Performance optimization and service life prediction method for water-cooled frequency converter under high and cold working conditions based on digital twinning

Abstract

The invention relates to the field of life prediction, and discloses a performance optimization and life prediction method for a water-cooled frequency converter under a high and cold working condition based on digital twinning, which is used for providing technical support for stable operation of wind power equipment in an extreme environment. Comprising the steps of generating accurate heat source data matched with a current working condition based on iterative calculation of real-time electrical parameters and initial junction temperature, eliminating static model errors, inputting the heat source data and low-temperature viscosity characteristics of cooling liquid into a simulation model, outputting multi-physical-field state data such as junction temperature, temperature distribution and stress of key parts, and obtaining a simulation result. And combining a stress change process and a damage accumulation model, quantifying accumulated damage of the key part, dynamically adjusting the rotating speed of a cooling water pump according to the junction temperature and the environment temperature, balancing anti-freezing protection and heat dissipation requirements, fusing damage data and future load prediction, and outputting a residual service life quantification result. The reliability, energy efficiency and economical efficiency of equipment under the high and cold working condition are remarkably improved, and technical support is provided for power electronic equipment in the fields of wind power, rail transit and the like.

Description

Technical Field

[0001] This invention relates to the field of lifespan prediction, and in particular to a method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins. Background Technology

[0002] With the global energy structure shifting towards renewable energy, wind power, as an important form of clean energy, has seen its installed capacity grow rapidly. Water-cooled frequency converters, as the core power conversion and control equipment in wind power systems, directly impact the power generation efficiency and economic benefits of wind farms due to their operational reliability and lifespan. Especially in high-altitude and cold regions, wind power equipment often faces extreme low-temperature environments, posing significant challenges to the performance optimization and lifespan prediction of water-cooled frequency converters.

[0003] In extremely cold operating conditions, water-cooled frequency converters need to address two major issues simultaneously: Low temperatures cause a significant increase in coolant viscosity and fluid resistance, which can easily lead to uneven flow distribution, local overheating, or coolant phase change and solidification, resulting in excessive junction temperature of power modules or system freezing damage. In low-temperature environments, the power module packaging material generates high stress due to the mismatch in thermal expansion coefficients. At the same time, the brittle transition temperature of the material increases, and the risks of fatigue damage and brittle fracture are superimposed, resulting in a significant reduction in equipment lifespan.

[0004] Therefore, we propose a method for optimizing the performance and predicting the lifespan of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins to solve the above problems. Summary of the Invention

[0005] This invention provides a method for optimizing the performance and predicting the lifespan of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins, which can provide technical support for the stable operation of wind power equipment in extreme environments.

[0006] The first aspect of this invention provides a method for performance optimization and lifespan prediction of a water-cooled frequency converter under high-altitude and cold-weather conditions based on digital twins. The method includes: iterative calculation based on the real-time operating electrical parameters and initial junction temperature data of the water-cooled frequency converter to obtain accurate heat source data matching the current junction temperature state; inputting the accurate heat source data and coolant low-temperature viscosity characteristics data into a simulation model, and outputting multi-physics field state data after calculation, including junction temperature data of each power module, radiator temperature distribution data, and stress data of key components; inputting the stress data of key components and its change process from the multi-physics field state data into a damage accumulation model to generate quantitative data of cumulative damage to key components; outputting optimized control commands for real-time adjustment of the cooling water pump speed based on the junction temperature data and ambient temperature data from the multi-physics field state data; and outputting predicted remaining service life data of the water-cooled frequency converter based on the quantitative data of cumulative damage to key components and predicted data of future load conditions.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the method includes: generating first loss distribution data based on the real-time operating electrical parameters and first junction temperature data; using the first loss distribution data as input conditions to perform thermal simulation of the water-cooled inverter and its heat dissipation system to obtain second junction temperature data; generating loss correction coefficient data based on the difference between the second junction temperature data and the first junction temperature data; correcting the first loss distribution data using the loss correction coefficient data to generate second loss distribution data; updating the first loss distribution data with the second loss distribution data and updating the first junction temperature data with the second junction temperature data, returning to perform thermal simulation, until the change in junction temperature data obtained from two adjacent thermal simulations is less than a preset threshold, and using the loss distribution data finally obtained in the current iteration as accurate heat source data.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the method includes: establishing dynamic fluid network resistance parameters based on the low-temperature viscosity characteristics data of the coolant and the geometric structure data of the water-cooled heat dissipation system piping; inputting the precise heat source data and the dynamic fluid network resistance parameters into a fluid and heat transfer co-simulation program to obtain the flow distribution data of each parallel branch in the entire cooling loop and the flow velocity distribution data of each microchannel inside the radiator; performing transient heat transfer calculations based on the flow distribution data, the flow velocity distribution data, and the precise heat source data to obtain the instantaneous junction temperature data of each power chip and the temperature field distribution data of the radiator substrate and cold plate; and obtaining the stress and strain data of the key welding layer and connection parts caused by thermal expansion mismatch based on the temperature field distribution data and its change history over time, combined with the thermal expansion coefficient data of each layer of packaging material in the power module.

