论文部分内容阅读
采用支持向量机回归(support vector regression,SVR)模型分别建立300MW燃煤锅炉的NOx排放特性模型和锅炉热损失模型。利用热态实炉试验数据对模型进行了训练和验证。结合NOx排放模型和锅炉热损失模型采用非支配排序遗传算法(nondominated sorting genetic algorithm,NSGA-II)对锅炉进行多目标优化,定量分析了优化参数对优化结果的影响。结果表明,支持向量机回归模型可以很好地预测锅炉的排放特性和锅炉的热损失特性,NSGA-II方法与SVR模型结合可以对锅炉燃烧实现有效的多目标寻优、得到理想的帕雷托分布,是对锅炉进行多目标优化的有效工具。
The model of NOx emission and the boiler heat loss model of a 300MW coal-fired boiler were set up respectively by the support vector regression (SVR) model. The model was trained and validated by using the hot experimental data. Combining with the NOx emission model and the boiler heat loss model, a nondominated sorting genetic algorithm (NSGA-II) was used to optimize the boiler’s multi-objective, and the influence of the optimization parameters on the optimization results was quantitatively analyzed. The results show that the support vector machine regression model can predict the boiler emissions characteristics and the boiler heat loss characteristics well. Combining the NSGA-II method with the SVR model can achieve effective multi-objective optimization of the boiler combustion, and the ideal Pareto Distribution, is an effective tool for multi-objective optimization of the boiler.