Electric control method and system for monitoring dangerous point part of motor rotor in real time

Abstract

The invention provides an electric control method and system for monitoring a dangerous point part of a motor rotor in real time, and the method comprises the steps: collecting simulation data of the service of a new energy automobile motor rotor under different working conditions, constructing a simulation data basic library, and obtaining a simulation result of the service of the rotor in a multi-damage and multi-load coupling state; according to the simulation result, identifying a dangerous point part of the rotor, and constructing a dangerous point part prediction model based on a deep neural network by utilizing simulation data accumulation of the dangerous point part; acquiring rotor service data in a target new energy automobile motor in real time through an electric control system, preprocessing the working condition data, and inputting the preprocessed working condition data into the dangerous point part prediction model to obtain single-cycle damage data; and performing real-time damage normalization state monitoring according to the one-week damage data, and automatically triggering early warning in response to monitoring that the dangerous point part has or is about to have damage.

Description

Technical Field

[0001] This invention relates to the field of electric motors for new energy vehicles, and in particular to an electronic control method and system for real-time monitoring of dangerous points on the motor rotor. Background Technology

[0002] In the field of new energy vehicles, the motor, as a core power component, directly affects the vehicle's power, economy, and reliability. The connection quality between the motor rotor and the shaft plays a decisive role in the stable operation of the motor. Currently, the industry widely adopts the interference fit method to connect the motor rotor and the shaft. This connection method has advantages such as simple structure, good centering, and strong load-bearing capacity, which can effectively ensure the NVH (noise, vibration, and harshness) performance of the motor during operation.

[0003] However, in actual operating conditions, the interference fit between the motor rotor and the shaft faces many complex challenges. When a new energy vehicle is in motion, the motor rotor rotates at high speed, generating a strong centrifugal force. Taking a high-speed permanent magnet synchronous motor for vehicles with a peak speed of 20,000 r / min as an example, the centrifugal force has a significant impact on the interference fit between the shaft and the iron core. Under the long-term action of this centrifugal force, the interference fit is prone to loosening, which in turn leads to the deterioration of the rotor's dynamic balance, resulting in abnormal vibration and noise during motor operation, and in severe cases, even motor failure.

[0004] Meanwhile, during the operation of new energy vehicles, factors such as road bumps subject the motor to continuous vibration loads. This vibration causes minute relative displacements between the interference fit surfaces, leading to fretting wear. Over time, fretting wear gradually intensifies, causing material damage to the mating surfaces, reducing fit precision, and significantly decreasing connection reliability. Furthermore, the motor generates heat during operation, causing the rotor and shaft temperatures to rise. Under high-temperature conditions, the mechanical properties of the materials change, resulting in creep. This gradually reduces the interference fit, further weakening the stability of the rotor-shaft connection.

[0005] To address these issues, the industry has undertaken numerous attempts. Some companies have optimized the materials used in the motor shaft and rotor core, hoping to improve their performance under complex operating conditions. For example, coating the motor shaft with a specific composite material utilizes the difference in thermal expansion between different materials to cause the gap between the rotor core and the motor shaft to change with temperature, thus maintaining the stability of the interference fit. However, this method requires extremely high standards in material selection and coating processes, is costly, and its effectiveness is limited by various factors, making widespread application difficult. Other companies have attempted to improve assembly processes, such as using a heat-shrink shaft process to enhance the connection strength between the rotor core and the shaft. However, due to the large shaft size of large motors, the heat-shrink shaft process can easily lead to problems such as core "shaft creep" when large pressure equipment is unavailable, resulting in motor malfunctions.

[0006] Currently, while some research and improvement measures exist for the interference fit between the motor rotor and shaft, they still have significant shortcomings in addressing damage issues under complex operating conditions. Existing technologies cannot accurately and in real-time identify potential damage points at the interference fit, making it difficult to achieve real-time monitoring and early warning of the motor's operating status. Once a motor malfunctions, repairs are often only possible after the fact, which not only causes vehicle downtime and significant inconvenience to users but may also result in high repair costs, severely hindering the development of the new energy vehicle industry. Therefore, there is an urgent need for an efficient and convenient technical solution to accurately identify and monitor potential damage points in real-time, ensuring the safe and stable operation of the motor. Summary of the Invention

[0007] In order to overcome the above-mentioned technical defects, the purpose of this invention is to provide an electronic control method and system for real-time monitoring of dangerous parts of motor rotors. Through electronic control, management, early warning and alarm can be realized, effectively solving the problem of damage to the rotor and shaft of new energy vehicle motors under complex working conditions due to interference fit.

