The topology identification of low-voltage distribution networks is an important foundation for the intelligence of low-voltage distribution networks. Its accuracy fundamentally determines the effectiveness of functions such as power system state estimation, operational control, optimization planning, and intelligent electricity consumption. The low-voltage distribution network is composed of transformers, lines, and end users. The key task of topology identification is to distinguish the connection relationship between distribution transformers, low-voltage lines, and phase sequence with end users, which can be divided into transformer user relationship, line user relationship, and phase user relationship. At present, the main methods of low-voltage network topology identification can be divided into signal injection method and data analysis method. The signal injection method requires a large number of additional terminal devices and is difficult to promote. The data analysis method combines the characteristics of switch state, voltage, current, electrical energy, and other data to perform topology analysis. The commonly used methods include correlation analysis and feature learning. Finally, typical problems that urgently need to be solved in topology recognition and representation were proposed, providing a reference for the research and development of low-voltage distribution network topology automatic recognition technology.

This study focuses on the maximum torque current ratio control of synchronous reluctance motors and proposes an optimized control method for the maximum torque current ratio of synchronous reluctance motors based on virtual signal injection. Firstly, the research on the maximum torque current ratio control of synchronous reluctance motors based on the virtual signal injection method is conducted, and the existing virtual unipolar square wave signal injection method is analyzed and studied. Secondly, a non-parametric maximum torque current ratio control strategy based on a synchronous reluctance motor combined with the virtual signal injection method is proposed. This strategy does not involve complex parameter calculations, and the control accuracy is not limited by the accuracy of the parameters in the model. The experimental results showed that under the control of virtual bipolar and unipolar square wave signal injection methods, the load torque was converted from 2 Nm to 6 Nm at t = 2:5 s, and there was a significant change in the current amplitude and waveform of the current vector. Under the control of the bipolar injection method, the current amplitude waveform of the motor was lower than that of the unipolar waveform, and the current was smaller. After the load suddenly changed, it could enter a stable state faster. After the load changed at t = 2:5 s, the phase angle of the current vector was quickly adjusted and stabilized under the control of the bipolar signal. The designed method has a good optimization effect compared to the traditional virtual signal injection method, and can achieve high-performance maximum torque current ratio optimization control on synchronous reluctance motors.