Akyol Sinem, Das Mehmet, Alatas Bilal
Software Engineering Department, Engineering Faculty, Firat University, Elazig 23279, Turkey.
Mechatronics Engineering Department, Engineering Faculty, Firat University, Elazig 23279, Turkey.
Biomimetics (Basel). 2023 Aug 30;8(5):397. doi: 10.3390/biomimetics8050397.
Refrigerant gases, an essential cooling system component, are used in different processes according to their thermophysical properties and energy consumption values. The low global warming potential and energy consumption values of refrigerant gases are primarily preferred in terms of use. Recently, studies on modeling properties such as compressor energy consumption, efficiency coefficient, exergy, and thermophysical properties of refrigerants in refrigeration systems with artificial intelligence methods has become increasingly common. In this study, a hybrid-optimization-based artificial intelligence classification method is applied for the first time to produce explainable, interpretable, and transparent models of compressor energy consumption in a vapor compression refrigeration system operating with R600a refrigerant gas. This methodological innovation obtains models that determine the energy consumption values of R600a gas according to the operating parameters. From these models, the operating conditions with the lowest energy consumption are automatically revealed. The innovative artificial intelligence method applied for the energy consumption value determines the system's energy consumption according to the operating temperatures and pressures of the evaporator and condenser unit. When the obtained energy consumption model results were compared with the experimental results, it was seen that it had an accuracy of 84.4%. From this explainable artificial intelligence method, which is applied for the first time in the field of refrigerant gas, the most suitable operating conditions that can be achieved based on the minimum, medium, and maximum energy consumption ranges of different refrigerant gases can be determined.
制冷剂气体是冷却系统的重要组成部分，根据其热物理性质和能耗值用于不同的过程。就使用而言，主要优先选择全球变暖潜值和能耗值较低的制冷剂气体。近年来，采用人工智能方法对制冷系统中制冷剂的压缩机能耗、效率系数、火用及热物理性质等特性进行建模的研究越来越普遍。在本研究中，首次应用基于混合优化的人工智能分类方法，为使用R600a制冷剂气体运行的蒸汽压缩制冷系统中的压缩机能耗生成可解释、可诠释且透明的模型。这一方法创新得到了根据运行参数确定R600a气体能耗值的模型。从这些模型中，能自动揭示出能耗最低的运行条件。应用于能耗值的创新人工智能方法根据蒸发器和冷凝器单元的运行温度和压力来确定系统的能耗。将获得的能耗模型结果与实验结果进行比较时，发现其准确率为84.4%。从这种首次应用于制冷剂气体领域的可解释人工智能方法中，可以确定基于不同制冷剂气体的最小、中等和最大能耗范围所能实现的最合适运行条件。