[0008] This invention discloses an electronic control method for real-time monitoring of dangerous points on a motor rotor, comprising the following steps: We collected simulation data of new energy vehicle motor rotor service under different working conditions, built a pure simulation data base library, and obtained simulation results of rotor service under multiple damage and multiple load coupling states. Based on the simulation results, dangerous points on the rotor are identified, and a prediction model for dangerous points based on deep neural networks is constructed using the accumulated simulation data of these dangerous points. The rotor service data of the target new energy vehicle motor is collected in real time by the electronic control system. After the data is preprocessed, it is input into the prediction model of the dangerous point to obtain the damage data of a single week. Based on the damage data from this single week, real-time damage normalization status monitoring is performed, and an early warning is automatically triggered in response to the detection of existing or impending damage at dangerous locations.

[0009] Preferably, the simulation data consists of different combinations of physical information parameters obtained by physical sensing during the service of the rotor in the motor of a new energy vehicle. These combinations of physical information parameters include, but are not limited to, centrifugal fatigue damage data and single-cycle vibration damage data obtained by elastic finite element analysis, as well as plastic / creep damage data affected by interference obtained by temperature field finite element analysis.

[0010] Preferably, identifying the dangerous locations of the rotor based on the simulation results includes the following steps: Based on the simulation results, through damage numerical quantification analysis, the parts or nodes with the largest damage values ​​of multiple damages and multiple load couplings within a single cycle are selected as dangerous points.

[0011] Preferably, the real-time acquisition of rotor service data in the target new energy vehicle motor via the electronic control system includes the following steps: The electronic control system uses dynamic sampling frequency logic to collect operating condition data and rotor service data. The rotor service data includes, but is not limited to, operating condition data. Different sampling frequencies are used in the motor start-up phase, steady state phase, and shutdown phase, and the electronic control system collects the data in real time through CAN / LIN bus or dedicated sensor interface. The operating condition data includes, but is not limited to, the operating condition parameters such as rotor initial temperature, speed increase rate, load increase rate, speed, torque, shaft power, and axial tension for each cycle. One cycle refers to the service process of the target new energy vehicle motor after completing one cycle from the start-up phase to the steady-state phase and from the steady-state phase to the shutdown phase.

[0012] Preferably, different sampling frequencies are used during the motor start-up phase, steady-state phase, and shutdown phase, including: The startup phase is set to the first 30 seconds of rotor startup, with a sampling frequency of 100Hz; the steady-state phase is set to a speed fluctuation of less than or equal to 5%, with the sampling frequency reduced to 20Hz; the shutdown phase is set to a speed drop to 0, with a sampling frequency of 80Hz.

[0013] Preferably, the preprocessing of the operating condition data includes, but is not limited to, data cleaning, data filtering, and standardization. The data cleaning includes, but is not limited to, removing outliers and filling in missing values. The data filtering includes, but is not limited to, using amplitude limiting filtering, first-order lag filtering, median filtering, and Kalman filtering.

[0014] Preferably, the real-time damage normalization status monitoring based on the weekly damage data, and the automatic triggering of an early warning in response to the detection of existing or impending damage at a dangerous location, includes the following steps: The real-time cumulative damage value of the weak point in the current cycle is obtained by using the multivariate linear damage accumulation method. One or more damage thresholds can be set. When the real-time cumulative damage value reaches the damage threshold, the electronic control system will trigger a vehicle alarm through hardware signals or bus messages.