[0009] Optionally, in a third implementation of the first aspect of the present invention, the method includes: processing the stress data of the key parts to obtain first-type stress intensity factor data based on preset defect feature size data; obtaining material fatigue strength data and material fracture resistance data corresponding to the current temperature state based on the temperature distribution data in the multi-physics state data; calculating the damage increment data of the current time step based on the rate of change of the stress data of the key parts, the first-type stress intensity factor data, the material fatigue strength data, and the material fracture resistance data; accumulating the damage increment data generated at each time step within the running time history to obtain the cumulative damage value from the start time to the current time, and outputting it as the cumulative damage quantification data of the key parts.

[0010] Optionally, in the fourth implementation of the first aspect of the present invention, local temperature data corresponding to the stress data of the key parts is obtained, and brittleness risk factor data is generated based on pre-stored material low-temperature brittle transition temperature data; based on the brittleness risk factor data, the proportional coefficient data used to characterize the relative weights of fatigue damage mechanism and brittle fracture mechanism in the damage evolution equation is dynamically adjusted; using the adjusted proportional coefficient data, the rate of change of the stress data of the key parts, the first type of stress intensity factor data, the material fatigue strength data, and the material fracture resistance data, the damage increment data of the current time step is recalculated, and the cumulative damage quantification data of the key parts is updated accordingly.

[0011] Optionally, in a fifth implementation of the first aspect of the present invention, the step of calculating and outputting an optimized control command for real-time adjustment of the cooling water pump speed based on the junction temperature data and ambient temperature data in the multiphysics state data includes: calculating the minimum safe circulation flow rate required to prevent the coolant from undergoing phase change and solidification in the pipeline based on the ambient temperature data and the physical property parameters of the coolant, and generating antifreeze flow rate threshold data; analyzing the dispersion and maximum value of the junction temperature data of each power module in the multiphysics state data, and calculating the theoretical heat dissipation flow rate required to maintain all power modules at a safe and uniform temperature level based on a preset junction temperature safety limit, and generating target heat dissipation flow rate data; establishing a decision function that includes the antifreeze flow rate threshold data as a constraint and approaches the target heat dissipation flow rate data as the optimization objective, solving the function to obtain the global optimal total flow rate demand data of the system under the current operating conditions; mapping the global optimal total flow rate demand data and the characteristic curve data of the water pump to obtain the corresponding target water pump speed, and generating an optimized control command for driving the physical water pump speed adjustment.

[0012] Optionally, in the sixth implementation of the first aspect of the present invention, the ambient temperature data is monitored in real time. When the ambient temperature data is lower than a preset extremely low temperature threshold, a system preheating start-up flag signal is generated. Based on the system preheating start-up flag signal and the rheological characteristics data of the coolant at extremely low temperatures, the safe preheating flow rate required to prevent water pump overload and ensure coolant flow during the preheating stage is calculated, and a water pump speed command for the preheating stage is generated. The water pump speed command for the preheating stage is executed, and the average temperature data of the main circuit of the coolant is continuously monitored until the average temperature data reaches the preset low temperature safe operation threshold. When the average temperature data reaches the low temperature safe operation threshold, the system preheating start-up flag signal is revoked, and an optimized control command is generated.

