换热网络优化改进及最优控制的理论和方法研究

发布时间：2018-07-08 15:11

本文选题：灵敏度分析 + 流程模拟　；参考：《青岛科技大学》2015年硕士论文

【摘要】：换热网络是能量回收的一个重要系统,灵敏度分析和弹性分析对换热网络的设计、优化改进及最优控制起着至关重要的作用。本文对换热网络的灵敏度分析和弹性分析进行深入系统的研究,建立灵敏度分析和弹性分析的理论方法和数学模型,以解决如何在最小程度上或以最少的投资对换热网络进行改造使网络拥有最优控制性能的优化问题。本文根据灵敏度系数、灵敏度和弹性指数的定义,建立灵敏度系数、灵敏度分析及弹性分析的数学模型,模型可自动确定决策变量、状态变量及参数,给出求解方法。将建立的模型与换热网络流程模拟有机结合,对需要优化的换热网络,将原始数据通过数学模型进行初步计算,然后进行流程模拟,如果不符合设计要求可返回进行迭代计算,再流程模拟,重复过程直到满足要求。此方法规避了计算复杂的非线性规划问题,减小了计算量并提高了计算精度。用Inter Visual Fortran语言对所建立的数学模型和求解方法编制计算机程序,通过程序的输出可以直接看到网络的变化并知晓如何优化网络。用工程实例进行考核,并将结果与经典的灵敏度分析和弹性分析的方法进行比较,得出年综合费用为30479美元,本方法与文献相比,费用小于其34376美元的年费用,同时本方法输入数据简单,即能节省运算时间,也能减少计算出错的几率。将编制的程序嵌入“换热网络优化节能系统”软件中,构成灵敏度分析及弹性分析模块,完善了软件功能,使用灵活方便,使软件功能更加强大。
[Abstract]:Heat transfer network is an important system of energy recovery. Sensitivity analysis and elastic analysis play an important role in the design, optimization and control of heat transfer network. In this paper, the sensitivity analysis and elastic analysis of heat transfer network are studied systematically, and the theoretical methods and mathematical models of sensitivity analysis and elastic analysis are established. In order to solve the problem of how to transform the heat exchanger network to the minimum degree or with the least investment, the optimal control performance of the network can be obtained. According to the definition of sensitivity coefficient, sensitivity and elastic index, a mathematical model of sensitivity coefficient, sensitivity analysis and elastic analysis is established in this paper. The model can automatically determine the decision variables, state variables and parameters, and give the solution method. Combining the established model with the flow simulation of heat transfer network, the original data is calculated through the mathematical model for the heat transfer network that needs to be optimized, and then the process simulation is carried out. If the model does not meet the design requirements, it can be returned to carry out iterative calculation. Reprocess simulation, repeat the process until it meets the requirements. This method avoids the complex nonlinear programming problem, reduces the computational complexity and improves the accuracy. The computer program is programmed with Inter Visual Fortran language for the mathematical model and solution method. Through the output of the program, the change of the network can be seen directly and how to optimize the network can be realized. The results are compared with the classical methods of sensitivity analysis and elastic analysis. The results show that the annual comprehensive cost is US $30479. Compared with the literature, the annual cost of this method is less than US $34376. At the same time, the method is simple to input data, which can save operation time and reduce the probability of calculation error. The program is embedded in the software of "Heat transfer Network Optimization and Energy-saving system", which constitutes a sensitivity analysis and elastic analysis module, which improves the function of the software, is flexible and convenient to use, and makes the software function more powerful.
【学位授予单位】：青岛科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ021.8

【参考文献】