Artificial neural network-based optimization of heat absorption process of phase change materials in a novel-designed finned-plate latent heat storage system

Journal of Energy Storage(2024)

引用 0|浏览3
暂无评分
摘要
As the world transitions to more renewable and clean energy sources, energy storage systems become crucial. They assist in storing surplus energy produced during peak production times and release it when there is a high energy demand. Phase change material (PCM) based latent heat storage systems are appropriate candidates for efficient thermal energy storage. Their ability to harness the latent heat of phase transitions provides energy density, temperature control, and reduced heat loss advantages, making them well-suited for various applications. In the current research, a novel plate latent heat storage system (P-LHS) inspired by plate heat exchangers was introduced to enhance heat absorption. Besides, to facilitate the limited thermal conductivity of the PCM and increase natural convection phenomena, several fins were applied within the P-LHS system. Artificial neural network (ANN) models were exerted to anticipate the duration needed for melting of 70 % (φ = 0.7) and 100 % (φ = 1) of the PCM. The input variables of the ANN models were the fin angle (β), the distance between two fins (D), and the fin's trail angle (α). This research examined different configurations of the fins within the P-LHS system to find the optimal one. In all the designed Tests, the heat absorption and temperature distribution within the PCM were accelerated; thereby, the melting time up to φ = 0.7 and φ = 1 was reduced remarkably compared to those in the fin-less P-LHS system. Besides, for the fastest full melting of the material in the P-LHS system, the optimal design 2 (OD2) was proposed, in which the fins should have the dimensions of α = 171.450°, β = 97.540°, and D = 27.545 mm. In OD2, the time needed until the melting of 70 % of the PCM was decreased by approximately 65.68 %. While, the necessary duration for the PCM to melt entirely was 1574 s, approximately 66.80 % less than the time needed for melting the substance in the system without fins (4741 s). This reduction in melting time was about 53 min, which is a significant amount of time in energy storage systems.
更多
查看译文
关键词
Sustainable energy,Energy storage systems,Phase change material,Artificial neural network,Genetic algorithm,Heat absorption optimization
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要