Integrated axial electric drive assembly state monitoring method and system based on cloud platform

Abstract

This invention provides an integrated axial electric drive assembly status monitoring method and system based on a cloud platform, relating to the field of data processing technology. The method includes: identifying key components, deploying sensors, constructing an end-side sensor network, establishing Q edge computing gateways, and connecting them to an RS485 bus; monitoring the axial electric drive assembly's operating data stream and mapping and transmitting it to the Q edge computing gateways; performing abnormal state assessment, outputting Q electric drive abnormal state parameters, and uploading them to the electric drive cloud platform; and performing coupled state assessment to obtain integrated axial electric drive assembly status parameters for early warning decision control. This invention solves the technical problem that existing electric drive assembly status monitoring technologies are typically limited to monitoring a single device or a local area, making it difficult to achieve cross-device and cross-regional data integration and comprehensive analysis, and failing to achieve global health management, thus leading to equipment failure propagation or system downtime.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to an integrated axial electric drive assembly status monitoring method and system based on a cloud platform. Background Technology

[0002] Integrated axial electric drive units (AEVs) are core drive devices widely used in electric vehicles, robots, wind turbines, and other fields. They integrate electric motors, transmissions, and controllers, enabling efficient power transmission while occupying a small space. With technological advancements and increasing application demands, the operating environment of integrated axial electric drive units is becoming increasingly complex. Especially under conditions of high load, high temperature, and complex environments, health monitoring and fault prediction for the electric drive unit become crucial.

[0003] However, traditional integrated axial electric drive assembly status monitoring is usually limited to monitoring a single device or local area, making it difficult to achieve cross-device and cross-regional data integration and comprehensive analysis. This makes it impossible to share fault information between different electric drive assemblies and makes it difficult to achieve overall equipment health management. Once a certain electric drive assembly fails, it may not be possible to identify its impact on other components or the entire system in a timely manner, thus leading to equipment failure spread or system shutdown. Summary of the Invention

[0004] This application provides an integrated axial electric drive assembly status monitoring method and system based on a cloud platform, aiming to solve the technical problems of existing electric drive assembly status monitoring, which is usually limited to monitoring a single device or local area, making it difficult to achieve cross-device and cross-regional data integration and comprehensive analysis, and failing to achieve global health management, thus leading to equipment failure propagation or system downtime.

[0005] The first aspect disclosed in this application provides a cloud platform-based method for monitoring the status of an integrated axial electric drive assembly. The method includes: identifying key components of the target integrated axial electric drive assembly, deploying sensors, constructing an end-side sensor network, and establishing Q edge computing gateways based on the end-side sensor network, connecting the Q edge computing gateways to an RS485 bus; monitoring the axial electric drive assembly's operating data stream through the end-side sensor network, activating the RS485 bus to map and transmit the axial electric drive assembly's operating data stream to the Q edge computing gateways; synchronously evaluating the abnormal status of the axial electric drive assembly's operating data stream based on the Q edge computing gateways, outputting Q electric drive abnormal status parameters, and uploading the Q electric drive abnormal status parameters to an electric drive cloud platform via a 5G communication module; performing a coupled status evaluation on the Q electric drive abnormal status parameters through the electric drive cloud platform to obtain integrated axial electric drive assembly status parameters, and performing early warning decision control based on the integrated axial electric drive assembly status parameters.

[0006] The second aspect disclosed in this application provides an integrated axial electric drive assembly status monitoring system based on a cloud platform. This system is used in the aforementioned integrated axial electric drive assembly status monitoring method based on a cloud platform. The system includes: an edge computing gateway construction unit, used for identifying key components of the target integrated axial electric drive assembly, deploying sensors, constructing an end-side sensor network, and establishing Q edge computing gateways based on the end-side sensor network, connecting the Q edge computing gateways to an RS485 bus; and a data mapping and transmission unit, used for monitoring the axial electric drive assembly's operating data stream through the end-side sensor network and activating RS485. The 485 bus maps and transmits the axial electric drive assembly operating data stream to the Q edge computing gateways; the abnormal state assessment unit is used to synchronously assess the abnormal state of the axial electric drive assembly operating data stream based on the Q edge computing gateways, output Q electric drive abnormal state parameters, and upload the Q electric drive abnormal state parameters to the electric drive cloud platform through a 5G communication module; the early warning decision control unit is used to perform coupled state assessment on the Q electric drive abnormal state parameters through the electric drive cloud platform to obtain integrated axial electric drive assembly state parameters, and perform early warning decision control based on the integrated axial electric drive assembly state parameters.

[0007] One or more technical solutions provided in this application have at least the following beneficial effects: By identifying key components and deploying sensors on the target integrated axial electric drive assembly, critical parts can be covered. Deploying an end-side sensor network ensures real-time and efficient data acquisition. The combination of RS485 bus and edge computing gateway enables continuous monitoring of the electric drive assembly's operating status, providing rich data support for subsequent analysis. During data transmission, a combination of RS485 bus and edge computing gateway is used. The RS485 bus provides stable, long-distance communication between devices, while the edge computing gateway processes data streams locally in real time, reducing the pressure on the central cloud platform. Data offloading and synchronous processing ensure efficient, low-latency system response, improving data processing speed. Through the edge computing gateway's anomaly assessment, instant analysis of the electric drive assembly's operating data is achieved. This assessment not only improves the response speed of fault detection but also effectively identifies potential abnormal or fault states. By promptly detecting abnormal states and outputting abnormal state parameters, it can provide early warnings before faults occur, thereby reducing downtime and maintenance costs. By coupling and assessing abnormal state parameters uploaded from multiple edge computing gateways through the electric drive cloud platform, it ensures comprehensive monitoring of the overall performance of the electric drive assembly, rather than just detecting local faults. This process helps achieve health monitoring of the entire electric drive system, improving system reliability from a global perspective. Through the integrated state assessment results of the electric drive cloud platform, early warning decision control is performed based on the state parameters of the electric drive assembly, realizing intelligent system management. Early warning decisions can respond quickly based on real-time data analysis, ensuring the safety and stability of the electric drive assembly and improving the system's adaptability and operating efficiency.

