Operational Hybrid Neural Network Model for NO$_{x}$ Forecast and Control in Real-World 2-GW Coal-Fired Power Plant

This study presents the development and implementation of an advanced hybrid neural network (HNN) model for predicting nitrogen oxide (NO$_{x}$) emissions and controlling ammonia (NH$_{3}$) injection in a 1-GW generator within a 2-GW operational coal-fired power plant. The HNN model, which integrates both endogenous and exogenous input features to effectively analyze complex relationships, shows significant improvement in accuracy with a forecast skill of 22% compared to multiple benchmark models. The real-world application of the HNN-based control strategy resulted in a slight increase in average outlet NO$_{x}$ concentration but remained well within the regulated limit of 50 ppm, while reducing the standard deviation from 9.7 to 4.9 ppm, indicating a more stable and controlled outlet NO$_{x}$ concentration. The successful deployment of the HNN model in an operational power plant demonstrates its practical applicability and effectiveness in large-scale industrial settings, ultimately supporting the transition toward a sustainable energy future.

I. Introduction

As Global efforts progress towards sustainability, conventional energy sources continue to dominate most electricity markets. For instance, as of 2022, coal-fired power plants still account for 70% of China's electricity generation, with a total capacity of 1070 GW [1]. Consequently, even minor improvements in the efficiency and environmental performance of conventional power plants can significantly contribute to global efforts of sustainable development. From a sustainability standpoint, the outlet nitrogen oxides (NO, primarily NO and NO) concentration of selective catalytic reduction systems in coal-fired power plants is critically important, as it is subject to strict regulation by government environmental protection agencies [2]. In China, for example, the emission standard for NO concentration in flue gas is limited to 50 ppm [3]. Thus, improving the accuracy of outlet NO control is essential for compliance with these regulations. The accuracy of outlet NO control largely depends on the injection characteristics of total NH. However, the operational characteristics of NO concentration in the inlet flue gas present considerable challenges for achieving satisfactory control of total NH [4]. As a result, there is an urgent need to optimize total NH control strategies based on accurate NO forecasting methods, particularly in the context of ultra-low emissions regulations for coal-fired power plants.