面向反应再生过程的量子粒子群多目标优化

针对催化裂化反应再生过程难以有效解决提升效率、降低损耗、减少排放的多目标优化问题,利用改进的多目标量子粒子群算法进行求解。建立轻油收率、焦炭产率和硫化物排量的多目标优化模型;引入拥挤熵排序更新外部档案,精确估计非支配解集分布性;构造自适应因子以动态调整吸引子,平衡算法的收敛性和多样性;再引入高斯变异进行分段式扰动,增强算法的局部搜索精度,最后求解该优化模型。对某厂催化裂化进行实验,得到轻质油吸收率76.22%,焦炭产率5.72%和硫化物排放量626 mg/m~3的结果,均优于其他比较算法,表明改进后的算法可以快速、准确地获得分布均匀的Pareto最优解,能有效解决反应再生过程多目标优化问题。     

基于动态优化的催化裂化装置再生器裕量分析与控制设计(Ⅱ) 求解方法与结果分析

用动态优化的方法求解催化裂化装置再生器的工艺裕量与控制设计。对于该多目标混合整数动态优化问题,通过ε-约束法处理多目标优化问题,将控制性能目标函数转化为控制性能约束;通过将0-1变量松弛化、引入附加等式约束求解混合整数动态优化问题。求解得到非劣解集,绘制关于控制性能指标和工艺指标的关系曲线,发现系统对控制器性能的要求愈高,所需要的裕量应愈大。由此进行折中处理,从而找到兼顾工艺要求和控制性能的催化裂化装置再生器优化设计方案。     

基于动态优化的催化裂化装置再生器裕量分析与控制设计（Ⅰ）动态优化的数学描述

从操作和控制的角度,通过动态优化的方法求解催化裂化装置再生器的工艺裕量与控制设计。该动态优化问题的数学描述包括催化裂化装置反应-再生系统和配套控制系统的动态模型、多回路PID控制器结构模型、过程不确定性模型、稳态起始点约束、生产要求和操作约束及目标函数等。该优化问题是一个带有0-1变量约束的多目标混合整数动态优化问题,具有分别描述控制性能指标和工艺指标的两个不相关目标函数。对两个优化目标进行折中处理,可以使设计结果既能满足工艺要求,又能实现良好的自动控制。     

芳烃抽提过程多目标优化

芳烃抽提是芳烃生产过程的重要环节,其生产调优对提高整个芳烃联合装置的效益具有重要意义。基于流程模拟及响应面分析方法,得到了芳烃抽提过程的产品纯度模型及能耗模型。建立了以产品纯度最大化及过程能耗最小化的多目标优化模型。提出了一种改进的自适应加权求和算法,并用于多目标优化模型的求解。求解结果表明新算法在Pareto最优解分布的均匀性上与原算法相当,但求解效率要高于原算法。给出了不同产品等级下的最佳操作参数,采用优化后的操作参数可有效地提高产品纯度并降低过程能耗。提出的多目标优化模型及求解算法用于芳烃抽提过程的操作调优,可有效地提高决策的准确性。     

高温短接触催化裂化与提升管反应器的优化

介绍了在降低焦炭产率基础上，提高原料中氢碳元素化学有效转化程度是优化重油催化裂化的重要途径，分析了高温短接触反应条件对重油催化裂化反应的影响以及近年来围绕实现高温短接触反应条件而发展起来的原料油雾化、反应控制、气固快分各项提升管反应优化技术。     

