Volume 19, Issue 4 (December 2023)
IJEEE 2023, 19(4): 101-116 |
Back to browse issues page
Abstract: (3403 Views)
This paper proposes a novel approach to analyzing and managing electricity consumption using a clustering algorithm and a high-accuracy classifier for smart meter data. The proposed method utilizes a multilayer perceptron neural network classifier optimized by an Imperialist Competitive Algorithm (ICA) called ICA-optimized MLP, and a CD Index based on Fuzzy c-means to optimally determine representative load curves. A case study involving a real dataset of residential smart meters is conducted to validate the effectiveness of the proposed method, and the results demonstrate that the ICA-optimized MLP method achieves an accuracy of 98.62%, outperforming other classification methods. This approach has the potential to improve energy efficiency and reduce costs in the power system, making it a promising solution for analyzing and managing electricity consumption.
Keywords: Smart Meter, Fuzzy C-Means, MLP Neural Network, ICA Algorithm, Residential Electricity Customers.
Type of Study: Research Paper |
Subject:
Artificial Intelligence Techniques
Received: 2023/04/08 | Revised: 2024/04/20 | Accepted: 2023/07/19
Received: 2023/04/08 | Revised: 2024/04/20 | Accepted: 2023/07/19
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |