| The content of this dissertation is an important part of "Mine ventilation and power system security condition monitoring and fault diagnosis of early warning systems"(No: 2007BAK29B05). It is based on the current situation of the short operational life of HV motor for main fan domestically, low condition monitoring reliability, imperfect diagnosis function, and lack of fault warning method, etc. The ventilation safety directly determines the safety production situation of the coal mine. Being the core of mine ventilation, the main ventilation system guarantees normal flow of air under mine, and dilutes gas concentration to eliminate the hidden danger of the coal mine. It's safe, reliable, and efficient operation is the key factor for the safety of mine production. However, due to long operation cycle, complex working condition, and harsh operating environment, the main ventilation will suffer from a variety of hidden trouble. It has been proved by practice that of all the failures, a large amount lay in the high voltage (HV) motor that drives the main ventilation.The failures, which could be insulation aging, the stator winding temperature abnormal, and the rotor bar breaking and the air gap eccentricity, etc., could lower the operating characteristics as well as affect the reliability of the main fan. Therefore, the research of online monitoring for the HV motor of the main ventilation has very important significance in safety production of coal mine for the reason that it can prevent sudden failure of the main fan and eliminate hidden dangers threatening the ventilation equipment.Currently, the adoption of the online monitoring system for large and medium-sized coal mine main ventilation system domestically can in some way reduce the blindness in equipment maintenance and waste of human and material resources in advanced repair. The reliability of the main fan system could be enhanced. However, in view of the particularity and complexity of the main ventilation system, the current monitoring system of mine main fan could only complete the monitoring on common parameters such as air volume, air pressure, wind speed and the vibration of the fan, etc..The hidden troubles in safety cannot be removed for there is no way to detect the real reason for abnormal operation of the drive motor,through monitoring the original parameter such as voltage, current, power and temperature, etc. and judging from the threshold value. In addition,parameters indicating the fault such as vibrating frequency and the fingerprint distribution of partial discharges, as well as the influence the source and the load has on fault parameters are not being considered.Aiming at these problems, in this dissertation, the electrical fault mechanism and parameters of HV motors were analyzed. The fault model of HV motors was found and the simulation was finished. Secondly, on the basis of fault characteristic quantity of multi information fusion technology,the fault early warning and online monitoring system with perfect monitoring function and reliable monitoring data was developed and debugged. Finally, the platform of the online early warning system for HV motors has been set up and the research of fault sample machine has been made. The system has been adopted on site for operating experiment after debugging and can provide technical support for the reliability of the HV motor of main ventilation in the coal mine. The main content is as follows:The reflected fault types are to be different due to the different fault establish process of HV motors. The fault mechanism of HV motor is analyzed based on the common fault types of HV motors. The fault space of insulation deterioration and damage of the motor stator winding is established. Changing laws of characteristics of voltage, current and vibration signal caused by the rotor bar fault are studied, and the rotor unbalance fault is analyzed to confirm the contact between characteristics of rotor unbalance fault and rotor bar broken fault. The effects that the quality of power supply and the load effect have on the fault characteristic quantity are put forward and the target parameters of the state of HV motor are determined, which provides the theoretical bases for the fault early warning system for HV motor.The extraction and signal processing methods of fault data is the basis for the accurate fault early warning. In this dissertation, the technology of wavelet packet frequency analysis for the signal acquisition, together with FFT analysis, is adopted to extract the early fault feature of the signal. The fault diagnosis and early warning of high voltage motor of main fan is realized by means of fuzzy pattern recognition with intelligent neural network as classifier of fault diagnosis.The realization of early warning methods relies on the monitoring and control of the data as well as the data processing platform. The parameter acquisition module of monitoring system is developed by using sensors with excellent performance. The feasible monitoring and early warning scheme is put forward to carry out acquisition of PD, insulation resistance,voltage, current, and vibration signal and the data will be provided to the online monitoring and early warning system as analysis and judgment basis.The overall structure of the software is decided according to overall requirements for diagnosis of early warning system combined with the characteristics of the hardware. The signal acquisition and processing program, features extraction procedure, man-machine interface program,and fault diagnosis and early warning procedures have been designed.Finally, the experiment platform has been found and the software and hardware of the system is debugged, which has proved the early warning function of the system.The experiment research is effective means referring to verify the correction of the fault mechanism and early warning methods. Associated with manufacture process, the fault models are made according to the fault operation conditions of HV motor. Considering the status of the HV motor,the slot discharge model and the corona discharge model are made to detect the PD signal taken place in the slot and the surface of the end arm.Several motor bars at specific locations are taken out to simulate the fault motor and can provide the fault data for the fault diagnosis and early warning of the winding of the rotor. Different characteristics representing the two faults are searched through fault motor with imbalanced rotor and broke bar destroyed from a new motor. The influence of load to the fault characteristic of the motor is analyzed through loading experiment. With the optimal sensor and fault diagnosis and early warning system, the experiment data acquired can provide test evidence for the industrial operation test system.Based on the current condition of the HV laboratory, the characteristics of the online early monitoring system has been tested. In addition, the system has been installed in Sihe No.2 Coal Mine Shanxi Coal Co.. The test result has shown that the system is of real-time in monitoring with reliable diagnosis and early-warning results. |
PDF Full Text Request