Predicting Power Generation from a Combined Cycle Power Plant Using Transformer Encoders with DNN

Abstract

With the development of the Smart Grid, accurate prediction of power generation is becoming an increasingly crucial task. The primary goal of this research is to create an efficient and reliable forecasting model to estimate the full-load power generation of a combined-cycle power plant (CCPP). The dataset used in this research is a subset of the publicly available UCI Machine Learning Repository. It contains 9568 items of data collected from a CCPP during its full load operation over a span of six years. To enhance the accuracy of power generation forecasting, a novel forecasting method based on Transformer encoders with deep neural networks (DNN) was proposed. The proposed model exploits the ability of the Transformer encoder to extract valuable information. Furthermore, bottleneck DNN blocks and residual connections are used in the DNN component. In this study, a series of experiments were conducted, and the performance of the proposed model was evaluated against other state-of-the-art machine learning models based on the CCPP dataset. The experimental results illustrated that using Transformer encoders along with DNN can considerably improve the accuracy of predicting CCPPs power generation (RMSE = 3.5370, MAE = 2.4033, MAPE = 0.5307%, and R2 = 0.9555).

Economic Dispatch of Combined Cycle Power Plant: A Mixed-Integer Programming Approach

“…Considering these factors is vital to prevent equipment damage, cut operating costs and enhance overall efficiency within integrated power systems. Below, we outline the contribution of this study, emphasizing the comparison with the results reported by the authors in [27,28,[32][33][34][35][36][37]:…”

“…The third approach is presented in Equation ( 5), which utilizes a piecewise method to depict the segments that form the cost curve. This representation differs from the mathematical models published in the literature [27,28,[32][33][34][35][36][37], we propose to incorporate a binary variable that simplifies the choice of a single operating state for the CCPP. It focuses exclusively on the two non-convex states of the CCPP, which enable a transition between them, and the binary variable facilitates its selection efficiently.…”