重油催化裂化装置产品分布调控与优化模拟分析

重油催化裂化焦炭产率较高,会加重再生器负荷,降低剂油比和轻质油收率。对此,在催化裂化装置基础上添加外取热器。通过采用外取热和外甩油浆相结合的方法,实现重油催化裂化轻质油收率的提高。外取热器的作用是为了快速有效将再生器部分过多的热量取走,达到再生催化剂降温的目的。外甩油浆的作用是为了降低焦炭产率,减少烧焦产生热量。热量的降低可以有效提高剂油比,增加轻质油收率。原料残炭值的大小对产品分布有直接影响。原料残炭值越大,催化裂化装置的反应器部分产生焦炭越多,待生催化剂上含碳量也会升高,到达再生器烧焦以后释放大量的热量,热量增加不仅会影响再生器的寿命,也会使催化裂化装置中的剂油比降低,从而降低轻质油收率。通过控制向量参数化方法对CO助燃剂、主风、外取热和外甩油浆进行了不同层次的调控与优化,结果发现,对于重油催化裂化,CO助燃剂、主风的优化影响效果有限,而外取热和外甩油浆相互促进可以有效提高剂油比和轻质油收率。     

煤化工废水脱酸脱氨流程的多目标优化

对某气化厂85 t/h煤化工废水单塔脱酸脱氨流程进行模拟，在此基础上建立以塔釜再沸器热负荷最小化和产物收率(氨气+酸性气)最大化为优化目标的多目标优化模型。由于化工流程的特殊性，对现有的多目标粒子群算法进行改进，加入边界检验以提升运算速度。运用改进后的算法对模型进行求解，得到一组Pareto解集。以目前工况为基准，选定2个Pareto解进行分析。结果表明，在维持当前塔釜再沸器热负荷基本不变的情况下，产物收率增加10.21%;在维持当前产物收率基本不变的情况下，塔釜再沸器热负荷降低16.98%。     

多目标过程系统优化的粒子群算法求解

化工领域的过程设计、生产控制、配方和计划等众多问题的数学模型,在考虑产品性能、单位成本、环境影响等诸多因素下,都是多目标优化问题;而求解多目标优化问题,目前还没有有效的方法;现今的做法是把多目标优化通过加权转化为单目标优化,再求解单目标优化问题,但这存在权数不易确定;还忽视了有效解集中存在一个其各目标的值与各目标的最优值距离最近的有效解的问题,称为理想有效解.理想有效解的求法一般分为两步,先求各目标的最优值、再求理想有效解,这将影响求解的速度;为此提出在PSO(粒子群优化)算法中加入惩罚项,同时对PSO算法中的个体极值与全局极值作调整,使PSO算法适用于求多目标优化问题理想有效解,该算法对多目标问题起到边优化边求理想有效解的功效;这使得在求解速度上加快.通过性能测试表明了算法的有效性,最后将算法用于求解多亚甲基多苯基多胺生产过程系统优化取得良好效果.     

粒子群算法求解边值固定的化工动态过程优化问题

针对化工过程系统优化中广泛存在着边值固定的动态优化问题,该问题的求解数学上还没有有效的方法,现今的方法之一是将问题转化为多目标优化问题.本文在粒子群优化(PSO)算法的基础上,提出在PSO算法中加入惩罚项,同时对局部极值与全局极值作进一步的调整,使PSO算法适用于求多目标优化问题理想有效解,该算法对多目标问题起到边优化边求理想有效解的功效;即只用一步即可求理想有效解,这使得在求解速度上大为加快.最后将其用于间歇反应器的最佳反应温度边值固定动态优化控制的实际运用中,取得良好效果.     

基于混合遗传算法的催化重整过程多目标优化

为实现催化重整过程生产指标的综合优化,基于已实现工业应用的催化重整17集总反应动力学模型和催化重整过程机理模型,考虑相应的多种约束条件,建立了以最大化总芳烃收率和最小化重芳烃收率为目标的多目标操作优化模型。提出了一种将遗传算法与局部优化方法相结合的多目标混合遗传算法HNAGA,并用于多目标操作优化模型的求解。现场工业数据的仿真研究表明,HNAGA在寻找Pareto最优解前沿方面比原遗传算法具有一定的优越性。将该多目标优化模型和求解方法应用于工业催化重整装置的操作优化,可以有效提高决策的准确性。     

Multi-objective operation optimization of olefin separation process for MTO plant

