تشخيص خطا با استفاده از مدل هاي يادگيري ماشين در اينترنت اشياء صنعتي

Abstract The rapid integration of the Internet of Things (IoT) in industrial environments, referred to as the Industrial Internet of Things (IIoT), has significantly improved operational efficiency an‎d decision-making processes. However, the increasing complexity an‎d scale of IIoT systems pose significant challenges for fault detection, which is critical to maintaining system reliability an‎d performance.In this Research, we present a model for fault detection in Industrial Internet of Things (IIoT) systems using advanced machine learning techniques. This research focuses on the detection an‎d analysis of three major types of faults: spike faults, stack faults, an‎d bias faults. The proposed model uses a variety of sensors such as temperature, pressure, an‎d flow sensors to collect data. A hybrid model based on CNN-LSTM is used for fault detection. The model is trained on a training dataset an‎d then eva‎luated on an eva‎luation dataset. After eva‎luation, the model is optimized to improve its performance. The overall accuracy of the proposed model is estimated at 95%.The ultimate goal of this research is to improve fault detection capabilities in IIoT systems an‎d provide reliable an‎d timely warnings to ensure the integrity an‎d optimal performance of the systems. It is expected that the proposed model can have a positive impact on improving the efficiency an‎d ensuring the correct operation of IIoT systems an‎d be used as an efficient tool in various industries. This research seeks to take an effective step in the advancement of monitoring an‎d control technologies in the field of Industrial Internet of Things by introducing an efficient approach. Keywords: Internet Of Things, Edge computing, Machine learning, Fault detection, Artificial intelligence.

5 فصل