基于PSO-控制变量参数化混合策略的间歇化工过程优化控制

doi:10.11949/j.issn.0438-1157.20181140

摘要/Abstract

摘要：

控制变量参数化方法作为一种化工过程动态优化的梯度搜索算法，其求解效率过于依赖初始给定轨迹。目前初始轨迹一般都是设定在边界值或中间值，缺乏科学依据，从而大大影响了算法的收敛速度。针对这一问题，提出了一种粒子群优化(PSO)与控制变量参数化方法混合的策略，首先利用粒子群优化对间歇化工过程最优控制量进行求解，结果作为控制变量参数化方法初始给定轨迹，进行二次优化。双层优化的混合策略提高了控制变量参数化方法的收敛速度和粒子群优化算法的求解精度。将混合策略应用于两个间歇化工过程优化控制实例，仿真结果表明了该算法对求解化工过程动态优化问题具有可行性和有效性。

关键词: 间歇式, 过程控制, 控制变量参数化, 最优控制, 优化

Abstract:

As a gradient search algorithm for dynamic optimization of chemical process, the efficiency of control vector parameterization depends on the initial given trajectory deeply. At present, the initial trajectory is usually set at the boundary value or the intermediate value, which does not have enough scientific reason and it affects the convergence speed of the algorithm. To solve this problem, a hybrid strategy combined with particle swarm optimization and control vector parameterization method is proposed in this paper, it uses particle swarm optimization to achieve the value of control variables before employing the method of control vector parameterization to reoptimize the process. The two-layer optimization hybrid strategy improves the convergence speed of the control vector parameterization method and the precision of the particle swarm optimization. The hybrid strategy is applied to two examples of batch chemical process optimization control, and the simulation results show that the algorithm is feasible and effective for solving dynamic optimization problems of chemical process.

Key words: batchwise, process control, control vector parameterization, optimal control, optimization

中图分类号:

TQ 021.8

石博文, 尹燕燕, 刘飞. 基于PSO-控制变量参数化混合策略的间歇化工过程优化控制[J]. 化工学报, 2019, 70(3): 979-986.

Bowen SHI, Yanyan YIN, Fei LIU. Optimal control strategies combined with PSO and control vector parameterization for batchwise chemical process[J]. CIESC Journal, 2019, 70(3): 979-986.

图/表 14

图1 分段常量参数化

Fig.1 Piecewise constant parameterization

图2 PSO-CVP算法流程图

Fig.2 Flowchart of PSO-CVP algorithms

图3 最优温度控制轨迹

Fig.3 Optimal temperature control trajectory

图4 最优状态变量轨迹

Fig.4 Optimal state variable trajectory

表1 PSO-CVP算法设置参数

Table 1 Setting parameters of PSO-CVP algorithm

Parameter	Value
quantity of swarm	100
dimension of particle	25
maximum number of iterations	100
accelerated factor c1	2
accelerated factor c2	2
inertia weight	0.8

表2 优化结果比较

Ref.	Method	Optimum
[14]	SQP	0.610775
[15]	ACSO	0.61045
[16]	OC	0.61
[17]	IACA(N=10)	0.61
[17]	IACA(N=20)	0.6104
this work	PSO-CVP	0.6105359

图5 Lee-Ramirez生化反应器

Fig.5 Lee-Ramirez biochemical reactor

表3 Lee-Ramirez反应器参数

Table 3 Lee-Ramirez reactor parameters

Parameter	Meaning	Value
$C n f$ /(g/L)	nutrients feed	100
k _s/(g²/L²)	matrix inhibition	111.5
Y	increased quantities of yield	0.51
k _CN/(g/L)	constant	14.35
f _IO/(g/L)	constant	0.005
k ₁₁/h^-1	constant	0.09
μ _max/h^-1	maximum growth rate	1.0
f _max/h^-1	maximum protein yield	0.233
k _cr/(g/L)	constant	0.22
k ₁ _x /(g/L)	constant	0.034
k_I /(g/L)	constant	0.022

表3 Lee-Ramirez反应器参数

Table 3 Lee-Ramirez reactor parameters

Parameter	Meaning	Value
$C n f$ /(g/L)	nutrients feed	100
k _s/(g²/L²)	matrix inhibition	111.5
Y	increased quantities of yield	0.51
k _CN/(g/L)	constant	14.35
f _IO/(g/L)	constant	0.005
k ₁₁/h^-1	constant	0.09
μ _max/h^-1	maximum growth rate	1.0
f _max/h^-1	maximum protein yield	0.233
k _cr/(g/L)	constant	0.22
k ₁ _x /(g/L)	constant	0.034
k_I /(g/L)	constant	0.022

