分离系统与换热网络同步优化方法的改进：神经元网络—遗传算法

采用神经元网络法和遗传算法，在过程系统用能一致性的基础上对分离系统与换热网络同步优化问题提出了改进的优化模型及优化策略。该方法不仅能够自动、迅速地同步得到分离序列与换热网络联合系统的流程结构与操作参数，而且具有获得全局最优解的能力。最后通过实例说明本方法的有效性。     

THE FUZZY NEURAL NETWORK CONTROLLER IN YEAST FED-BATCH FERMENTATION

On the basis of the optimized substrate feeding rate of glucose acquired from an artificial neural network model of yeast fed-batch fermentation, an adaptive fuzzy neural network controller (FNNC) was developed, and on-line experiments were performed to examine the feasibility of the controller for actual use. The objective is to obtain a high concentration of cell mass in the shortest fed-batch culture time. Experimental results reveal that FNNC-controlled on-line feeding of substrate performs well.     

Application of Fractal Theory for Prediction of Shrinkage of Dried Kiwifruit Using Artificial Neural Network and Genetic Algorithm

In current research, fractal theory has been applied for estimation of shrinkage of osmotically dehydrated and air-dried kiwifruit using a combination of neural network and genetic algorithm. Kiwifruits were dehydrated at different conditions and digital images of final dried products were taken. Kiwifruit-background interface lines were detected using a threshold combined with an edge detection approach and their corresponding fractal dimensions were calculated based on a box counting method. A neural network was constructed using fractal dimension and moisture content as inputs to predict shrinkage of dried kiwifruit and a genetic algorithm was applied for optimization of the neural network's parameters. The results indicated good accuracy of optimal model (correlation coefficient of 0.95) and high potential application of fractal theory and described intelligent model for shrinkage estimation of dried kiwifruit.     

基于神经网络和遗传算法的塑料热压成型多目标优化

论述了人工神经网络和遗传算法在塑料热压成型工艺优化中的应用,首先利用人工神经网络建立热压成型工艺参数与零件性能之间关系的数学模型,然后用遗传算法对工艺参数优化。根据多目标函数优化问题的单目标化思想,对优化后的单目标进行分解,得到最优工艺参数条件下的塑料热压产品性能,从而为建立和控制塑料热压成型工艺参数提供了一种行之有效的方法。     

广义回归神经网络的改进及在延迟焦化建模中的应用

广义回归神经网络（GRNN）具有明确的概率意义,其参数大多能自动确定,仅光滑因子参数需优化估值.采用优进遗传算法（EGA）,将确定性与随机性寻优操作相融合,实现了高效全局搜优,它所基于的优进策略包括设计Powell寻优算子、改进交叉算子、自适应地调整交叉率和变异率等.以推广能力作为优化目标,所建的GRNN有很强的非线性拟合能力和优良的预报性能,将其成功地为延迟焦化过程建模,与径向基网络（RBFN）等相比,显示了明显的优势.     

基于遗传神经网络的微滤膜通量的预测

针对BP神经网络在寻优过程中容易陷入局部极小值的缺点,将遗传算法和BP神经网络相结合,构造了一种基于遗传算法的反向传播（GABP）神经网络。网络的训练分为两步：首先利用遗传算法群体寻优策略,采用遗传算法对网络权值和阈值进行全局搜索,保证其落入全局最优点的邻域;然后再用梯度法对网络权值进行细化训练以进一步减小误差,使其收敛于全局最优解或性能很好的近似最优解。网络训练时使用的数据是在不同操作条件 （温度、压力、浓度）下,用标准孔径为0.1 μm 的亲水聚偏氟乙烯微滤膜过滤牛血清白蛋白(BSA)溶液时得到的膜通量实验数据,用训练好的GABP神经网络对微滤膜过滤BSA的膜通量预测研究结果表明,与传统的BP算法相比,GABP神经网络算法改善了网络收敛速度以及膜通量预测的准确度。     

应用群能量恒定粒子群优化算法的批次流加发酵过程批次间优化(英文)

