19 / 2021-09-23 10:02:29

An imputation method based on dummy variable and unsupervised learning for electricity consumption data with missing values

electricity consumption, imputation method, dummy variable, unsupervised learning, missing values

AI-driven technology > Intelligent Anomaly Detection, Fault Diagnosis and Prognostics

Due  to the  fault  of smart  meters, system maintenance, data  storage, and other  reasons,  part  of  the electricity  consumption  data  is  missing,  which  brings  some difficulties to the behavior analysis of electricity consumption data. And deleting the missing samples will cause a loss of electricity information  and  detection  model accuracy. Aiming  at  these problems, this paper proposes an imputation method based on dummy variable  and unsupervised learning,  and impute the missing values of electricity consumption data without affecting the  quality of data.  First, the  dummy  variable  considers  the missing information mode in the users’ electricity consumption and saves the important missing information. Combined with the deep learning model to learn the potential feature representation of electricity  consumption data, and  effectively  realizes the complex  relationship  between  variables  through  nonlinear transformation. The generated network and discriminant network are employed to generate the missing values and discriminate the imputation values to reduce the model error, thus providing a great imputation model with missing information to estimate the missing values of electricity consumption  data. The root  mean square error (RMSE) of different imputation models based on different  datasets  and  different  missing  rates  verifies  that  the proposed  missing  imputation  model  can  more  accurately  and efficiently impute the missing values of electricity consumption data.