[0008] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the cloud-based integrated axial electric drive assembly status monitoring method provided in the embodiments of this application.

[0010] Figure 2 This is a schematic diagram of the structure of the cloud-based integrated axial electric drive assembly condition monitoring system provided in the embodiments of this application.

[0011] Figure labeling: Edge computing gateway construction unit 10, data mapping and transmission unit 20, abnormal state assessment unit 30, early warning decision control unit 40. Detailed Implementation

[0012] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0013] Example 1, as Figure 1 As shown in the embodiments of this application, an integrated axial electric drive assembly condition monitoring method based on a cloud platform is provided, the method comprising: Key components of the target integrated axial electric drive assembly are identified and sensors are deployed to construct an end-side sensor network. Based on the end-side sensor network, Q edge computing gateways are established and connected to an RS485 bus.

[0014] By analyzing the design data, material properties, and actual application scenarios of the target integrated axial electric drive assembly, performance analysis is conducted to identify key components that may experience abnormalities during operation. These key components include bearings, motors, battery cells, transmission systems, and temperature-sensitive parts. Sensors are deployed at these key components to collect various data such as vibration, temperature, pressure, stress, current, and voltage. These sensors can monitor the operating status of the electric drive assembly in real time and provide accurate sensing data. The sensors deployed in different locations are connected wirelessly or via wired connections to form an efficient data acquisition network, namely an end-side sensor network.

[0015] Based on the edge sensor network, Q edge computing gateways are established. These gateways are responsible for receiving data collected by the sensors and performing preliminary processing and analysis. These edge computing gateways are connected via an RS485 bus. RS485 is an industry-standard communication protocol that supports communication between multiple devices and has strong anti-interference capabilities, making it suitable for long-distance transmission in industrial environments.

[0016] The axial electric drive assembly operating data stream is monitored through the end-side sensor network, and the RS485 bus is activated to map and transmit the axial electric drive assembly operating data stream to the Q edge computing gateways.

[0017] The electric drive assembly's operational data, including vibration, temperature, current, and voltage data, is monitored in real time via an edge sensor network, covering all key performance parameters. Once the RS485 bus is activated, the axial electric drive assembly's operational data stream is transmitted to each edge computing gateway via the RS485 bus. All data transmission is synchronous; that is, while different sensors collect data, the RS485 bus maintains a real-time connection with each edge computing gateway, ensuring real-time synchronous transmission of the data stream.

[0018] Based on the Q edge computing gateways, the abnormal status of the axial electric drive assembly operation data stream is evaluated synchronously, and Q electric drive abnormal status parameters are output. The Q electric drive abnormal status parameters are then uploaded to the electric drive cloud platform through the 5G communication module.

[0019] Each edge computing gateway receives the axial electric drive assembly operation data stream from the edge sensor network and performs local data synchronization and preliminary anomaly assessment. Each gateway analyzes data from multiple sensors to determine if any anomalies exist. Edge computing gateways can employ various algorithms, such as machine learning-based methods like LSTM networks and neural networks, to process the data and detect abnormal patterns. Anomaly assessment involves comparing historical data, setting thresholds, and pattern recognition. When the operation of an electric drive assembly exceeds the set normal range, the edge computing gateway generates abnormal drive status parameters, such as excessive temperature, excessive vibration, or abnormal current. These parameters indicate the current abnormal condition of the electric drive assembly. The assessed abnormal drive status parameters are uploaded to the electric drive cloud platform via a 5G communication module. Due to the high speed and low latency of the 5G network, real-time data transmission is possible, ensuring the cloud platform can quickly receive the electric drive assembly status information. The electric drive cloud platform receives abnormal drive status parameters from multiple edge computing gateways and further performs comprehensive analysis, prediction, and decision-making.

[0020] The electric drive cloud platform performs coupled state evaluation on the Q electric drive abnormal state parameters to obtain the integrated axial electric drive assembly state parameters, and performs early warning decision control based on the integrated axial electric drive assembly state parameters.

[0021] The purpose of coupled status assessment on Q abnormal state parameters of the electric drive system on the electric drive cloud platform is to evaluate the overall health of the electric drive assembly by integrating data uploaded from various edge gateways. Coupled assessment involves combining the abnormal state parameters of the electric drive from each gateway with the overall operating mode of the electric drive assembly to determine whether there are more complex system-level faults. For example, a partial failure of a component could have a cascading effect on the performance of the entire system. This coupled analysis includes integrated analysis of different data sources, using statistical methods or machine learning models to assess the overall health of the system. Through coupled status assessment, comprehensive integrated axial electric drive assembly status parameters are obtained, including the overall health of the electric drive assembly, operating efficiency, and the presence of potential fault risks.

[0022] Based on the status parameters of the integrated axial electric drive assembly, early warning decision control is implemented. This means that when any potential fault or performance degradation is detected, an early warning is immediately issued and corresponding control measures are taken. Early warning decision control may include: sending alarms to operators indicating a current fault in the electric drive assembly; initiating predetermined maintenance procedures, such as automatically adjusting parameters or shutting down certain equipment to prevent damage; and conducting further fault diagnosis to identify the root cause of the problem and take action, such as shutdown, repair, or replacement of the faulty component. Ultimately, this achieves intelligent monitoring, fault warning, and adaptive control of the integrated axial electric drive assembly.

[0023] Furthermore, constructing an edge-side sensor network includes: Based on the structural design data and material property data of the target integrated axial electric drive assembly, finite element modeling is performed to generate the axial electric drive assembly finite element model; according to the application scenario of the electric drive assembly, simulated operating condition parameters of the electric drive assembly are constructed; the simulated operating condition parameters of the electric drive assembly are mapped to the axial electric drive assembly finite element model for operating condition simulation testing, and the operating test data of the electric drive assembly is recorded; based on the operating test data of the electric drive assembly, key parts of the target integrated axial electric drive assembly are identified, sensors are deployed, and an end-side sensor network is constructed.

