Coal mining machine fault prediction method based on time sequence prediction method

Abstract

The invention discloses a coal mining machine fault prediction method based on a time sequence prediction method, and relates to the technical field of computer technology and fault prediction cross research; a support vector machine method in a nonlinear method is used, a predictor is trained according to a training sample, and a subsequent time sequence is predicted; according to the specific scheme, the method comprises the following steps: obtaining a penalty parameter C and a kernel function parameter, enabling a particle swarm algorithm to start to calculate from n random particles, eachparticle serves as a candidate solution, and each particle is an m-dimensional vector; the method also comprises the steps of space initialization, particle evaluation, particle updating and result verification. The particle swarm optimization algorithm is widely applied to solving the extreme value optimal problem, the particle swarm optimization algorithm calculates the fitness value accordingto all possible particles in the constraint range, and the corresponding extreme value optimal solution is finally obtained through continuous iteration.

Description

technical field [0001] The invention relates to the research technical field of the intersection of computer technology and fault prediction, more particularly, it relates to a shearer fault prediction method based on a time series prediction method. Background technique [0002] The problems of coal mine machinery failure have the problems of complex situation, high failure cost and difficult to identify effectively. For coal mining machinery, its operating state usually undergoes an evolution process from intact to faulty state. The state at this time is called an abnormal state. If the abnormal operation of the equipment can be found during this period, and timely measures can be taken to eliminate the fault in the bud, the occurrence of faults or accidents can be avoided. [0003] It can be seen from this that, compared with the fault diagnosis function, in a sense, state prediction is a function that is more urgently needed in field production. Therefore, it is neces...

AI Technical Summary

Problems solved by technology

However, these methods also have some technical problems to be solved, such as the lack of intelligent judgment of the working conditions of the coal mining machine; under the condition of too few negative samples, it is difficult to determine the steady state of the coal mining machine; Hard-to-sample learning, etc.

[0005] Time series prediction can be divided into linear methods and nonlinear methods. Linear methods need to establish corresponding models for specific situations, and have high requirements for data, which are rarely met in actual situations.

Images