CAJ | 학술논문

冶金除尘风机广泛应用于冶金工业中，如果能及时检测诊断出除尘风机的故障，就能够很好地提高风机的安全性。针对除尘风机故障，提出一种基于差分入侵杂草算法优化支持向量机( DIWO-SVM)的模型。差分入侵杂草算法( DIWO)是一种具有较强鲁棒性、自适应性和随机性等特点的数值优化算法，用于支持向量机( SVM)的参数优化，能有效提高SVM分类准确率，以准确地识别除尘风机常见的故障。通过比较，验证了这种算法是可行的，并且具有很好的诊断能力。
야금제진풍궤엄범응용우야금공업중，여과능급시검측진단출제진풍궤적고장，취능구흔호지제고풍궤적안전성。침대제진풍궤고장，제출일충기우차분입침잡초산법우화지지향량궤( DIWO-SVM)적모형。차분입침잡초산법( DIWO)시일충구유교강로봉성、자괄응성화수궤성등특점적수치우화산법，용우지지향량궤( SVM)적삼수우화，능유효제고SVM분류준학솔，이준학지식별제진풍궤상견적고장。통과비교，험증료저충산법시가행적，병차구유흔호적진단능력。
The metallurgical dust removal air blowers are widely applied in metallurgical industry. To ensure the operation security of air blower, it is necessary to detect and diagnose the faults of the air blower timely. Aiming at the faults of the air blower, the model based on differential invasive weed optimization support vector machine ( DIWO-SVM ) is proposed. The DIWO algorithm is a numerical optimization algorithm which features strong robustness, good adaptability and stochastic performance;it is used to optimize the parameters of SVM, thus the classification accuracy of SVM can be effectively enhanced, to precisely identify the common faults of the dust removal air blower. The comparison verifies the feasibility and perfect diagnosis capability of this algorithm.