[0024] Finite element modeling involves discretizing the design data of the target integrated axial electric drive assembly, dividing the structure of the electric drive assembly into a finite number of elements, such as triangles and tetrahedrons, and simulating the behavior of the overall structure through the physical properties of each element. In this process, the model is built based on the structural design data of the electric drive assembly, such as dimensions, geometry, and assembly methods, as well as material property data, such as strength, stiffness, and thermal conductivity. This data provides the foundation for understanding the mechanical behavior, heat conduction, stress, and strain of the electric drive assembly.

[0025] Based on the application scenarios of electric drive assemblies, such as electric vehicles, wind turbines, and autonomous driving systems, operating parameters for the electric drive assembly under specific environments are constructed. These parameters describe various operating conditions that the electric drive assembly may encounter during operation. For example, load conditions define the operating states of the electric drive assembly under different load conditions, such as full-load and light-load conditions, simulating the impact of different loads on the electric drive assembly; environmental conditions include changes in environmental factors such as temperature, humidity, and wind speed, which affect the heat dissipation and operational stability of the electric drive assembly; vibration conditions define the operating states of the electric drive assembly under different vibration conditions, which have a significant impact on structural stability and lifespan; and speed conditions set operating parameters at different speeds based on the usage frequency and operating mode of the electric drive assembly. These operating parameters are used to simulate the performance of the electric drive assembly in practical applications, helping to study its performance changes and potential problems under different conditions.

[0026] The defined simulation parameters of the electric drive assembly are input into the finite element model. These parameters serve as input conditions for the model, simulating the response of the electric drive assembly under different operating conditions. The simulation is run and key data during the simulation process are recorded, such as stress distribution, temperature field, and vibration response. By simulating different operating conditions, the performance of the electric drive assembly under different conditions is recorded, generating electric drive assembly operation test data.

[0027] By analyzing the operational test data of the electric drive assembly, key components that have a significant impact on equipment safety and operational stability were identified. These key components include: high-stress areas, such as bearings, gears, and joints, which are easily affected by mechanical stress; heat-sensitive areas, such as batteries and motors, which are easily affected by heat, leading to performance degradation; and vibration sources, such as rotors and drive shafts, which are prone to vibration and affect system stability.

[0028] Based on the identification results of key components, appropriate types of sensors, such as temperature sensors, vibration sensors, and stress sensors, are selected and installed in these key locations. Sensor deployment is optimized according to the characteristics of each key component; for example, stress sensors are installed more frequently in load-bearing areas, while temperature sensors are mainly deployed in heat-sensitive areas. All sensors are connected via a network to form an end-side sensor network, ensuring real-time collection of status information from all components of the electric drive assembly.

[0029] Furthermore, based on the operational test data of the electric drive assembly, key components of the target integrated axial electric drive assembly are identified, sensors are deployed, and an end-side sensor network is constructed, including: Based on the operational test data of the electric drive assembly, stress, temperature, and vibration performance analysis was performed on the target integrated axial electric drive assembly to obtain multi-dimensional performance distribution parameters of the electric drive assembly. According to these multi-dimensional performance distribution parameters, key components of the target integrated axial electric drive assembly were identified, determining multiple key components. Sensor selection and installation coverage analysis were performed on each of these key components to obtain sensor parameter sets for each key component. Based on these sensor parameter sets, sensors were deployed on the target integrated axial electric drive assembly to construct an end-side sensor network.

[0030] By analyzing the operational test data of the electric drive assembly, the focus is on its stress, temperature, and vibration performance under different operating conditions. These factors are key indicators of the electric drive assembly's operational safety and stability. Specifically, the stress conditions of various components of the electric drive assembly under load and power are monitored, such as bearings, gears, and connecting parts. Excessive stress may lead to fatigue damage or structural failure. Temperature changes in various parts of the electric drive assembly during operation are measured, especially in heat-sensitive components such as the battery and motor. Excessive temperature can affect performance and may even lead to overheating failure. The vibration generated by the electric drive assembly during operation is monitored to analyze whether there are imbalances, resonances, or other abnormal vibration modes. Excessive vibration can cause structural damage or affect other components of the system. The multidimensional performance distribution parameters of the electric drive assembly refer to the performance of each component under multiple dimensions such as stress, temperature, and vibration. These parameters reflect the comprehensive performance of the electric drive assembly under different operating conditions.

[0031] Critical component identification aims to identify those components from the electric drive assembly that significantly impact its safety, reliability, and performance during normal operation. These critical components are highly susceptible to factors such as stress, temperature, and vibration. By analyzing the stress, temperature, and vibration distribution of each component during operation, components with poor performance or deviations from normal operating ranges are identified. Examples include: high-stress areas prone to fatigue, fracture, or deformation; high-temperature areas, such as the motor, control unit, and battery, which may overheat due to poor heat dissipation; and high-vibration areas, such as the rotor and bearings, which are prone to excessive vibration due to imbalance or wear. Based on the performance analysis results, several critical components are identified. These critical components require special attention because they may play a decisive role in the operational safety and stability of the entire electric drive assembly.

[0032] Sensor selection is based on the characteristics of each critical component, choosing appropriate sensor types and specifications. For example, for high-temperature components, temperature sensors such as thermocouples and thermistors are selected; for vibration-sensitive components, vibration sensors such as accelerometers and displacement sensors are selected; and for components subjected to significant stress, stress sensors such as strain gauges and force sensors are selected. After sensor selection, an installation coverage analysis is performed, ensuring that sensors for each critical component can cover all important locations within the component. For example, temperature sensors need to cover the heat source area of ​​the motor, and vibration sensors need to cover the vibration source area of ​​rotating components. For comprehensive monitoring of multi-dimensional performance, it is essential to ensure that each component has a sufficient number of sensors for monitoring. For each critical component, based on its functional requirements and installation location, a set of sensor parameters is obtained, including the specifications, performance indicators, installation methods, and operating parameters of all selected sensors.

