融合温度场信息的锅炉燃烧优化专家系统研究

发布时间：2018-05-29 08:17

本文选题：温度场 + 燃烧优化　；参考：《东南大学》2016年硕士论文

【摘要】：当前锅炉的燃烧优化控制大多还是运行人员根据经验手动进行调整,由于锅炉运行状况复杂多变,运行人员很难从众多的信息中及时对锅炉燃烧状况做出准确判断而进行相应的调整,容易造成燃烧调整的延误,进而导致锅炉运行偏离最佳燃烧状况。温度场能快速反映炉膛的燃烧状况,在对锅炉配煤、配风方式进行调整时,炉内的温度场亦跟着快速变化,研究融合温度场信息的锅炉燃烧优化方法能部分克服燃烧系统延迟造成的影响。目前,基于人工智能的锅炉燃烧优化方法得到了日益广泛的关注,本文研究融合温度场信息的锅炉燃烧优化专家系统,主要研究内容如下：1、本文利用声学测温系统获取沿炉膛高度相关截面的温度分布信息,结合锅炉运行数据,建立了NOx排放和锅炉效率的最小二乘支持向量机模型。针对粒子群算法易陷入局部最优和出现早熟收敛的缺点,采用一种改进的混合PSO/DE优化算法实现模型参数的寻优。相比无断面温度信息的NOx排放和锅炉效率模型,考虑温度场信息的模型预测精度更高。基于所建模型,采用混合PSO/DE优化算法进行锅炉高效率和低NOx排放的优化计算,可获得各工况下的最优操作变量,实现锅炉燃烧优化,并为数值型最优知识库的建立奠定了基础。2、在数值型最优知识库的基础上,结合模糊C均值聚类算法,提出了构建锅炉燃烧优化知识库的方法。在所建模糊知识库的基础上,采用CLIPS开发工具构造锅炉燃烧优化专家系统的推理机,建立了基于CLIPS的模糊推理模型,并采用一种反向追踪方法进行推理解释。有效性检验结果表明,基于专家系统求解出的优化结果与直接进行非线性优化所得结果相比误差很小,且计算量更小。3、完成了锅炉燃烧优化专家系统的整体架构设计,以及专家系统各个功能模块的开发。通过调用ClipsNET组件实现将CLIPS项目整合到.Net框架中,应用OPC通信技术实现了专家系统软件对DCS数据以及声学测温系统数据的读取,最后对锅炉燃烧优化专家系统各功能界面进行设计,为进一步的工程应用打下了基础。
[Abstract]:At present, most of the boiler combustion optimization control is manually adjusted by the operator according to the experience, because the boiler operation condition is complex and changeable, It is difficult for operators to accurately judge the combustion condition of boiler from many information, which can easily cause the delay of combustion adjustment and lead to boiler running deviating from the optimal combustion condition. The temperature field can quickly reflect the combustion state of the furnace, and the temperature field in the furnace changes rapidly when the coal blending and air distribution mode of the boiler are adjusted. The study of boiler combustion optimization method with temperature field information can partly overcome the influence of combustion system delay. At present, the boiler combustion optimization method based on artificial intelligence has been paid more and more attention. In this paper, a boiler combustion optimization expert system based on temperature field information is studied. The main research contents are as follows: 1. In this paper, the temperature distribution information along the furnace height correlation section is obtained by using acoustic temperature measurement system, and the least square support vector machine model of NOx emission and boiler efficiency is established in combination with boiler operation data. Aiming at the shortcomings of particle swarm optimization (PSO) which is prone to local optimization and premature convergence, an improved hybrid PSO/DE optimization algorithm is used to optimize the model parameters. Compared with the model of NOx emission and boiler efficiency without cross-section temperature information, the prediction accuracy of the model considering temperature field information is higher than that of the model without cross-section temperature information. Based on the established model, a hybrid PSO/DE optimization algorithm is used to calculate the high efficiency and low NOx emission of the boiler. The optimal operating variables can be obtained and the combustion optimization of the boiler can be realized. It lays a foundation for the establishment of numerical optimal knowledge base. On the basis of numerical optimal knowledge base and combined with fuzzy C-means clustering algorithm, the method of constructing boiler combustion optimization knowledge base is put forward. On the basis of the established fuzzy knowledge base, the inference engine of boiler combustion optimization expert system is constructed by using CLIPS development tool, and the fuzzy inference model based on CLIPS is established, and a reverse tracing method is used for reasoning interpretation. The results of validity test show that the optimization results based on expert system are less error than those obtained by direct nonlinear optimization, and the calculation amount is smaller than that of expert system, and the whole structure design of boiler combustion optimization expert system is completed. And the development of each function module of expert system. The CLIPS project is integrated into .net framework by calling ClipsNET component, and the expert system software is used to realize the reading of DCS data and acoustical temperature measurement system data by using OPC communication technology. Finally, the function interface of boiler combustion optimization expert system is designed, which lays a foundation for further engineering application.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM621.2

【相似文献】