TLSCA-SVM Fault Diagnosis Optimization Method Based on Transfer Learning

An analog circuit fault diagnosis method using improved sparrow search algorithm and support vector machine

In analog circuits, component tolerances and circuit nonlinearity pose obstacles to fault diagnosis. To solve this problem, a soft fault diagnosis method based on Sparrow Search Algorithm (SSA) and Support Vector Machine (SVM) is used. In this study, ISSA is obtained by optimization using four strategies for SSA deficiency. Twenty-three benchmark functions are used for optimization experiments, and ISSA converges faster, more accurately, and with better robustness than other swarm intelligence algorithms. Finally, ISSA is used to optimize the SVM parameters and establish the ISSA-SVM fault diagnosis model. In the Sallen-key test circuit diagnosis experiments, the correct fault diagnosis rates of SSA-SVM and ISSA-SVM are 97.41% and 98.15%, respectively. The results show that the optimized ISSA-SVM model has a good analog circuit fault diagnosis with an increase in diagnostic accuracy.

A research report on the phase-domain trajectories of fault recording and their mathematical models

In currently known theories and algorithms of fault recording analysis and application, there is little literature where fault feature parameters are mined deeply from electrical physical quantities themselves. In this report the calculation method of the phase angles of the reference point and sample points is obtained, which ensures the correspondence of the time-domain waveform of digital fault recording with its phase-domain trajectory. The relationship between the initial phase angle of a sinusoid and the position of its trajectory, the characteristics of the trajectories of three-phase short-circuit currents containing dc components and the mathematical models of these trajectories are analyzed. Taking a rectangular wave current and a triangle wave current as example, the characteristics of the trajectories of non-sinusoidal and periodical waveforms containing harmonics are analyzed as well as their mathematical models. The research results show that the relationship between the initial phase angle of a sinusoid and the position of its circular trajectory is definite; the dc components have no impact on the positions of the trajectories of the short-circuit currents, but on their sizes and shapes; the harmonics have an impact on the shapes of the trajectories, and their positions are decided by the initial phase angles of the ac fundamental components in the waveforms. Subsequent study of the algorithm of fault recording analysis will be spread based on the contents in this report.

Study on the algorithm of fault recording analysis combining its time-domain waveforms with phase-domain trajectories

The untimely handling of faults in a power system has a negative impact on its operation and even the national economy, and this requires coordination in the functions of protective relaying as well as supervisory & control devices, where digital fault recorders are used to record fault waveforms of electrical physical quantities. The fault recording of a simulated current is taken as the research object in this article, and it is transformed from the time-domain waveform into a phase-domain trajectory, which is used to analyze fault feature parameters and then reformulate the waveform. The original waveform of the current will be substituted by the reformulated one with fault features to realize functions in the power system. The algorithm of reformulating fault recording, the correlativity of the reformulated waveform and its original one, and errors produced in the research process are researched. The high correlation coefficient between the reformulated waveform and its original one shows that the algorithm studied in the article offers a simple and convenient option for fault recording analysis.

Special Issue on “Modelling, Monitoring, Control and Optimization for Complex Industrial Processes”

Industrial automation systems, such as chemical processes, manufacturing processes, power networks, transportation systems, sustainable energy systems, wireless sensor networks, robotic systems, and biomedical systems, are becoming more complex [...]