[0033] Based on sensor parameter sets from multiple key components, sensors are deployed to various critical parts of the electric drive assembly. Multiple sensors are installed at each key component to collect the necessary operational data. All installed sensors are connected wirelessly or via wired connections, forming an edge sensor network. This network can transmit the data collected by the sensors to an edge computing gateway in real time for further analysis.

[0034] Furthermore, establish Q edge computing gateways, including: Based on the type and quantity of parameters collected by the edge sensor network, processing requirements are assessed to obtain the edge sensor data processing requirements. According to the monitoring scale and data processing objectives of the target integrated axial electric drive assembly, the number of edge gateways to be deployed, Q, is determined. Edge gateway deployment rules are obtained, including spatial proximity, processing requirement satisfaction, and data processing balance. Using the edge sensor data processing requirements and the number of edge gateways to be deployed, Q, as constraints, edge gateways are configured for the target integrated axial electric drive assembly based on the edge gateway deployment rules, establishing Q edge computing gateways.

[0035] The data collected by the end-side sensor network includes different types of signals such as temperature, vibration, and stress. These signals change in real time according to the operating status of the electric drive assembly, so each type of data needs to be monitored and processed in real time. The number of sensors is closely related to the key parts of the electric drive assembly and the monitoring requirements. Different components will deploy different numbers of sensors, and the amount of data collected will also vary. For example, one component needs multiple temperature sensors to cover different areas, while another component needs more vibration sensors to monitor vibrations in different directions.

[0036] Based on the type and quantity of collected parameters, the data processing requirements are assessed. Considering the real-time nature and complexity of the data, the edge computing gateway needs to possess corresponding processing capabilities. These processing requirements include: data preprocessing, such as data cleaning, noise reduction, and missing value imputation; real-time analysis, performing preliminary analysis of the data to check for anomalies, such as excessively high temperature, excessive stress, or excessive vibration; and data transmission preparation, transmitting the data to the cloud platform in an appropriate format and frequency. This processing requirement assessment aims to ensure efficient data processing at the edge computing gateway, thereby reducing the computational burden on the cloud platform and ensuring data real-time performance.

[0037] Monitoring scale refers to the number and distribution of sensors that need to be monitored throughout the entire electric drive assembly. Electric drive assemblies with a larger monitoring scale require more edge computing gateways to share the data processing tasks. Monitoring scale also includes the area to be covered, such as multiple electric drive assembly components or dispersed devices, as well as the number and density of sensors in each monitoring area. Data processing objectives include real-time processing of collected data, anomaly detection, and early warning triggering. The complexity of these data processing objectives determines the performance requirements and deployment quantity of edge computing gateways. For example, in situations with high data traffic and high computational demands, more edge gateways are needed to share the processing load.

[0038] Based on the monitoring scale and data processing objectives, the number Q of edge computing gateways needs to meet requirements such as processing capacity, real-time performance, and coverage. Specifically, when calculating Q, the following needs to be met: the maximum amount of sensor data that each edge gateway can process; and the load balancing requirements among the edge gateways.

[0039] Spatial proximity means that edge computing gateways should be deployed as close as possible to the sensor network to reduce data transmission latency and bandwidth consumption. For example, if sensors are distributed across different electric drive assembly components, the edge gateway needs to be deployed physically close to these components to ensure timely reception of sensor data. Processing demand satisfaction means that each edge computing gateway should have sufficient computing power to meet the processing needs of the sensor data. Each edge gateway must be able to process the sensor data stream it receives, performing data preprocessing and preliminary analysis. Data processing balance means that when deploying multiple edge computing gateways, the processing tasks undertaken by each gateway should be relatively balanced to avoid some edge gateways being overloaded with data processing tasks while others remain idle.

[0040] Based on the preceding analysis, Q edge computing gateways are configured. These gateways are deployed according to edge gateway deployment rules, and each edge computing gateway is assigned appropriate tasks based on its required processing capacity to ensure that data processing tasks can be completed in real time, avoiding data backlog or delays. After configuration, each edge computing gateway can connect to the edge sensor network and realize data transmission and processing functions. At the same time, the edge computing gateway can upload the processing results to the cloud platform for further analysis and decision-making.

[0041] Furthermore, activating the RS485 bus to map and transmit the axial electric drive assembly's operating data stream to the Q edge computing gateways includes: Assign an RS485 address to each of the Q edge computing gateways according to the RS485 bus, and construct a Q edge gateway-RS485 mapping address table; associate and match the axial electric drive assembly operating data stream with the Q edge computing gateways to determine a matching edge computing gateway set; perform self-organizing network analysis on the matching edge computing gateway set according to the Q edge gateway-RS485 mapping address table to activate the RS485 self-organizing communication network; and map and transmit the axial electric drive assembly operating data stream to the matching edge computing gateway set based on the RS485 self-organizing communication network.

[0042] RS485 is a widely used communication standard for industrial control and data transmission. It supports multiple device connections and boasts advantages such as strong anti-interference capabilities, long-distance transmission, and parallel communication among multiple devices. Each edge computing gateway transmits data via the RS485 bus; therefore, each edge computing gateway needs to be assigned a unique RS485 address. This address helps identify and correctly transmit data to the corresponding gateway. A mapping address table records each edge computing gateway with its corresponding RS485 address. This table helps ensure correct data forwarding based on the edge computing gateway's address during subsequent communication.

[0043] The axial electric drive assembly's operating data stream is associated with Q assigned edge computing gateways. Each edge computing gateway is responsible for processing sensor data from different parts of the electric drive assembly. A mapping address table determines which sensor data streams each edge computing gateway will process. This matching process ensures that each edge gateway only processes its corresponding data stream, thereby reducing data processing conflicts. Based on the sensor distribution and data stream requirements, a set of edge computing gateways associated with each data stream is determined; that is, each data stream corresponds to one edge computing gateway, ensuring that data can be processed quickly and accurately.

[0044] Ad hoc network analysis refers to optimizing the network topology based on a set of matched edge computing gateways. This ensures the RS485 communication network operates efficiently and stably. In an RS485 bus, multiple devices share the same communication line; therefore, the layout of the edge computing gateways needs to be analyzed to optimize data transmission paths and avoid signal interference and transmission delays. After completing the ad hoc network analysis and determining the optimal network topology, the RS485 ad hoc communication network is activated. This means that all edge computing gateways will establish communication links through the RS485 bus and begin synchronous operation.

