Computer State Monitoring and Fault Intelligent Identification of Monitoring Electronic Equipment in Highway Environment

Abstract

ABSTRACT The monitoring system can improve the effective use of the expressway. It is a necessary part of ensuring the expressway to achieve high-speed, safety and comfort functions. It is also a necessary means to ensure the normal operation of the expressway. The construction of the monitoring system directly affects the operation management system and operation management mode. This article analyses the reasonable construction and function of the monitoring management system, expounds the reasonable selection and layout of vehicle detectors and cameras, and conducts in-depth research and analysis on the layout of variable information boards and variable speed limit signs. Aiming at the problem of the difference in the distribution of monitoring data caused by the fluctuation of the equipment working condition, which leads to the failure of the original diagnosis model and the reduction of the fault intelligent recognition rate, a deep domain adaptive method is proposed. Through deep feature mapping, interference factors of working conditions are eliminated, and insensitive features of working conditions are excavated. In addition, a method for determining the parameters of a deep mapping network is proposed to achieve the best domain adaptation effect. The test results show that, compared with artificial features, the representation features of deep network mining are more robust to data fluctuations; compared with shallow transfer learning, deep networks have stronger domain adaptability and cross-working fault diagnosis correct rate is also higher.