AI intelligent motor driving device of new energy overhead working truck and control method

Abstract

The invention discloses an AI intelligent motor driving device of a new energy overhead working truck and a control method. The AI intelligent motor driving device comprises a main controller, a sensor, a power driving module and an AI coprocessor. A convolutional neural network model used for real-time working condition recognition, a long and short-term memory network model used for energy consumption prediction and a deep reinforcement learning model used for generating adaptive control are loaded and run in the AI coprocessor. The attitude of the vehicle, the real-time state of an operation arm and the environment information of a moving path can be processed at the same time, fault symptoms such as abnormal mechanical vibration of a motor and degradation of electrical parameters can be recognized in time, and control strategies adaptive to various complex working conditions such as climbing, heeling, wind loads and path obstacles can be automatically generated; therefore, the driving stability and the operation safety of the vehicle in the moving process are effectively enhanced, and meanwhile, the energy consumption of the whole machine is reduced through energy efficiency optimization. The system also has fault early warning, redundancy switching control and remote algorithm maintenance capabilities, and the comprehensive reliability of the equipment is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the field of new energy aerial work platforms, and in particular to an AI intelligent motor driving device and control method of a new energy aerial work platform. BACKGROUND

[0002] Aerial work platforms are important engineering machinery, and their driving stability, operation safety and energy consumption economy are widely concerned. With the popularization of new energy technology, more and more aerial work platforms are driven by electricity. However, the operation environment of such vehicles is complex, and they often need to cope with high-altitude wind, inclined road, overload in the process of transfer and mechanical failure of the walking system and other challenges. The controllers commonly used at present are mostly based on fixed parameter setting, and their adaptive adjustment capability is limited when coping with the above-mentioned complex dynamic conditions.

[0003] Although intelligent control is attempted in the prior art, the core computing task is usually deployed on a remote server or an upper computer, resulting in a large control loop delay and difficulty in ensuring real-time performance. Some other solutions are processed locally, but only use microcontrollers with limited computing power, which cannot support complex AI algorithms for deep learning and predictive control.

[0004] In addition, the traditional system also lacks a sensor integration scheme for the special needs of the aerial work platform during movement, such as comprehensive collection and fusion of real-time attitude of the whole vehicle, safety state of the operation arm, and visual information of the surrounding environment. It also fails to establish an effective intelligent fault management mechanism, and cannot realize early warning of mechanical failure of the walking system and real-time judgment of mobile safety risks.

[0005] Therefore, it is necessary to realize real-time processing and intelligent decision-making of multi-source sensor information, and on this basis, to build an integrated driving system with fault prediction and risk mitigation capability, so as to improve the safety, reliability and overall intelligence level of the aerial work platform during movement. SUMMARY

[0006] In order to solve the above technical problems, the application adopts the technical scheme of an AI intelligent motor driving device and control method of a new energy aerial work platform, comprising:

[0007] In order to solve the above technical problems, the application adopts the technical scheme of an AI intelligent motor driving device and control method of a new energy aerial work platform, comprising: a main controller, which communicates with a whole vehicle controller and a battery management system of the aerial work platform through a CAN bus; The sensor, whose signal output terminal is connected to the main controller, includes an integrated sensor module and an external sensor module. The integrated sensor module collects information including vehicle attitude and position, while the external sensor module collects information including boom attitude, load, and path obstacles. The power drive module is controlled by the main controller to drive the motor through PWM signals according to parameter adjustment instructions; The AI ​​coprocessor communicates with the main controller via a data bus and performs real-time fusion analysis on sensor data to generate optimized decision instructions for the motor. The AI ​​coprocessor loads and runs a convolutional neural network model for real-time operating condition identification, a long short-term memory network model for energy consumption prediction, and a deep reinforcement learning model for generating adaptive control.

