TY  - JOUR
AU  - V Prathyusha
AU  - Bolishetti Ananya Varma
AU  - Rachel R
PY  - 2026
DA  - 2026/06/15
TI  - Fuzzy Logic in Predictive Models: A Study on Improving Accuracy in Energy Demand Forecasting
JO  - Global Journal of Engineering Innovations and Interdisciplinary Research
VL  - 6
IS  - 4
AB  - Accurate energy demand forecasting is essential for ensuring the efficient operation of modern power
systems, smart grids, and renewable energy infrastructures. Reliable prediction of future electricity
consumption enables utility providers to optimize power generation, reduce operational costs, improve
load balancing, and enhance energy distribution efficiency. However, energy demand is influenced by
numerous uncertain and nonlinear factors including weather conditions, seasonal variations, population
growth, economic activities, consumer behavior, and industrial operations. Conventional statistical
forecasting models often struggle to represent these uncertainties, resulting in reduced forecasting
accuracy and inefficient energy management. This paper presents a comprehensive study on the
application of Fuzzy Logic in Predictive Models for improving energy demand forecasting accuracy. The
proposed framework integrates fuzzy inference systems with machine learning techniques to effectively
model uncertain relationships among multiple energy consumption variables. Historical energy
consumption data, temperature, humidity, population density, time of day, and seasonal information
are processed using fuzzy membership functions and rule-based inference mechanisms to generate
accurate energy demand predictions. The framework combines fuzzy reasoning with predictive learning
algorithms to enhance forecasting performance under uncertain operating conditions. Experimental
evaluation was conducted using publicly available smart grid and electricity consumption datasets
collected from residential, commercial, and industrial sectors. Comparative analysis demonstrates
that the proposed fuzzy logic-based predictive model significantly improves forecasting accuracy while
reducing prediction error compared with conventional regression and machine learning techniques.
The proposed framework offers an intelligent solution for next-generation energy management systems
supporting sustainable power distribution and smart grid optimization.
SN  - 3066-1226
UR  - https://dx.doi.org/10.33425/3066-1226.1311
DO  - 10.33425/3066-1226.1311