A method for predicting the remaining mechanical life of a universal circuit breaker contact system based on deep learning

A technology of contact system and mechanical life, which is applied in the direction of circuit breaker testing, machine/structural component testing, instruments, etc., and can solve the problems of undeveloped contact system life prediction, low prediction accuracy, and lack of generalization ability , to achieve the effect of enriching the mechanical properties

CN113125095BActive Publication Date: 2022-03-22HEBEI UNIV OF TECH
6 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2022-03-22

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention is a method for predicting the remaining mechanical life of a contact system of a universal circuit breaker based on deep learning. This method uses the deep learning method to carry out life prediction research. Firstly, the concept of effective segment of vibration signal for mechanical life prediction is proposed; secondly, VMD algorithm and double threshold based on short-term energy are introduced to automatically calibrate its interval; again, multi-channel convolutional self-encoding is constructed. Machine network (MCCAE), trained by unsupervised learning method, extracts deep degraded temporal features of effective segments; finally builds long short-term memory neural network (LSTM), takes temporal features as input, and uses supervised learning training method to complete prediction and other steps, The remaining mechanical life prediction of the universal circuit breaker contact system can be effectively completed.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The technical solution of the present invention relates to the technical field of forecasting the remaining life of a contact system of a circuit breaker, specifically a method for predicting the remaining mechanical life of a contact system of a universal circuit breaker based on deep learning. Background technique

[0002] As the end of the entire power system, the low-voltage power distribution system is the link closest to the user, and is inseparable from the safety of people's lives and the stable operation of society. As the key power equipment of the low-voltage power distribution system, the universal circuit breaker, on the one hand, as the dispatching control equipment of the power system, implements the input or removal of specific lines according to the operation needs of the power grid, and on the other hand, it plays a protective role in the power system. In the event of a short-circuit fault in the system, the universal circuit breaker ...

Examples

Embodiment 1

[0191] In this embodiment, the contact system installed on the DW15-1600 universal circuit breaker is used as the test object. As the key actuator for circuit breaker opening and closing, the contact system is mainly composed of main contacts and arc contacts. The closing sequence is first arc and then main. In the specific implementation mode, the effective mechanical life vibration signal fragment of the contact system is obtained. And use it as an input to build a life prediction model. This section verifies the validity of the above theory.

[0192] The residual life prediction method of the universal circuit breaker contact system based on deep learning is used to predict the remaining life of the universal circuit breaker contact system. The specific steps are as follows:

[0193] The first step is to collect data. Use the universal circuit breaker contact system life test system to collect vibration signals, contact status signals and closing accessory current signals...