Nonlinear Data Classification of Power Internet of Things Considering Transient and Steady State

The fluctuating nonlinear data from the power Internet of Things results in inaccurate classification outcomes. To tackle this issue, a novel intelligent method for classifying nonlinear data in the power Internet of Things is proposed. Results of acquiring dynamic and stable-state nonlinear data from the power Internet of Things are obtained, and a threshold function is utilized to denoise the power load data. The disparity between the actual and observed values of the nonlinear data is calculated to determine the optimal classification attribute result. A standardization factor is determined, leading to the classification outcome of nonlinear data in the power Internet of Things. Experimental findings illustrate that the proposed approach yields robust classification results, with minimal average error rates of 3.38% and 1.26% for transient and stable-state scenarios, respectively. These high-accuracy results satisfactorily fulfill the requirements for classifying the data.