The fuzzy logic method evaluation of the motor conditio […]
The fuzzy logic method evaluation of the motor condition monitoring results, we all know that the three-phase asynchronous motor will have some faults, but what are the characteristics of the fault? ?
First, the characteristics and diagnosis method of three-phase asynchronous motor fault
The practice of daily maintenance and repair of electric motors proves that the main causes of electric faults are electromagnetic vibration, motor misalignment, rotor imbalance, bearing wear, poor lubrication, insufficient foundation rigidity and mechanical looseness.
Electrical faults of the motor can be judged by measuring the running current, motor insulation and winding DC resistance; mechanical faults often use vibration monitoring and diagnosis, shock pulse method, current diagnosis method and spectrum analysis method.
According to the motor vibration speed standard ISO2373, the position of the six measuring points of the motor, the characteristics of vibration acceleration, speed, displacement amplitude, and the function of using the instrument, determine the vibration monitoring item as the vibration amplitude, velocity value and acceleration at each monitoring point. The high-frequency values Hi and L cancel the measurement of the axial direction of the non-axial extension according to the structure of the small and medium-sized motor.
There are many uncertain factors in the operation of the motor, so there is a certain ambiguity in the fault diagnosis, which is mainly manifested in: (1) the ambiguity of the diagnostic parameters. The frequency range of the vibration parameters used for fault diagnosis has a certain ambiguity, the vibration values of different parameters, locations and directions exceed the standard, and the fault state reflected is ambiguous; (2) the ambiguity of the diagnostic method. The cause of motor failure is often not single. Faults such as bearing damage and loose foundation are often caused by electromagnetic vibration, misalignment or dynamic imbalance. Therefore, it is ambiguous to use which method or methods to comprehensively diagnose faults; 3) The ambiguity of the diagnostic criteria. Diagnostic criteria are the basis for fault diagnosis. Which standard is used for diagnosis is more accurate. When the operating conditions such as temperature and load change, how to correct the diagnostic criteria will affect the accuracy of the diagnosis and there is a large ambiguity.
Therefore, the fuzzy vibration diagnosis method using multi-parameter comprehensive analysis is in line with the actual situation of motor fault diagnosis, and can receive good results.
Second, the establishment of fuzzy diagnosis model
Since the diagnostic criteria were established for each motor, pattern recognition was performed using a direct method.
The multi-factor comprehensive evaluation fuzzy vector has the formula Y=R·X', where: the row vector of the fuzzy relation matrix R represents the vibration fault feature, the column vector represents the fault cause, and “·” represents the fuzzy logic operator.
Since the ordinary matrix multiplication method can avoid the loss of fault information, the fuzzy logic operator uses the rule of multiplication by ordinary matrix.
The membership function reflects the degree of likelihood of failure. The degree of membership of the symptom domain can be determined by a method similar to the evaluation score, that is, the monitored parameter values are separated from their corresponding fault diagnosis criteria, and appropriate corrections are made to determine the degree of membership. The method is simple and practical, and has been verified to meet the diagnostic requirements for common vibration faults of electric motors. Its calculation method is as shown in formula (1)
Where: Ux = membership degree;
Xi = the actual value of the fault characteristic parameter;
Sx = diagnostic criteria for the corresponding monitoring parameters of the diagnosed motor;
Mx = correction factor.
The use of expert knowledge to build a knowledge base is the key to fuzzy reasoning. The fault diagnosis fuzzy matrix reflects the relationship between the cause of the fault and the symptom of the fault. The reason and the fault are complicated. In order to find out the cause from the symptom, it is necessary to pre-determine the degree of correlation between the symptom and the cause, that is, the fuzzy diagnosis matrix.
By summarizing and analyzing the daily maintenance data of the motor and the corresponding state monitoring data, the fault type of all faulty motors that are monitored by the state is identified, and the membership value of each motor diagnostic standard is calculated as the standard membership degree. The corresponding membership degree of the monitoring parameters is greater than the number of the corresponding standard membership degree, and the percentage of each monitoring parameter is used as the weight coefficient between the symptom and the cause, thereby forming a fuzzy diagnosis matrix.