基于信赖域算法的油田开发规划模型研究

根据油田开发的各种成本投资形式,可以建立多种油田开发的非线性模型,结合求解非线性规划的信赖域方法,以最低开采成本规划模型为例,提出一种油田开发规划模型的求解算法,为合理开发油田提供了定量依据,并运用MATLAB进行了算例分析,得出规划模型的最优参数。     

精馏塔联立优化的简约空间SQP算法

引言采用面向开放式方程的联立算法求解精馏塔优化问题,可以使模拟计算和寻优搜索在同一层迭代中完成,避免了双层法每一步优化搜索都需进行模型收敛计算的缺点,从而可以大大提高计算效率．而且原则上可任意指定自由变量．无论是模拟计算、操作优化还是设计优化都可以利用这个灵活的计算结构进行解算.高效、可靠的非线性约束最优化算法是实现精馏塔联立优化的关键．在流程模拟和过程系统优化中,序贯二次规划法（SequentialQuadraticProgramming,SQP)因其较高的计算效率而被广泛应用．SQP算法中,每一个迭代步都需要求解一个QP子问题…     

基于PSO的精馏塔稳态优化

本文主要研究了精馏塔的稳态优化问题。采用基于平衡级模型的矩阵法,对一个脱甲烷塔进行了模拟计算,精度满足要求,用来获得精馏塔的若干工作状态,利用回归分析得到精馏操作的简化模型作为优化的等式机理约束。基于对过程和实际需要的分析,提出了精馏塔稳态优化问题,最后用微粒群(PSO)优化算法求解目标函数并得到最优化解,优化结果为精馏塔的操作提供了参考指导。     

面向方程联立求解的精馏塔模拟与优化一体化算法

在精馏塔机理模型的基础上,提出了面向方程联立求解的精馏塔模拟与优化一体化算法.精馏塔模型方程的特殊结构使得可以进行一定的变量分解和降维,采用面向方程联立求解的技术,使模拟计算和寻优搜索在同一层迭代中完成,避免了双层法每一步优化搜索都需进行模型收敛计算的缺点.从而大大提高了计算效率.同时,这样的联立法提供了一种开放式的统一框架.无论是模拟计算、操作优化还是设计优化都可以毫无困难地利用这个灵活的计算结构进行解算.文中用两个计算实例说明了这个方法的有效性.     

基于EGO算法的常压塔能量优化

常减压装置能量消耗约占炼厂总用能的25%～30%,在保证产品产量与质量的条件下,优化常减压蒸馏塔操作条件,可有效降低能耗.为了避免随机优化算法对常压塔机理模型进行操作优化时,存在计算资源消耗大、效率低的问题,文中采用基于代理模型的全局优化方法优化常压塔的余热回收过程,在优化迭代过程中用Kriging代理模型来代替耗时的精确模型评估.实验表明模型调用次数相较于粒子群优化算法减少了90%,优化时间减少了85%,实现了能量优化并且保证了侧线产品之间的分离精度.     

基于粒子群算法的隔离壁精馏塔的综合与设计

隔离壁精馏塔（dividing-wall distillation column,DWDC）是提高两个或者多个传统精馏塔热力学效率的有效手段。由于隔离壁精馏塔内部结构复杂、相互作用强,传统的序贯优化方法计算时间长,很难达到全局最优解。标准粒子群算法应用广泛、易于实现,但易于早熟、易于陷入局部极值点。因此,本工作采用改进的元胞粒子群算法对Kaibel隔离壁精馏塔进行综合与设计研究。元胞粒子群算法通过改进粒子的学习策略,采用元胞邻域的方法可有效地将粒子分散在多个子空间。对比标准粒子群优化和元胞粒子群优化两种方法的50次优化效果,结果表明,两种粒子群算法能够对内部结构复杂、相互作用强的四组分Kaibel隔离壁精馏塔这一复杂分离系统进行优化,优化效果显著。     

基于简约空间序列二次规划算法和混合求导方法的精馏塔操作优化

以开放式方程建立精馏塔严格机理模型,对精馏塔的操作优化进行了研究.根据精馏塔操作优化命题自由度低、模型结构稀疏且导数难以解析求解的特点,提出了基于简约空间序列二次规划算法（SQP）和混合求导方法相结合的精馏塔操作优化计算方法.在该方法中优化命题采用简约空间SQP算法求解,求解过程中需要的导数信息采用解析表达和预处理的自动微分技术求取.计算结果表明,本文方法在计算效率上大大高于基于差分求导的标准SQP算法,有利于在线实时优化的实施.另外本文优化结果也表明,精馏塔操作优化对提高产品产量、提高综合经济效益具有明显作用.     

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。     

基于过程模型的常压蒸馏装置在线能量优化

提出了一种基于严格机理模型和塔板水力学模型的常压蒸馏过程在线能量优化策略,在某炼油厂先进的计算机网络和数据库技术基础上,开发和实现了常压蒸馏过程能量在线优化系统,取得了良好的效果。     

基于实时采集指令域示波器的数控机床加工工艺参数优化

数控机床加工工艺参数一般都会处于线下进行优化设计,于是为了提高工艺参数优化的便捷性和有效性,文章提出了一种基于实时采集指令域示波器的机床加工工艺参数优化方式.这种方式能够在数控机床加工过程中,通过对各种相关数据的收集和存储,然后对收集的大数据进行分析,于是可以实现在线状态上对加工工艺参数进行优化.这种优化方式比较简单,...     