[0015] Preferably, the multivariate linear damage accumulation method is as follows: Based on the single-cycle damage data di, the cumulative damage value D at the current moment of the current cycle N is calculated using the following formula; Where D represents the cumulative damage value at the current moment. For single-cycle vibration fatigue damage, For vibration fatigue damage, For plastic or creep damage, N is the number of cycles of the high-temperature rotor at the current moment, i is the number of cycles, di is the damage data per cycle, and Xi is the pre-processed working condition data under the service conditions corresponding to the number of cycles.

[0016] Preferably, the setting of multi-level damage thresholds includes setting two levels of damage thresholds, which include a warning level damage threshold and a fault level damage threshold. The warning level damage threshold is 0.8 to 0.95 times the rotor design life, and the fault level damage threshold is greater than or equal to 0.95 times the rotor design life.

[0017] This invention also discloses an electrical control system for real-time monitoring of dangerous points on a motor rotor, comprising: The simulation data acquisition module is used to collect simulation data of the rotor of the new energy vehicle motor under different working conditions, build a pure simulation data base library, and obtain simulation results of rotor service under multiple damage and multiple load coupling states. The model building module is used to identify dangerous parts of the rotor based on the simulation results, and to build a prediction model of dangerous parts based on deep neural networks by accumulating simulation data of dangerous parts. The damage data acquisition module is used to collect rotor service data in the motor of the target new energy vehicle in real time through the electronic control system. After preprocessing the working condition data, it is input into the prediction model of the dangerous point to obtain single-week damage data. The real-time monitoring and early warning module is used to perform real-time damage normalization status monitoring based on the damage data of this single week, and automatically triggers an early warning in response to the detection of damage existing or about to exist at dangerous points.

[0018] Compared with existing technologies, the above technical solution has the following advantages: 1. One embodiment of the present invention constructs a high-precision multiphysics coupling simulation model, namely a dangerous point prediction model. Through the simple and replaceable engineering model, it can accurately simulate the influence of complex working conditions such as centrifugal force, vibration load, and creep on the interference fit, realize the accurate location and quantitative analysis of the dangerous point of damage, and provide a scientific basis for subsequent monitoring. Moreover, the present invention does not require the construction of a digital twin model, the hardware adaptation cost of this solution is reduced by 50%, and it can be directly deployed in the vehicle electronic control system, with a wide range of application scenarios. 2. One embodiment of the present invention realizes multi-factor fusion analysis, which fully considers the synergistic effect of factors such as centrifugal force, vibration load, and creep, and breaks through the limitation of isolated analysis of a single factor in traditional research. It more realistically restores the actual working conditions of the motor during operation and significantly improves the accuracy and reliability of damage hazard point identification. The hazard point identification in this solution is based on the coupling effect of multiple damages and multiple loads, and is not a virtual sensing that only identifies a single damage type. The identification accuracy is improved to over 98%. 3. One embodiment of the present invention establishes an integrated system combining simulation and real-time monitoring. This system integrates the simulation model with a sensor data acquisition system and a real-time analysis module. It can both predict potential damage risks based on simulation results and dynamically correct the simulation model through real-time monitoring data, achieving dynamic and accurate monitoring of the interference fit between the motor rotor and shaft. When an anomaly is detected, the system can quickly issue an early warning and provide fault diagnosis and maintenance suggestions based on simulation analysis, forming a complete technical chain of "identification-monitoring-early warning-diagnosis," effectively improving the safety and operational efficiency of new energy vehicle motors. 4. One embodiment of the present invention adopts an electronic control method that combines dynamic sampling, lightweight model quantization, and multi-level early warning logic to achieve real-time performance of the entire process of acquisition, calculation, and early warning (calculation per cycle ≤10ms), meeting the real-time control requirements of vehicles. Attached Figure Description

[0019] Figure 1 This is a flowchart of the steps of the electrical control method for real-time monitoring of dangerous points on the motor rotor disclosed in this invention; Figure 2 This is a structural framework diagram of the electrical control method for real-time monitoring of dangerous points on the motor rotor disclosed in this invention.

[0020] Figure labeling: Simulation data acquisition module-30; Model building module-40; Damage data acquisition module-50; Real-time monitoring and early warning module-60. Detailed Implementation

[0021] The advantages of the present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments.

