Optimisation of Industrial Oil-Flooded Screw Compressors: A Comparative Analysis of Conventional and Soft Computing Approaches

Screw compressors are widely used in industrial and commercial applications due to their high efficiency, reliability, and durability. However, there is still room for optimisation of such machines to improve energy efficiency and performance. This paper presents a comparative analysis of different optimisation methods for screw compressors, including conventional methods such as the simplex method and soft computing methods such as genetic algorithms, particle swarm optimisation, and Bayesian optimisation. The focus of the paper is on the development, application and evaluation of a novel Bayesian optimisation method for oil-flooded screw compressors. Optimisation targets are rotor profile parameters including the interlobe clearance, radial clearance, and axial clearance. The multi-objective function includes specific power consumption, adiabatic efficiency, volumetric efficiency, and flow. The results of the comparative analysis show that the Bayesian optimisation method outperforms other optimisation methods in terms of time and computational effort, while also achieving a better objective function using a weighted-sum approach. This provides a deeper understanding of the trade-offs between the different objective functions and enables engineers to make more informed decisions about the design and operation of oil-flooded screw compressors.