لطفا منتظر بمانید ...
0% Complete
صفحه اصلی
/
یازدهمین کنفرانس بین المللی کنترل ، ابزار دقیق و اتوماسیون
A Comparative Analysis of Machine Learning Approaches for Electricity Consumption Forecasting: Performance Evaluation, Failure Analysis, and Hybrid Model Development
نویسندگان :
Beheshte Sadeghi Sabzevari
1
1- پژوهشگر آزاد
کلمات کلیدی :
Deep Learning،Electricity Consumption Forecasting
چکیده :
This study presents a comparative evaluation of nine forecasting models—Linear Regression, ARIMA, Prophet, XGBoost, CatBoost, LSTM, GRU, ConvLSTM, and LSTM+XGBoost—applied to Spain Electric Power Generation and Consumption dataset, spanning different geographical regions and time periods. The dataset was preprocessed to address missing values. Then systematic failure case analysis was conducted to identify limitations under various consumption patterns. Results demonstrate that the CatBoost model achieves superior performance (RMSE=386.40, R²=0.9927). Detailed computational efficiency analysis reveals trade-offs between accuracy and execution time in different models. We identify specific scenarios where deep learning models fail, including sudden consumption changes and low-variability periods, and propose targeted improvements. This work provides actionable guidance for practitioners in smart grid applications, highlighting model selection criteria based on deployment constraints, data characteristics, and performance requirements.
لیست مقالات
لیست مقالات بایگانی شده
Virtual Synchronous Generator Controller for Power Quality Enhancement in Microgrids
Delnia Hosseini - Mobin Naderi - Qobad Shafiee - Mehdi Savaghebi - Hassan Bevrani
AI-BASED ROBUST LINEAR QUADRATIC REGULATOR CONTROLLER FOR QUADCOPTER SYSTEM
Ali Taleb obud alisari - Shoorangiz Shams Shamsabad Farahani - Roohollah Barzamini - Mohammad Arabian
مدلسازی و کنترل غیرخطی تکنیک ضربه پرسرعت با جلوی راکت در ورزش بدمینتون
سروناز نیساری - رامین وطن خواه - سید آرش حق پناه
شبیه سازی بازوی 5urfd در متلب و v-rep
رضا جلالی رومی - علیرضا نقی پور - فراز جباری ایلخچی
Multi-Derivative Portfolio Management with Data-Driven Nonlinear Stochastic Control Policies in Retail Energy Markets
Mahdi Rezaei Bahrmand - Hamid Khaloozadeh
Design and Implementation of a ROS-Based Intelligent Surveillance System with Real-Time Facial Recognition and Data Logging
Ali Godarzvand chegini - Mohammad Arabian - Mahdi Siavash
UAV Lateral Near-optimal Control Based on Model Reduction
Saman Zaker - Hassan Zarabadipour
Synthetic to Real Framework based on Convolutional Multi-Head Attention and Hybrid Domain Alignment
Mohammadreza Ghorvei - Mohammadreza Kavianpour - Mohammad TH Beheshti - Amin Ramezani
Controller Education: Learning with the Help of Control Characters
Amirhosein Mansouri - Roghayeh Gavagsaz-Ghoachani - Matheepot Phattanasak
A Comparative Study of 4G and 5G Impact on the Performance of Networked Control Systems
Ehsan Badfar - Babak Tavassoli
بیشتر
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 44.9.5