Research On The Acquisition Of High Voltage Frequency Inverter Output Signal And Fault Detection System Based On SOC

With the development of the modern technology, the technology of variable frequency has been widely used, and the economic loss caused by the high voltage frequency converter fault has attracted people’s high attention. Because the high voltage frequency converters are various and complicated, it is hard to complete the task of an intelligent fault detection system by using the characteristics of amplitude spectrum and phase spectrum to detect the fault information, after transforming fault information in time domain to frequency domain by using the traditional Fourier transform. Although it has the Walsh transform and artificial neural network diagnosis algorithm to the high voltage frequency converter, if we use the traditional CPU to complete the fault detection of a high voltage inverter system as a result of the fault characteristics that the high voltage inverter has, not only it will have a great deal of effort, but also real-time tasks will be wanting.For the frequency converter, the main part is inverter circuit. So the thesis mainly describe the detection of fault information about inverter circuit of series and multilevel high voltage frequency converter.This thesis will finish the detection system of the high voltage frequency converter fault based on the SOC. With the aid of data processed by parallel, the characteristic of SOC, the aspect of processing the real-time task of high voltage frequency converter fault has certain guarantee. The work of this thesis include building the simulink simulation model about inverter circuit on series multi-level high voltage inverter in the matlab platform. In algorithm, firstly, by using one-dimensional, multi-scale wavelet transform, it process the wavelet decomposition about the three-phase current and three-phase voltage signal of the high voltage frequency converter. To identify the fault types, we need analyze the low frequency energy of the transformed voltage and current signals to find out the location of the fault, In addition, to confirm the feasibility of the algorithm, it is necessary to write a program using the matlab. The system hardware design and software design are completed eventually.Using xilinx XC7Z020 as the core processor, this thesis will design a high voltage frequency converter based on the SOC. The hardware system of fault detection mainly include power managing circuits, signal conditioning circuit, analog to digital converting circuit, signal processing circuit and signal transmitting circuit. In software, this thesis will design the ADC controller Intellectual Property core based on AXI-Stream interface, complete the Linux driver transplanted, program the Linux driver about the ADC controller IP core of AXI-Stream interface, display the waveform based on C# program language and design the software modules about the wavelet transform. At last the design of the whole system will be completed.