Transformer-based Drum-level Prediction in a Boiler Plant with Delayed Relations among Multivariates

TL;DR

This paper explores the use of Transformer models to accurately predict steam drum water levels in boilers, addressing complex delays and interrelations to improve safety and efficiency in power plant operations.

Contribution

It introduces a novel Transformer-based predictive framework tailored for boiler water level forecasting, incorporating delay inference and variable augmentation techniques.

Findings

Transformer models outperform traditional methods in prediction accuracy.

The proposed pipeline effectively captures delayed relations among variables.

Enhanced prediction stability supports proactive plant control.

Abstract

The steam drum water level is a critical parameter that directly impacts the safety and efficiency of power plant operations. However, predicting the drum water level in boilers is challenging due to complex non-linear process dynamics originating from long-time delays and interrelations, as well as measurement noise. This paper investigates the application of Transformer-based models for predicting drum water levels in a steam boiler plant. Leveraging the capabilities of Transformer architectures, this study aims to develop an accurate and robust predictive framework to anticipate water level fluctuations and facilitate proactive control strategies. To this end, a prudent pipeline is proposed, including 1) data preprocess, 2) causal relation analysis, 3) delay inference, 4) variable augmentation, and 5) prediction. Through extensive experimentation and analysis, the effectiveness of Transformer-based approaches in steam drum water level prediction is evaluated, highlighting their potential to enhance operational stability and optimize plant performance.