameters, and temperature parameters. The power parameter refers to the operating power, the flow parameter refers to the actual gas production, the pressure parameter includes the exhaust pressure and the pipeline pressure difference, and the temperature parameter refers to the exhaust temperature.

[0119] The optimized instruction set, based on real-time network differential pressure and load rate, performs root cause diagnosis, specifically including the following steps:

[0120] When the energy efficiency deviation coefficient δ is greater than 0.1, determine whether the pipeline pressure difference is greater than 0.5 bar. If yes, it is determined to be a pipeline fault; if no, it is determined to be a compressor body fault.

[0121] When the load rate is less than 40%, it is determined that the no-load loss is too high. When the load rate is greater than or equal to 40% but less than or equal to 90%, it is determined that the operating parameters are out of balance. If the load rate is greater than 90%, it is determined that the equipment is at risk of overload.

[0122] The pipeline differential pressure is set at 0.5 bar based on the design resistance and allowable safety margin under the rated operating conditions of the system. This threshold can effectively filter out normal pressure fluctuations and accurately capture real pipeline flow resistance anomalies. The 40% load rate lower limit is a critical critical point for the economic operation of the compressor. When the load rate is below 40%, the compressor is in or frequently enters the "no-load" state most of the time, which is extremely uneconomical. The 90% load rate lower limit is a boundary that reserves safety margin and prevents equipment overload. Long-term operation at a load rate exceeding 90% will cause the compressor to approach its maximum working capacity, which will shorten the equipment life and increase the risk of failure.

[0123] S402: Optimize the instruction set and send it to the physical device for execution. The pressure adjustment rate is less than or equal to 0.3 bar / min. Multiple device actions are interlocked.

[0124] It should be noted that, in this embodiment, the pressure adjustment rate is less than or equal to 0.3 bar / min, and the multi-device action interlocking specifically includes the following steps:

[0125] (1) Collision detection:

[0126] Verify whether there are any action conflicts between the newly issued instructions and the instructions being executed.

[0127] (2) Priority determination:

[0128] Safety protection commands are enforced with priority, and pipeline maintenance commands take precedence over pressure regulation commands and frequency regulation commands.

[0129] (3) Instruction serialization execution:

[0130] High-conflict instructions enter the queuing mechanism and are executed after the preceding instructions are completed and resources are released;

[0131] (4) Status synchronization monitoring:

[0132] All equipment provides real-time feedback on pressure and load status to ensure that the pipeline pressure differential is less than or equal to 0.1 bar;

[0133] (5) Interlock release:

[0134] Once the instruction has been executed and the status verification has passed, control of the device is released.

[0135] Setting the pipeline differential pressure threshold to 0.1 bar ensures that the system is highly sensitive to changes in supply and demand, and can trigger the regulation mechanism during periods of small pressure fluctuations, thereby effectively avoiding system instability and production interruption caused by drastic pressure fluctuations or severe supply and demand imbalances.

[0136] S403: Collect energy efficiency parameters after execution in real time and update the state of the digital twin model.

[0137] It should be added that the energy efficiency parameters after execution include pressure parameters, power parameters, temperature parameters, flow parameters, and status parameters. The pressure parameters include exhaust pressure and pipeline pressure difference, the power parameters refer to the operating power, the temperature parameters refer to the exhaust temperature, the flow parameters refer to the actual exhaust volume, and the status parameters include the operating status and the number of hours of operation.

[0138] S404: Perform closed-loop verification and decision-making based on the updated energy efficiency deviation coefficient.

[0139] The closed-loop verification and decision-making in S404 includes the following steps:

[0140] (1) If the updated energy efficiency deviation coefficient δ is less than or equal to 0.05, terminate the dynamic optimization closed-loop control process;

[0141] If the exhaust pressure is greater than 110% of the rated exhaust pressure or the exhaust temperature is greater than 20°C of the rated temperature, the optimization will be forcibly stopped and the alarm protocol will be triggered.

