US-Egypt Cooperative Research: Online Fault Diagnostics for Induction Motor Drive Systems Through Electronic Signals and Artificial Intelligence Techniques

该奖项旨在支持威斯康星州密尔沃基市马奎特大学电气与计算机工程系Nabeel Demerdash博士和埃及开罗电子研究所Faeka Khater博士的一项合作研究。他们计划通过电子信号和人工智能技术研究感应电机驱动系统的在线故障诊断。目标是利用电子信号，如电机端子电压和电流，结合创新的信号处理诊断技术和现代人工智能方法，如神经网络和高斯混合模型，开发在线故障诊断技术。所研究的驱动系统是一种不需要速度传感装置的无传感器场定向控制（F.O.C）驱动系统。该技术能够诊断故障的发生并识别故障的严重程度。建模和分析算法将用于预测性能，并使诊断系统在驱动控制器中实现。在一台5马力的感应电机上进行的实验工作和对记录结果的分析将用于评估和验证所开发的故障诊断技术。智能优势：现代电机驱动系统（可调/变速驱动）的故障诊断问题是一个具有挑战性和技术难度的问题，特别是对于面向场（矢量控制）闭环的电机驱动系统。在这类电机驱动器中，每当电机出现故障或缺陷时，在驱动控制系统中就会发生相应的补偿动作。这种作用通常倾向于掩盖断层的存在，特别是在断层的早期阶段。本研究将解决这一补偿故障掩蔽问题，并开发新的诊断方法和手段来检测存在驱动控制补偿动作的此类故障。研究结果将对电机驱动故障检测和诊断技术的发展做出重大贡献。更广泛的影响：现代交流电机驱动系统广泛应用于整个制造，加工和服务行业，包括医疗领域的辅助和生命支持系统等关键服务。在此类驱动器中采用的复杂控制使得在故障演变为灾难性或严重设备故障之前诊断此类系统故障的任务变得复杂。在存在闭环控制的情况下，揭露此类故障将构成在实际应用中避免危险故障和相关后果的重大改进，这些应用越来越依赖于交流电机驱动系统。本文的研究结果对未来电驱动系统的在线故障诊断与检测系统的设计具有重要的指导意义。该项目涉及高水平的研究、教学和应用，通过美国和埃及的研究人员和学生的共同努力，在文化交流方面具有相当的广度。两名马凯特大学的研究生将参与该项目。该项目得到了美国-埃及联合基金项目的支持，该项目向两国的科学家和工程师提供赠款，以开展这些合作活动。

0609731DemerdashDescription: This award is to support a cooperative research by Dr. Nabeel Demerdash, Department of Electrical and Computer Engineering, Marquette University, Milwaukee, Wisconsin and Dr. Faeka Khater, Electronic Research Institute, Cairo, Egypt. They plan to investigate on-line fault diagnostics for induction motor drives-system through electronic signals and artificial intelligence techniques. The goal is development of online fault diagnostic techniques using electronic signals such as motor terminal voltages and currents in conjunction with innovative signal processing diagnostic techniques and modern artificial intelligence methods such as Neural Networks and Gaussian Mixture Models. The drive system to be studied is a sensor-less Field Oriented Controlled (F.O.C) one without the need for a speed sensing device. The proposed technique will diagnose failure occurrences and identify the severity of the fault. Modeling and analytical algorithms will be used to predict the performance and enable the implementation of the diagnostic system in the drive controller. Experimental work on a 5-hp induction machine and analysis of recorded results will be used to evaluate and validate the developed fault diagnostic techniques.Intellectual Merit: The problem of fault diagnostics in modern motor-drive systems (adjustable/variable speed drives) is a challenging and technically difficult one, especially if the motor-drive system is of the field oriented (vector control) closed-loop class. In such a class of motor-drives, whenever a fault or defect materializes in the motor, a corresponding compensating action takes place within the drive control system. This action usually tends to mask the existence of a fault particularly in its early stages. This research will address this compensating fault masking problem, and development of new diagnostic methods and means to detect such faults in the presence of drive control compensation actions. The results would constitute a major original contribution to the state of the art in motor-drive fault detection and diagnostics.Broader Impact: Modern AC motor-drive systems are extensively used throughout the manufacturing, processing, and service industries, including critical services such as auxiliaries and life support systems in the medical field. The sophistication of the controls employed in such drives complicates the task of diagnosing faults in such systems before faults evolve into catastrophic or serious equipment failure. The unmasking of such faults in the presence of closed-loop controls would constitute a significant improvement in avoiding dangerous failures and associated consequences in practical applications, which increasingly rely on AC motor-drive systems. The results of the proposed work could have a great impact on the design of future electric drive systems to have an online fault diagnosis and detection system. The project involves high-level research, teaching, and application, as well as considerable breadth in culture exchange through the joint effort by researchers and students from the US and Egypt. Two Marquette University graduate students will participate in the project.This project is being supported under the US-Egypt Joint Fund Program, which provides grants to scientists and engineers in both countries to carry out these cooperative activities.