To address the shearer drum's cutting trajectory problems in complex coal-rock interfaces significantly affected by geological disturbances such as faults, folds, and collapse columns, which lead to path instability, inefficient cutting, and increased energy consumption. Current optimization methods including GA, PSO, and SA face challenges like local optima, poor path continuity, and low convergence efficiency, especially in highly disturbed regions.
This paper introduces a novel three-stage collaborative optimization method combining Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA) Employs a functionally decoupled architecture to achieve exploration of the search space, acceleration of convergence, and local refinement of cutting paths The three algorithms are coordinated through staged control and information sharing, enhancing stability and robustness The optimized paths are further smoothed using cubic spline interpolation to improve trajectory continuity and curvature control.
Simulation experiments demonstrate that the proposed approach effectively generates smooth, constraint-compliant paths under varying disturbance conditions, offering significant improvements in stability and adaptability.
This research addresses a critical engineering challenge in coal mining automation by developing an innovative hybrid optimization algorithm that significantly improves shearer cutting path planning in geologically complex environments. The novel three-stage GA-PSO-SA collaborative approach overcomes limitations of traditional single-algorithm methods, achieving superior convergence efficiency and path quality. By integrating functional decoupling architecture with cubic spline smoothing, the method ensures practical applicability while maintaining computational efficiency. This work contributes to advancing intelligent mining technologies and provides a robust framework for multi-objective optimization in constrained engineering systems, with potential applications extending beyond mining to other path-planning domains in complex environments.