[0013] Optionally, in the seventh implementation of the first aspect of the present invention, the step of calculating and outputting the remaining service life prediction data of the water-cooled frequency converter based on the cumulative damage quantification data of the key components and the prediction data of future load conditions includes: generating virtual wind speed time series data based on the long-term meteorological statistical characteristics of the target wind field; converting the virtual wind speed time series data into virtual long-term power load sequence data corresponding to the water-cooled frequency converter according to the power characteristic curve of the wind turbine; inputting the virtual long-term power load sequence data into a twin model to obtain the thermal stress sequence data experienced by the key components under the virtual load; inputting the thermal stress sequence data into a damage accumulation model to perform damage increment calculation and accumulation process, simulating the damage accumulation process data from the current moment under future virtual operating conditions; using the cumulative damage quantification data of the key components as the initial state, superimposing it with the damage accumulation process data until the total accumulation reaches a preset failure damage threshold, recording the virtual operating time experienced from the current moment to reaching the threshold, and outputting the remaining service life prediction data.

[0014] Optionally, in the eighth implementation of the first aspect of the present invention, the method further includes: arranging temperature sensors at designated locations on the physically water-cooled inverter to obtain actual temperature measurement data at key monitoring points; comparing the theoretical temperature data at the corresponding locations generated by the twin model simulation with the actual temperature measurement data to generate temperature deviation data; based on the temperature deviation data, adjusting the preset undetermined parameters related to heat dissipation performance in the twin model through a parameter identification algorithm to generate updated model calibration parameters; and loading the updated model calibration parameters into the twin model for subsequent performance simulation and life prediction calculations.

[0015] Beneficial effects: Eliminates the deviation of heat source data caused by ignoring real-time operating condition changes in traditional methods, improves simulation accuracy, accurately reflects nonlinear thermal behavior in cold environments, and provides a reliable foundation for subsequent multiphysics field analysis; Quantitative analysis of the impact of increased viscosity at low temperatures on flow distribution, junction temperature uniformity, and stress distribution reveals the underlying mechanisms of equipment failure under cold and harsh conditions; it also helps to identify potential risks in advance, providing a basis for structural optimization and maintenance strategies. It automatically balances antifreeze protection and heat dissipation needs at extremely low temperatures, avoiding freezing blockage due to insufficient flow or pump overload due to excessive flow; it significantly reduces the failure rate of equipment caused by thermal management failure and extends continuous operation time. Accurately capture the dynamic impact of material brittle transition on damage mechanisms in cold environments, solve the problem of overestimation of life caused by traditional models ignoring low-temperature brittleness, and achieve accurate prediction of remaining life through virtual load sequence mapping, providing quantitative support for operation and maintenance decisions and reducing the risk of unplanned downtime. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an embodiment of the method for performance optimization and life prediction of water-cooled frequency converters under high-altitude and cold conditions based on digital twins in this invention. Figure 2 This is a schematic diagram of an embodiment of the device for optimizing the performance and predicting the lifespan of a water-cooled frequency converter under high-altitude and cold-weather conditions based on digital twins in this invention. Figure 3 This is a schematic diagram of multiphysics state data simulation. Detailed Implementation

[0017] This invention provides a method for optimizing the performance and predicting the lifespan of water-cooled frequency converters under extreme cold conditions based on digital twins, providing technical support for the stable operation of wind power equipment in extreme environments. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 and Figure 3 One embodiment of the method for performance optimization and life prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins in this invention includes: 101. Based on the real-time operating electrical parameters of the water-cooled frequency converter and the initial junction temperature data, iterative calculations are performed to obtain accurate heat source data that matches the current junction temperature state after correction. It is understood that the executing entity of this invention can be a device for optimizing the performance and predicting the lifespan of a water-cooled frequency converter under high-altitude and cold-weather conditions based on digital twins, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0019] Specifically, based on real-time operating electrical parameters and first junction temperature data, first loss distribution data is generated through calculation using a semiconductor physical characteristic model; Using the first loss distribution data as input, perform thermal simulation of the water-cooled inverter and its heat dissipation system to obtain the second junction temperature data, which includes the junction temperature values ​​of each power module. Based on the difference between the second junction temperature data and the first junction temperature data, a dimensionless loss correction coefficient is generated using a correction function based on the physical relationship between semiconductor carrier mobility and threshold voltage temperature drift. The first loss distribution data is corrected using the loss correction coefficient data to generate the second loss distribution data; Update the first loss distribution data with the second loss distribution data, and update the first junction temperature data with the second junction temperature data. Return to execute thermal simulation until the change in junction temperature data obtained from two adjacent thermal simulations is less than a preset threshold. Output the loss distribution data finally obtained in the current iteration as accurate heat source data.

