论文部分内容阅读
高压共轨柴油机可控燃油喷射参数的增加,在使对燃烧的控制更加灵活的同时也带来标定和优化工作量显著增加的问题。为适应高效率的需要,提出并研究了基于模型的标定优化,即采用神经网络在一些工况点上建立模型,再通过自适应神经模糊推理系统(ANFIS)进行插值,将模型由这些工况点扩展到所需工况空间。模型精度由对象、建模所用数据量及模型参数调整共同决定。试验在一台六缸高压共轨柴油机上进行。理论分析和试验结果表明:该方法可以在保证精度的同时有效减少标定优化的试验工作量。
The addition of controllable fuel injection parameters for high-pressure common-rail diesel engines brings with them a marked increase in calibration and optimization workload while making the control of combustion more flexible. In order to meet the needs of high efficiency, the model-based calibration optimization is proposed and studied. That is to say, the neural network is used to establish the model at some operating points and then interpolated by adaptive neuro-fuzzy inference system (ANFIS) Point expansion to the required working space. The accuracy of the model is determined by the object, the amount of data used for modeling, and the adjustment of the model parameters. The test was carried out on a six-cylinder high-pressure common-rail diesel engine. Theoretical analysis and experimental results show that this method can effectively reduce the workload of calibration optimization while ensuring the accuracy.