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0023] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0025] In the description of this invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0026] In the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, or internal connections between two components. They can be direct connections or indirect connections through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0027] In the following description, suffixes such as "module," "part," or "unit" used to denote elements are used only for the convenience of the description of the invention and have no specific meaning in themselves. Therefore, "module" and "part" can be used interchangeably.

[0028] like Figure 1 As shown, to achieve the above objectives, the present invention discloses an electronic control method for real-time monitoring of dangerous points on a motor rotor, comprising the following steps: Step S101: Collect simulation data of new energy vehicle motor rotor service under different working conditions, build a pure simulation data base library, and obtain simulation results of rotor service under multiple damage and multiple load coupling states. Step S102: Based on the simulation results, identify the dangerous points of the rotor, and use the accumulated simulation data of the dangerous points to construct a prediction model of the dangerous points based on a deep neural network. Step S103: The rotor service data of the target new energy vehicle motor is collected in real time through the electronic control system. The preprocessed data is then input into the prediction model of the dangerous point to obtain the single-cycle damage data. Step S104: Real-time damage normalization status monitoring is performed based on the damage data of this single week. In response to the detection of damage existing or about to exist at the dangerous point, an early warning is automatically triggered.

[0029] Specifically, by collecting simulation data of rotor operation under different interference fits, speeds, vibrations, and temperatures, and based on simulation results of multi-damage and multi-load coupled states of rotor service in new energy vehicle motors, where multi-damage includes centrifugal fatigue, vibration fatigue, and plastic / creep damage, and multi-load includes speed, torque, and temperature, simulations are conducted under the interaction and action of multiple damages and multiple loads to identify dangerous parts of the rotor. Based on this, a deep neural network dangerous part prediction model based on big data model under multi-damage and multi-load coupling is built. The electronic control system collects rotor service condition data in real time under driving conditions. Then, the electronic control data processing unit preprocesses the rotor service condition data in real time. Finally, the electronic control system monitors the damage normalization state in real time, and if necessary, responds with early warnings through the vehicle virtual monitoring platform.

[0030] Preferably, in step S101, the simulation data is a combination of different physical information parameters obtained by physical sensing of the rotor in the new energy vehicle motor during service. The combination of physical information parameters includes, but is not limited to, centrifugal fatigue damage data and single-cycle vibration damage data obtained by elastic finite element analysis, as well as plastic / creep damage data affected by interference obtained by temperature field finite element analysis.

[0031] Preferably, in step S102, identifying the dangerous locations of the rotor based on the simulation results includes the following steps: Based on the simulation results, through damage numerical quantification analysis, the parts or nodes with the largest damage values ​​of multiple damages and multiple load couplings within a single cycle are selected as dangerous points.

[0032] Specifically, identifying dangerous points on the rotor is used to further accumulate simulation data of these dangerous points, thereby establishing a predictive model for the dangerous points on the rotor. This completes the mapping from input operating parameters to single-cycle damage output, and the weight parameters of the predictive model for dangerous points are transmitted from the cloud or on-board server to the electronic control memory.

[0033] Preferably, in step S103, the real-time acquisition of rotor service data in the target new energy vehicle motor through the electronic control system includes the following steps: The electronic control system uses dynamic sampling frequency logic to collect operating condition data and rotor service data. The rotor service data includes, but is not limited to, operating condition data. Different sampling frequencies are used in the motor start-up phase, steady state phase, and shutdown phase, and the electronic control system collects the data in real time through CAN / LIN bus or dedicated sensor interface. The operating condition data includes, but is not limited to, the operating condition parameters such as rotor initial temperature, speed increase rate, load increase rate, speed, torque, shaft power, and axial tension for each cycle. One cycle refers to the service process of the target new energy vehicle motor after completing one cycle from the start-up phase to the steady-state phase and from the steady-state phase to the shutdown phase.

[0034] In one embodiment of the present invention, the initial temperature of the rotor is obtained by a temperature sensor, the rotational speed is obtained by an encoder, and the torque is obtained by a torque sensor.

