Research and Development of an AM5728 DSP + ARM-based Online Fault Monitoring System for Wind Turbines

In recent years, the total installed capacity and size of wind turbines in China have continuously increased. Mechanical failures in wind turbines are becoming more frequent, and operation and maintenance (O&M) costs are rising, becoming a major factor affecting the cost of the wind power industry. Therefore, the safe, efficient, and reliable operation of wind turbine equipment is particularly crucial. To monitor equipment operating status in real-time, detect hidden faults, and take preventive measures in advance, the development of an online fault monitoring system for wind turbines is of paramount importance. This paper primarily designs an AM5728 DSP+ARM-based online fault monitoring system for wind turbines. The system uses TI's AM5728 as the core controller to achieve multi-channel vibration signal acquisition, fault analysis of vibration signals, human-machine interface (HMI) display, and host computer communication functions. By installing sensors at appropriate locations on the wind turbine equipment, vibration signals from critical components are acquired. The acquired signals from critical components are then processed using amplitude modulation and order analysis algorithms to determine the fault status of these components. The status information of the wind turbine's critical components is sent to the host computer's HMI display via serial communication, allowing real-time monitoring of their operating conditions and enabling necessary preventive measures to be taken based on actual situations.

This paper first analyzes and researches domestic and international wind turbine condition monitoring systems, conducting detailed studies on the vibration characteristics, fault types of wind turbine gearbox gears and bearings, and gearbox fault diagnosis methods, and then determines the fault diagnosis algorithm adopted in this paper. Secondly, the hardware and software of the wind turbine online fault monitoring system are designed. The hardware part mainly includes core control system design, power supply design, communication circuit design, multi-channel signal acquisition circuit design, data storage circuit design, and human-machine interface display design. The software part mainly includes signal acquisition program design, fault diagnosis program design, LCD display program design, serial communication program design, data storage program design, and host computer HMI display interface design. Finally, the system's various functions are tested, and experimental data results are analyzed. MATLAB simulation software is used to assist in verifying the system's fault diagnosis results, demonstrating the feasibility and superiority of this system in wind turbine fault monitoring.