In modern coal processing industries, methanol-to-olefins (MTO) is an important equipment. Its olefin separation process is facing with problems such as the change of raw materials, the loss of olefin products and the high consumption of utilities. Operation optimization is required to achieve maximum benefits under the circumstance of quality assurance and requirements. This article takes the pre-depropanized olefin separation process of Lummus as the research object. And the optimization objectives are the total yield of ethylene and propylene as well as the total energy consumption. Modeling, simulation and multi-objective optimization of the process are conducted. Non-dominated sorting genetic algorithm (NSGA-II) is used to solve multi-objective optimization problem. The simultaneous optimization of 15 operational variables is achieved. Under the current yield, the optimal operation point is found by reducing the reflux ratio of low pressure depropanizer, deethanizer and 1^# propylene tower and so on. The results show that the optimal operating point can reduce energy consumption by 20 MW compared with the existing operating point. The optimization interval of each operation variable corresponding to different trade-off points is determined by the comprehensive analysis of decision variables. It is also found that distillation equipment can operate in different optimal operation intervals.     

Multi-objective modeling and optimization for scheduling of cracking furnace systems

Cracking furnace is the core device for ethylene production. In practice, multiple ethylene furnaces are usual y run in parallel. The scheduling of the entire cracking furnace system has great significance when multiple feeds are simultaneously processed in multiple cracking furnaces with the changing of operating cost and yield of product. In this paper, given the requirements of both profit and energy saving in actual production process, a multi-objective optimization model contains two objectives, maximizing the average benefits and minimizing the average coking amount was proposed. The model can be abstracted as a multi-objective mixed integer non-linear programming problem. Considering the mixed integer decision variables of this multi-objective problem, an improved hybrid encoding non-dominated sorting genetic algorithm with mixed discrete variables (MDNSGA-I ) is used to solve the Pareto optimal front of this model, the algorithm adopted crossover and muta-tion strategy with multi-operators, which overcomes the deficiency that normal genetic algorithm cannot handle the optimization problem with mixed variables. Finally, using an ethylene plant with multiple cracking furnaces as an example to illustrate the effectiveness of the scheduling results by comparing the optimization results of multi-objective and single objective model.     

MTO烯烃分离过程的多目标操作优化

现代煤化工中，甲醇制烯烃 （MTO） 是一个非常重要的装置。其烯烃分离过程面临着原料变动大、烯烃产品损失以及较高的公用工程消耗等问题。这就需要在满足产品规格和需求的情况下，优化操作条件以实现最大效益。以Lummus前脱丙烷的烯烃分离工艺为研究对象，以增加乙烯与丙烯的总收率和降低总能耗为优化目标，对该工艺流程进行建模模拟与多目标优化。采用非支配排序遗传算法（NSGA-II）进行多目标优化的求解，实现了15个操作变量的同时优化。在维持产品收率不变的前提下，可通过降低脱丙烷塔、脱乙烷塔和1^#丙烯精馏塔的回流比等优化措施找到了当前最优操作点。结果表明，该最优操作点与现有操作点相比可降低20 MW能耗。通过对决策变量的综合分析，确定了不同目标权衡下对应的各个操作变量的优化区间，发现精馏塔可以在多个最佳操作区间内运行。     

非支配排序进化策略求解煤气化多目标优化问题

应用非支配排序进化策略（non-dominated sorting evolution strategy,NSES）对煤气化多目标优化问题进行求解。通过解两个经典测试函数,并与NSGA-2算法进行比较,表明了非支配排序进化策略的有效性和优势。应用Aspen Plus流程模拟软件对煤气化过程进行了模拟计算。在此基础上,以氧煤比、水煤比、气化炉的压力为操作变量,分别对冷煤气效率和有效气产出率两个目标进行灵敏度分析。分析结果表明,3个变量对气化结果评价指标均有不同程度的影响。将非支配进化策略用于煤气化过程的多目标优化模型的求解,得到了Pareto最优前沿面,为确定冷煤气效率和有效气产出率两个目标的协调提供了依据。     

A multi-objective invasive weeds optimization algorithm for solving multi-skill multi-mode resource constrained project scheduling problem