[0035] Furthermore, the start-up phase is set to the first 30 seconds of rotor startup, with a sampling frequency of 100Hz; the steady-state phase is set to a speed fluctuation of less than or equal to 5%, with the sampling frequency reduced to 20Hz; and the shutdown phase is set to a speed drop to 0, with a sampling frequency of 80Hz.

[0036] Preferably, in step S103, the preprocessing of the working condition data includes, but is not limited to, data cleaning, data filtering, and standardization of the working condition data. The data cleaning includes, but is not limited to, removing outliers and filling in missing values. The data filtering includes, but is not limited to, using amplitude limiting filtering, first-order lag filtering, median filtering, and Kalman filtering.

[0037] Specifically, data acquisition and preprocessing are performed by integrating sensor control and data filtering algorithms. The electronic control unit dynamically adjusts the sampling frequency according to the rotor operating conditions (e.g., high-frequency sampling during startup and low-frequency sampling during steady state) to optimize the balance between data volume and computing resources. Algorithms such as first-order hysteresis filtering and Kalman filtering are embedded in the electronic control software to smooth high-frequency fluctuation data such as speed and torque and avoid misjudgment.

[0038] Preferably, in step S104, real-time damage normalization status monitoring is performed based on the single-week damage data, and an early warning is automatically triggered in response to the detection of existing or impending damage at a dangerous location, including the following steps: The real-time cumulative damage value of the weak point in the current cycle is obtained by using the multivariate linear damage accumulation method. One or more damage thresholds can be set. When the real-time cumulative damage value reaches the damage threshold, the electronic control system will trigger a vehicle alarm through hardware signals or bus messages.

[0039] Specifically, when the virtual monitoring platform detects that the cumulative damage at the dangerous points of the rotor has reached the set damage threshold, it immediately issues a warning signal to the operator through various means such as sound, light, and electricity, and provides detailed fault information, including but not limited to the fault type, possible causes, and suggested handling measures.

[0040] Furthermore, this multivariate linear damage accumulation method is specifically as follows: Based on the single-cycle damage data di, the cumulative damage value D at the current moment of the current cycle N is calculated using the following formula; Where D represents the cumulative damage value at the current moment. For single-cycle vibration fatigue damage, For vibration fatigue damage, For plastic or creep damage, N is the number of cycles of the high-temperature rotor at the current moment, i is the number of cycles, di is the damage data per cycle, and Xi is the pre-processed working condition data under the service conditions corresponding to the number of cycles.

[0041] Furthermore, the setting of multi-level damage thresholds includes setting two levels of damage thresholds, namely a warning level damage threshold and a fault level damage threshold. The warning level damage threshold is 0.8 to 0.95 times the rotor design life, and the fault level damage threshold is greater than or equal to 0.95 times the rotor design life.

[0042] Specifically, D is set as the cumulative damage value at the current moment. When D=1, the rotor structure fails. Setting two levels of damage thresholds enables damage trend prediction and graded response, avoiding false alarms / missed alarms. For example, setting a safety factor of 0.95, when... When the value is ≥0.95, the electronic control unit triggers the vehicle alarm system to initiate a warning via a hard-wired signal (such as a PWM pulse) or a bus message (such as CANopen), and outputs the cause of the damage based on the proportion of different damages. It should be noted that the rotor design life is the rated life of the rotor at the time of manufacture.

[0043] Specifically, this invention achieves the deployment of predictive models in electronic control systems through lightweight model deployment, cumulative damage calculation, multi-level early warning strategies, and integrated fault diagnosis. Specifically, the predictive model for hazardous locations is quantized and compressed to match the computing power of the electronic control MCU, achieving real-time damage calculation per cycle and enabling lightweight model deployment. In one specific embodiment, INT8 quantization is used to quantize and compress the predictive model for hazardous locations, adapting it to the computing power of the nfineon AURIX series electronic control MCU, achieving a real-time damage calculation per cycle of less than or equal to 10ms / cycle. The electronic control system maintains a real-time damage counter (updated every cycle) based on a multivariate linear damage accumulation criterion and compares it with a preset threshold (e.g., 0.8 times the design life) to trigger early warning logic and achieve cumulative damage calculation. The electronic control system classifies early warning levels according to the degree of damage, as shown in Table 1. Finally, the electronic control unit outputs the fault type, possible causes, and handling suggestions to the vehicle display screen via the OBD interface to achieve integrated fault diagnosis. The fault types include centrifugal fatigue and creep damage, the possible causes include insufficient interference fit and excessive temperature rise, and the handling suggestions include reduced speed operation and maintenance appointment.

