基于改进黏菌算法与B样条曲线拟合的采煤机过断层截割路径优化方法

Cutting path optimization method for shearer fault crossing based on improved slime mold algorithm and B-spline curve fitting

  • 摘要: 传统采煤机截割路径优化方法在面对断层起伏变化大煤层时适应性较差,难以保障截割路径的连续性与稳定性。针对上述问题,提出了一种基于改进黏菌算法与B样条曲线拟合的采煤机过断层截割路径优化方法。首先,根据煤层地质模型获取断层特征信息,并结合B样条曲线规划出采煤机过断层截割轨迹控制点;然后,以采煤机截割轨迹与煤岩界面曲线之间的方差最小化为优化目标,利用引入人工蜂群搜索和邻域搜索策略的改进黏菌算法对B样条曲线规划的截割轨迹控制点进行优化,获取采煤机截割轨迹控制点的最佳权值和阈值;最后,结合过断层割岩率、路径平滑性、截割轨迹端点等约束条件进行B样条曲线光滑拟合,获得采煤机过断层最优截割路径。实验结果表明,与B样条−粒子群和B样条−蜉蝣方法相比,所提方法的截割路径优化结果在整体上能更好地接近实际煤岩界面,且均方根误差(RMSE)分别降低了46.87%和42.05%,决定系数(R2)分别提升了95.24%和60.78%,截割路径优化精度更高。

     

    Abstract: Traditional shearer cutting path optimization methods show poor adaptability when dealing with coal seams significantly affected by fault-induced undulations, and it is difficult to ensure the continuity and stability of the cutting path. To address this problem, a cutting path optimization method for shearer fault crossing based on improved slime mold algorithm and B-spline curve fitting was proposed. First, fault feature information was obtained based on the coal seam geological model, and the control points of the shearer cutting trajectory across faults were planned using B-spline curves. Then, by taking the minimization of the variance between the shearer cutting trajectory and the coal-rock interface curve as the optimization objective, the control points of the cutting trajectory planned by B-spline curves were optimized using an improved slime mold algorithm incorporating artificial bee colony search and neighborhood search strategies, and the optimal weights and thresholds of the control points were obtained. Finally, B-spline curve smoothing fitting was performed by incorporating constraints such as the rock-cutting rate during fault crossing, path smoothness, and cutting trajectory endpoints, and the optimal cutting path for the shearer across faults was obtained. The experimental results showed that, compared with the B-spline–particle swarm and B-spline–mayfly methods, the optimization results of the proposed method were overall closer to the actual coal–rock interface, the Root Mean Square Error (RMSE) decreased by 46.87% and 42.05%, respectively, and the Coefficient of Determination (R2) increased by 95.24% and 60.78%, respectively, indicating higher accuracy of cutting path optimization.

     

/

返回文章
返回