基于支持向量机的飞机空中结冰严重程度识别

为了保证飞机的飞行安全,必须对飞机空中结冰的严重程度作出较准确的判断。针对飞机空中结冰状况的复杂性,提出将支持向量机与二分法相结合的飞机空中结冰严重程度识别的算法模型。仿真结果表明,虽然该训练样本较少且为多参量分类识别,但是由于建立了多支持向量机且采用二分法的概率抉择能找到最佳的建立支持向量机的分类方式,所以找到了最佳的分类方式,提高了分类准确率,而且可以较准确地识别飞机空中结冰的严重程度。可见该方法可以在训练样本较少的情况下对飞机空中结冰严重程度作出较好的识别效果。

Aircraft Air Icing Severity Recognition Based on Support Vector Machine

In order to guarantee the aircraft flight safety,must on the severity of the aircraft air icing a more accurate judgment.Aiming at the complexity of the aircraft air icing conditions,the aircraft air icing severity recognition algorithm model was built which mixed SVM（support vector machine） and method of bisection.The simulation results show that,although the training sample is not enough and there are many parameters,but due to not only one support vector machine and with dichotomy of probability choices can find the best build SVM classification approach,and found the perfect classification method,improve the classification accuracy,but also can accurately identify the aircraft air icing severity.So this method make good recognition effect of the aircraft air icing severity even there are not enough the training sample.