基于改进蚁群算法的催化裂化分馏塔在线多目标优化

为了对催化裂化分馏塔多目标问题在线智能优化,本文选择可以在线催化裂化分馏塔多目标的优化模型,同时提出一种新的改进蚁群算法,作为智能优化手段.对分馏塔的优化过程作单位化和数字位的处理,通过研究优化过程中参数的取值对优化的影响,确定一组合适的参数组合,去优化分馏塔.实验数据表明:优化结果较好并大大缩短了优化时间,为在线优化分馏塔提供行之有效的方法.     

动态电源管理的随机切换模型与在线优化

考虑系统参数未知情况下的动态电源管理问题，提出一种基于强化学习的在线策略优化算法. 通过建立事件驱动的随机切换分析模型，将动态电源管理问题转化为带约束的Markov 决策过程的策略优化问题. 利用此模型的动态结构特性，结合在线学习估计梯度与随机逼近改进策略，提出动态电源管理策略的在线优化算法．随机切换模型对电源管理系统的动态特性描述精确，在线优化算法自适应性强，运算量小，精度高，具有较高的实际应用价值．     

加热炉加热过程优化控制系统的研究与应用

针对鞍钢线材加热炉存在的问题,建立了在线实时专家系统主控模型和基于传感器的在线实时最佳燃烧状态、空燃比例、炉膛炉压等辅助动态优化模型.针对加热炉各种工程情况下的主要过程参数,设计了加热炉专家模型数据库,绘制了控制系统结构及工作原理图;经过多次实际操作、反复修正,确定系统能够达到预期的功能,实现了加热炉加热过程优化控制;改善了加热炉的生产状况,提高了产品的质量,具有一定的推广价值.     

发酵过程建模与优化控制的研究

针对发酵过程的时变性和强非线性,本研究采用神经网络对发酵过程建模,在线预测某些重要生物参数,并以这些生物参数的预测值为指导,使用遗传算法(GA)在线优化主要的环境参数,确定各参数的最优轨线来控制发酵过程,并构建了由PC机和单片机组成的发酵优化控制的监控系统.实验表明,该控制系统应用于实际发酵生产过程能有效提高发酵生产效率.     

氧化铝蒸发浓度的自适应加权LSSVR预测

针对氧化铝蒸发过程的工业现场出口料液浓度在线检测困难、操作参数具有时变性以及传统离线预测所存在的不足等特点,提出了一种多输入多输出系统的自适应加权最小二乘支持向量回归,并用于氧化铝蒸发过程出口料液浓度的在线预测.谊方法根据模型预测效果自适应在线调整建模的训练样本集,利用主元分析提取主元作为分段加权支持向量回归模型的输入,采用网格搜索和交叉验证法对多输入多输出模型参数进行优化.采用工业现场的实测数据进行实验分析,计算结果表明;该方法能够很好地在线预测氧化铝蒸发过程出口料液浓度,相比基于最小二乘支持向量回归以及基于BP神经网络的浓度预测模型,谊方法具有更高的预测精度和更好的泛化性能,满足实际工业生产在线优化控制要求.     

一种有效的稳态自适应优化方法

０　引　言工业过程,尤其是化工过程的建模大多属于“黑箱子”,其稳态建模往往使用统计建模的方法。一般说来,模型结构并不能准确地反映实际过程。因此,模型作为目标函数进行优化求解,显然得不到实际的最优解。为解决这一问题,通常采用的是用在线优化控制输入修正模型参数,然后用修正后的模型进行优化求解,如此重复进行的二步方法。但是参数估计和优化之间存在关联,只有在模型和实过程对输入的微分匹配时,这种方法才能得到真实最优解［１］。在实过程数学关系未知的情况下,这个梯度条件是不满足的,因而该方法得不到真实最优解。…     

企业过程模型参数优化技术研究*

通过分析企业过程模型参数对过程性能的影响,将其分为对象属性参数和调度策略两类。其中将对象属性参数又分为产品属性参数、资源属性参数和活动属性参数。针对基于仿真的企业过程模型优化方法,确定了优化参数及优化解空间,探讨了各种优化参数的优化方法以及实际应用时需要注意的问题。最后对各类优化参数进行了比较,分析了对其进行优化的重要性差异,指出了采用复杂系统全局优化算法的必要性。     

针铁矿法沉铁过程Fe2+浓度预测建模

针铁矿法沉铁过程中反应器出口的二价铁离子浓度无法在线检测,影响到生产过程的实时优化控制.为此需要建立二价铁离子浓度预测模型,以系统入口参数和控制参数等输入条件为依据,实现预测反应器出口二价铁离子的浓度值.基于化学反应动力学理论与实验研究确立预测模型的结构;对于模型中难以确定的参数,采用蕴含大量工况信息的实际生产数据辨识得到.通过参数输出灵敏度矩阵分析参数模型的敏感度,并将灵敏度信息引入模型参数辨识的目标函数,以提高模型参数的辨识精度.仿真结果表明:预测模型具有较为明确的物理意义,预测结果能很好地跟踪现场生产数据的波动,模型的精度可以满足生产需要.     