[0045] Based on the activated RS485 self-organizing communication network, the axial electric drive assembly operation data stream is transmitted to the corresponding matching edge computing gateway set via the RS485 bus. At this time, the data stream will be transmitted in a specified format and frequency to ensure that each edge computing gateway can obtain the sensor data it is responsible for in a timely manner.

[0046] Furthermore, output Q electric drive abnormal state parameters, including: Based on the Q edge computing gateways, the axial electric drive assembly operation data stream is mapped and split synchronously to obtain Q gateway electric drive assembly operation data streams; according to the Q edge computing gateways, the electric drive assembly status analysis module is called; based on the electric drive assembly status analysis module, the abnormal status of the Q gateway electric drive assembly operation data streams is evaluated synchronously, and Q electric drive abnormal status parameters are output.

[0047] Mapping and offloading refers to distributing the axial electric drive assembly operating data stream according to the load capacity and responsibilities of each edge computing gateway. Specifically, each edge computing gateway receives the data stream from its corresponding sensor and performs offloading processing. Since multiple edge computing gateways operate simultaneously, the processing tasks of each gateway are synchronized. This means that each gateway can process its own data stream without interfering with the tasks of other gateways, thus achieving parallel processing. Synchronization ensures that all edge computing gateways process data according to predetermined time steps or timestamps. This maintains the consistency of the data stream throughout the system, meaning that each edge computing gateway starts simultaneously and processes data synchronously. After offloading, the resulting Q gateway electric drive assembly operating data streams are transmitted to their respective edge computing gateways, and each gateway receives, processes, and generates results related to its assigned portion.

[0048] The electric drive assembly status analysis module is responsible for analyzing the incoming electric drive assembly operation data stream and assessing the health status of the electric drive assembly. This module uses predefined algorithms, such as machine learning models, rule engines, and threshold judgments, to analyze the data stream and identify anomalies and potential faults.

[0049] Each gateway's electric drive assembly's operating data stream undergoes anomaly assessment in the electric drive assembly status analysis module. This assessment is performed by checking the data stream for any abnormalities that deviate from the preset normal operating range. Anomalies include: abnormal temperature (temperature exceeding safe limits, potentially causing overheating or damage); abnormal vibration (excessive vibration indicating mechanical problems such as imbalance or damage); current / voltage fluctuations (fluctuations in current or voltage indicating electrical faults); and abnormal stress (excessive stress on components could lead to physical damage). The assessment results output Q electric drive anomaly status parameters, each representing the anomaly status of an electric drive assembly processed by an edge computing gateway.

[0050] Furthermore, calling the electric drive assembly status analysis module includes: The integrated axial electric drive assembly historical operation dataset is acquired through the end-side sensor network; the integrated axial electric drive assembly historical operation dataset is analyzed and processed to construct an electric drive assembly data preprocessing program and an electric drive assembly state anomaly evaluator; based on the electric drive assembly data preprocessing program and the electric drive assembly state anomaly evaluator, an electric drive assembly state analysis module is formed, and the electric drive assembly state analysis module is stored in the Q edge computing gateways respectively.

[0051] The integrated axial electric drive assembly historical operating dataset refers to the collection of all operating data acquired by sensors over a past period. This data includes, but is not limited to, temperature, vibration, pressure, current, and voltage, covering all important performance indicators of the electric drive assembly. This data is stored in time-series format, with each data point representing a sensor reading at a specific time or under a specific operating condition. By integrating this historical data, the long-term performance of the electric drive assembly can be analyzed.

[0052] By processing, cleaning, transforming, and classifying the historical operating dataset of the integrated axial electric drive assembly, key features are extracted, and noise and irrelevant information are removed. The electric drive assembly data preprocessing procedure cleans, normalizes, imputes missing values, and removes outliers from the input historical data. This procedure ensures data quality, making it suitable for further analysis. After preprocessing, the data becomes more standardized, aiding in subsequent anomaly assessment and fault detection. The preprocessing procedure includes data cleaning, missing value imputation, and smoothing operations. The electric drive assembly status anomaly evaluator, based on the preprocessed data, uses specific algorithms to evaluate the operating status of the electric drive assembly. The core task of the anomaly evaluator is to determine whether the electric drive assembly has operational anomalies or faults. Specifically, it employs data-driven analysis methods, such as statistical methods and machine learning algorithms, for fault diagnosis. For example, by analyzing patterns in historical data, a model is built to determine whether new data deviates from normal operating conditions. For instance, the annotation information of historical data is used to train a classification model to identify which data indicates normal operation and which indicates an abnormal state. The evaluator can further output relevant status parameters based on the analysis results, which reflect the health status of the electric drive assembly.

[0053] Based on the electric drive assembly data preprocessing program and the electric drive assembly status anomaly evaluator, an electric drive assembly status analysis module is constructed. This module receives real-time operating data from the electric drive assembly, performs data preprocessing, and applies the anomaly evaluator to determine whether the electric drive assembly is in an abnormal state. This electric drive assembly status analysis module is deployed to each edge computing gateway because each edge computing gateway needs to independently process local sensor data; therefore, the status analysis module must run on each gateway. Through this deployment, the edge computing gateways can quickly process and analyze real-time data from the electric drive assembly locally, reducing latency in data transmission to the cloud platform.

[0054] Furthermore, the electric drive assembly data preprocessing program and the electric drive assembly status anomaly evaluator are constructed, including: Based on the electric drive assembly data application standard, an electric drive assembly data preprocessing program is constructed. The electric drive assembly data preprocessing program is used to preprocess and label the historical operation dataset of the integrated axial electric drive assembly to obtain an abnormal operation sample set of the electric drive assembly. The abnormal operation sample set of the electric drive assembly is trained and optimized using an LSTM network to construct an electric drive assembly state anomaly evaluator.

