论文部分内容阅读
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。
In order to ensure the flight safety of the aircraft, a more accurate judgment on the severity of airborne icing on the aircraft must be made. In allusion to the complexity of airborne icing conditions, an algorithm model of aircraft icing severity identification based on support vector machine and dichotomy is proposed. The simulation results show that although the training samples are few and multi-parameter classification and identification, but because of the establishment of a multi-support vector machine and the dichotomy probability choice can find the best way to establish support vector machine classification, so find the best Classification, improve the classification accuracy, but also can more accurately identify the severity of airborne icing. It can be seen that this method can better recognize the severity of airborne icing in aircraft with less training samples.