[0020] 102. Input the precise heat source data and the low-temperature viscosity characteristics data of the coolant into the simulation model, and output multi-physics state data including junction temperature data of each power module, heat sink temperature distribution data and stress data of key parts after calculation. Specifically, based on the low-temperature viscosity characteristics of the coolant and the geometric structure data of the water-cooled heat dissipation system piping, dynamic fluid network resistance parameters characterizing the flow resistance of each branch in relation to temperature and flow velocity are established. By inputting precise heat source data and dynamic fluid network resistance parameters into the fluid and heat transfer joint simulation program, the flow distribution data of each parallel branch in the entire cooling loop and the flow velocity distribution data of each microchannel inside the radiator are calculated. Based on flow distribution data, velocity distribution data, and precise heat source data, transient heat transfer calculations are performed to obtain the instantaneous junction temperature data of each power chip and the temperature field distribution data of the heat sink substrate and cold plate. Based on the temperature field distribution data and its change over time, combined with the thermal expansion coefficient data of each layer of packaging material in the power module, the stress and strain data of the key welding layer and connection parts caused by thermal expansion mismatch were calculated.

[0021] 103. Input the stress data and its change process of key parts in the multiphysics state data into the damage accumulation model, and generate quantitative data of cumulative damage of key parts after calculation. Specifically, the stress data of key parts are processed, and based on the preset defect feature size data, the first type of stress intensity factor data used to characterize the local stress field intensity is calculated. Based on the temperature distribution data in the multiphysics state data, obtain the material fatigue strength data and material fracture resistance data corresponding to the current temperature state; Based on the rate of change of stress data in key parts, the first type of stress intensity factor data, material fatigue strength data, and material fracture resistance data, the damage increment data at the current time step is calculated by fusing the damage evolution equation of fatigue damage mechanism and brittle fracture mechanism. The incremental damage data generated at each time step within the running time history are accumulated to obtain the cumulative damage value from the start time to the current time, which serves as the quantitative data of cumulative damage to key parts.

[0022] Furthermore, local temperature data corresponding to stress data in key locations are acquired, and a brittleness risk factor is calculated based on pre-stored material low-temperature brittle transition temperature data to characterize the degree of brittle fracture risk at the current temperature. Based on the brittleness risk factor data, the proportional coefficient data used to characterize the relative weights of fatigue damage mechanism and brittle fracture mechanism in the damage evolution equation is dynamically adjusted. Using the adjusted proportional coefficient data, the rate of change of stress data in key locations, the first type of stress intensity factor data, material fatigue strength data, and material fracture resistance data, the damage increment data at the current time step is recalculated, and the cumulative damage quantification data of key locations is updated accordingly.

[0023] 104. Based on the junction temperature data and ambient temperature data in the multiphysics state data, calculate and output the optimized control command for real-time adjustment of the cooling water pump speed; Specifically, based on ambient temperature data and coolant physical property parameters, the minimum safe circulation flow rate required to prevent coolant from undergoing phase change and solidification in the pipeline is calculated, and antifreeze flow rate threshold data is generated. Based on the junction temperature data of each power module in the multiphysics state data, we analyze its dispersion and maximum value, and combined with the preset junction temperature safety limit, we calculate the theoretical heat dissipation flow required to maintain all power modules at a safe and uniform temperature level, and generate target heat dissipation flow data. Establish a decision function that includes antifreeze flow threshold data as a constraint and aims to approach the target heat dissipation flow data as the optimization objective. Solve the function to obtain the global optimal total flow demand data of the system under the current operating conditions. Based on the global optimal total flow demand data and the characteristic curve data of the water pump, the corresponding target water pump speed is obtained by mapping, and an optimized control command for driving the physical water pump speed regulation is generated.