[0044] In one embodiment of the present invention, a specific application of the method is provided, wherein the vehicle is controlled and monitored through a virtual AR monitoring platform, including the following steps: Step S201: Real-time collection and preprocessing of operating condition data using Internet of Things (IoT) technology: The virtual AR monitoring platform uses IoT technology to collect various monitoring data from the rotor of the new energy vehicle motor in real time, including but not limited to key parameters such as vibration, temperature, and current, and processes the above operating condition data to obtain damage data. Step S202, Construct a three-dimensional model of the rotor: Using high-precision three-dimensional scanning or CAD modeling technology, construct an accurate three-dimensional model of the rotor. This model not only has high geometric accuracy, but also contains key information such as the rotor's material properties and structural characteristics. Step S203: Map the damage data to the 3D model data: Accurately map the damage data after real-time processing to the rotor 3D model, so that maintenance personnel can directly view the real-time status of the rotor on the 3D model and intuitively display the location and severity of the damage. Step S204, Identification and visualization of hazardous points: The algorithm built into the virtual AR monitoring platform analyzes real-time monitoring data, identifies potential hazardous points on the rotor, and uses AR technology to highlight the hazardous points on the 3D model. The severity of the damage is displayed intuitively through color coding or numerical labeling.

[0045] Step S205, Interactive Operation: Maintenance personnel can view more detailed information about damage points on the 3D model through interactive methods such as touch and gesture recognition, such as damage type and development trend.

[0046] Step S206, Damage Trend Analysis and Early Warning Trigger: The platform uses machine learning algorithms to perform in-depth analysis of real-time monitoring data, predict the damage development trend of the rotor, and automatically trigger early warnings of different levels based on the damage trend. Early warning levels typically include immediate alarms, preventive maintenance recommendations, etc.

[0047] Step S207, Real-time Alarm and Preventive Maintenance Suggestions: When serious damage or impending failure of the rotor is detected, the platform will immediately issue an alarm to remind maintenance personnel to take emergency measures; in the case of minor damage or slow development, the platform will provide preventive maintenance suggestions, such as adjusting operating parameters or replacing parts, to extend the service life of the rotor.

[0048] It should be noted that steps 201-207 above are not shown in the attached figures.

[0049] In another specific embodiment of the present invention, the multivariate linear damage accumulation method may further be: Based on the rotor service data, the rotor load type is decomposed, and the uniaxial stress is obtained by using the multiaxial stress equivalent method. The actual stress is obtained by local stress correction based on the uniaxial stress. Based on Miner's cumulative damage rule, single-load level damage is calculated according to actual stress and material parameters, and the actual cumulative damage value is obtained by superimposing the single-load level damage. The rotor service data also includes load data, material data, and structural data. The load data includes, but is not limited to, variable amplitude load spectrum and stress amplitude corresponding to each load. The material data includes, but is not limited to, material SN curves.

[0050] Specifically, structural parameters include, but are not limited to, rotor diameter, slot size, stress concentration factor Kt, and material density; material parameters include, but are not limited to, material SN curve, elastic modulus E, and Poisson's ratio μ; and load parameters include, but are not limited to, variable amplitude load spectrum (number of start-steady-stop cycles N1, number of steady-state vibration load cycles N2, etc.), and stress amplitudes σa1, σa2, etc. corresponding to each load.

[0051] Furthermore, based on rotor service data, the rotor load types are decomposed, and uniaxial stress is obtained using a multiaxial stress equivalence method. The actual stress is then obtained by performing local stress correction based on the uniaxial stress, including: The load is decomposed into three types: start-stop cycle, steady-state operation vibration cycle, and sudden impact load. For each load type, calculate the stress for each load type, and use the von Mises equivalent stress method to unify and equate the stress of each load type to uniaxial stress; Based on the stress concentration effect, the equivalent stress is corrected to obtain the actual stress.

