Molecular Electrical Strength Prediction Method Based on Machine Learning

Sulfur hexafluoride (SF6) has excellent insulation properties and is widely used in the field of electrical equipment. However, SF6 exhibits an extremely strong greenhouse effect, with a global warming potential of 25,200 times that of carbon dioxide. In response to environmental concerns, finding alternative gases to SF6 is a crucial issue that needs to be addressed in the field of electrical power equipment. The use of machine learning algorithms in this context is instrumental in accelerating the development of eco-friendly alternatives to SF6. In this paper, we propose a method using machine learning and neural network techniques to predict the electrical strength of alternative gases. By establishing a neural network model, which learns the complex relationship between the known molecular descriptors and the electrical insulation strength, the electrical properties of various gases can be accurately predicted. The flexibility and versatility of this method can be used to guide the design of novel, environmentally friendly gases, and can quickly and efficiently screen a large number of molecular candidate materials with unknown insulation electrical strength.