[0008] Furthermore, the integrated sensor module includes a current sensor, a temperature sensor, a BeiDou / GPS positioning unit, a slope angle detection sensor, and a six-axis inertial measurement unit; Current sensors are used to monitor the drive current of the motor and the system bus current; Temperature sensors are used to collect the temperature of components and the environment; The slope angle detection sensor is used to measure the tilt angle of the entire vehicle;

[0009] The six-axis inertial measurement unit is used to detect the three-dimensional acceleration and angular velocity of the entire vehicle in order to sense the vehicle's attitude and vibration status.

[0010] Furthermore, the external sensor module includes vibration sensors, displacement sensors, angle sensors, pressure sensors, infrared sensors, and vision sensors; The vibration sensor is mounted on the motor housing to collect the vibration spectrum; Displacement sensors are used to measure the extension and retraction length of the boom relative to the vehicle body; Angle sensors are used to detect the rotation angle of the boom; Pressure sensors are installed at the lifting cylinder or support hinge of the boom to monitor load pressure. Infrared sensors and vision sensors work together to detect people and obstacles in the vehicle's path.

[0011] Furthermore, the AI ​​coprocessor analyzes vibration spectrum data from vibration sensors using a convolutional neural network model to identify mechanical faults in the walking drive motor; The AI ​​coprocessor analyzes time-series data of current and temperature through a long short-term memory network model to predict electrical fault risks and energy consumption trends. The AI ​​coprocessor generates control strategies based on the state of the working arm, the vehicle's posture, and path obstacle information through a deep reinforcement learning model.

[0012] Furthermore, the AI ​​coprocessor also performs real-time fusion analysis and credibility assessment of the received sensor data; when the AI ​​coprocessor identifies anomalies in the sensor data, it provides the assessment results to the main controller, which then enables redundant data sources or data generated based on estimation algorithms according to the assessment results.

[0013] Furthermore, the main controller, integrated sensor module, power drive module, and AI coprocessor are integrated into the same housing, which is formed with electromagnetic shielding and heat dissipation structure. The external sensor module is connected to the main controller via CAN bus.

[0014] Furthermore, it also includes an integrated 5G communication unit. The 5G communication unit is implemented by forming a communication link, realizing remote data monitoring, modifying operating parameters, and updating / optimizing the algorithm model in the AI ​​coprocessor online. The communication link formed by the 5G communication unit supports bidirectional data transmission, allowing users to issue commands through a host computer.

[0015] A control method for an AI intelligent motor drive device of a new energy aerial work platform vehicle, characterized by comprising the following steps: S1. Real-time acquisition of vehicle and environmental status data through integrated sensor modules and external sensor modules; S2. The main controller preprocesses the collected data to form standardized data packets; S3. The main controller sends standardized data packets to the AI ​​coprocessor; Based on the received data, the S4.AI coprocessor performs real-time inference through its internally loaded AI algorithm model to generate driving motor control optimization decision instructions; S5. The main controller outputs an adjustment signal to the power drive module based on the control optimization decision command; S6. The power drive module outputs a corresponding PWM drive signal to control the motor operation based on the adjustment signal.

[0016] Furthermore, the main controller monitors the operating status and communication link of the AI ​​coprocessor in real time. When it determines that the communication has failed or the communication has timed out, the main controller will switch the control to the traditional control mode in which the main controller operates independently.

[0017] An AI-powered intelligent motor drive device and control method for a new energy aerial work platform can simultaneously process information on the vehicle's own posture, the real-time status of the boom, and the environmental information of the movement path. It can promptly identify fault symptoms such as abnormal motor mechanical vibration and deterioration of electrical parameters, and autonomously generate control strategies adapted to various complex working conditions, including climbing, tilting, wind loads, and obstacle courses. This effectively enhances the vehicle's driving stability and operational safety during movement, while also reducing overall energy consumption through energy efficiency optimization. The system also features early fault warning, control redundancy switching, and remote algorithm maintenance capabilities, improving the overall reliability of the equipment. Attached Figure Description

[0018] Fig. 1 This is a structural block diagram of the device of the present invention.

