梁勇, 费红姿, 刘冰鑫, 等. 高压共轨燃油系统喷油量的预测模型[J]. 中国舰船研究, 2021, 16(4): 164–172. doi: 10.19693/j.issn.1673-3185.02128
引用本文: 梁勇, 费红姿, 刘冰鑫, 等. 高压共轨燃油系统喷油量的预测模型[J]. 中国舰船研究, 2021, 16(4): 164–172. doi: 10.19693/j.issn.1673-3185.02128
LIANG Y, FEI H Z, LIU B X, et al. Prediction model of injection quantity of high pressure common rail fuel system[J]. Chinese Journal of Ship Research, 2021, 16(4): 164–172. doi: 10.19693/j.issn.1673-3185.02128
Citation: LIANG Y, FEI H Z, LIU B X, et al. Prediction model of injection quantity of high pressure common rail fuel system[J]. Chinese Journal of Ship Research, 2021, 16(4): 164–172. doi: 10.19693/j.issn.1673-3185.02128

高压共轨燃油系统喷油量的预测模型

Prediction model of injection quantity of high pressure common rail fuel system

  • 摘要:
      目的  为了给精确控制柴油机燃油喷射量提供基础支撑,针对预−主喷模式,提出一种以间隔时间、共轨压力和主喷脉宽为输入参数的主喷油量波动预测模型的构建方法。
      方法  运用AMESim仿真模型进行模拟试验并采集数据,利用获取的数据得到单一工况下的间隔时间与主喷油量之间的关系;在上述基础上,考虑共轨压力和主喷脉宽对模型系数的影响,构建出完整的多工况主喷油量预测模型;使用多组不同组合的输入参数对所构建预测模型的准确性和可靠性进行验证。
      结果  结果表明,构建的多工况主喷油量预测模型均方根误差为1.443 mm3,所需试验量可由n3量级降至n2量级。
      结论  所提的预测模型具有较高的准确度,对基于模型的喷油量控制具有工程应用意义。

     

    Abstract:
      Objectives  In order to provide the basis for accurate control of diesel engine fuel injection quantity, a method was proposed to establish the main injection quantity prediction model with input parameters as dwell time, common rail pressure and main injection pulse width for the pre-main injection mode.
      Methods  AMESim simulation model was used to conduct simulation experiments and then collect data. Based on the data, the relationship between the dwell time and the main injection volume under single working condition was obtained. On this basis, the influence of common rail pressure and main injection pulse width on the model coefficient was introduced to build a complete multi-working condition main injection quantity prediction model. Multiple input parameter combinations were used to verify the accuracy and reliability of the prediction model of main injection quantity fluctuation.
      Results  The results show that the root-mean-square error (RMSE) of the scheme of multi-working condition main injection quantity prediction model is 1.443 mm3. It can reduce the experiment amount required for modeling from n3 orders of magnitude to n2 orders of magnitude.
      Conclusions  The proposed prediction model has sufficient accuracy, which is of great significance for engineering application of model-based fuel injection control.

     

/

返回文章
返回