The on-load tapping transformer is one of the important equipment in the power system.It plays an important role in stabilizing the load center voltage,adjusting the reactive power flow and increasing the flexibility of the grid dispatching.The on-load tap changer(OLTC)is the only movable part in the on-load tap changer transformer.It relies on the cooperation between the operating mechanism components and the operation to complete the voltage adjustment action,because of its complex structure and operation process Often lead to electrical failures and mechanical failures during equipment operation.The mechanical performance of OLTC is directly related to the safe and reliable operation of the power system,and it is of great significance to study OLTC state recognition and mechanical fault diagnosis in depth.Based on the in-depth analysis of the basic structure,working principle,failure causes and types of OLTC,this paper selects the vibration and driving motor current parameters during the operation process as the research object.According to the characteristics of the signal,it focuses on the denoising processing and characteristics of multiple signals.Extraction and OLTC mechanical fault diagnosis method.First of all,combining the vibration signal generated by the OLTC contact shifting process and the driving motor current signal can characterize the characteristics of the mechanical state change,build an OLTC monitoring information collection system composed of multi-signal sensors,data acquisition cards and microcomputers,and design a state diagnosis plan.Secondly,use complementary set empirical mode decomposition(CEEMD)and wavelet to denoise the collected signals respectively.In view of the chaotic characteristics of the vibration signal,the phase space reconstruction theory is used to extract the dynamic characteristics of the vibration signal;the Hilbert transform is used to extract the characteristics of the time domain characteristics of the motor current signal envelope;The feature vector composed of electro-vibration data is input into the gravitational optimization learning vector quantization network(GSA-LVQ)for learning and training,and a mechanical fault diagnosis method of OLTC driving mechanism based on the joint analysis of electro-vibration signal characteristics is proposed.Then,for the OLTC switching vibration signal,CEEMD is used to remove the noise and the energy entropy characteristics of the inherent modal components are extracted,and a differential evolution and firework algorithm joint optimization support vector machine(DE-FWA-SVM)algorithm is proposed to realize the mechanical fault identification of the switching mechanism.Finally,the typical mechanical faults of the tap changer of the on-load tap-changing transformer are simulated,and the validity and feasibility of the OLTC mechanical state recognition algorithm given in this paper are verified through experimental data. |