IoT devices and data availability optimization by ANN and KNN

Abstract

Abstract To improve the availability of IoT devices and data, research has been conducted on rapid prediction of instantaneous fault rates and temperatures. An IoT device and data availability optimization scheme based on artificial neural networks and K-nearest Neighbo drivers is proposed, using artificial neural network algorithms and K-nearest Neighbo driven neural network algorithms. The preliminary algorithm for achieving availability optimization is selected, and the objectives are divided into data optimization and device optimization. Applicable models are constructed separately, and the proposed optimization model is solved using the K-neighborhood driven neural network algorithm. The validation results showed that the proposed scheme reduced the maximum temperature to 2.0750 ℃ compared to the benchmark method, availability forward fault-tolerant method, and heuristic optimization algorithm. Compared with the first three methods, the improved method can improve the average availability of IoT devices by 27.03%, 15.76%, and 10.85%; The instantaneous fault rates of the three algorithms reached 100%, 87.89%, and 84.4%. This optimization algorithm has high efficiency in eliminating fault signals and optimizing the prediction of time limited satisfaction, and has strategic foresight in the decision plans of decision implementers.