煤气压缩机数据建模

发布时间：2018-04-25 02:25

本文选题：煤气系统 + 离心式压缩机　；参考：《东北大学》2011年硕士论文

【摘要】：随着工业规模的不断扩大和工业过程复杂性的日益增加,离心压缩机被广泛应用于冶金、石油、化工、天然气输送以及制冷等工业部门。离心压缩机虽有很多优点,但也有其本身难以消除的缺点,如不适用于气量小的场合、稳定工况区较窄、易喘振等等。喘振是离心压缩机的固有特性,出现喘振是很危险的,严重时会危及压缩机的安全,为了保证离心压缩机稳定、长周期运行,必须快速判断使其脱离喘振工况。本文以宝钢集团的燃气—蒸汽联合循环发电机组的煤气系统为研究对象。针对上述出现的一些问题,从系统实际运行的角度出发,通过数据建模和混合建模的方法建立煤气系统中离心压缩机的数学模型。由于系统具有多变量、非线性的特性,传统的数据建模方法在处理非线性问题上精度不高。本文采用基于核函数的偏最小二乘(KPLS)方法,利用核函数将低维空间的非线性回归转化为高维空间的线性回归,KPLS方法可以克服实际生产工艺中非线性因素对模型的不利影响。本文利用现场的数据,以煤气成分、质量流量、入口压力、入口温度等作为输入变量,建立基于KPLS方法的离心压缩机模型。针对机理模型误差较大、精度不高的问题,建立了离心压缩机混合模型。混合模型是由机理模型和误差补偿器两部分并联组成,采用基于KPLS方法训练误差补偿器。仿真分析表明基于KPLS方法的压缩机模型准确可靠、精度较高。
[Abstract]:With the continuous expansion of industrial scale and the increasing complexity of industrial processes, centrifugal compressors are widely used in metallurgy, petroleum, chemical, natural gas transportation and refrigeration industries. Although centrifugal compressor has many advantages, it also has some disadvantages, such as it is not suitable for the situation of small volume of gas, the stable working condition is narrow, the surge is easy, and so on. Surge is the inherent characteristic of centrifugal compressor. Surge is very dangerous and will endanger the safety of compressor. In order to ensure the stability and long period operation of centrifugal compressor, it must be judged quickly to get rid of surge condition. In this paper, the gas system of the gas-steam combined cycle generator set of Baosteel Group is taken as the research object. In view of the problems mentioned above, the mathematical model of centrifugal compressor in gas system is established by the method of data modeling and mixed modeling from the point of view of the actual operation of the system. Because the system is multivariable and nonlinear, the traditional data modeling method is not accurate in dealing with nonlinear problems. In this paper, the kernel function based partial least squares (KPLS) method is used to transform the nonlinear regression in low dimensional space into linear regression in high dimensional space. The KPLS method can overcome the adverse effects of nonlinear factors on the model in practical production process. In this paper, the centrifugal compressor model based on KPLS method is established with gas composition, mass flow rate, inlet pressure and inlet temperature as input variables. In order to solve the problem that the error of mechanism model is large and the precision is not high, the mixed model of centrifugal compressor is established. The hybrid model is composed of mechanism model and error compensator in parallel. The error compensator is trained based on KPLS method. Simulation results show that the compressor model based on KPLS method is accurate and accurate.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH452

【参考文献】