[0024] Furthermore, the system monitors ambient temperature data in real time. When the ambient temperature falls below a preset extremely low temperature threshold, a system preheating start signal is generated. Based on the system preheating start signal and the rheological characteristics of the coolant at extremely low temperatures, the system calculates the safe preheating flow rate required to prevent pump overload and ensure coolant flow during the preheating phase, and generates a preheating phase pump speed command. The preheating phase pump speed command is executed, and the average temperature data of the coolant's main circuit is continuously monitored until the average temperature data reaches a preset low-temperature safe operating threshold. Once the average temperature data reaches the low-temperature safe operating threshold, the system preheating start signal is revoked, and the system switches to generating optimized control commands.

[0025] 105. Based on the quantitative data of cumulative damage to key components and the prediction data of future load conditions, calculate and output the prediction data of the remaining service life of the water-cooled frequency converter.

[0026] Specifically, based on the long-term meteorological statistical characteristics of the target wind field, a virtual wind speed time series data that conforms to its wind speed probability distribution and covers the next few years is generated. Based on the power characteristic curve of the wind turbine, the virtual wind speed time series data is converted into virtual long-term power load series data corresponding to the water-cooled frequency converter. The virtual long-term power load sequence data is input into the twin model to obtain the thermal stress sequence data experienced by key parts under the virtual load. The thermal stress sequence data is input into the damage accumulation model to perform damage increment calculation and accumulation process, and simulate the damage accumulation process data from the current moment under future virtual operating conditions. The accumulated damage quantification data of key parts is used as the initial state and superimposed with the damage accumulation process data until the total accumulated amount reaches the preset failure damage threshold. The virtual running time from the current moment to reaching the threshold is recorded and output as the remaining service life prediction data.

[0027] 106. Install temperature sensors at designated locations on the physically water-cooled inverter to obtain actual temperature measurement data at key monitoring points; The theoretical temperature data at the corresponding location generated by the twin model simulation is compared with the actual temperature measurement data to generate a set of temperature deviation data; Based on temperature deviation data, the pre-set, undetermined parameters related to heat dissipation performance in the twin model are adjusted through parameter identification algorithms to generate a set of updated model calibration parameters. The updated model calibration parameters are loaded into the twin model for subsequent performance simulation and lifetime prediction calculations.

[0028] In this embodiment of the invention, accurate heat source data is obtained through iterative calculation and correction. Compared with traditional methods, this more accurately reflects the actual heat source state of the water-cooled frequency converter, providing a reliable foundation for subsequent performance analysis and life prediction, and improving the accuracy and reliability of the overall analysis. By comprehensively considering multiple physical field factors such as heat, fluid, and stress, the working state of the water-cooled frequency converter under extremely cold conditions can be fully analyzed, revealing problems that are difficult to detect with single physical field analysis, providing a more comprehensive basis for performance optimization. In the damage calculation of key components, brittleness risk factors are considered and the parameters of the damage evolution equation are dynamically adjusted, enabling more accurate simulation of the material damage process under different temperature and stress conditions, adapting to the complex and variable operating environment under extremely cold conditions, and improving the accuracy of life prediction. Real-time data calculation optimizes cooling water pump speed control commands and sets preheating start-up controls at extremely low temperatures, effectively preventing issues such as coolant phase change solidification and pump overload. This ensures the safe and stable operation of water-cooled frequency converters in cold conditions, extending equipment lifespan. Based on future load condition predictions, remaining lifespan is calculated, providing a scientific basis for maintenance personnel to plan equipment maintenance and replacement in advance. This avoids production interruptions and economic losses due to sudden equipment failures, improving the economy and reliability of equipment operation. Calibration of the digital twin model using actual temperature data makes the model more closely resemble the actual operating conditions of the physical equipment, improving the accuracy of the model in performance simulation and lifespan prediction, and enhancing the practicality and effectiveness of digital twin technology in engineering applications.

[0029] Figure 2 This is a schematic diagram of a digital twin-based water-cooled inverter performance optimization and lifespan prediction device for cold-weather conditions, provided in an embodiment of the present invention. This digital twin-based water-cooled inverter performance optimization and lifespan prediction device 200 can vary significantly depending on its configuration or performance. It may include one or more central processing units (CPUs) 210 (e.g., one or more processors) and a memory 220, and one or more storage media 230 (e.g., one or more mass storage devices) for storing application programs 233 or data 232. The memory 220 and storage media 230 can be temporary or persistent storage. The program stored in the storage media 230 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the digital twin-based water-cooled inverter performance optimization and lifespan prediction device 200. Furthermore, the processor 210 can be configured to communicate with the storage medium 230 and execute a series of instructions stored in the storage medium 230 on the digital twin-based water-cooled inverter high-altitude and cold-weather performance optimization and life prediction device 200.