[0142] (2) If the real-time pipeline differential pressure is greater than 0.8 bar or the differential pressure change rate is greater than 30% / min, the current optimization process is stopped and the pipeline maintenance subprocess is started;

[0143] (3) If the number of consecutive optimizations is greater than or equal to 3 and the energy efficiency deviation coefficient decrease rate is less than 5%, a freeze command is issued and a system fault diagnosis report is generated.

[0144] In this embodiment, it should be noted that the upper limit of the energy consumption deviation coefficient of 0.1 is based on industrial measurements, the lower limit of the energy consumption deviation coefficient of 0.05 is based on international standards, the exhaust pressure threshold is 110% of the rated exhaust pressure because the safety factor of the compressor's pressure-bearing components is 1.5, 110% is the elastic deformation limit, and pressure vessel regulations require that the melting value be less than or equal to 1.1 times the design pressure, the exhaust temperature threshold is the rated temperature + 20℃, which is a scientific choice based on comprehensive thermodynamics, material safety, industrial standards and measured data, the pipeline pressure difference threshold is 0.8 bar based on the pipeline energy consumption model, the pressure difference change rate threshold is 30% / min based on pipeline leakage characteristics, the maximum number of iterations is 3 times based on the industrial PLC control cycle limit and field verification, and the energy efficiency deviation coefficient threshold is 0.05 based on the convergence judgment of the optimization algorithm and the measurement noise influence range.

[0145] The forced termination of optimization and triggering of the alarm protocol specifically includes the following steps:

[0146] Terminate all optimization commands, switch to safety control mode, and take corresponding measures according to the alarm level: when a single parameter exceeds the limit, it is judged as a level one alarm, and an audible and visual alarm is triggered to notify the maintenance personnel for on-site handling.

[0147] Manually fine-tune the relevant parameters until the out-of-limit parameters fall back below the safe limit and stabilize for a period of time.

[0148] Only after manual authorization by maintenance personnel can the system switch back to "Energy Efficiency Optimization Mode" and resume automatic operation.

[0149] When both parameters exceed the limit simultaneously, it is judged as a level two alarm, and the unit dispatching command is executed to shut down the unit urgently. Professional maintenance personnel will conduct a comprehensive inspection and maintenance of the equipment. After the maintenance is completed, a manual power-on reset operation must be performed on site. Only after multiple safety confirmations can the system be allowed to be put back into manual start-up. And it must run stably for a period of time before it can be applied to be put back into optimization mode.

[0150] The term "single parameter exceeding limit" refers to either the exhaust pressure or the exhaust temperature exceeding its corresponding safety limit. The term "simultaneous exceeding of two parameters" refers to both the exhaust pressure and the exhaust temperature exceeding their respective safety limits. The preset pressure safety limit is determined based on the pressure-bearing design specifications of the compressor body and pipelines, and the safety valve setting pressure. The preset temperature safety limit is determined based on the flash point or thermal decomposition temperature of the compressor lubricating oil.

[0151] The specific steps of starting the pipeline maintenance sub-process are as follows:

[0152] (1) The digital twin generates intelligent maintenance instructions based on real-time pressure difference abnormal data. The core objective is to reduce the compressor discharge pressure in stages and smoothly to the preset safe pressure range.

[0153] (2) The system performs graded pressure reduction operation by coordinating the opening of the vent valve and the compressor unloading solenoid valve, and strictly monitors the pressure change rate to not exceed the safe rate threshold during the process.

[0154] (3) Verify the differential pressure status of the pipeline network after pressure reduction in real time, and exit the subprocess when the differential pressure returns to normal.

[0155] The generation of the system fault diagnosis report specifically includes the following steps:

[0156] (1) The digital twin model automatically retrieves the current abnormal state data and compares it with the historical fault feature database in multiple dimensions, and identifies the fault mode through the decision tree algorithm;

[0157] (2) Root cause analysis is performed based on real-time data and simulation results to accurately determine the location and severity of the fault.

