基于自适应遗传算法的离心压缩机模型参数辨识

发布时间：2018-03-08 06:10

本文选题：离心式压缩机　切入点：机理模型　出处：《东北大学》2012年硕士论文　论文类型：学位论文

【摘要】：离心式压缩机是工业生产中的关键设备,它具有排气压力高、输送流量小的优点,在航天、能源、化工及冶金等行业中发挥着十分重要的作用。为了适应循环经济发展的需要,合理利用钢铁冶炼过程中的富余煤气,节约能源,同时减少煤气放散带来的环境污染,宝钢投产建成兼具经济效益和社会效益的燃气—蒸汽联合循环发电工程。离心压缩机为燃气轮机提供高压煤气,是整个发电工艺流程的重要设备。针对实际生产过程中离心式压缩机系统的性能分析和研究,建立压缩机系统的机理模型是主要的技术手段,而结合系统的实际运行数据确定模型中的参数是一个亟待解决的问题。本文以宝钢联合循环发电机组离心压缩机为对象,研究多级离心式压缩机模型参数辨识问题。通过分析离心压缩机的工作机理,利用压缩机对煤气做功过程中的能量守恒以及煤气质量守恒关系,建立离心压缩机的机理模型。结合现场大量实测数据,采用基于sigmoid函数的自适应遗传算法辨识机理模型中的四个非线性参数,得到离心压缩机的准确模型,并利用现场数据对所建立的模型与标准遗传算法和余弦改进型自适应遗传算法拟合效果进行了对比分析验证。仿真结果表明,基于sigmoid函数的自适应遗传算法收敛速度更快,稳定性更好,适合于处理传统搜索算法解决不好的复杂的和非线性问题,具有全局优化能力,且参数辨识后的模型输出与实际测量值的相对误差更小。模型验证结果表明,辨识后的参数能够准确反应离心式压缩机的实际工作特性,得到较高精度的模型。
[Abstract]:Centrifugal compressor is a key equipment in industrial production. It has the advantages of high exhaust pressure and low flow rate. It plays a very important role in aerospace, energy, chemical and metallurgical industries. Make rational use of the surplus gas in the process of iron and steel smelting, save energy and reduce the environmental pollution caused by the gas emission at the same time, Baosteel put into operation a gas-steam combined cycle power generation project with both economic and social benefits. Centrifugal compressors provide high pressure gas for gas turbines. It is an important equipment in the whole power generation process. In view of the performance analysis and research of centrifugal compressor system in actual production process, establishing the mechanism model of compressor system is the main technical means. It is an urgent problem to determine the parameters of the model by combining the actual operation data of the system. In this paper, taking the centrifugal compressor of Baosteel combined cycle generating set as the object, the model parameter identification of the multistage centrifugal compressor is studied, and the working mechanism of the centrifugal compressor is analyzed. The mechanism model of centrifugal compressor is established by using the conservation of energy and gas quality in the process of work done by compressor on gas. The accurate model of centrifugal compressor is obtained by using adaptive genetic algorithm based on sigmoid function to identify four nonlinear parameters in the model of mechanism. The simulation results show that the adaptive genetic algorithm based on sigmoid function converges faster than the conventional genetic algorithm and cosine modified adaptive genetic algorithm. It is more stable and suitable for solving complex and nonlinear problems which are not well solved by traditional search algorithms. It has the ability of global optimization, and the relative error between the model output and the actual measured value after parameter identification is smaller. The model verification results show that, The identified parameters can accurately reflect the actual working characteristics of centrifugal compressor and obtain a higher precision model.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH452;TP18

【参考文献】