Absorption-refrigeration-cycle-based exhaust gas heat recovery technology is effective in improving the thermal efficiency and fuel economy of marine diesel engines. However, the absorption refrigeration system is inflexible in the start-stop operation, and this cannot fulfil the fluctuating demand of refrigeration. This paper presents both the theoretical and experimental investigations of an absorption refrigeration and freezing pre-desalination-based marine engine exhaust gas heat recovery system. The energy storage subcycle is introduced to overcome the energy underutilisation and balance the excessive refrigerating output of the absorption refrigeration cycle. Seawater is utilised as the phase-change material and it is pre-desalinated in the energy storage subcycle. A mathematical model of the system is established and experimental investigation is conducted. Furthermore, the theoretical and experimental performances are compared, and an economic analysis of seawater desalination is performed to evaluate its economy. The results show that the total refrigeration output of the system ranges from 6.1 kW to 9.9 kW, and the system COP (Coefficient of Performance) can reach 16% under the experimental operating conditions. Additionally, the salinity of pre-desalinated seawater can be reduced to below 10 ppt. Moreover, the cost of RO (Reverse Osmosis) seawater desalination can be reduced by 26% through the predesalination process of seawater.

The output of the absorption refrigeration system driven by exhaust gas is unstable and the efficiency is low. Therefore, it is necessary to keep the performance of absorption refrigeration systems in a stable state. This will help predict the dynamic parameters of the system and thus control the output of the system. This paper presents a machine-learning algorithm for predicting the key parameters of an ammonia–water absorption refrigeration system. Three new machine-learning algorithms, Elman, BP neural network (BPNN), and extreme learning machine (ELM), are tested to predict the system parameters. The key control parameters of the system are predicted according to the exhaust gas parameters, and the cooling system is adjusted according to the predicted values to achieve the goal of stable cooling output. After comparison, the ELM algorithm has a fast learning speed, good generalization performance, and small test set error sum, so it is selected as the final optimal prediction algorithm.