Investigation on thermochemical energy network for efficient waste heat recovery

TL;DR

This study explores a thermochemical fluid-based energy network for recovering waste heat and managing thermal energy efficiently.

Contribution

The paper introduces an AI-based simulator and evaluates performance under various heating profiles and operating conditions.

Findings

Higher air flow rates significantly enhance total energy recovery with effectiveness up to 30%.

Increasing heating temperature improves moisture recovery and reduces sensitivity to L/G ratio variations.

The Gaussian heating profile provides the best water removal to heat supplied ratio at lower L/G ratios.

Abstract

The performance of a thermochemical fluid (TCF)-based energy network is investigated for waste heat recovery and sustainable thermal management. An experimental TCF energy network was developed and tested under three different waste heating profiles, i.e. Gaussian, steady, and regenerative thermal oxidiser (RTO), across a range of air and solution flow rates and regeneration temperatures. An artificial intelligence-based multi-layer perceptron simulator was also developed to map the TCF energy network performance. Results demonstrate that higher air flow rates significantly enhance total energy recovery across a wide range of solution flow rates, with potential energy recovery effectiveness reaching around 30%. Increasing the heating temperature significantly improves the moisture recovery performance of the TCF network, while simultaneously reducing the sensitivity of the network to variations in the liquid-to-gas flow rate (L/G) ratio. At higher regeneration temperatures, humidity ratio differences up to 4.3 g/kgda are achieved and the performance differences between L/G ratios become less pronounced. Across all profiles, the water removal to heat supplied (W/H ratio) decreases as the L/G ratio increases, indicating a consistent decline in performance at higher desiccant flow rates. The Gaussian heating profile offers the highest W/H ratio at lower L/G ratios compared to steady and RTO heating profiles. Further, the simulator demonstrates strong predictive accuracy for the TCF-based energy network at lower L/G ratios and under Gaussian and steady heating profiles, with low overall prediction errors. These findings provide essential insights for operating the TCF energy network, emphasising the importance of optimising working fluid operating conditions and regeneration temperatures.

Figures