Research on the strategy of locating abnormal data in internet of things management platform based on improved modified particle swarm optimization convolutional neural network algorithm

The Internet of Things (IOT) management platform is used to manage and transmit data from a variety terminal devices in the power system. In terms detecting abnormal data, existing IOT has low processing efficiency high error rate. addition, optimal selection determination structural parameters convolutional neural network (CNN) have substantial effect on its prediction performance. On this basis, paper proposes decision algorithm for locating anomalous an integrated using CNN global optimization key improved particle swarm (APSO) algorithm. Initially, index model developed identify whether obtained abnormal. Second, structure CNN-based anomaly detection approach investigated. Next, designed optimize CNN, APSO-CNN with higher performance localization constructed. Using Adam optimizer, accuracy, feasibility, established method were assessed. results demonstrate that APSO-CNN-based offers significant advantages precision execution speed, potentially intriguing application potential.