图6 最优控制轨迹

Fig.6 Optimal control trajectory

图7 最优状态轨迹

Fig.7 Optimal state variable trajectory

图8 基质x 3的状态轨迹

Fig.8 Substrate state variable trajectory of x 3(t)

图9 外源蛋白的变化量

Fig.9 Mass of foreign protein

表4 PSO-CVP算法设置参数

Table 4 Setting parameters of PSO-CVP algorithm

Parameter	Value
quantity of swarm	200
dimension of particle	30
maximum number of iterations	500
accelerated factor c1	2
accelerated factor c2	2
inertia weight	0.8

表5 优化结果比较

Ref.	Method	Optimum
[19]	FIDP	6.16
[20]	GA	6.1504
[21]	biogeographic algorithm	6.15
[22]	CVP	6.15123355
[23]	multiple shooting	6.15153759
this work	PSO-CVP	6.15154923

参考文献 23

1	庄绪君, 李宏光 . 基于遗传算法与迭代动态规划混合策略的青霉素发酵过程优化控制[J]. 计算机与应用化学, 2013, 30(9): 1051-1054.
	Zhuang X J , Li H G . Optimal control of penicillin fermentation process based on genetic algorithm and iterative dynamic programming[J]. Computer and Applied Chemistry, 2013, 30(9):1051-1054.
2	Pollard J P , Sargent R W H . Off line computation of optimum controls for a plate distillation column[J]. Automatica, 1970, 6(1): 59-76.
3	Sargent R W H , Sullivan G R . The development of an efficient optimal control package[J]. Nonlinear and Stochastic Programming, 1979, 13(7): 158-168.
4	Vassiliadis V S , Sargent R W H , Pantelides C C . Solution of a class of multistage dynamic optimization problems(Ⅱ): Problems with path constraints[J]. I&EC Research, 1994, 33(9): 2123-2133.
5	Martin S , Klaus S , Thomas B , et al . Dynamic optimization using adaptive control vector parameterization[J]. Computers & Chemical Engineering, 2005, 29(8): 1731-1751.
6	Wang P , Tian X M . Enhanced control vector parameterization method and its application in dynamic optimization[J]. Control & Decision, 2009, 24(11): 1757-1760.
7	Chen X , Du W L , Huaglory T , et al . Dynamic optimization of industrial processes with nonuniform discretization-based control vector parameterization[J]. IEEE Transactions on Automation Science & Engineering, 2014, 11(4): 1289-1299.
8	Teo K L , Jennings L S , Lee H W J , et al . The control parametrization enhancing transform for constrained optimal control problems[J].J. Austral. Math. Soc. Sen. B, 1999, 40: 314-335.
9	Li G D , Liu P , Liu X G . A control parameterization approach with variable time nodes for optimal control problems[J]. Asian Journal of Control, 2016, 18(3): 976-984.
10	Binder T , Cruse A , Cruz Villar C A , et al . Dynamic optimization using a wavelet based adaptive control[J]. Computers and Chemical Engineering, 2000, 24: 1201-1207.
11	Liu P , Li G D , Liu X G . Fast engineering optimization: a novel highly effective control parameterization approach for industrial dynamic processes[J]. Isa Transactions, 2015, 58: 248-254.
12	Sun F , Zhong W M , Cheng H , et al . Novel control vector parameterization method with differential evolution algorithm and its application in dynamic optimization of chemical processes[J]. Chinese Journal of Chemical Engineering , 2013, 21(1): 64-71.
13	王东风, 孟丽 . 粒子群优化算法的性能分析和参数选择[J]. 自动化学报, 2016, 42(10): 1552-1561.
	Wang D F , Meng L . Performance analysis and parameter selection of particle swarm optimization algorithm [J]. Journal of Automation, 2016, 42(10): 1552-1561.
14	Logsdon J S , Biegler L T . A relaxed reduced space SQP strategy for dynamic optimization problems[J]. Computers & Chemical Engineering, 1993, 17(4): 367-372.
15	Rajesh J , Gupta K , Kusumakar H S , et al . Dynamic optimization of chemical processes using ant colony framework[J]. Computers & Chemistry, 2001, 25(6): 583-595.
16	Renfro J G , Morshedi A M , Asbjornsen O A . Simultaneous optimization and solution of systems described by differential algebraic equations[J]. Computers & Chemical Engineering, 1987, 11(5): 503-517.
17	Zhang B , Chen D Z , Zhao W . Iterative ant-colony algorithm and its application to dynamic optimization of chemical process[J]. Computers & Chemical Engineering, 2005, 29(10): 2078-2086.
18	Lee J , Ramirez W F . Optimal fed-batch control of induced foreign protein production by recombinant bacteria[J]. AIChE Journal, 1994, 40(5): 899-907.
19	Tholudur A , Ramirez W F . Obtaining smoother singular arc policies using a modified iterative dynamic programming algorithm[J]. International Journal of Control, 1997, 68(5): 1115-1128.
20	Sarkar D , Modak J M . Optimization of fed-batch bioreactors using genetic algorithm: multiple control variables[J]. Computers & Chemical Engineering, 2004, 28(5): 789-798.
21	Nikumb S , Ghosh S , Jayaraman V K , et al . Biogeography-based optimization for dynamic optimization of chemical reactors[M]// Applications of Mataheuristics in Process Engineering, Springer ,2014: 201-216.
22	Hirmaje T , Balsacanto E , Banga J R . DOTcvpSB, a software toolbox for dynamic optimization in systems biology[J]. BMC Bioinformatics, 2009, 10: 199.
23	黎向宇 . 基于时间网格重构的多重打靶最优控制策略研究[D]. 杭州: 浙江大学, 2017.
	Li X Y . Study on optimal control strategy for multiple targets based on time grid reconstruction[D]. Hangzhou: Zhejiang University, 2017.

