Linear Regression and Gradient Descent Method for Electricity Output Power Prediction

• 作者: Yuanliang Liao
• 来源期刊: 电脑和通信(英文)
• 学科: 经济
• 关键词: Machine Learning LINEAR ALGEBRA LINEAR Regression GRADIENT DESCENT Error Analysis
• 年，卷（期）: 2019,（12）
• 页码范围: 31-36
• 页数: 6页
• 分类号: F42

摘要

Regulating the power output for a power plant as demand for electricity fluctuates throughout the day is important for both economic purpose and the safety of the generator. In this work, gradient descent method together with regularization is investigated to study the electricity output related to vacuum level and temperature in the turbine. Ninety percent of the data was used to train the regression parameters while the remaining ten percent was used for validation. Final results showed that 99% accuracy could be obtained with this method. This opens a new window for electricity output prediction for power plants.