An AI-Based Fast Design Method for New Centrifugal Compressor Families

Limiting global warming’s effects requires a sudden reduction of greenhouse gas emissions to pursue net-zero carbon growth in the next decades. Along with this energy transition, drastic and rapid changes demand are expected many sectors, including one for centrifugal compressors. In context, new aerodynamic design processes exploiting know-how existing impeller families generate novel compressors could quickly react variations ensure companies’ success. Modifying characteristics using 1D single-zone model is fast way exploit know-how. Besides, artificial intelligence be useful highlight relationships between geometrical parameters performance, thus facilitating achievement optimized machines applications. Although scientific literature shows several studies on mono-dimensional approaches, joint use an neural network designing impellers from pre-engineered ones remains understudied. Such was provided paper. An application case study expander–compressor family derived other natural liquefaction presented. Results proved that proposed enabled developing one, improving performance while containing time computational efforts.