Font Size: a A A

The Research On Cycloconverter Intelligent Fault Diagnosis

Posted on:2006-02-19Degree:MasterType:Thesis
Country:ChinaCandidate:J YuFull Text:PDF
GTID:2132360155472239Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
The character of Cycloconverter is without DC link, small volume and so on. Then cycloconverter is widely used in the condition of low rotate speed and high capacity, as it's used in the drive system of rolling mill. Due to the important function of cycloconverter in the system, once the fault of cycloconverte happens, if the position and reason of the fault can't be found in time, it will induce great economic loss. At present the intelligent fault diagnosis for frequency converter is not perfectly achieved, and the researches on AC-DC-AC converter fault diagnosis are more, then the research is very essential and practical to set up the cycloconverter fault diagnosis system. The thesis associates the actual condition of steel industry in country, summarizes the composing and frame of the system, emphasizes the analysis of work method of cycloconverter in sychchro-motor speed governing system, and introduces the common cycloconverter faults. Aiming at the actual work condition, the speed governing system is simulated by MATLAB. Based on the simulation model, the common fault is simulated, for example SCR short or open condition, then the experiment data of three phase output current is got and transformed by FFT. The transformed data is taken as the basic fault diagnosis information. By studying and analyzing the various Neural Network and intelligent fault diagnosis, associating the difficulty which is to measure the interior signal of cycloconverter, the thesis puts forward the cycloconverter fault diagnosis which bases on Neural Network and takes cycloconverter input voltage and output current as the input of NN. According to the self-learning and non-linearity of NN, it can confirm the position of SCR fault. At last aiming at the shortage of BP NN, the thesis introduces the recurrent neural network with bias(RNNB). By comparison, the good generalization and fitting of RNNB are proved. The simulation experiment and research indicate that the method can quickly diagnose the fault and confirm the position of SCR fault, efficiently reduce the time of searching the fault, enhancing the reliability of cycloconverter, assure the good production, put forward a new idea of research on cycloconverter intelligent fault diagnosis.
Keywords/Search Tags:Cycloconverter, Simulation, Fault Diagnosis, RNNB
PDF Full Text Request
Related items