[0158] (3) Automatically generate structured diagnostic reports and automatically create maintenance work orders and assign them to the mobile terminals of the relevant responsible personnel.

[0159] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0160] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing and processing energy efficiency data of industrial equipment based on digital twins, characterized in that, include: S1: System localization and twin construction: Based on the topology and characteristic parameters of physical devices, a dynamically mapped digital twin model is constructed, and a dynamic energy efficiency optimization target threshold is set in the digital twin model. S2: Real-time collection of core energy efficiency parameters: Sensors deployed by physical devices collect parameters directly related to energy efficiency in real time. These parameters include pressure parameters, flow parameters, power parameters, state parameters, and environmental parameters. S3: Dynamic Energy Efficiency Assessment Analysis: Performing a multi-dimensional energy efficiency assessment in a digital twin model includes the following steps: S301: Calculate real-time energy efficiency parameters: specific power SP and isentropic efficiency. ; S302: This combines specific power SP and isentropic efficiency. The energy efficiency deviation coefficient is generated by comparing it with the dynamic energy efficiency optimization target threshold and using a weighted algorithm. , This is a normalized weighted average of the deviation values ​​of multiple parameters; S4: Dynamic Data Optimization Processing When the energy efficiency deviation coefficient exceeds the set upper limit threshold, the dynamic optimization closed-loop control process is initiated, which includes the following steps: S401: Simulate and generate an optimized instruction set in a digital twin model, the instruction set performing root cause diagnosis based on real-time pipeline differential pressure and load rate; S402: The optimized instruction set is sent to the physical device for execution, the pressure adjustment rate does not exceed the safe rate threshold, and multiple devices are interlocked. S403: Real-time acquisition of energy efficiency parameters after execution, and updating of the digital twin model status; S404: Perform closed-loop verification and decision-making based on the updated energy efficiency deviation coefficient.

2. The method for optimizing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The target threshold for dynamic energy efficiency optimization is dynamically calibrated based on ambient temperature and equipment aging rate.

3. The method for optimizing and processing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The dynamic energy efficiency optimization target threshold includes the specific power dynamic optimization target threshold and the isentropic efficiency dynamic optimization target threshold.

4. The method for optimizing and processing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The comparison process is as follows: The real-time specific power value is compared with the specific power dynamic optimization target threshold to generate a specific power deviation value; The real-time isentropic efficiency is compared with the dynamic optimization target threshold of the isentropic efficiency to generate an isentropic efficiency deviation value.

5. The method for optimizing and processing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The energy efficiency deviation coefficient The calculation formula is: ， in, This is the actual value of the specific power. This represents the actual value of isentropic efficiency. The specific power target value, The target value for isentropic efficiency is... and Let be the weight coefficient, and satisfy... .

6. The method for optimizing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The optimized instruction set includes pressure regulation instructions, unit scheduling instructions, frequency regulation instructions, pipeline maintenance instructions, and safety protection instructions.

7. The method for optimizing and processing energy efficiency data of industrial equipment based on digital twins according to claim 1, characterized in that, The closed-loop verification and decision-making in S404 includes the following steps: (1) If the updated energy efficiency deviation coefficient is less than or equal to the lower limit of the set threshold, terminate the dynamic optimization closed-loop control process. (2) If the exhaust pressure exceeds the preset pressure safety limit or the exhaust temperature exceeds the preset temperature safety limit, the optimization will be forcibly stopped and the alarm protocol will be triggered. (3) If the real-time pipeline differential pressure exceeds the preset differential pressure abnormality threshold or the differential pressure change rate exceeds the preset sudden change rate threshold, the current optimization process is stopped and the pipeline maintenance subprocess is started. (4) If the number of consecutive optimizations reaches the preset maximum number of iterations and the energy efficiency deviation coefficient decrease rate is lower than the preset minimum convergence rate, a freeze command is issued and a system fault diagnosis report is generated.

Images