Neural Network Based Situation Detection and Service Provision in the Environment of IoT

Abstract

The safety production in coal mine has attracted considerable research attentions due to the frequently occurred mining accidents. In order to ensure the safety production in coal mine, technology of the Internet of Things (IoT) is widely used to detect the situation in coal mine. Here the situation is composed of several elements. Existing solutions for such situation detection are mainly based on the directed graph or automatic machine. These methods are only effective when few situation element change simultaneously or the change(s) can be determined clearly. However, when the situation comprises a lot of elements or the element's change is ambiguous, these methods cannot effectively determine the situation. In this paper, we propose a situation detection method based on neural network. Trained neural network can detect the situation well, especially when multiple situation elements change at the same time or changes are ambiguous.