T. Agheb, I. Ahmadi, A. Zakariazadeh,
Volume 17, Issue 3 (9-2021)
Abstract
Optimal placement and sizing of distributed renewable energy resources (DER) in distribution networks can remarkably influence voltage profile improvement, amending of congestions, increasing the reliability and emission reduction. However, there is a challenge with renewable resources due to the intermittent nature of their output power. This paper presents a new viewpoint at the uncertainties associated with output powers of wind turbines and load demands by considering the correlation between them. In the proposed method, considering the simultaneous occurrence of real load demands and wind generation data, they are clustered by use of the k-means method. At first, the wind generation data are clustered in some levels, and then the associated load data of each generation level are clustered in several levels. The number of load levels in each generation level may differ from each other. By doing so the unrealistic generation-load scenarios are omitted from the process of wind turbine sizing and placement. Then, the optimum sizing and placement of distributed generation units aiming at loss reduction are carried out using the obtained generation-load scenarios. Integer-based Particle Swarm Optimization (IPSO) is used to solve the problem. The simulation result, which is carried out using MATLAB 2016 software, shows that the proposed approach causes to reduce annual energy losses more than the one in other methods. Moreover, the computational burden of the problem is decreased due to ignore some unrealistic scenarios of wind and load combinations.
Somayeh Talebzadeh, Reza Radfar, Abbas Toloei Ashlaghi,
Volume 21, Issue 3 (8-2025)
Abstract
The accurate prediction of electricity demand is crucial for efficient energy management and grid operation. However, the complexities of demand patterns, weather variability, and socioeconomic factors make it challenging to forecast demand with high accuracy. To address this challenge, this research proposes a novel hybrid machine-learning approach for predicting electricity demand. In this research, first, different regression methods were investigated to solve the problem, the results showed that the multi-layer perceptron (MLP) regression model has the best performance in predicting electricity demand. Furthermore, the proposed system, BIMLP (Bagging-Improved MLP), is designed to iteratively improve its parameters using a binary search algorithm and reduce the learning error using bagging, a technique for ensemble learning. The proposed system was applied to the Electric Power Consumption data set and achieved a value of 0.9734 in the r2 criterion. The results of implementing and evaluating the proposed system demonstrate its satisfactory performance compared to existing techniques.
