关键词: 焊接碳排放, GRNN网络, 遗传算法, 参数选择

Abstract:

The carbon emission function of CO2 shielded welding was proposed based on the analysis of its carbon emissions behavior by synthesizing the material carbon emission,energy carbon emission and process carbon emission.In order to optimize its carbon emissions while guaranteeing its cost and quality,a comprehensive evaluation model of carbon emission performance was then established by taking advantage of the GRNN to approximate the relationship between welding parameters and the carbon emission,quality and cost.Furthermore,genetic algorithm was applied to resolve this model.Finally,the model was validated by the investigation of welding process of loader fuel tank and the results show that the model can reduce the carbon emissions while guaranteeing the quality and cost welding.

Key words: carbon emission of welding process, general regression neural network(GRNN), genetic algorithm, parameter selection