Equipment evaluation method and system based on LSTM (Long Short Term Memory)

Abstract

The invention relates to an LSTM-based equipment evaluation method and system, and relates to the field of intelligent transportation. The method comprises the following steps: constructing an input feature vector of an LSTM model, wherein the input feature vector comprises state data of the equipment; constructing a training set, a verification set and a test set based on historical data; constructing an initial LSTM (Long Short Term Memory) model; training, verifying and testing the initial LSTM model, and generating a final LSTM model; and based on equipment state data acquired in real time, evaluating the equipment state and the abnormality type through the LSTM model. According to the equipment evaluation method and system based on the LSTM, the running state of the future equipment is predicted based on the historical state sequence data of the equipment, such as the current, the voltage, the chip temperature, the idle memory size, the fault time, the fault type and the network packet loss rate, so that the real-time performance and the accuracy of equipment state monitoring are improved.

Description

technical field [0001] This application relates to the field of intelligent transportation, in particular to an LSTM-based equipment evaluation method and system. Background technique [0002] In the vehicle-road coordination system, the roadside equipment RSU broadcasts MAP, SPAT, RSI, RSM and other information to the vehicle-mounted terminal OBU through PC5, and the real-time nature of road traffic and the demand for continuous and uninterrupted operation, RSU must maintain 7*24 hours uninterrupted operation. For such equipment with continuous operation services, the service status is generally monitored through the platform, and the service quality is evaluated. Existing RSU status detection methods generally poll the RSU status through cloud software, and when the RSU status is found to be abnormal, the operation and maintenance personnel are notified to perform maintenance in a timely manner. The above method achieves the RSU operation guarantee problem to a certain e...

AI Technical Summary

Problems solved by technology

However, the main problem in establishing an expert system and performing matching queries lies in the establishment of a rule base, including the establishment of a rule base that fully describes various abnormal situations, which is difficult to completely "exhaust" by expert manpower. The experience summed up by different experts is based on personal knowledge, experience There is a certain conflict in the difference, etc.

Images