Customer Information System Data Pre-Processing with Feature Selection Techniques for Non-Technical Losses Prediction in an Electricity Market

Abstract

Non-technical losses (NTL) identification and prediction are important tasks for many utilities. Data from customer information system (CIS) can be used for NTL analysis. However, in order to accurately and efficiently perform NTL analysis, the original data from CIS need to be pre-processed before any detailed NTL analysis can be carried out. In this paper, we propose a feature selection based method for CIS data pre-processing in order to extract the most relevant information for further analysis such as clustering and classifications. By removing irrelevant and redundant features, feature selection is an essential step in data mining process in finding optimal subset of features to improve the quality of result by giving faster time processing, higher accuracy and simpler results with fewer features. Detailed feature selection analysis is presented in the paper. Both time-domain and load shape data are compared based on the accuracy, consistency and statistical dependencies between features.

Power Utility Nontechnical Loss Analysis With Extreme Learning Machine Method

“…The approach proposed in the present paper extends the research work of [15] to include more advanced data-mining techniques that deliver improved accuracy and efficiency. An important step in this proposed NTL detection framework is to classify the customer behavior data into two classes, namely those with typical load profiles and those with untypical load profiles that are highly correlated with potential NTL activities.…”

What is the importance of selecting features for non-technical losses identification?

“…As far as we know, only Nizar et al [4] proposed a study to select a subset of samples in order to make the classif cation more accurate. In this direction, we also believe that to identify the features that best describe possible illegal consumers is as much important as to recognize them.…”

NTL detection of electricity theft and abnormalities for large power consumers In TNB Malaysia

“…Such losses mainly occur due to meter tampering, meter malfunction, illegal connections and billing irregularities [2].The problem of NTLs is not only faced by the least developed countries in the Asian and African regions, but also by developed countries such as the United States of America and the United Kingdom. Specifically, high rates of NTL activities have been reported in the majority of developing countries in the Association of South East Asian Nations (ASEAN) group, which include Malaysia, Indonesia, Thailand, Myanmar and Vietnam [1].…”