基于神经网络对电控柴油机运行特性影响因素仿真研究

基于AVL-BOOST软件仿真平台建立某船用四缸柴油机仿真模型,标定后的模型进行柴油机全工况仿真计算.仿真出来的3 200组数据作为人工神经网络输入数据,采用贝叶斯统计方法对网络进行训练建立2层的反馈神经网络仿真模型.并分别通过实验、AVL-BOOST和神经网络数据曲线的对比分析,验证人工神经网络预测的准确性.利用验证好的人工神经网络模型预测进排气压力对柴油机转矩的影响,以及预测压缩比和供油定时对柴油机排放性能和动力性能的影响,最后利用扰动法分析不同工况下柴油机各个参数对柴油机性能的影响程度.

Investigating the influence factors on performance of electronic controlled diesel engine using neural network

A four cylinder diesel engine simulation model was built based on AVL-BOOST software simulation platform,and the validated model was carried out to simulate the whole working condition of diesel engine.3 200 sets of data were simulated as input data of artificial neural network,and the two layer feedback neural network simulation model which was trained by Bayesian statistical method,was established the network.The correctness of neural network is verified by the comparison of experiments,AVL-BOOST and neural network data.The influence of intake and exhaust pressure on the diesel engine torque is predicted by the ANN model,and the effects of compression ratio and fuel injection timing on the emission performance and dynamic performance of the diesel engine are predicted.Finally,the relative importance of each parameter on different engine performance and emission characteristics are investigated using perturbation method and most influential parameters on different outputs are obtained.