Читать книгу Design and Development of Efficient Energy Systems - Группа авторов - Страница 45

Compared to hardware equipment, sensors are getting simpler to collect data and less expensive. Whether using sounds, vibrations, images, electrical signals or accelerometer or other kinds of sensor data, richer analytics can be built by teaching a machine to detect and classify events happening in real time, at the edge, using an inexpensive microcontroller for processing—even with noisy, high variation data.

Sensor datas such as sounds, vibrations, images, electrical signals or accelerometer or other kinds of sensor datas are collected and trained by various classifiers. The classifier Support Vector Machine (SVM), Neural Networks (NN) and Convolution Neural Network (CNN) can be used to train the data.

The proposed system estimates the risk factor in sewer gas leakage. Hence it provides a cheaper solution than other existing mechanisms and is well suited for real-time applications.