A new multi-skill multi-mode resource constrained project scheduling problem with three objectives is studied in this paper. The objectives are: (1) minimizing project's makespan, (2) minimizing total cost of allocating workers to skills, and (3) maximizing total quality of processing activities. A meta-heuristic algorithm called multi-objective invasive weeds optimization algorithm (MOIWO) with a new chromosome structure guaranteeing feasibility of solutions is developed to solve the proposed problem. Two other meta-heuristic algorithms called non-dominated sorting genetic algorithm (NSGA-II) and multi-objective particle swarm optimization algorithm (MOPSO) are used to validate the solutions obtained by the developed MOIWO. The parameters of the developed algorithms are calibrated using Taguchi method. The results of the experiments show that the MOIWO algorithm has better performance in terms of diversification metric, the MOPSO algorithm has better performance regarding mean ideal distance, while NSGA-II algorithm has better performance in terms of spread of non-dominance solution and spacing metrics.     

Multi-objective optimization of fuel oil blending using the jumping gene adaptation of genetic algorithm

Production and marketing of heavy fuel oil (HFO) are an easy, effective and economical way to dispose off certain very heavy refinery streams such as short residue (SR, available from the bottom of vacuum distillation units) and clarified liquid oil (CLO, available from the bottom of the main fractionators of fluidized-bed catalytic crackers). Certain lighter streams such as heavy cycle oil (HCO), light cycle oil (LCO) and kerosene, are added to the heavy residual stock to improve its quality in terms of fluidity, combustibility, etc., to be marketed as fuel oil. The present study aims at optimization of the fuel oil blending process to maximize profit, minimize quality give-away, maximize production, minimize use of lighter products such as LCO and kerosene, and maximize the calorific value, etc. Several multi-objective optimization problems have been formulated comprising of two and three-objective functions and solved using the elitist non-dominated sorting genetic algorithm (NSGA-II). This evolutionary technique produces a set of non-dominating (equally good) Pareto optimal solutions from which the operator can choose the one that is most suitable (preferred point). Also, a fixed-length macro–macro mutation operator, inspired by jumping genes in natural genetics, has been used with NSGA-II to solve this problem. This modified algorithm leads to a significant reduction in the computational effort. Indeed, this adaptation can be of immense use in reducing the computational effort for other problems in chemical engineering.     

基于改进NSGA-Ⅱ算法的FCC分离系统多目标优化

应用流程模拟软件HYSYS对某炼厂65万吨/年重油催化裂化装置进行全流程模拟。以最大目的产品（汽油+液化气）收率与最小分离系统能耗为优化目标,采用改进的非支配排序遗传算法（NSGA-Ⅱ）对FCC分离系统进行多目标优化。通过对决策变量的择优选取、最优算法参数的研究（包括种群规模、P_c与P_m自适应策略、遗传代数）、最优点的决策建立了一套缜密的多目标优化方法。优化结果显示,目的产品（汽油+液化气）收率提升4.32个百分点,分离系统能耗降低16.88%,为FCC分离系统的操作变量优化提供了重要的数据支持与优化建议。     

Combining support vector regression and cellular genetic algorithm for multi-objective optimization of coal-fired utility boilers

Support vector regression (SVR) was employed to establish mathematical models for the NO_x emissions and carbon burnout of a 300 MW coal-fired utility boiler. Combined with the SVR models, the cellular genetic algorithm for multi-objective optimization (MOCell) was used for multi-objective optimization of the boiler combustion. Meanwhile, the comparison between MOCell and the improved non-dominated sorting genetic algorithm (NSGA-II) shows that MOCell has superior performance to NSGA-II regarding the problem. The field experiments were carried out to verify the accuracy of the results obtained by MOCell, the results were in good agreement with the measurement data. The proposed approach provides an effective tool for multi-objective optimization of coal combustion performance, whose feasibility and validity are experimental validated. A time period of less than 4 s was required for a run of optimization under a PC system, which is suitable for the online application.