Abstract: Based on the analysis of the research status and development trend of fault diagnosis and protection theory for large electric motors, this paper discusses the conventional protection of electric motors based on overcurrent protection and the online monitoring and fault diagnosis of large electric motors based on the novel integrated protection method of symmetrical component method. A method for adaptive current protection using wavelet transform and artificial neural network (ANN) is proposed. This method fully utilizes the powerful time-frequency decomposition capability and excellent singularity detection capability of wavelet transform and the powerful adaptive, learning, and pattern recognition capabilities of artificial neural network to identify various fault conditions in electric motors. The network is trained, and the results show that the method is reliable, superior, and feasible. Keywords: Symmetrical component method; Wavelet transform; Adaptive; Artificial neural network; Fault diagnosis; Overcurrent Introduction Electric motors are the most widely used power equipment in various industries, consuming approximately 90% of the electrical energy in power supply systems. The value of a single large electric motor can reach millions of yuan. Due to outdated protection technology, their burn-out is severe, resulting in huge direct economic losses. Fault diagnosis and protection of large electric motors has long been a research hotspot and a difficult problem in academia and engineering. In the past decade or so, research in this field has mainly focused on two aspects: firstly, seeking breakthroughs in protection theory; and secondly, developing implementation methods. 1. Motor Protection Theory Based on Symmetrical Component Method According to the symmetrical component method theory, when an asymmetrical fault occurs, the motor current can be decomposed into positive-sequence, negative-sequence, and zero-sequence current components. The positive-sequence component reflects the degree of overcurrent in the motor, while the negative-sequence and zero-sequence components are absent or very small during normal operation. Therefore, monitoring the negative-sequence and zero-sequence current components to identify various asymmetrical faults should have high sensitivity and reliability. Common faults in high-voltage motors can be divided into two main categories: symmetrical and asymmetrical. Symmetrical faults include symmetrical overload, locked rotor, and short circuit; asymmetrical faults in motors are numerous, such as phase loss, unbalanced phase-to-phase short circuit, and ground fault. The main characteristic of a symmetrical fault in a motor is an increase in current amplitude, while the main characteristic of an asymmetrical fault is the appearance of negative-sequence and zero-sequence current components. Based on this conclusion, this paper decomposes the protection of high-voltage motors into three parts: overcurrent protection, negative-sequence current protection, and zero-sequence current protection. For details, please click: Integrated Motor Protection Based on Wavelet Transform and Neural Network