[0019] Fig. 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0021] like Figs. 1-2 As shown in the figure, this embodiment relates to the AI ​​intelligent motor drive device and control method for new energy aerial work vehicles, which realizes adaptive optimization control and intelligent fault management of the walking drive system under complex working conditions, especially for the aerial work vehicle's high-altitude working environment, vehicle dynamic balance requirements and mobile safety during the movement process.

[0022] The AI ​​intelligent motor drive device for new energy aerial work platforms includes: The main controller communicates with the vehicle controller and battery management system of the aerial work platform via the CAN bus. The main controller is a high-performance multi-core microcontroller that meets automotive-grade standards, such as the Infineon AURIX™ TC3xx series. It has multiple CAN-FD interfaces to coordinate internal module data exchange and communicate with the vehicle controller and battery management system through the vehicle's CAN network.

[0023] The sensor, whose signal output terminal is connected to the main controller, includes an integrated sensor module and an external sensor module. The integrated sensor module collects information including vehicle attitude and position, while the external sensor module collects information including boom attitude, load, and path obstacles. Preferably, the integrated sensor module includes a current sensor, a temperature sensor, a Beidou / GPS positioning unit, a slope angle detection sensor, and a six-axis inertial measurement unit; the current sensor is used to monitor the drive current of the motor and the system bus current; the temperature sensor is used to collect the temperature of the components and the ambient temperature; the slope angle detection sensor is used to measure the tilt angle of the entire vehicle; and the six-axis inertial measurement unit is used to detect the three-dimensional acceleration and angular velocity of the entire vehicle to sense the attitude and vibration state of the entire vehicle. Preferably, the external sensor module includes a vibration sensor, a displacement sensor, an angle sensor, a pressure sensor, an infrared sensor, and a vision sensor. The vibration sensor is mounted on the motor housing and is used to collect the vibration spectrum. The displacement sensor is used to measure the extension length of the boom relative to the vehicle body. The angle sensor is used to detect the rotation angle of the boom. The pressure sensor is mounted at the lifting cylinder or support hinge point of the boom and is used to monitor the load pressure. The infrared sensor and vision sensor work together to detect personnel and obstacles in the vehicle's movement path. Specifically, a high-definition camera (visible light / infrared) is used, combined with AI image recognition, to detect personnel, equipment, and cables in the vehicle's movement path, enabling warnings of dangerous driving areas. It should be noted that a proximity switch is also included to confirm whether the boom is in the safe retracted position.

[0024] The power drive module uses an IPM (Intelligent Power Module) as the power switching device. It integrates IGBTs and drive and protection circuits. It is controlled by the main controller to drive the motor through PWM signals according to parameter adjustment instructions. At the same time, the power drive module integrates analog and digital interfaces to collect temperature sensor signals and high-precision photoelectric encoder signals installed on the walking drive motor in real time, and uploads these data to the main controller to form the basis of closed-loop control of current loop, speed loop and position loop.

[0025] The AI ​​coprocessor communicates with the main controller via a data bus, performs real-time fusion analysis on sensor data to generate optimized decision instructions for the motor. The AI ​​coprocessor uses a dedicated computing chip with an integrated neural network processing unit, such as the Huawei Ascend310 series core module. The AI ​​coprocessor loads and runs a convolutional neural network model for real-time operating condition recognition, a long short-term memory network model for energy consumption prediction, and a deep reinforcement learning model for generating adaptive control. Preferably, the AI ​​coprocessor analyzes vibration spectrum data from vibration sensors using the convolutional neural network model to identify mechanical faults in the walking drive motor; the AI ​​coprocessor analyzes time-series data of current and temperature using the long short-term memory network model to predict electrical fault risks and energy consumption trends; and the AI ​​coprocessor generates control strategies based on the state of the working arm, the vehicle's posture, and path obstacle information using the deep reinforcement learning model.

