论文部分内容阅读
本文使用MATLAB/Simulink建模软件与Qt类库,建立了基于径向基函数(RBF)神经网络的直升机模拟器发动机起动过程实时仿真模型。将模型连入直升机模拟器,利用试飞数据对该模型进行了验证。结果表明该模型可实时接收发动机控制开关的变化,实现两台发动机相互独立的开车过程的实时仿真;具有动态性好,精度高的优点,模型平均延迟时间仅为1.98ms,平均误差为0.028%,仿真精度满足高等级直升机模拟器发动机启动性能的鉴定要求。为后续自主研发完整的高性能模拟器发动机奠定基础。
In this paper, real-time simulation model of helicopter simulator engine starting process based on radial basis function (RBF) neural network is established by using MATLAB / Simulink modeling software and Qt class library. The model is connected to the helicopter simulator, and the model is validated by flight test data. The results show that the model can receive real-time changes of the engine control switch and realize the real-time simulation of the two engines running independently. The model has the advantages of good dynamics and high precision. The average delay time of the model is only 1.98ms and the average error is 0.028% , Simulation accuracy to meet the high level helicopter simulator engine start performance identification requirements. Lay the foundation for the subsequent independent research and development of a complete high-performance simulator engine.