基于遗传算法的柴油机喷油系统模型参数辨识

Parameter identification of diesel fuel injection system model using genetic algorithms

柴油机喷油系统中压力的变化会引起气泡的产生和溃灭,而气泡的变化会对燃油的流动和压力波的传播产生影响.为了更准确预测喷油系统中的初始含气率以及变化情况,提出了应用改进遗传算法对柴油机喷油系统模型中的柱塞腔、出油阀紧帽腔、高压油管和针阀腔初始含气率进行辨识的新方法.基于气泡模型,建立了气-液两相条件下的柴油机喷油系统模型,通过仿真数据和试验数据对比,构造了适应度函数,实现了对柴油机喷油系统模型的参数辨识,并得到了参数优化后的柴油机喷油系统模型.仿真结果与试验数据的比较验证了采用改进遗传算法对气-液两相条件下的柴油机喷油系统模型进行参数辨识的可行性.

The changing of pressure in the diesel injection system results in the growth and collapse of gas bubbles,which affects the diesel flowing and pressure.In order to predict the initial void fraction and changes in the diesel injection system,the method using improved genetic algorithms in parameter identification for the model of diesel injection system was presented,including the initial void fraction of plunger chamber,delivery valve chamber,high-pressure oil pipe and needle valve chamber.The model of diesel injection system under the condition of gas-liquid was established based on the gas model,fitness function was built by comparing with the simulation results and the experimental data,so the parameter optimization of the model of diesel injection system was realized,and the model of diesel injection system of parameter identification was carried out.Comparisons between simulation results and experimental data show that the improved genetic algorithms are capable of estimating unknown parameters in the model of diesel injection system.