An optimized scheduling strategy combining prediction models and optimization algorithms to reduce electricity consumption costs for Liquefied natural gas regasification system

Manual-based peak-shaving operation has been adopted for Liquefied natural gas (LNG) regasification system to save electricity. However, this approach faces challenges of low efficiency and high randomness. In this research, it is tried to use intelligent optimal scheduling strategy in such a way that the electricity consumption costs are optimized. The proposed scheduling strategy classifies into two main parts. The first part employs the prediction models of both Kriging model and artificial neural network (ANN) for the forecast of electricity costs. The result shows that the Kriging model yields a lower relative error than that of ANN approach. The second part develops a novel improved geyser inspired algorithm (IGEA) with a chaotic mutation method, aiming to solve the electricity cost predictive model and obtain the optimal scheduling scheme. The proposed IGEA is evaluated on 18 benchmark functions, and is compared with other four popular optimization algorithms. Simulation results indicate the superior performance of the tent map on balancing the exploration and the exploitation. Furthermore, IGEA is implemented to solve the optimal scheduling problem of LNG regasification. The experimental results suggest that IGEA can benefit the scheduling scheme by reducing electricity costs of $4.63 \% \sim 15.81 \%$, compared with manual-based strategies. Therefore, we can conclude that IGEA is a promising and effective tool to solve the optimal scheduling problems and guide the manual operation for LNG regasification.