[0055] Electric drive assembly data application standards refer to the unified rules and standards used to process electric drive assembly monitoring data. These standards cover data format, data cleaning, missing value handling, time synchronization, and unit standardization. For example, the outputs of all temperature sensors need to be converted to a unified temperature unit, such as degrees Celsius; vibration data needs to be filtered to remove high-frequency noise; and all data needs to be standardized to the same range or scale. Based on these application standards, an electric drive assembly data preprocessing program is constructed. This preprocessing program adopts different data processing methods according to the specific application scenario and data requirements of the electric drive assembly.

[0056] An electric drive assembly data preprocessing program was used to process the historical operating dataset of the integrated axial electric drive assembly. The goal of the processing was to transform the raw data into a clean, standardized, and normalized form, making it suitable for subsequent anomaly detection and analysis. During preprocessing, abnormal states in the historical data were marked according to preset standards or manually labeled data. These abnormal states included: temperature exceeding safety thresholds, excessive vibration, and current fluctuations exceeding normal ranges. The labeled abnormal states could be defined based on expert experience, historical fault records, or by comparing existing data with normal data patterns. After preprocessing and labeling, the data containing abnormal states was collected into an electric drive assembly abnormal operation sample set, which was used for subsequent model training and testing.

[0057] LSTM (Laser-Side Module) networks are deep learning network architectures suitable for processing time-series data. LSTMs can effectively learn complex patterns in time-series data by memorizing long-term dependencies. In the monitoring of electric drive assemblies, LSTM networks can analyze historical operating data, identify long-term operating patterns, and predict anomalies. Since the operating data of electric drive assemblies is time-series data, LSTMs are very effective in this type of task. Using a sample set of abnormal operating data from electric drive assemblies, which provides labeled data for both normal and abnormal states, the LSTM network is trained. The LSTM network learns features and patterns in the data to detect abnormal states in future real-time data. During training, by inputting the sample set into the LSTM network, the network adjusts its internal weights and learns how to predict abnormal states from the input data. LSTM training is conducted through supervised learning, using labeled abnormal samples to train the network. Through cross-validation and parameter optimization, such as adjusting the number of LSTM layers, learning rate, and activation function, the model performance is continuously optimized to ensure good predictive accuracy when processing the operating data of electric drive assemblies. After training and tuning, the LSTM network becomes an evaluator for the abnormal state of an electric drive assembly. This evaluator can assess whether the electric drive assembly is in an abnormal state based on new real-time data.

[0058] Furthermore, the state parameters of the integrated axial electric drive assembly are obtained, including: Anomaly coupling analysis is performed on the finite element model of the axial electric drive assembly using the electric drive cloud platform to construct anomaly coupling model of the axial electric drive assembly. Based on the anomaly coupling model of the axial electric drive assembly, the state integration evaluation of the Q abnormal state parameters of the electric drive is performed to obtain integrated axial electric drive assembly state parameters.

[0059] The finite element model of the axial electric drive assembly can simulate its performance under different operating conditions and provide a foundation for further anomaly analysis. Within the electric drive cloud platform, anomaly coupling analysis is performed on the finite element model of the axial electric drive assembly. Anomaly coupling analysis refers to the comprehensive evaluation of abnormal state parameters of various components of the electric drive assembly, analyzing how these abnormal states influence and interact with each other. The goal is to detect existing system-level anomalies through the coupling relationships between the various components and sensor data of the electric drive assembly. For example, an anomaly in one component, such as overheating or excessive vibration, can affect the normal operation of other components, leading to a larger system failure. After completing the anomaly coupling analysis, an anomaly coupling model of the axial electric drive assembly is constructed. This model integrates abnormal data from different electric drive components and sensors, better reflecting the overall state of the system and helping to identify potential faults.

[0060] State integration assessment refers to the aggregation and integration of Q abnormal state parameters of electric drives on the electric drive cloud platform. This data is then comprehensively analyzed using an axial electric drive assembly anomaly coupling model. By evaluating the abnormal state parameters uploaded by Q edge computing gateways, the cloud platform transforms these local anomaly information into global system state parameters. The integration assessment process involves weight evaluation and correlation analysis of each abnormal state parameter to determine the degree of anomaly in each electric drive and their impact on the overall system operation. Through integration assessment, a comprehensive integrated axial electric drive assembly state parameter is obtained. This state parameter includes the overall performance of all electric drive anomalies, as well as their interactions and influences. For example, if one electric drive experiences a serious failure, other electric drives will also be affected; therefore, the integrated state parameter incorporates this coupling effect.

[0061] Example 2, based on the same inventive concept as the cloud-based integrated axial electric drive assembly condition monitoring method in the foregoing examples, such as... Figure 2 As shown in the embodiment of this application, an integrated axial electric drive assembly condition monitoring system based on a cloud platform is provided. The system includes: Edge computing gateway construction unit 10 is used to identify key parts of the target integrated axial electric drive assembly, deploy sensors, construct an end-side sensor network, and establish Q edge computing gateways based on the end-side sensor network, connecting the Q edge computing gateways to an RS485 bus; data mapping and transmission unit 20 is used to monitor the axial electric drive assembly operating data stream through the end-side sensor network, activate the RS485 bus to map and transmit the axial electric drive assembly operating data stream to the Q edge computing gateways; abnormal state evaluation unit 30 is used to synchronously evaluate the abnormal state of the axial electric drive assembly operating data stream based on the Q edge computing gateways, output Q electric drive abnormal state parameters, and upload the Q electric drive abnormal state parameters to the electric drive cloud platform through a 5G communication module; early warning decision control unit 40 is used to perform coupled state evaluation on the Q electric drive abnormal state parameters through the electric drive cloud platform to obtain integrated axial electric drive assembly state parameters, and perform early warning decision control based on the integrated axial electric drive assembly state parameters.

