Random resonant current weak information identification method based on linear searching strategy

Abstract

The invention provides a random resonant current weak information identification method based on a linear searching strategy. A signal-to-noise ratio is used as a condition for generating random resonance by a control system, and an advantage of maximizing the output signal-to-noise ratio of a noise strength control system is realized. Furthermore a maximal signal-to-noise ratio max(SNR) is used as a system optimization target function. Through estimating a weight function of a system structure parameter a and a noise strength D in each iteration process, and then correcting the weight function in an iteration process, the system finally outputs the maximal signal-to-noise ratio, thereby improving calculating capability in determining the system structure parameter through optimizing the signal-to-noise ratio (SNR), and realizing adaptive random resonance of the current signal. Because system searching difficulty and weight function complexity are reduced in an iteration process, calculation time is reduced and the method is particularly suitable for periodical signal detection of the current.

Description

Technical field [0001] The invention belongs to the technical field of motor fault diagnosis and monitoring, and specifically relates to a method for identifying weak information of stochastic resonance current based on a linear search strategy. Background technique [0002] As a driving device, the motor is widely used in various fields. Because its fault is related to the load, environment or other operating conditions changes and manufacturing defects, the detection and fault diagnosis of the operating state are complicated, which makes the traditional method to detect the motor. There are certain difficulties in the state. In recent years, a new technology that uses current signals as motor state detection has gradually been applied in the industrial field. Current detection can directly reflect the motor system information, with non-contact, high performance-to-noise ratio, and can directly reflect its transmission information. Although the current signal has obvious techni...

AI Technical Summary

Problems solved by technology

As one of the main methods to deal with weak signals, the stochastic resonance method mainly adds a certain intensity of noise to the nonlinear system to realize the stochastic resonance of the system when detecting weak signals, and finally obtains the characteristic signal, but this method cannot be directly applied In the current signal in the background of strong noise

Since the stochastic resonance method takes the single target of structural parameters as the optimization object, it ignores the influence of noise intensity on the system, thus affecting the measured results output by the stochastic resonance system

Images