基于邻域-克隆选择学习算法的分馏装置负荷优化

在免疫克隆选择和人工免疫网络算法基础上,采用了Agent的思想,提出了一种邻域-克隆选择学习全局优化算法(N-Clonalg)。不同于其他人工免疫算法,N-Clonalg定义了网格化的邻域操作环境,其主要搜索算子有N-克隆选择、N-竞争和自学习算子,能有机结合全局与局部搜索,多峰测试函数表明能较好地克服克隆选择算法(Clonalg)的早熟及人工免疫网络算法(Opt-aiNet)收敛速度慢问题。分馏装置负荷优化实例应用表明,算法具有较好的最优解搜索性能,能较好地实现化工中的寻优问题。     

基于有效能分析的复杂精馏塔优化设计

提出了以有效能消耗最小为目标的复杂精馏塔优化设计新方法。给出了精馏塔优化设计模型及最佳进料位置、适宜理论板数NT、塔内换热器简约的确定方法,并以两组分正庚烷和乙苯的混合物为例进行设计与讨论。优化后的精馏塔含有中间换热器,与传统的简单塔有本质的区别,平衡线和操作线均处于相对平均的位置,更节能。     

Nonlinear adaptive control algorithm for a multicomponent batch distillation column

In this article, a nonlinear adaptive control strategy is proposed for a multicomponent batch distillation column. The hybrid control scheme consists of a generic model controller (GMC) and a nonlinear adaptive state estimator (ASE). In the first part of the study, an adaptive observer is designed aiming to estimate the partially known parameters based on the measured compositions in the presence of process/predictor mismatch. The open-loop dynamic behavior of the developed ASE estimator is investigated under initialization error, disturbance, and uncertain parameters. In the subsequent part, the adaptive GMC-ASE controller (GMC control structure in conjunction with ASE estimator) has been synthesized for the example distillation column. A simulation-based comparative study has been conducted between the derived nonlinear GMC-ASE control algorithm and a gain-scheduled proportional integral (GSPI) law in terms of constant composition control. The proposed adaptive control scheme is shown to be quite promising due to the exponential error convergence capability of the ASE estimator in addition to the high-quality performance of the GMC controller.     

Automatic design of conventional distillation column sequence by genetic algorithm

Synthesis of the optimum conventional (with non‐sharp separations) distillation column sequence (DCS) is a challenging problem, in the field of chemical process design and optimization, due to its huge search space and combinatorial nature. In this paper, a novel procedure for the synthesis of optimum Conventional Distillation Column Sequence is proposed. The proposed method is based on evolutionary algorithms. The main criterion used to screen alternative DCS's is the Total Annual Cost (TAC). In order to estimate the TAC of each DCS alternative all columns that exist in the DCS are designed using short‐cut methods. The performance of the proposed method and other alternatives are compared based on the results obtained for four standard benchmark problems used by researchers working in this area. Based on the results of the comparison, the proposed method outperforms the other methods and is also more flexible than other existing methods.     

Model based control of a high purity distillation column

In this paper, the problem of dual product composition control of an industrial high purity distillation column, a deisohexanizer (DIH), is addressed using a Generic Model Control framework. A dynamic simulation of the DIH was performed for preliminary studies of the performance of different controller strategies/algorithms. The performance of Generic Model Control incorporating different process models was studied. Process models are presented ranging from simple first order approximations to mechanistic short cut distillation models where a tradeoff between model complexity and model adaptivity is investigated. The different controllers were implemented and compared using a dynamic simulation of an industrial deisohexanizer (DIH) to select the best condidate controller. A controller using a nonlinear process model emerged as the best controller and was implemented on the actual process, resulting in improved performance over the original controller. Simulation results and industrial plant data are presented.     

基于集合算法的高纯度精馏塔分段建模方法

多级汽液平衡过程的严格模型变量维数高,用于构建优化命题时不易收敛且计算速度较慢。采用集合算法中基于吸收因子和理论塔板数计算全塔段组分回收率的组分分配特征方程代替逐板相平衡,加入物料衡算等基本方程与若干简化关系匹配模型自由度,建立可用于联立计算的逐段降维GM模型。同时根据高纯度精馏塔吸收因子沿塔分布的强非线性特征,提出分段线性拟合求解组分回收率的线性GM模型。以空分装置的粗氩塔为例进行高纯度精馏塔的严格模型、降维GM模型和线性GM模型的对比计算。结果表明,降维GM模型的计算变量由逐板模型的2377个减少到34个,同样平台下收敛时间由6136 s缩短至1 s左右,而关键组分纯度模拟与严格机理模型一致。     

Generalized short-cut distillation column modeling for superstructure-based process synthesis

We introduce a generalized Underwood method to calculate the minimum vapor flowrate for distillation in superstructure-based process synthesis. Our method addresses two important limitations of current methods, namely, separation tasks where (1) the flow rates of some components may be zero, and (2) the key components may be unknown prior to optimization. This is achieved through the introduction of binary variables and mixed-integer constraints to implicitly determine the active roots and the root loci. In addition, we develop strategies tailored to improve the numerical stability and computational performance when using the proposed method.     

Optimization-based design of reactive distillation columns using a memetic algorithm

The design optimization of reactive distillation columns (RDC) is characterized by complex nonlinear constraints, nonlinear cost functions, and the presence of many local optima. The standard approach is to use MINLP solvers that work on a superstructure formulation where structural decisions are represented by discrete variables and lead to an exponential increase in the computational effort. The mathematical programming (MP) methods which solve the continuous sub-problems provide only one local optimum which depends strongly on the initialization. In this contribution a memetic algorithm (MA) is introduced and applied to the global optimization of four different formulations of a computational demanding real-world design problem. An evolution strategy addresses the global optimization of the design decisions, while continuous sub-problems are efficiently solved by a robust MP solver. The MA is compared to MINLP techniques. It is the only algorithm that finds the global solution in reasonable times for all model formulations.