[0026] Preferably, the AI ​​coprocessor also performs real-time fusion analysis and credibility assessment on the received sensor data; when the AI ​​coprocessor identifies anomalies in the sensor data, it provides the assessment results to the main controller, which then enables redundant data sources or data generated based on estimation algorithms according to the assessment results.

[0027] Based on the above structure, the main controller, integrated sensor module, power drive module and AI coprocessor are integrated in the same housing. The housing is formed in a way that has electromagnetic shielding and heat dissipation structure. The external sensor module is connected to the main controller through CAN bus.

[0028] It also includes an integrated 5G communication unit, which is implemented by forming a communication link, enabling remote data monitoring, modification of operating parameters, and online updating / optimization of algorithm models in the AI ​​coprocessor. The communication link formed by the 5G communication unit supports bidirectional data transmission, allowing users to issue commands through a host computer.

[0029] In addition, the main controller monitors the operating status and communication link of the AI ​​coprocessor in real time. When it determines that the communication has failed or the communication has timed out, the main controller will switch the control to the traditional control mode in which the main controller operates independently.

[0030] The control method for the AI ​​intelligent motor drive device of the new energy aerial work platform vehicle is as follows:

[0031] After the device in this embodiment is powered on and initialized, the integrated sensor module, external sensor module and power drive module begin to collect real-time and synchronous status, environmental information and driving motor operation data related to vehicle movement. The main controller performs fusion preprocessing on the collected multi-channel raw data. This preprocessing includes timestamp alignment of all sensor data based on a unified time reference, Kalman filtering of IMU, current, vibration and other signals to suppress noise, temperature compensation and zero bias calibration of sensor data, and encapsulation of all data into standard data frames with timestamps and data source identifiers.

[0032] The main controller sends the pre-processed standardized data packets to the AI ​​coprocessor via a high-speed link. The AI ​​coprocessor first performs real-time fusion analysis and confidence assessment on the received multi-source sensor data. By comparing the redundancy or correlation information from different sensors, it identifies whether the data of a specific sensor has abnormal drift or failure. When an abnormal data is detected, it sends an evaluation result with confidence level to the main controller.

[0033] The AI ​​coprocessor uses a convolutional neural network model to analyze vibration sensor signals from the walking drive motor, identifying changes in specific frequency components caused by bearing wear, rotor imbalance, etc., to achieve early warning and location of mechanical faults. At the same time, the AI ​​coprocessor combines a long short-term memory network model to analyze time-series data such as current and temperature, predicting the development trend of progressive faults such as motor overheating and efficiency degradation. The AI ​​coprocessor uses convolutional neural networks to extract and fuse features from multi-source input data, identifying and classifying the specific driving conditions of the vehicle in real time, such as constant speed driving on flat roads, steep slope climbing, heavy-load transfer, obstacles in the movement path, and side tilt risk warning. The AI ​​coprocessor utilizes a long short-term memory network to combine current real-time state data with historical operation sequence data to predict energy consumption trends and expected energy consumption values ​​in the short term. Using deep reinforcement learning algorithms, with the comprehensive goals of optimal global energy efficiency, system stability, and safety, the AI ​​coprocessor dynamically generates optimal control command parameters for the drive motor based on operating condition identification, fault diagnosis, and energy consumption prediction information. When a potential or existing fault is detected, the model can generate specific mitigation strategies, such as proactively limiting output torque when an overheating risk is predicted for the drive motor, smoothly decelerating in advance when obstacles or people are detected in the movement path, and adjusting the torque distribution of the drive motor in real time to maintain driving stability when strong winds cause the entire vehicle to sway, as detected by the IMU.

