This research aims to describe the hybrid algorithm's effectiveness in predicting and optimizing the abrasive water jet machining (AWJM) parameter on flexible sheep wool reinforced polyester composites. Selected five parameters are transverse speed (TS), water jet pressure (WJP), nozzle stand-off distance (NSoD), reinforcement weight percentage (wt%) and abrasive size (AS). In contrast, Surface Roughness (R_a) and Kerf Angle (K_a) are output performances. Multi objective optimization by ratio analysis (MOORA) is a tool is used for selecting and optimizing control variables. The most influential control variables are AS, WJP, TS, wt%, and NSoD, according to MOORA–Entropy feature selection results. The support vector machine algorithm (SVM) represents the AWJM process, and the model's performance is compared to SVM hybrid models. The differential evolutionary (DE) algorithm and the Entropy idea create a hybrid model. An SVM model is compared with the Hybrid SVM—Entropy model; hybrid improves prediction performance by 21.6%. When the MOORA—SVM—Entropy hybrid model is compared to the SVM model, it is revealed that the MOORA—SVM—Entropy hybrid model's prediction performance improves by 38.7%. According to the MOORA—Entropy approach, the optimal control variables are A₂, B₁, C₁, D₃, and E₁.

中文翻译：

---

一种新的羊毛增强聚酯复合材料的预测方法：表面质量和混合建模

本研究旨在描述混合算法在预测和优化柔性羊毛增强聚酯复合材料的磨料水射流加工 (AWJM) 参数方面的有效性。选择的五个参数是横向速度 (TS)、水射流压力 (WJP)、喷嘴间隔距离 (NSoD)、增强体重量百分比 (wt%) 和磨料尺寸 (AS)。相比之下，表面粗糙度 (R _a ) 和切口角 (K _a) 是输出性能。比率分析多目标优化 (MOORA) 是一种用于选择和优化控制变量的工具。根据 MOORA-Entropy 特征选择结果，影响最大的控制变量是 AS、WJP、TS、wt% 和 NSoD。支持向量机算法（SVM）代表AWJM过程，将模型的性能与SVM混合模型进行比较。差分进化（DE）算法和熵思想创建了一个混合模型。将 SVM 模型与 Hybrid SVM—Entropy 模型进行比较；hybrid 将预测性能提高了 21.6%。MOORA-SVM-Entropy 混合模型与 SVM 模型相比，MOORA-SVM-Entropy 混合模型的预测性能提高了 38.7%。根据 MOORA-Entropy 方法，最优控制变量为 A₂、 B ₁、 C ₁、 D ₃和 E ₁。