夹点分析法在换热网络优化中的应用

夹点技术是一种卓有成效的过程能量综合方法，本文介绍了夹点技术的基本思想、及其所遵循的规则。准确地找出换热网络中的夹点，可以有效地进行换热网络的最优设计，最大限度地利用能量。     

Multi-period design of heat exchanger networks

Heat exchanger networks are an integral part of chemical processes as they recover available heat and reduce utility consumption, thereby improving the overall economics of an industrial plant. This paper focuses on heat exchanger network design for multi-period operation wherein the operating conditions of a process may vary with time. A typical example is the hydrotreating process in petroleum refineries where the operators increase reactor temperature to compensate for catalyst deactivation. Superstructure based multi-period models for heat exchanger network design have been proposed previously employing deterministic optimisation algorithms, e.g. (0005 and 0180). Stochastic optimisation algorithms have also been applied for the design of flexible heat exchanger networks recently (0110 and 0115). The present work develops an optimisation approach using simulated annealing for design of heat exchanger networks for multi-period operation. A comparison of the new optimisation approach with previous deterministic optimisation based design approaches is presented to illustrate the utilisation of simulated annealing in design of optimal heat exchanger network configurations for multi-period operation.     

Optimum design of heat exchanger networks

The optimum design of heat exchanger networks is considered in two stages. In the first, the optimum configuration for fixed values of the continuous variables is determined using an implicit enumeration algorithm. In the second stage the optimization of continuous variables of the network is performed solving a large scale nonlinear programme. The proposed method for discrete variables lessens considerably the computational effort for solving the combinatorial problem in the first stage. The results show the importance of optimizing the continuous variables of these networks.     

The pinch design method for heat exchanger networks

A novel method is presented for the design of heat exchanger networks. The method is the first to combine sufficient simplicity to be used by hand with near certainty to identify “best” designs, even for large problems. “Best” designs feature the highest degree of energy recovery possible with a given number of capital items. Moreover, they feature network patterns required for good controllability, plant layout, intrinsic safety, etc.     

基于子网络强制进化的大规模换热网络优化

换热网络优化是典型的混合整数非线性规划（MINLP）问题,此问题的非线性约束以及到处存在的局部极值,使得最优解的获得尤其困难。特别是对于大规模网络来说,当物流数目增加时,可行的结构数目呈指数增长,目前还没有一种有效的算法来解决此类问题。应用改进的混合遗传算法,首先对换热网络进行初始优化,对初步优化结果进行子网络的划分,然后进行基于官能团（子网络）的重组、分解和交叉操作,获得了很好的结果。     

基于热损失的换热网络夹点设计法

现行的换热网络目标方法,在进行换热网络价格估算时,未考虑热物流的热损失。在真实的设计中,换热器壳体保温后仍与环境温度相差较大,则其热损失不可忽略。文章提出了一个新的基于热损失的换热网络夹点设计法,该方法首先以综合费用最小为目标确定出最小温差,然后建立问题表格确定出夹点位置及最小公用工程消耗,最后再进行换热网络设计。文中采用某石油常减压换热网络系统为典型算例对该方法的前两步进行了分析研究,论证了该方法的必要性及可行性。结果表明:该方法与基本Linnhoff夹点技术法估算的投资费用有较大的差距之外,在一定的最小温差下,其与基本Linnhoff夹点技术法确定的夹点位置不同,公用工程消耗也有较大的差距。     

基于热容流率的换热网络夹点设计法

以Linnhoff换热网络夹点技术理论为基础,提出了在进行换热网络设计时,考虑热容流率随温度发生变化时确定夹点位置的方法,并结合实例进行分析计算。该法对夹点位置、最小温差、公用工程用量的确定更加准确,对投资费用的估算也更接近于实际。     

An area targeting algorithm for the synthesis of heat exchanger networks

A new algorithm for the prediction of area requirements in heat exchanger networks (HEN) is presented. The method uses a diverse pinch diagram as a basis, similar to the one suggested by Rev and Fonyo (Chemical Engineering Science, 46 (7), 1623). A numerical application for a problem involving streams with significant differences in their heat transfer coefficient values is included to show how the proposed algorithm provides better estimates for minimum area requirements in HEN than the widely-used Bath formula and the algorithm by Rev and Fonyo.     

Globally optimal synthesis of heat exchanger networks. Part II: Non-minimal networks

In Part I of this work, the synthesis of minimal heat exchanger networks using the isothermal mixing stage-wise superstructure was presented. In this Part II, an extension of the algorithm presented in Part I is made to consider networks that allow multiple solutions regarding heat allocation, that is, they have energy loops where heat loads can be rearranged without changing the overall energy consumption. We extend the strategy of our Part I to use a set of nested loops to enumerate the number of units, the structure of matches, the energy consumption, different values of exchanger minimum approximation temperature (EMAT), and different locations where EMAT is active. All the models used are linear and are aided by capital cost evaluations. As in Part I, we claim global optimality based on our conjectures. Literature examples are solved to show the effectiveness of the algorithm.     

Globally optimal synthesis of heat exchanger networks. Part I: Minimal networks

This article introduces the concept of minimal structure (MSTR) and presents an enumeration algorithm for the synthesis of heat exchanger networks based on MSTR. Minimal Structures refer to a class of heat exchanger networks featuring acyclic heat transfer networks without energy loops. The enumerations used are either exhaustive or smart with a stopping criterion. Without loss of generality we use the isothermal mixing Synheat model, that is, the method applies identically to other superstructures, with likely variations in the optimization models associated to each step. A conjecture is used to state that the algorithm renders solutions that are globally optimal. Literature examples are used to demonstrate the capabilities of the enumeration algorithm. Most of our solutions compare favorably with the best reported ones in literature, with exceptions where the reported solution is not minimal.     