[0034] The AI ​​coprocessor returns the generated control optimization instructions to the main controller, which then dynamically corrects the internal settings and outputs corresponding adjustment signals to the power drive module. If the main controller receives an alarm command, it sends fault information to other modules in the vehicle via the CAN bus, displays the error code, and triggers an audible and visual alarm. Simultaneously, it displays detailed error information on the host computer via the 5G communication unit. The power drive module precisely controls the output torque and speed of the drive motor by adjusting the duty cycle and characteristics of its output PWM signal, completing one intelligent control cycle.

[0035] During operation, the device in this embodiment continuously collects data on the execution effect of control commands and forms an experience dataset. The AI ​​coprocessor can periodically or during idle periods use this new data to incrementally learn the model. Remote engineers can also send commands to the device via a secure 5G communication link through host computer software to adjust parameters, optimize algorithms, or update and replace the complete model version of the algorithm model in the AI ​​coprocessor.

[0036] The typical application scenarios of this embodiment are as follows: Taking the complex working conditions of a high-altitude work vehicle during relocation, involving potential mechanical failures and sudden environmental risks, as an example, the system detects abnormal harmonics in the bearing frequency at the drive end of the travel drive motor through vibration sensors. The convolutional neural network fault diagnosis model of the AI ​​coprocessor identifies this as an early sign of bearing wear. Simultaneously, the visual sensor detects a worker suddenly entering the vehicle's path. The main controller packages this multi-source information and sends it to the AI ​​coprocessor. The convolutional neural network condition recognition model of the AI ​​coprocessor confirms that there is a safety risk in the travel path and that the travel system is in a state of early mechanical failure warning. The long short-term memory network model predicts that if the vehicle continues to travel under the current load and speed, bearing failure is likely. This will accelerate the deterioration; subsequently, the deep reinforcement learning adaptive control strategy generator decides on the optimal strategy that integrates safety, equipment lifespan, and transfer efficiency, namely, immediately outputting power reduction and speed limit commands to limit the peak torque and maximum vehicle speed of the drive motor, thereby reducing the bearing load and reserving a longer safe braking distance, while triggering an audible and visual alarm to remind the driver to pay attention; this strategy is converted into specific control parameters and returned to the main controller, which adjusts the output accordingly, and sends an early wear warning error code of the drive motor bearing to the display unit via the CAN bus, and reports it to the host computer via the 5G communication unit and displays detailed information and logs of early wear of the drive motor bearing and intrusion of personnel on the path ahead.

[0037] This application discloses an AI-powered intelligent motor drive device and control method for a new energy aerial work platform vehicle. It can simultaneously process information on the vehicle's own posture, the real-time status of the boom, and the environmental information of the movement path. It can promptly identify fault symptoms such as abnormal motor mechanical vibration and deterioration of electrical parameters, and autonomously generate control strategies adapted to various complex working conditions, including climbing, tilting, wind loads, and obstacle courses. As a result, the vehicle's driving stability and operational safety are effectively enhanced during movement, while energy efficiency optimization reduces overall energy consumption. The system also features early fault warning, control redundancy switching, and remote algorithm maintenance capabilities, improving the overall reliability of the equipment.

[0038] The above embodiments are not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the technical solution of the present invention are also within the protection scope of the present invention.

Claims

1. An AI intelligent motor driving device of a new energy aerial work platform vehicle, characterized in that it comprises: a main controller which communicates with a vehicle controller and a battery management system of the aerial work platform vehicle through a CAN bus; a sensor whose signal output end is connected with the main controller, the sensor comprising an integrated sensor module and an external sensor module, the integrated sensor module collecting vehicle posture and position information, and the external sensor module collecting working arm posture, load and path obstacle information; a power driving module which is controlled by the main controller to drive the motor to operate according to parameter adjustment instructions through PWM signals; and an AI coprocessor which is communicatively connected with the main controller through a data bus, performs real-time fusion analysis on data of the sensor to generate an optimized decision instruction for the motor, and has loaded and runs therein a convolutional neural network model for real-time working condition recognition, a long short-term memory network model for energy consumption prediction, and a deep reinforcement learning model for generating adaptive control.