[1]	杨欣, 王文, 徐凯, 马凡华. 高压氢气加注过程中温度特征仿真分析[J]. 化工学报, 2023, 74(S1): 280-286.
[2]	康飞, 吕伟光, 巨锋, 孙峙. 废锂离子电池放电路径与评价研究[J]. 化工学报, 2023, 74(9): 3903-3911.
[3]	何松, 刘乔迈, 谢广烁, 王斯民, 肖娟. 高浓度水煤浆管道气膜减阻两相流模拟及代理辅助优化[J]. 化工学报, 2023, 74(9): 3766-3774.
[4]	陈哲文, 魏俊杰, 张玉明. 超临界水煤气化耦合SOFC发电系统集成及其能量转化机制[J]. 化工学报, 2023, 74(9): 3888-3902.
[5]	曹跃, 余冲, 李智, 杨明磊. 工业数据驱动的加氢裂化装置多工况切换过渡状态检测[J]. 化工学报, 2023, 74(9): 3841-3854.
[6]	齐聪, 丁子, 余杰, 汤茂清, 梁林. 基于选择吸收纳米薄膜的太阳能温差发电特性研究[J]. 化工学报, 2023, 74(9): 3921-3930.
[7]	陈国泽, 卫东, 郭倩, 向志平. 负载跟踪状态下的铝空气电池堆最优功率点优化方法[J]. 化工学报, 2023, 74(8): 3533-3542.
[8]	刘文竹, 云和明, 王宝雪, 胡明哲, 仲崇龙. 基于场协同和耗散的微通道拓扑优化研究[J]. 化工学报, 2023, 74(8): 3329-3341.
[9]	邢雷, 苗春雨, 蒋明虎, 赵立新, 李新亚. 井下微型气液旋流分离器优化设计与性能分析[J]. 化工学报, 2023, 74(8): 3394-3406.
[10]	张曼铮, 肖猛, 闫沛伟, 苗政, 徐进良, 纪献兵. 危废焚烧处理耦合有机朗肯循环系统工质筛选与热力学优化[J]. 化工学报, 2023, 74(8): 3502-3512.
[11]	诸程瑛, 王振雷. 基于改进深度强化学习的乙烯裂解炉操作优化[J]. 化工学报, 2023, 74(8): 3429-3437.
[12]	吴文涛, 褚良永, 张玲洁, 谭伟民, 沈丽明, 暴宁钟. 腰果酚生物基自愈合微胶囊的高效制备工艺研究[J]. 化工学报, 2023, 74(7): 3103-3115.
[13]	汤晓玲, 王嘉瑞, 朱玄烨, 郑仁朝. 基于Pickering乳液的卤醇脱卤酶催化合成手性环氧氯丙烷[J]. 化工学报, 2023, 74(7): 2926-2934.
[14]	文兆伦, 李沛睿, 张忠林, 杜晓, 侯起旺, 刘叶刚, 郝晓刚, 官国清. 基于自热再生的隔壁塔深冷空分工艺设计及优化[J]. 化工学报, 2023, 74(7): 2988-2998.
[15]	江锦波, 彭新, 许文烜, 门日秀, 刘畅, 彭旭东. 泵出型螺旋槽油气密封泄漏特性及参数影响研究[J]. 化工学报, 2023, 74(6): 2538-2554.