To address the critical challenge of balancing high-fidelity representation with low-dimensional parameterization in turbomachinery blade design, this paper introduces a novel Modular Constructive Geometry Parameterization method. This method integrates three distinct camber line formulations-parabola, double cubic polynomials, and slope integral-with a thickness distribution governed by double cubic polynomials. A Blade Matching method is developed to establish correlations between characteristic parameters and discrete geometric data, validated through a 30-profile dataset, achieving an averaged mean squared error near \(10^{-5}\) . Hybrid optimization strategies, specifically combining Genetic Algorithm with Sequential Quadratic Programming and Genetic Algorithm with Particle Swarm, demonstrate superior performance compared to the standalone Genetic Algorithm strategy. Results indicate that the combination of Genetic Algorithm and Sequential Quadratic Programming excels for the slope integral camber, while Genetic Algorithm and Particle Swarm outperform for parabola and double cubic polynomial cambers. Furthermore, a comparative study against the classic clamped B-spline method validates the parameter efficiency of the proposed framework. Under identical low-dimensional constraints, the proposed method exhibits superior robustness for high-turning turbine profiles, reducing fitting errors by approximately 40%. Comprehensive analysis reveals that double cubic polynomial and slope integral camber types deliver higher fitting accuracy than parabola, particularly in high-camber-angle applications. A design criterion is established: slope integral camber prioritizes high-camber-angle profiles, double cubic polynomial camber offers parameter efficiency, while parabola camber proves non-optimal for performance-critical designs. This methodology enables experience-independent blade database construction with quantified characteristic parameters.
