Turbomachines are essential for converting renewable, chemical, or potential energy into propulsion, mechanical force, or electrical power. As the demand for efficiency, reliability, compactness, and affordability increases, there is a need to improve the accuracy of design methodologies. These improvements are necessary to keep pace with the introduction of new materials, fluids, and fuels, as well as advances in manufacturing techniques and technologies. This chapter discusses recent advances in design strategies that leverage the combined strengths of artificial intelligence (AI) and precise simulation techniques. These strategies benefit from the integration of design information from a variety of sources and the improvement of design validation accuracy. These methodologies are effective in areas such as aerodynamic and thermal design, as well as the critical domains of materials and fluid mechanics. In the future, the integrated application of machine learning with detailed simulation processes could enable researchers to conduct reliable virtual experiments during the design stages, thereby reducing the time and risk associated with design.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Artificial Intelligence as a New Frontier in Turbomachinery Design and Applications

  • Ryoichi S. Amano,
  • Md. Tarif Raihan

摘要

Turbomachines are essential for converting renewable, chemical, or potential energy into propulsion, mechanical force, or electrical power. As the demand for efficiency, reliability, compactness, and affordability increases, there is a need to improve the accuracy of design methodologies. These improvements are necessary to keep pace with the introduction of new materials, fluids, and fuels, as well as advances in manufacturing techniques and technologies. This chapter discusses recent advances in design strategies that leverage the combined strengths of artificial intelligence (AI) and precise simulation techniques. These strategies benefit from the integration of design information from a variety of sources and the improvement of design validation accuracy. These methodologies are effective in areas such as aerodynamic and thermal design, as well as the critical domains of materials and fluid mechanics. In the future, the integrated application of machine learning with detailed simulation processes could enable researchers to conduct reliable virtual experiments during the design stages, thereby reducing the time and risk associated with design.