基于SSA-TSVR的飞机状态预测方法研究

Research on Aircraft State Prediction Method Based on SSA-TSVR

为了构建地面飞行安全态势监测系统,针对飞机状态数据向地面传输过程中出现数据传输异常情况而导致无法对飞机状态进行实时监控的问题,提出一种基于SSA-TSVR的飞机状态预测方法,使用随机森林算法对真实飞行数据进行特征重要度分析,筛选与待预测飞机状态参数关系密切的重要参数,获得待预测参数与飞行数据间重要度关系;通过孪生支持向量回归算法建立预测模型,对缺失的关键飞行状态参数进行预测;并运用飞鼠搜索算法对孪生支持向量回归模型进行优化,根据不同预测对象选择对应的最优核函数,提高了模型预测精度;以飞行高度、速度为预测对象进行实验验证,预测模型实现了利用不完整飞行数据对飞机状态进行准确预测,对飞机飞行状态监测有着重要意义。

In order to build ground flight safety situation monitoring systems,solve the problem that abnormal data transmission occurs during the transmission process of aircraft status data to the ground,resulting in the inability to the real-time monitoring of aircraft status data,an aircraft status prediction method based on squirrel search algorithm and twin support vector regression(SSA-TSVR)was proposed.the random forest method was used to analyze the feature importance of real flight data,screen and predict the important parameters closely related to the aircraft s state,and obtain the importance relationship between the parameters to be predicted and the flight data.The twin support vector regression algorithm was used to build the prediction model,and predict the missing key flight state parameters.The twin support vector regression model was optimized by using the SSA,and the optimal kernel function was selected according to different prediction objects,improving the prediction accuracy of the model.With flight altitude and speed as prediction objects,the prediction model realizes the accurate prediction of aircraft state by using incomplete flight data,which is of great significance for aircraft flight state monitoring.