糖厂澄清工段生产指标建模及操作参数优化

针对亚法糖厂澄清工段清汁色值和清汁残硫量难以在线测量的问题,提出了一种基于人工蜂群优化的在线极限学习机软测量方法;先用核主元分析法确定影响清汁质量的关键参数,建立基于在线极限学习机的软测量模型;同时利用人工蜂群算法对在线极限学习机的隐层参数进行寻优,优化所建模型;最后,使用带约束的粒子群对软测量模型进行优化求解,得到典型工况下的最优操作设定值,为后续工况操作提供参考依据;仿真结果表明,基于人工蜂群优化的在线极限学习机模型能够准确地预测清汁色值和残硫量,同时基于此模型优化的操作参数设定值能够达到期望的指标.     

基于模糊聚类和矩阵分解的模糊辨识方法

通过改进模糊聚类方法确定模糊模型的前件结构,然后对模糊推理关系矩阵进行QR分解,通过分析秩的亏损来确定聚类规则的有效性,然后采用基于矩阵UD分解最小二乘确定模糊模型的后件参数,实现模糊模型的结构和参数的优化.该方法成功地应用于Box-Jenkins煤气炉数据系统建模.     

Adaptive on-line steady-state optimization of slow dynamic processes

A software package OLIOPT was developed for the on-line optimization of the steady-state behaviour of slow dynamic processes in a relatively short time period. In the starting phase, the independently variable inputs are changed according to a special test signal. A nonlinear dynamic process model is identified on-line. Based on the static part of the model and the known inputs, the gradients of the performance index are calculated. An optimization algorithm changes the inputs towards their optimal values. On-line identification of the nonlinear model continues and the prediction of the optimum improves. In the last phase, the inputs take their optimal values and the process follows, feedforward controlled, to its optimal steady-state. The method is suited for industrial processes with one or more variable inputs, where a small gain in efficiency turns out to give a relatively large financial return. Results are shown for the on-line optimization of a thermal pilot process.     

Advanced control and on-line process optimization in multilayer structures

The paper demonstrates the place, role and mutual interaction of advanced control algorithms and on-line set-point optimization in process control structures. First, a multilayer control structure resulting from a functional decomposition is briefly presented. The role and selected realizations of advanced control algorithms, in particular mostly applied now model predictive control (MPC) ones, at direct control and supervisory constraint control layers is discussed. Then possible solutions to on-line set-point optimization, depending of disturbance dynamics, are presented: dynamic set-point optimization including involved structures based on temporal decomposition, and steady-state set-point optimization for cases with disturbance dynamics both much slower than and comparable with the process dynamics. For the last case, important in industrial practice, different structures of interaction and even integration of MPC and steady-state optimization are discussed. The topics are illustrated by briefly presented examples, selected from given references.     

基于滑动时间窗的支持向量机软测量建模研究

提出了一种基于滑动时间窗的最小二乘支持向量机软测量建模方法，并针对某一工业共沸精馏塔成分估计问题，建立了合适的工业软测量模型。介绍了最小二乘支持向量机计算法和滑动时间窗的建立。并利用滑动时间窗内的数据进行系统的在线优化和周期性模型更新，提高了工业软测量模型的在线估计精度。研究结果表明，LS—SVM是一种有效的软测量建模方法。     

汽驱工业过程优化监控系统设计

以辽河油田汽驱工业过程为对象，设计开发了一套优化监控系统。该系统由操作站、工程师站构建的小型局域网组成，建立了湿蒸汽干度的软测量模型，实现了指标变量的在线检测，并采用遗传算法进行优化计算，实现了对整个汽驱过程的优化操作指导。系统具有良好的人机界面，同时具有较好的可靠性和稳定性，系统各部分之间的数据通讯是基于OPC接口进行的。目前该监控系统已成功运行于辽河油田的注气3站和5站。     

ProOpter: An advanced platform for production analysis and optimization

This paper presents the prototype of an advanced platform for production analysis and optimization, referred to as ProOpter. The platform was developed to support the recently derived concept of holistic production control (HPC), which relies on model-based control. The prototype is comprised of a set of off-line and on-line modules. The off-line modules support the definition of key performance indicators (KPIs), the selection of the most influential input (manipulative) variables, and the identification of a simple production model from historical data. The on-line modules enable KPI prediction and suggest actions to the production manager, employing model-based production control and/or optimization techniques. In this way, a new decision-support reasoning based on historical production data can be introduced. ProOpter has a modular design and can be used as an add-on to existing production IT systems since it relies on established industrial communication standards. The use of the platform is validated on the well-known Tennessee Eastman benchmark simulation process and on two industrial case studies.     

