【摘要】 针对深度调峰下运行工况频繁变动使锅炉燃烧优化参数调整难度增大的问题，提出了一种基于KPCA-FCM工况精简的燃烧优化方法。首先对锅炉实际历史运行数据提取稳态工况后，通过核主成分分析法(KPCA)进行降维，选取贡献率较大的运行参数利用模糊聚类算法(FCM)进行分析完成工况划分，实现对工况的精简。然后对不同的燃烧工况匹配对应的工况簇，调整燃烧参数到该类的最佳运行参数。为了验证该方法的合理性，采用最小二乘支持向量机辨识锅炉燃烧热效率模型。以高低两个工况区间为例进行仿真验证，结果表明提取到的最优运行参数目标值可以使锅炉热效率最高提升0.2%。因此，提出的工况精简方法可有效选取最优运行目标值，为现场运行人员调整运行参数提高锅炉效率提供了合理的数据参考。更多还原

【Abstract】 Aiming at the problem that the frequent changes of operating conditions under deep peak shaving make it more difficult to adjust the optimization parameters of boiler combustion, this paper proposed a combustion optimization method based on KPCA-FCM to simplify operating conditions. Firstly, after extracting the steady-state working conditions from the actual historical operation data of the boiler, dimensionality reduction was performed by Kernel Principal Component Analysis(KPCA). We selected the operating parameters with high contribution rates and used the fuzzy clustering algorithm(FCM) to analyze, classify and achieve working condition simplification. Secondly, the corresponding working condition clusters were matched for different combustion conditions, and the combustion parameters were adjusted to the best operating parameters of this type to achieve the purpose of improving thermal efficiency. In order to verify the rationality of the method, the least squares support vector machine was used to identify the boiler combustion thermal efficiency model. Taking the high and low working condition intervals as an example, the simulation verification was carried out. The results show that the extracted target value of the optimal operating parameters can increase the thermal efficiency of the boiler by up to 0.2%. Therefore, the proposed method for streamlining operating conditions can effectively select the optimal operating target value and provide a reasonable data reference for field operators to adjust operating parameters.更多还原