Search

Traditional fault diagnosis methods have disadvantages such as incomplete coding and overly absolute coding boundaries, which are difficult to meet the actual needs of power grid operation and maintenance. Using the gas generated when a power transformer fails to diagnose the transformer fault is currently a popular research area for smart grid condition detection. However, the frequency of various types of faults in transformers varies greatly, which may result in incomplete fault sample information and insufficient data, and the traditional convolutional neural net- work models have problems such as unstable training process, low training accuracy and long time. Based on the transformer fault diagnosis technology of one dimensional convolutional neural network, this study proposes a new method of data enhancement while keeping the original data features unchanged, transforming expanded one dimensional data into two dimensional pictures to input into the two dimensional convolutional neural network diagnosis model, and improve the Adam optimization algorithm in the convolutional neural network model architecture. Diagnostic results indicate that the accuracy of network training reaches 96. 20%. At the same time, it has higher convergence speed and generalization ability than the traditional one dimensional convolutional neural network fault diagnosis method (92.12%). [ABSTRACT FROM AUTHOR]

The calculation of motor temperature field is an essential part of motor design. The temperature rise of motor is related to the safety and stability of the whole system and related equipment. According to the parameters of the motor, the electromagnetic model of the motor is established to calculate the loss, and the loss is loaded into the two-dimensional model in the form of heat source density to calculate the motor temperature field. In order to improve the accuracy of temperature rise calculation, the equivalent thermal conductivity method is used to process the heat exchange in the air gap of the motor according to the structure characteristics of the moto. The accuracy of the equivalent thermal conductivity method is verified by comparing the simulation results with the experimental data. Fluent software is adopted to solve the two-dimensional temperature field model in a unidirectional coupling way. Finally, the temperature field data obtained by the experiment is compared with the simulation calculation results to verify the accuracy of the two-dimensional model to calculate the steady-state temperature field of the experimental motor. [ABSTRACT FROM AUTHOR]

In view of the calculation of starting characteristics of small asynchronous motor in different starting modes, the two-dimensional electromagnetic model is constructed based on the finite element method to simulate the starting characteristics of the motor. Considering the effect of the motor end on the overall electromagnetic simulation in the two-dimensional model, the parameterization of the resistance and leakage inductance of the stator and rotor end are equivalently processed to the model by the method of field circuit combination. The power models of direct starting, Y-Δ step-down starting and constant frequency voltage ratio starting are established by MagNet and Simulink software according to the starting circuit structure. The starting time, starting current and starting speed of the motor under different starting modes are compared and analyzed. The results show that the model can accurately simulate the characteristics of asynchronous motors under common starting methods. [ABSTRACT FROM AUTHOR]