[0062] Furthermore, the edge computing gateway construction unit 10 is used to perform the following operation steps: Based on the structural design data and material property data of the target integrated axial electric drive assembly, finite element modeling is performed to generate the axial electric drive assembly finite element model; according to the application scenario of the electric drive assembly, simulated operating condition parameters of the electric drive assembly are constructed; the simulated operating condition parameters of the electric drive assembly are mapped to the axial electric drive assembly finite element model for operating condition simulation testing, and the operating test data of the electric drive assembly is recorded; based on the operating test data of the electric drive assembly, key parts of the target integrated axial electric drive assembly are identified, sensors are deployed, and an end-side sensor network is constructed.

[0063] Furthermore, the edge computing gateway construction unit 10 is used to perform the following operation steps: Based on the operational test data of the electric drive assembly, stress, temperature, and vibration performance analysis was performed on the target integrated axial electric drive assembly to obtain multi-dimensional performance distribution parameters of the electric drive assembly. According to these multi-dimensional performance distribution parameters, key components of the target integrated axial electric drive assembly were identified, determining multiple key components. Sensor selection and installation coverage analysis were performed on each of these key components to obtain sensor parameter sets for each key component. Based on these sensor parameter sets, sensors were deployed on the target integrated axial electric drive assembly to construct an end-side sensor network.

[0064] Furthermore, the edge computing gateway construction unit 10 is used to perform the following operation steps: Based on the type and quantity of parameters collected by the edge sensor network, processing requirements are assessed to obtain the edge sensor data processing requirements. According to the monitoring scale and data processing objectives of the target integrated axial electric drive assembly, the number of edge gateways to be deployed, Q, is determined. Edge gateway deployment rules are obtained, including spatial proximity, processing requirement satisfaction, and data processing balance. Using the edge sensor data processing requirements and the number of edge gateways to be deployed, Q, as constraints, edge gateways are configured for the target integrated axial electric drive assembly based on the edge gateway deployment rules, establishing Q edge computing gateways.

[0065] Furthermore, the data mapping and transmission unit 20 is used to perform the following operation steps: Assign an RS485 address to each of the Q edge computing gateways according to the RS485 bus, and construct a Q edge gateway-RS485 mapping address table; associate and match the axial electric drive assembly operating data stream with the Q edge computing gateways to determine a matching edge computing gateway set; perform self-organizing network analysis on the matching edge computing gateway set according to the Q edge gateway-RS485 mapping address table to activate the RS485 self-organizing communication network; and map and transmit the axial electric drive assembly operating data stream to the matching edge computing gateway set based on the RS485 self-organizing communication network.

[0066] Furthermore, the abnormal state assessment unit 30 is used to perform the following operation steps: Based on the Q edge computing gateways, the axial electric drive assembly operation data stream is mapped and split synchronously to obtain Q gateway electric drive assembly operation data streams; according to the Q edge computing gateways, the electric drive assembly status analysis module is called; based on the electric drive assembly status analysis module, the abnormal status of the Q gateway electric drive assembly operation data streams is evaluated synchronously, and Q electric drive abnormal status parameters are output.

[0067] Furthermore, the abnormal state assessment unit 30 is used to perform the following operation steps: The integrated axial electric drive assembly historical operation dataset is acquired through the end-side sensor network; the integrated axial electric drive assembly historical operation dataset is analyzed and processed to construct an electric drive assembly data preprocessing program and an electric drive assembly state anomaly evaluator; based on the electric drive assembly data preprocessing program and the electric drive assembly state anomaly evaluator, an electric drive assembly state analysis module is formed, and the electric drive assembly state analysis module is stored in the Q edge computing gateways respectively.

[0068] Furthermore, the abnormal state assessment unit 30 is used to perform the following operation steps: Based on the electric drive assembly data application standard, an electric drive assembly data preprocessing program is constructed. The electric drive assembly data preprocessing program is used to preprocess and label the historical operation dataset of the integrated axial electric drive assembly to obtain an abnormal operation sample set of the electric drive assembly. The abnormal operation sample set of the electric drive assembly is trained and optimized using an LSTM network to construct an electric drive assembly state anomaly evaluator.

[0069] Furthermore, the early warning decision control unit 40 is used to perform the following operation steps: Anomaly coupling analysis is performed on the finite element model of the axial electric drive assembly using the electric drive cloud platform to construct anomaly coupling model of the axial electric drive assembly. Based on the anomaly coupling model of the axial electric drive assembly, the state integration evaluation of the Q abnormal state parameters of the electric drive is performed to obtain integrated axial electric drive assembly state parameters.

[0070] Through the foregoing detailed description of the cloud-based integrated axial electric drive assembly condition monitoring method, those skilled in the art can clearly understand the cloud-based integrated axial electric drive assembly condition monitoring system in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to in the method section.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A cloud-based integrated axial electric drive assembly condition monitoring method, characterized in that, The method includes: Key components of the target integrated axial electric drive assembly are identified and sensors are deployed to construct an end-side sensor network. Based on the end-side sensor network, Q edge computing gateways are established and connected to an RS485 bus. The axial electric drive assembly operating data stream is monitored through the end-side sensor network, and the RS485 bus is activated to map and transmit the axial electric drive assembly operating data stream to the Q edge computing gateways. Based on the Q edge computing gateways, the abnormal status assessment of the axial electric drive assembly operation data stream is performed synchronously, and Q electric drive abnormal status parameters are output. The Q electric drive abnormal status parameters are then uploaded to the electric drive cloud platform through the 5G communication module. The electric drive cloud platform performs coupled state evaluation on the Q electric drive abnormal state parameters to obtain the integrated axial electric drive assembly state parameters, and performs early warning decision control based on the integrated axial electric drive assembly state parameters.

2. The cloud-based integrated axial electric drive assembly condition monitoring method as described in claim 1, characterized in that, Constructing an edge-side sensor network includes: Finite element modeling is performed based on the structural design data and material property data of the target integrated axial electric drive assembly to generate a finite element model of the axial electric drive assembly. Based on the application scenarios of the electric drive assembly, simulated operating parameters of the electric drive assembly are constructed. The simulated operating parameters of the electric drive assembly are mapped to the finite element model of the axial electric drive assembly for operating condition simulation testing, and the operating test data of the electric drive assembly are recorded. Based on the operational test data of the electric drive assembly, key components of the target integrated axial electric drive assembly are identified, sensors are deployed, and an end-side sensor network is constructed.

