Unsupervised Detection of Abnormal Electricity Consumption Behavior Based on Feature Engineering


In recent years, detecting anomalous electricity usage behavior has become increasingly important. Existing research, on the other hand, frequently concentrates on algorithm improvement while ignoring the process of obtaining characteristics. The best feature set, which reflects consumers' electricity usage patterns, has a big impact on the final detection results. Furthermore, obtaining datasets containing label information is difficult. This research proposes a feature engineering-based strategy for unsupervised identification of anomalous electricity consumption behavior. In the feature engineering process, the original feature set is first developed via brainstorming. The best feature set, which reflects the consumers' electricity usage habits, is then created by selecting features based on their variance and similarity. Then, in the abnormal detection step, a density-based clustering algorithm is utilized to detect abnormal electricity consumption behaviors, with the best clustering parameters obtained through iteration and assessment, paired with unsupervised clustering evaluation indexes. Finally, many conventional feature techniques are used to the load dataset of an industrial park in order to compare with the feature engineering described in this paper. The label information of abnormal behaviors is obtained by merging the original electricity consumption behavior detection findings with abnormal data injections in order to execute the evaluation. The proposed abnormal detection method has shown to be successful and generalizable, outperforming traditional feature strategies.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here

PROJECT TITLE : Using Cost-Sensitive Learning and Feature Selection Algorithms to Improve the Performance of Imbalanced Classification ABSTRACT: The problem of unbalanced data is common in network intrusion detection, spam filtering,
PROJECT TITLE : Financial Latent Dirichlet Allocation (FinLDA) Feature Extraction in Text and Data Mining for Financial Time Series Prediction ABSTRACT: Many financial time series predictions based on fundamental analysis have
PROJECT TITLE : OFS-NN An Effective Phishing Websites Detection Model Based on Optimal Feature Selection and Neural Network ABSTRACT: Phishing attacks have become a major menace to people's daily lives and social networks. Attackers
PROJECT TITLE : Robust Unsupervised Multi-view Feature Learning with Dynamic Graph ABSTRACT: By modeling the affinity associations with a graph to lower the dimension, graph-based multi-view feature learning algorithms learn a
PROJECT TITLE : A Grassmannian Graph Approach to Affine Invariant Feature Matching ABSTRACT: Feature matching in 2D and 3D for affine invariant 2D and 3D objects is a long-standing challenge in computer vision. There are two stages

Ready to Complete Your Academic MTech Project Work In Affordable Price ?

Project Enquiry