一种新型Powell粒子群算法同步综合换热网络

针对换热网络同步综合方法的不足,本文提出了一种新型Powell粒子群算法,具有传统确定性方法的高精度以及启发式方法的高效率。同时针对群体智能算法优化换热网络问题时存在的不足,提出了云记忆体和个体对立策略,有效地避免算法发生早熟现象,扩大搜索范围。为处理整型变量而提出的两条整型变量优化策略与Powell粒子群算法结合,实现了连续变量与整型变量的同步优化。最后,选取两个经典算例验证算法的性能,均获得了优于文献的结果,表明算法能够找到更优的换热网络结构,是一种处理混合整数非线性问题的有效方法。     

换热网络旁路优化设计

提出了一种可以降低已有换热网络运行费用的旁路优化设计方法,使得网络控制自由度增大,运行费用明显降低。该优化设计方法将网络运行总费用作为目标函数,同时考虑结垢热阻对换热器换热效果的影响,经过多次优化计算确定旁路位置和开度,并根据现场具体情况选择合适的旁路开度下限,确定旁路数目,最终实现网络旁路优化设计。通过示例说明了本文所提方法的有效性和实用前景。     

A grid-based facilities allocation approach with safety and optimal heat exchanger networks synthesis

A new approach is presented to determine optimal layout of facilities where toxic releases may occur in an existing or new facility. The land area is divided in equally sized rectangular grids, where each grid contains up to one facility surrounded by streets. Some facilities may produce hot and/or cold streams and the associated heat exchangers network (HEN) is simultaneously optimized with the layout problem. The three dimensions of geographical allocation points for each generated stream are included in the model. No additional cost for geographical allocation of heating and cooling services is considered since every facility is expected to contain these services regardless of their use in the HEN. The toxic effect is estimated via probit functions and its associated risk reduction results in providing safety to the combined HEN-facility layout problem. The grid-based allocation eliminates numerical difficulties appearing with conventional non-overlapping and Euclidian distance equations.     

微分进化算法应用于换热网络全局最优化

前言换热网络是过程系统中实现能量回收和高效利用的重要环节,其性能直接关系到整个系统的能量利用水平。目前,换热网络优化设计的方法主要分为三类,分别是夹点法[1-2]、数学规划法[3-4]与启发式方法[5-7]。     

一种新的用于换热网络全局优化的强制进化随机游走算法

应用启发式方法在换热网络全局优化上的优点,提出了一种全新的强制进化随机游走算法（random walk algorithm with compulsive evolution,RWCE）,算法以目标函数减小为强制方向,通过各换热单元面积的随机扩大或缩小,同时实现了整型变量（换热单元数）和连续变量（换热单元面积）的同步优化。另外,算法能够以一定的概率选择接受差解,使其具备极强的跳出局部最优解的能力和全局搜索能力。算例验证表明,RWCE算法相比于其他启发式方法具有程序简单、更易实现、算法适应性及全局搜索能力更强的优点,使优化质量得到进一步提升。     

换热网络控制分析与设计研究进展

回顾了近年来对换热网络可控性方面的研究,即分析了当前换热网络的两种控制策略:控制性能指数法和旁路优化控制方法。阐述了这些方法的优点和尚待改进的问题,并指出换热网络的设计应同时考虑工艺与控制两方面。提出了工艺与控制集成的设计方法的具体研究思路,并指出换热网络工艺与控制集成设计的方法将成为今后换热网络设计与控制方面研究的发展方向。     

换热网络操作夹点分析与旁路优化控制

换热网络夹点设计法是从设计的角度,针对某一给定的典型操作条件而进行的,而炼油化工过程的生产条件经常在一定范围内波动。在实际的生产中,换热网络的操作夹点和最小温差与设计值往往不尽相同,为换热网络的优化控制带来了一定困难。因而近年来对于换热网络夹点技术以及旁路优化控制方面的研究不断深入,但将夹点技术与换热网络控制集成的方法仍不成熟。本文从操作的角度求解并分析换热网络结构已定或网络正在运行情况下的操作夹点,定性分析操作夹点的变化规律,并提出在操作夹点附近设置旁路实现网络的旁路优化控制,从而将夹点技术应用于换热网络旁路优化控制中。实例仿真表明,这一旁路优化控制方法在满足控制要求的同时明显降低了网络的总公用工程,验证了其有效性。     

Two-tier control structure design methodology applied to heat exchanger networks

Because of its paramount importance in the successful industrial control strategy of a given heat exchanger network(HEN), the control structure designs for providing appropriate manipulated variable(MV)and controlled variable pairings have received considerable attention. However, quite frequently HENs with such control structures face the problem of hard constraints, typically holding the HENs at less controlled operating space. So both the MV pairings and the above control pairings should be c...     

微分进化算法优化换热网络的性能

通过分析换热网络换热器面积与综合费用之间的关系,证明了连续变量引起的换热网络非线性及非凸等特性,这些特性造成换热网络局部最优解众多。而传统的基于梯度的优化算法受初始点影响很大,容易陷入局部最优解,是换热网络最优化的障碍所在。针对优化换热网络问题难以稳定收敛的角度出发,在换热网络分级超结构模型的基础上,研究了微分进化算法求解换热网络问题的鲁棒性。通过具体的算例,在固定结构下与经典的牛顿法比较,结果表明,该方法优化结果不受初始点影响,具有很好的鲁棒性,并且年综合费用得到进一步降低。用于工业生产过程中,可以稳定有效地节约成本。