[0030] The digital twin-based water-cooled frequency converter performance optimization and lifespan prediction device 200 for cold-weather operation may also include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input / output interfaces 260, and / or one or more operating systems 231, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 2 The illustrated structure of the water-cooled frequency converter performance optimization and life prediction device based on digital twins for cold-weather conditions does not constitute a limitation on the water-cooled frequency converter performance optimization and life prediction device based on digital twins for cold-weather conditions. It may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0031] The present invention also provides a device for optimizing the performance and predicting the lifespan of a water-cooled frequency converter under cold conditions based on digital twins. The device includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the method for optimizing the performance and predicting the lifespan of a water-cooled frequency converter under cold conditions based on digital twins in the above embodiments.

[0032] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the method for optimizing the performance and predicting the lifespan of a water-cooled frequency converter under high-altitude and cold-weather conditions based on digital twins.

[0033] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0034] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0035] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins, characterized in that, include: Based on the real-time operating electrical parameters of the water-cooled frequency converter and the initial junction temperature data, iterative calculations are performed to obtain accurate heat source data that matches the current junction temperature state. The precise heat source data and the low-temperature viscosity characteristics data of the coolant are input into the simulation model, and after calculation, multi-physics state data are output, including junction temperature data of each power module, heat sink temperature distribution data and stress data of key parts. The stress data and its change process of key parts in the multiphysics state data are input into the damage accumulation model to generate quantitative data of cumulative damage of key parts. Based on the junction temperature data and ambient temperature data in the multiphysics state data, an optimized control command for real-time adjustment of the cooling water pump speed is output. Based on the quantitative data of cumulative damage to the key components and the prediction data of future load conditions, the remaining service life prediction data of the water-cooled frequency converter is output.

2. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins as described in claim 1, is characterized in that, include: First loss distribution data is generated based on the real-time operating electrical parameters and the first junction temperature data. Using the first loss distribution data as input, perform thermal simulation of the water-cooled inverter and its heat dissipation system to obtain the second junction temperature data. Based on the difference between the second junction temperature data and the first junction temperature data, loss correction coefficient data is generated; The first loss distribution data is corrected using the loss correction coefficient data to generate the second loss distribution data; The first loss distribution data is updated with the second loss distribution data, and the first junction temperature data is updated with the second junction temperature data. The thermal simulation is then performed until the change in junction temperature data obtained from two adjacent thermal simulations is less than a preset threshold. The loss distribution data finally obtained in the current iteration is then used as the accurate heat source data.

3. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 2, characterized in that, include: Based on the low-temperature viscosity characteristics of the coolant and the geometric structure data of the water-cooled heat dissipation system piping, dynamic fluid network resistance parameters are established. The precise heat source data and the dynamic fluid network resistance parameters are input into the fluid and heat transfer joint simulation program to obtain the flow distribution data of each parallel branch in the entire cooling loop and the flow velocity distribution data of each microchannel inside the radiator. Based on the flow distribution data, the flow velocity distribution data, and the precise heat source data, transient heat transfer calculations are performed to obtain the instantaneous junction temperature data of each power chip and the temperature field distribution data of the heat sink substrate and the cold plate. Based on the temperature field distribution data and its change over time, combined with the thermal expansion coefficient data of each layer of packaging material in the power module, stress and strain data of key welding layers and connection parts caused by thermal expansion mismatch are obtained.

4. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 3, is characterized in that, include: The stress data of the key parts are processed, and the first type of stress intensity factor data is obtained based on the preset defect feature size data. Based on the temperature distribution data in the multiphysics state data, the material fatigue strength data and material fracture resistance data corresponding to the current temperature state are obtained. Based on the rate of change of stress data at the key locations, the first type of stress intensity factor data, the material fatigue strength data, and the material fracture resistance data, the damage increment data at the current time step is calculated. The incremental damage data generated at each time step within the running time history is accumulated to obtain the cumulative damage value from the start time to the current time, and the output is the quantitative data of cumulative damage of key parts.

5. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 4, characterized in that, Obtain the local temperature data corresponding to the stress data of the key parts, and generate brittleness risk factor data based on the pre-stored material low-temperature brittle transition temperature data; Based on the brittleness risk factor data, dynamically adjust the proportional coefficient data in the damage evolution equation used to characterize the relative weights of fatigue damage mechanism and brittle fracture mechanism; Using the adjusted proportional coefficient data, the rate of change of stress data in the key parts, the first type of stress intensity factor data, the material fatigue strength data, and the material fracture resistance data, the damage increment data at the current time step is recalculated, and the cumulative damage quantification data of the key parts is updated accordingly.

6. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 4, characterized in that, The step of calculating and outputting optimized control commands for real-time adjustment of cooling water pump speed based on junction temperature data and ambient temperature data from the multiphysics state data includes: Based on the ambient temperature data and the physical property parameters of the coolant, the minimum safe circulation flow rate required to prevent the coolant from undergoing phase change and solidification in the pipeline is calculated, and antifreeze flow rate threshold data is generated. Based on the junction temperature data of each power module in the multiphysics state data, analyze its dispersion and maximum value, and combine it with the preset junction temperature safety limit to calculate the theoretical heat dissipation flow required to maintain all power modules at a safe and uniform temperature level, and generate target heat dissipation flow data. Establish a decision function that includes the antifreeze flow threshold data as a constraint and aims to approach the target heat dissipation flow data as the optimization objective. Solve the function to obtain the global optimal total flow demand data of the system under the current operating conditions. Based on the global optimal total flow demand data and the characteristic curve data of the water pump, the corresponding target water pump speed is mapped to generate an optimized control command for driving the physical water pump speed regulation.

7. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 6, characterized in that, The system monitors the ambient temperature data in real time, and generates a system preheating start signal when the ambient temperature data is lower than the preset extreme low temperature threshold. Based on the system preheating start flag signal and the rheological characteristics data of the coolant at extremely low temperatures, the safe preheating flow rate required to prevent water pump overload and ensure coolant flow during the preheating stage is calculated, and the water pump speed command for the preheating stage is generated. The pump speed command during the preheating stage is executed, and the average temperature data of the main circuit of the coolant is continuously monitored until the average temperature data reaches the preset low temperature safe operation threshold. Once the average temperature data reaches the low-temperature safe operation threshold, the system preheating start signal is revoked, and the optimized control command is switched to be generated.

8. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 6, characterized in that, The calculation and output of the remaining service life prediction data of the water-cooled frequency inverter based on the cumulative damage quantification data of the key components and the prediction data of future load conditions includes: Based on the long-term meteorological statistical characteristics of the target wind field, virtual wind speed time series data is generated; Based on the power characteristic curve of the wind turbine, the virtual wind speed time series data is converted into virtual long-term power load series data corresponding to the water-cooled frequency converter. The virtual long-term power load sequence data is input into the twin model to obtain the thermal stress sequence data experienced by key parts under the virtual load. The thermal stress sequence data is input into the damage accumulation model to perform damage increment calculation and accumulation process, and to simulate and obtain damage accumulation process data from the current moment under future virtual operating conditions. The accumulated damage quantification data of the key parts is used as the initial state and superimposed with the damage accumulation process data until the total accumulated amount reaches the preset failure damage threshold. The virtual running time from the current moment to reaching the threshold is recorded, and the remaining service life prediction data is output.

9. The method for performance optimization and lifespan prediction of water-cooled frequency converters under high-altitude and cold-weather conditions based on digital twins according to claim 1, characterized in that, Also includes: Temperature sensors are placed at designated locations on the physically water-cooled frequency converter to obtain actual temperature measurement data at key monitoring points. The theoretical temperature data at the corresponding location generated by the twin model simulation is compared with the actual temperature measurement data to generate temperature deviation data; Based on the temperature deviation data, the preset undetermined parameters related to heat dissipation performance in the twin model are adjusted by the parameter identification algorithm to generate updated model calibration parameters. The updated model calibration parameters are loaded into the twin model for subsequent performance simulation and lifetime prediction calculations.

Images