An early gas leakage intelligent visual early warning system

Abstract

The invention relates to an early gas leakage intelligent visual early warning system, including a detected area, a data collection and processing module, and a central control room; the data collection and processing module is responsible for monitoring video real-time storage and leakage detection and positioning, and the central control room is responsible for organic gas leakage display and Early warning; the organic gas leakage detection and positioning module used in the present invention belongs to the deep learning hybrid model, and the trained module is embedded into the embedded information processor to realize the automatic detection of organic gas leakage in the monitoring video. The model used in the present invention includes an unsupervised self-encoder model and a supervised target recognition model. The former avoids the problem that supervised model training is difficult to collect and label data sets, and the training set required by the latter is only for the data set at the source of leakage. , this type of data set is easy to collect and label, which ensures a high detection rate and low false alarm rate for the detection and positioning of organic gases, and provides security for the petrochemical field.

Description

technical field [0001] The invention relates to an early gas leakage intelligent visual early warning system, which belongs to the technical field of organic gas leakage monitoring. Background technique [0002] In petrochemical and other process industries, the leakage of organic gas will cause safety accidents such as explosion, poisoning and environmental pollution. Therefore, early detection of on-site organic gas leakage should be carried out. In terms of organic gas leakage monitoring, traditional sensors have a high rate of false alarms when detecting gas leakage, are greatly affected by the site environment, and cannot accurately determine the location of the leakage, so there is a high risk. Compared with the sensor monitoring method, the infrared camera monitoring method can realize the visual monitoring of organic gas leakage, but it needs to manually observe the real-time monitoring video and judge whether organic gas leakage occurs. , Misreporting situation. ...

AI Technical Summary

Problems solved by technology

However, the network model constructed by deep learning requires a large amount of abnormally labeled data for training. Organic gas leakage is a rare event. In reality, it is difficult to collect a large amount of labeled data that characterizes the abnormal characteristics of organic gas leakage. Detects data similar to leaked gas anomalies as leaked gas anomalies, leading to abnormal false alarms in the system

The network model based on unsupervised deep learning theory can characterize the abnormal spatio-temporal characteristics of non-leakage gas without a large amount of anomaly labeling data, so as to overcome the defects of deep supervised learning detection model, but the model cannot intuitively give the abnormal location of leaking gas

Images