[0052] Specifically, complex loads are decomposed into three categories: "start-stop cycle (low-frequency large load)," "steady-state operation vibration cycle (high-frequency small load)," and "sudden impact load (occasional large load)." Using the von Mises equivalent stress formula, the bending stress σb, torsional stress τ, and centrifugal stress σc of different load types are equivalent to uniaxial stress σeq, i.e. Based on the stress concentration effect, the actual stress is calculated as σ = Kt × σeq, where the stress concentration factor Kt is determined by the structural dimensions and manufacturing process.

[0053] Furthermore, based on Miner's cumulative damage rule, the single-load level damage is calculated according to the actual stress and material parameters, and the actual cumulative damage value is obtained by superimposing the single-load level damage, including the following steps: The single-load level damage caused by each load type is calculated based on the material's SN curve. The damage from all load types is then superimposed to obtain the actual cumulative damage value.

[0054] Specifically, the single-load level damage calculation involves calculating the fatigue life Ni corresponding to the stress amplitude based on the material's SN curve; then, the single-load level damage Di = actual number of cycles ni / Ni is obtained, where ni is the actual number of cycles of the load, and Ni is the fatigue life corresponding to the stress amplitude; subsequently, the damage from all load levels is superimposed to obtain the actual cumulative damage value D = ΣDi = Σ(ni / Ni). It should be noted that other parts not described in this embodiment are prior art and will not be elaborated upon.

[0055] like Figure 2 As shown, to achieve the above objectives, one embodiment of the present invention also discloses an electronic control system for real-time monitoring of dangerous points on a motor rotor, comprising: The simulation data acquisition module 30 is used to collect simulation data of the rotor of the new energy vehicle motor under different working conditions, build a pure simulation data base library, and obtain simulation results of rotor service under multiple damage and multiple load coupling states. The model building module 40 is used to identify dangerous parts of the rotor based on the simulation results, and to build a dangerous part prediction model based on deep neural network by accumulating simulation data of dangerous parts. Damage data acquisition module 50 is used to collect rotor service data in the motor of the target new energy vehicle in real time through the electronic control system, and input the preprocessed working condition data into the prediction model of the dangerous point to obtain single-week damage data. The real-time monitoring and early warning module 60 is used to perform real-time damage normalization status monitoring based on the damage data of the single week, and automatically triggers an early warning in response to the detection of damage existing or about to exist at dangerous points.

[0056] It should be noted that this system corresponds to an electronic control method for real-time monitoring of dangerous points on a motor rotor; for other unspecified parts, please refer to the content of that method.

[0057] It should be noted that the embodiments of the present invention have better implementability and are not intended to limit the present invention in any way. Any person skilled in the art may use the above-disclosed technical content to change or modify it into equivalent effective embodiments. However, any modifications or equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for real-time monitoring of dangerous points on a motor rotor, characterized in that, Includes the following steps: Simulation data of new energy vehicle motor rotor service under different working conditions were collected, a basic simulation data library was constructed, and simulation results of rotor service under multiple damage and multiple load coupling states were obtained. Based on the simulation results, dangerous points on the rotor are identified, and a prediction model for dangerous points based on deep neural networks is constructed using the accumulated simulation data of these dangerous points. The system collects rotor service data in the motor of the target new energy vehicle in real time through the electronic control system. After preprocessing the operating data, it is input into the prediction model of the dangerous point to obtain single-week damage data. Real-time damage normalization status monitoring is performed based on the weekly damage data, and an early warning is automatically triggered in response to the detection of existing or impending damage at dangerous locations.

2. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 1, characterized in that, The simulation data consists of different combinations of physical information parameters obtained by physical sensing of the rotor in the motor of a new energy vehicle during its service life. These combinations of physical information parameters include, but are not limited to, centrifugal fatigue damage data and single-cycle vibration damage data obtained by elastic finite element analysis, as well as plastic / creep damage data affected by interference obtained by temperature field finite element analysis.

3. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 1, characterized in that, The step of identifying the dangerous locations of the rotor based on the simulation results includes the following steps: Based on the simulation results, through damage numerical quantification analysis, the parts or nodes with the largest damage values ​​of multiple damages and multiple load couplings within a single cycle are selected as dangerous points.

4. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 1, characterized in that, The real-time acquisition of rotor service data in the target new energy vehicle motor through the electronic control system includes the following steps: The electronic control system uses dynamic sampling frequency logic to collect operating condition data and rotor service data. The rotor service data includes, but is not limited to, operating condition data. Different sampling frequencies are used in the motor start-up phase, steady state phase, and shutdown phase, and the electronic control system collects the data in real time through CAN / LIN bus or dedicated sensor interface. The operating condition data includes, but is not limited to, the operating condition parameters such as rotor initial temperature, speed increase rate, load increase rate, speed, torque, shaft power, and axial tension in each cycle. One cycle refers to the service process of the rotor in the target new energy vehicle motor after completing one cycle from the start-up phase to the steady-state phase and from the steady-state phase to the shutdown phase.

5. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 4, characterized in that, The method of using different sampling frequencies during the motor start-up, steady-state, and shutdown phases includes: The startup phase is set to the first 30 seconds of rotor startup, with a sampling frequency of 100Hz; the steady-state phase is set to a speed fluctuation of less than or equal to 5%, with the sampling frequency reduced to 20Hz; and the shutdown phase is set to a speed drop to 0, with a sampling frequency of 80Hz.

6. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 1, characterized in that, The preprocessing of the operating data includes, but is not limited to, data cleaning, data filtering, and standardization. The data cleaning includes, but is not limited to, removing outliers and filling in missing values. The data filtering includes, but is not limited to, using amplitude limiting filtering, first-order lag filtering, median filtering, and Kalman filtering.

7. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 1, characterized in that, The real-time damage normalization status monitoring based on the weekly damage data, and the automatic triggering of an early warning in response to the detection of existing or impending damage at a dangerous location, includes the following steps: The real-time cumulative damage value of the weak point in the current cycle is obtained by using the multivariate linear damage accumulation method. One or more damage thresholds are set. When the real-time cumulative damage value reaches the damage threshold, the electronic control system triggers a vehicle alarm through hardware signals or bus messages.

8. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 7, characterized in that, The setting of multi-level damage thresholds includes setting two levels of damage thresholds, namely a warning level damage threshold and a fault level damage threshold. The warning level damage threshold is 0.8 to 0.95 times the rotor design life, and the fault level damage threshold is greater than or equal to 0.95 times the rotor design life.

9. The electrical control method for real-time monitoring of dangerous points on a motor rotor as described in claim 7, characterized in that, The multivariate linear damage accumulation method is specifically as follows: Based on the single-cycle damage data di, the cumulative damage value D at the current moment of the current cycle N is calculated using the following formula; Where D represents the cumulative damage value at the current moment. For single-cycle vibration fatigue damage, For vibration fatigue damage, For plastic or creep damage, N is the number of cycles of the high-temperature rotor at the current moment, i is the number of cycles, di is the damage data per cycle, and Xi is the pre-processed working condition data under the service conditions corresponding to the number of cycles.

10. An electrical control system for real-time monitoring of dangerous points on a motor rotor, characterized in that, include: The simulation data acquisition module is used to collect simulation data of the rotor of the new energy vehicle motor under different working conditions, build a pure simulation data base library, and obtain simulation results of rotor service under multiple damage and multiple load coupling states. The model building module is used to identify dangerous parts of the rotor based on the simulation results, and to build a prediction model of dangerous parts based on deep neural networks by accumulating simulation data of dangerous parts. The damage data acquisition module is used to collect rotor service data in the motor of the target new energy vehicle in real time through the electronic control system, and input the preprocessed operating data into the prediction model of the dangerous point to obtain single-week damage data. The real-time monitoring and early warning module is used to perform real-time damage normalization status monitoring based on the weekly damage data, and automatically triggers an early warning in response to the detection of existing or impending damage at dangerous locations.

Images