Institute of Electrical Engineering, Anhui Technical College of Mechanical and Electrical Engineering, Wuhu 241002, Anhui, China.
Comput Intell Neurosci. 2022 Jul 1;2022:1447333. doi: 10.1155/2022/1447333. eCollection 2022.
To solve these uncertain problems by studying the motor fault diagnosis technology, so as to ensure the normal operation of the motor equipment is the primary problem to be solved in the field of motor fault diagnosis. The traditional DC motor is one of the most widely used motors at present. It has excellent speed regulation performance and is easy to control. It is widely used in applications that require high motor startup and speed regulation characteristics. This research mainly discusses DC motor control technology. Evidence theory can combine various fault information at different levels to enhance mutual support between pieces of evidence, thereby improving the accuracy of motor fault detection. Based on the steps of signal processing, feature extraction, feature dimensionality reduction, and state recognition, the research on the state recognition method of belt conveyor drive motor based on multisource information fusion is carried out. By studying the multisource information fusion, this paper proposes a two-stage belt conveyor drive motor information fusion model based on the optimal D-S evidence theory. The correct identification rate of broken rotor bars during fault monitoring is 99.8%. This method divides the specific motor fault feature set into multiple fault subspaces and uses different diagnostic neural networks and different fault feature parameters for local diagnosis, respectively. The scheme designed in this study significantly improves the recognition accuracy of the motor in the same working condition and under variable working conditions. The drive motor state recognition and intelligent decision-making system designed by combining the results of multisource information fusion can effectively describe the fault type and has strong operability.
为了解决这些不确定问题，需要研究电机故障诊断技术，以确保电机设备的正常运行，这是电机故障诊断领域需要解决的首要问题。传统的直流电机是目前应用最广泛的电机之一，它具有优良的调速性能和易于控制的特点，广泛应用于需要高电机启动和调速特性的应用场合。本研究主要讨论直流电机控制技术。证据理论可以结合不同层次的各种故障信息，增强证据之间的相互支持，从而提高电机故障检测的准确性。基于信号处理、特征提取、特征降维和状态识别的步骤，对基于多源信息融合的带式输送机驱动电机状态识别方法进行了研究。通过研究多源信息融合，提出了一种基于最优 D-S 证据理论的两级带式输送机驱动电机信息融合模型。故障监测中断条故障的正确识别率为 99.8%。该方法将特定电机故障特征集划分为多个故障子空间，并分别使用不同的诊断神经网络和不同的故障特征参数进行局部诊断。本研究设计的方案显著提高了电机在相同工作条件下和变工况下的识别精度。结合多源信息融合的结果设计的驱动电机状态识别和智能决策系统可以有效地描述故障类型，具有很强的可操作性。