工业大系统双层结构预测控制的集中优化与分散控制策略

为降低工业大系统模型预测控制(Model predictive control,MPC)在线计算复杂度,同时保证系统的全局优化性能,提出一种集中优化、分散控制的双层结构预测控制策略.在稳态目标计算层(Steady-state target calculation, SSTC),基于全局过程模型对系统进行集中优化,将优化结果作为设定值传递给动态控制层;在动态控制层,将大系统划分为若干个子系统,每个子系统分别由基于各自子过程模型的模型预测控制进行控制,为减少各子系统之间的相互干扰,在各个子系统之间添加前馈控制器对扰动进行补偿,提高系统的总体动态控制性能.该策略的优点在于能确保系统全局最优性的同时降低了在线计算量,提高了工业大系统双层结构预测控制方法的实时性.仿真实例验证该方法的有效性.     

基于区间数的黄金湿法冶炼过程建模与优化研究

针对黄金湿法冶炼生产过程中某些关键变量不能准确在线测量，导致局部工序无法定量建模、难以基于定量模型实现过程优化控制的问题，提出一种基于区间数的过程分层优化方法.在对黄金湿法冶炼生产过程特点进行分析的基础上，提出了基于区间数的过程分层优化框架.基于专家知识和现场操作人员经验等信息，建立了调浆过程的模糊定性模型.结合氰化浸出和置换等工序的定量模型及调浆过程的定性模型，建立了以综合经济效益最大为优化目标的黄金湿法冶炼生产过程优化模型.针对模糊定性模型的每一输出模态，利用区间数代替无法检测关键变量，提出了基于区间优化和分层优化思想相结合的优化方法，实现了黄金湿法冶炼过程的优化.与传统全流程优化方法的仿真对比实验表明，所提方法在具有不确定性的流程工业生产过程优化中具有一定的应用价值.     

Plant-wide hierarchical optimization and control of an industrial hydrocracking process

Hydrocracking is a crucial refinery process in which heavy hydrocarbons are converted to more valuable, low-molecular weight products. Hydrocracking plants operate with large throughputs and varying feedstocks. In addition the product specifications change due to varying economic and market conditions. In such a dynamic operating environment, the potential gains of real-time optimization (RTO) and control are quite high. At the same time, real-time optimization of hydrocracking plants is a challenging task. A complex network of reactions, which are difficult to characterize, takes place in the hydrocracker. The reactor effluent affects the operation of the fractionator downstream and the properties of the final products. In this paper, a lumped first-principles reactor model and an empirical fractionation model are used to predict the product distribution and properties on-line. Both models have been built and validated using industrial data. A cascaded model predictive control (MPC) structure is developed in order to operate both the reactor and fractionation column at maximum profit. In this cascade structure, reactor and fractionation units are controlled by local decentralized MPC controllers whose set-points are manipulated by a supervisory MPC controller. The coordinating action of the supervisory MPC controller accomplishes the transition between different optimum operating conditions and helps to reject disturbances without violating any constraints. Simulations illustrate the applicability of the proposed method on the industrial process.     

A hybrid intelligent model for order allocation planning in make-to-order manufacturing

This paper investigated a multi-objective order allocation planning problem in make-to-order manufacturing with the consideration of various real-world production features. A novel hybrid intelligent optimization model, integrating a multi-objective memetic optimization (MOMO) process, a Monte Carlo simulation technique and a heuristic pruning technique, is developed to tackle this problem. The MOMO process, combining a NSGA-II optimization process with a tabu search, is proposed to provide Pareto optimal solutions. Extensive experiments based on industrial data are conducted to validate the proposed model. Results show that (1) the proposed model can effectively solve the investigated problem by providing effective production decision-making solutions; (2) the MOMO process has better capability of seeking global optimum than an NSGA-II-based optimization process and an industrial method.     

基于网络结构搜索的工业过程自动故障诊断方法

针对现有基于深度神经网络的工业过程故障诊断方法存在网络结构设计烦琐及参数寻优耗时等问题,提出了一种基于网络结构搜索的工业过程自动故障诊断方法(automatic fault diagnosis, AutoFD),该方法采用AutoFD网络结构搜索算法,来自动完成卷积神经网络的网络结构设计和网络参数寻优。在此基础上,首先通过在原始数据上施加操作生成新通道;接着利用表现预测加速获取通道适应性排序的过程;然后依据通道适应性排序,通过表现预测来快速选取最优卷积通道数;最终根据最优卷积通道来搜索表现最优的多通道卷积神经网络模型用于工业过程自动故障诊断。采用田纳西—伊斯曼(Tennessee Eastman, TE)工业过程和数值系统对提出方法进行验证,结果表明该方法可以实现网络结构自动设计及网络参数的自动寻优,并且具有优良的故障诊断性能。