3. The cloud-based integrated axial electric drive assembly condition monitoring method as described in claim 2, characterized in that, Based on the operational test data of the electric drive assembly, key components of the target integrated axial electric drive assembly are identified, sensors are deployed, and an end-side sensor network is constructed, including: Based on the operational test data of the electric drive assembly, stress, temperature and vibration performance analysis was performed on the target integrated axial electric drive assembly to obtain multidimensional performance distribution parameters of the electric drive assembly. Based on the multi-dimensional performance distribution parameters of the electric drive assembly, the key components of the target integrated axial electric drive assembly are identified, and multiple key components of the electric drive assembly are determined. Sensor selection and installation coverage analysis were performed on key components of the multiple electric drive assemblies to obtain sensor parameter sets for multiple key components. Based on the sensor parameter sets of the multiple key components, sensors are deployed on the target integrated axial electric drive assembly to construct an end-side sensor network.

4. The integrated axial electric drive assembly condition monitoring method based on a cloud platform as described in claim 1, characterized in that, Establish Q edge computing gateways, including: Based on the type and quantity of parameters collected by the end-side sensor network, the processing requirements are assessed to obtain the end-side sensor data processing requirements. The number of edge gateways to be deployed is determined based on the monitoring scale and data processing objectives of the target integrated axial electric drive assembly; Obtain edge gateway deployment rules, which include spatial proximity, processing requirement satisfaction, and data processing balance; Using the data processing requirements of the end-side sensors and the number of edge gateways Q as constraints, the target integrated axial electric drive assembly is configured with edge gateways based on the edge gateway deployment rules to establish Q edge computing gateways.

5. The integrated axial electric drive assembly condition monitoring method based on a cloud platform as described in claim 1, characterized in that, Activating the RS485 bus to map and transmit the axial electric drive assembly's operating data stream to the Q edge computing gateways includes: According to the RS485 bus, each of the Q edge computing gateways is assigned an RS485 address, and a Q edge gateway-RS485 mapping address table is constructed. The axial electric drive assembly operating data stream is associated and matched with the Q edge computing gateways to determine the matching edge computing gateway set; According to the Q edge gateway-RS485 mapping address table, perform self-organizing network analysis on the matching edge computing gateway set and activate the RS485 self-organizing communication network. The axial electric drive assembly operation data stream is mapped and transmitted to the matching edge computing gateway set based on the RS485 self-organizing communication network.

6. The cloud-based integrated axial electric drive assembly condition monitoring method as described in claim 1, characterized in that, Output Q electric drive abnormal status parameters, including: Based on the synchronous mapping and splitting of the axial electric drive assembly operation data stream by the Q edge computing gateways, Q gateway electric drive assembly operation data streams are obtained; Based on the Q edge computing gateways, the electric drive assembly status analysis module is invoked; Based on the electric drive assembly status analysis module, abnormal status assessments are performed on the operating data streams of the Q gateway electric drive assemblies simultaneously, and Q electric drive abnormal status parameters are output.

7. The cloud-based integrated axial electric drive assembly condition monitoring method as described in claim 6, characterized in that, Call the electric drive assembly status analysis module, including: The historical operating dataset of the integrated axial electric drive assembly is obtained through the end-side sensor network. The historical operating dataset of the integrated axial electric drive assembly is analyzed and processed to construct an electric drive assembly data preprocessing program and an electric drive assembly state anomaly evaluator; Based on the electric drive assembly data preprocessing program and the electric drive assembly status anomaly evaluator, an electric drive assembly status analysis module is formed, and the electric drive assembly status analysis module is stored in the Q edge computing gateways respectively.

8. The integrated axial electric drive assembly status monitoring method based on a cloud platform as described in claim 7, characterized in that, Construct an electric drive assembly data preprocessing program and an electric drive assembly status anomaly evaluator, including: Based on the electric drive assembly data application standard, a preprocessing program for electric drive assembly data is constructed. The integrated axial electric drive assembly historical operation dataset is preprocessed and anomaly status is labeled using the electric drive assembly data preprocessing program to obtain an abnormal operation sample set of the electric drive assembly; An LSTM network is used to train and optimize the abnormal operation sample set of the electric drive assembly to construct an abnormal state evaluator for the electric drive assembly.

9. The cloud-based integrated axial electric drive assembly condition monitoring method as described in claim 2, characterized in that, The integrated axial electric drive assembly state parameters are obtained, including: An abnormal coupling analysis was performed on the finite element model of the axial electric drive assembly using the electric drive cloud platform to construct an abnormal coupling model of the axial electric drive assembly. Based on the abnormal coupling model of the axial electric drive assembly, the state integration evaluation of the Q abnormal state parameters of the electric drive is performed to obtain the integrated axial electric drive assembly state parameters.

10. An integrated axial electric drive assembly condition monitoring system based on a cloud platform, characterized in that, For implementing the cloud-based integrated axial electric drive assembly condition monitoring method according to any one of claims 1-9, the system comprises: An edge computing gateway construction unit is used to identify key parts and deploy sensors for the target integrated axial electric drive assembly, build an end-side sensor network, and establish Q edge computing gateways based on the end-side sensor network, and connect the Q edge computing gateways to an RS485 bus. The data mapping and transmission unit is used to monitor the axial electric drive assembly operation data stream through the end-side sensor network, and activate the RS485 bus to map and transmit the axial electric drive assembly operation data stream to the Q edge computing gateways. An abnormal state assessment unit is used to assess the abnormal state of the axial electric drive assembly operating data stream synchronously based on the Q edge computing gateways, output Q electric drive abnormal state parameters, and upload the Q electric drive abnormal state parameters to the electric drive cloud platform through a 5G communication module. The early warning decision control unit is used to perform coupled state evaluation on the Q abnormal state parameters of the electric drive through the electric drive cloud platform to obtain the integrated axial electric drive assembly state parameters, and to perform early warning decision control based on the integrated axial electric drive assembly state parameters.

Images