Collaborative optimization of multi-stage machining parameters for compressor blades

The machining accuracy of compressor blades directly affects their aerodynamic performance, and optimizing machining parameters is a mainstream approach to improve their precision. Existing studies predominantly utilize a phased and isolated optimization approach, which fails to account for the collaborative optimization of multi-stage machining parameters, resulting in less-than-ideal outcomes. To tackle this limitation, this paper proposes a collaborative optimization method for multi-stage machining parameters of compressor blades that considers error propagation. Firstly, the maximum machining error in the rough machining stage is calculated by integrating the adaptive kernel density estimation function with the small probability event principle (AKDE-SPE). Secondly, a data-driven predictive model for multi-stage machining errors is developed, and a multi-objective optimization framework is established to minimize finish machining errors and total machining time. Subsequently, the model is solved using the multi-objective dung beetle optimization algorithm (MODBO), producing a high-quality Pareto front solution set. On this basis, to reduce reliance on manual experience, a combined weighting-TOPSIS method is proposed for optimal parameters selection. Finally, a case study is conducted to demonstrate the efficacy of the proposed method. Experimental results demonstrate that, compared to the traditional empirical method, the collaborative optimization approach reduces contour error by 28.88 %, position error by 28.29 %, torsion error by 7.26 %, and machining time by 6.05 %. In comparison with the independent optimization method, it reduces contour error by 8.12 %, position error by 2.68 %, torsion error by 0.86 %, and machining time by 6.39 %.