2. The AI intelligent motor driving device of the new energy aerial work platform vehicle according to claim 1, characterized in that: the integrated sensor module comprises a current sensor, a temperature sensor, a Beidou / GPS positioning unit, a slope angle detection sensor and a six-axis inertial measurement unit; the current sensor is used for monitoring driving current and system bus current of the motor; the temperature sensor is used for collecting component and environmental temperature; the slope angle detection sensor is used for measuring the inclination angle of the vehicle; and the six-axis inertial measurement unit is used for detecting three-dimensional acceleration and angular velocity of the vehicle to perceive the vehicle posture and vibration state.

3. The AI intelligent motor driving device of the new energy aerial work platform vehicle according to claim 2, characterized in that: the external sensor module comprises a vibration sensor, a displacement sensor, an angle sensor, a pressure sensor, an infrared sensor and a vision sensor; the vibration sensor is installed on the housing of the motor and used for collecting vibration frequency spectrum; the displacement sensor is used for measuring the telescopic length of the working arm relative to the vehicle body; the angle sensor is used for detecting the rotation angle of the working arm; the pressure sensor is installed at the lifting cylinder or support hinge point of the working arm and used for monitoring load pressure; and the infrared sensor and the vision sensor cooperate to detect personnel and obstacles on the moving path of the vehicle.

4. The AI intelligent motor driving device of the new energy aerial work platform vehicle according to claim 3, characterized in that: the AI coprocessor analyzes vibration frequency spectrum data from the vibration sensor through the convolutional neural network model to identify mechanical faults of the walking driving motor; the AI coprocessor analyzes time series data of current and temperature through the long short-term memory network model to predict electrical fault risk and energy consumption trend; and the AI coprocessor generates a control strategy based on the state of the working arm and the vehicle posture and path obstacle information through the deep reinforcement learning model. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 5. The AI intelligent motor drive device of the new energy aerial work vehicle according to claim 1, characterized in that: The AI coprocessor also implements real-time fusion analysis and credibility evaluation on the received sensor data; when the AI coprocessor identifies abnormal data of the sensor, it provides the evaluation result to the main controller, and the main controller enables a redundant data source or generates data based on the evaluation algorithm according to the evaluation result.

6. The AI intelligent motor drive device of the new energy aerial work vehicle according to claim 1, characterized in that: The main controller, integrated sensor module, power drive module and AI coprocessor are integrated in the same housing, the housing is formed in a way with electromagnetic shielding and heat dissipation structure, the external sensor module is connected with the main controller through CAN bus.

7. The AI intelligent motor drive device of the new energy aerial work vehicle according to claim 6, characterized in that: It also includes an integrated 5G communication unit, which implements communication link formation, remote data monitoring, operating parameter modification and online updating / optimization of algorithm models in the AI coprocessor. The communication link formed by the 5G communication unit supports bidirectional data transmission, allowing users to issue instructions through the host computer.

8. A control method of the AI intelligent motor drive device of the new energy aerial work vehicle according to any one of claims 1-7, characterized in that, The method comprises the following steps: S1. Collecting vehicle and environmental state data in real time through the integrated sensor module and external sensor module; S2. The main controller preprocesses the collected data to form a standardized data packet; S3. The main controller sends the standardized data packet to the AI coprocessor; S4. The AI coprocessor generates walking drive motor control optimization decision instructions based on the received data through its internally loaded AI algorithm model for real-time inference; S5. The main controller outputs an adjustment signal to the power drive module according to the control optimization decision instruction; S6. The power drive module outputs corresponding PWM drive signals to control the motor operation according to the adjustment signal.

9. The control method of the AI intelligent motor drive device of the new energy aerial work vehicle according to claim 8, characterized in that: The main controller monitors the running state and communication link of the AI coprocessor in real time, and when it is determined that the communication is invalid or the communication is timed out, the main controller switches the control to the traditional control mode independently run by the main controller.

Images