Numerical optimisation of the diffuser in a typical turbocharger compressor using the adjoint method

In the automotive industry, the demand for fuel economy and emission reduction has resulted in engine downsizing, with turbochargers playing a key role in compensating for the performance loss. To be effective, a turbocharger’s compressor must be accurately designed to match the engine’s requirements. This study presents a novel non-parametric optimisation of the turbocharger compressor diffuser based on the compressor efficiency. The numerical models are based on the validation and mesh dependency study against experimental data from three points on each speed line of 150,000 (rpm) and 80,000 (rpm). The geometry and case data are related to the TD025-05T4 compressor from the 1.2-L Renault Megane passenger car. The turbocharger compressor diffuser geometry was optimised using the adjoint solver method within ANSYS FLUENT 2019 R1. The adjoint solver provides a gradient-based optimisation that can automatically create a series of iterations of a design, so that the mesh gradually deforms into an optimal shape to achieve a single target, the compressor efficiency in this study. The study considers a total of six operating cases on the compressor map to optimise the full and partial load compressor operations, leading to a real-world drive cycle. These cases are the three cases (closer to surge, stable midpoint, and closer to the choke point) on each of the speed lines. A typical result for mid-stable operation on a 150,000 (rpm) speed line shows a gradual increase in efficiency up to a maximum of 2.6% improvement. The optimal diffuser geometry impacts the overall car engine efficiency for real-world drive cycles, increasing power output and improving thermal efficiency.