A machine learning-based early warning method and device for chemical industry parks

Abstract

The invention discloses a machine learning-based early warning method and device for a chemical industry park. By acquiring the historical sensor data of the target location, the historical sensor data is classified according to the early warning status, and finally the classified data are respectively input into the pre-established Mathematical models are trained to obtain predictive models. Since each historical sensor data is real and has a corresponding warning state, in addition, the model does not train a single data, so it can better reflect the correlation between the historical sensor data, Make the early warning model more accurate. After the prediction model is obtained, if the real-time data is input into the prediction model, the accuracy of the prediction result obtained will be higher. In addition, the calculation process of this method is relatively simple, and no manual calculation is required, which saves labor costs.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a machine learning-based early warning method and device for a chemical industry park. Background technique [0002] In order to achieve intelligent early warning, most of the current chemical parks have installed a large number of sensor devices, such as liquid level sensors, pressure sensors, temperature sensors, smoke sensors, etc., by comparing the data collected by these sensors with the threshold to determine whether to carry out early warning of accidents . [0003] Although the existing method can realize accident warning, the cause of the accident often includes many factors, and it is difficult to judge through a threshold, and the threshold is also set according to human experience. If the threshold is set too high, it will not be timely. Early warning of upcoming accidents; setting too low will bring false early warnings. Once the frequency of false early wa...

AI Technical Summary

Problems solved by technology

[0003] Although the existing method can realize accident warning, the cause of the accident often includes many factors, and it is difficult to judge through a threshold, and the threshold is also set according to human experience. If the threshold is set too high, it will not be timely. Early warning of upcoming accidents; setting too low will bring false early warnings. Once the frequency of false early warnings increases, it will reduce the awareness of monitoring personnel, and it will also lead to failure to take emergency measures in time when real accidents come.

Images