An iterative optimization strategy for fed-batch fermentation process is presented by combining a run-to-run optimization with swarm energy conservation particle swarm optimization (SEC-PSO). SEC-PSO, which is designed with the concept of energy conservation, can solve the problem of premature convergence frequently appeared in standard PSO algorithm by partitioning its population into several sub-swarms according to the energy of the swarm and is used in the optimization strategy for parameter iden-tification and operation condition optimization. The run-to-run optimization exploits the repetitive nature of fed-batch processes in order to deal with the optimal problems of fed-batch fermentation process with inaccurate process model and unsteady process state. The kinetic model parameters, used in the operation condition optimization of the next run, are adjusted by calculating time-series data obtained from real fed-batch process in the run-to-run optimization. The simulation results show that the strategy can adjust its kinetic model dynamically and overcome the instability of fed-batch process effectively. Run-to-run strategy with SEC-PSO provides an effective method for optimization of fed-batch fermentation process.     

前馈神经网络与遗传算法相结合解决曲轴中心缩孔

引言缩痕与孔洞是塑件成型时一种常见的缺陷。当制件外层材料冷却固化后,心部材料开始冷却,心部收缩把制件表层拉向制件内部,引起缩痕,如果制件表层的刚度足够大,则将在制件内部产生孔洞而不是缩痕[1]。关于塑件成型过程中的缩痕、翘曲、收缩等缺陷问题,许多学者提出了工艺参数设计的单目标和多目标优化模型,如Kriging模型、BP神经网络、响应面法、支持向量回归等,针对这些模型,采用的优化求解算法主要有:遗传算法、粒子群法、蚁群算法等。申长雨等[2]采用神经网络与混合遗传算法结合优化注塑成型工艺,改善了制品的体收缩     

热偶精馏过程模拟优化方法的改进——人工神经网络-遗传算法

采用人工神经网络和遗传算法对热偶精馏分离过程提出了一种新的建模方法和优化算法,该方法不仅能够有效地求解热偶精馏过程的数学模型,迅速地得到优化变量和目标函数的解,而且具有获得全局最优解的能力.最后通过实例说明了本方法的有效性.     

利用人工神经网络和遗传算法对热偶精馏过程进行模拟优化

In this paper, a new approach using artificial neural network and genetic algorithm for the optimization of the thermally coupled distillation is presented. Mathematical model can be constructed with artificial neural network based on the simulation results with ASPEN PLUS. Modified genetic algorithm was used to optimize the model. With the proposed model and optimization arithmetic, mathematical model can be calculated, decision variables and target value can be reached automatically and quickly. A practical example is used to demonstrate the algorithm.     

Performance evaluation of an ANN–GA aided experimental modeling and optimization procedure for enhanced synthesis of marine biosurfactant in a stirred tank reactor

BACKGROUND: An improved resilient back‐propagation neural network modeling coupled with genetic algorithm aided optimization technique was employed for optimizing the process variables to maximize lipopeptide biosurfactant production by marine Bacillus circulans. RESULTS: An artificial neural network (ANN) was used to develop a non‐linear model based on a 2⁴ full factorial central composite design involving four independent parameters, agitation, aeration, temperature and pH with biosurfactant concentration as the process output. The polynomial model was optimized to maximize lipopeptide biosurfactants concentration using a genetic algorithm (GA). The ranges and levels of these critical process parameters were determined through single‐factor‐at‐a‐time experimental strategy. Improved ANN‐GA modeling and optimization were performed using MATLAB v.7.6 and the experimental design was obtained using Design Expert v.7.0. The ANN model was developed using the advanced neural network architecture called resilient back‐propagation algorithm. CONCLUSION: Process optimization for maximum production of marine microbial surfactant involving ANN‐GA aided experimental modeling and optimization was successfully carried out as the predicted optimal conditions were well validated by performing actual fermentation experiments. Approximately 52% enhancement in biosurfactant concentration was achieved using the above‐mentioned optimization strategy. © 2012 Society of Chemical Industry     

Optimal Temperature and pH Control for a Batch Simultaneous Saccharification and Co-Fermentation Process

Crisp and fuzzy optimization approaches were applied to design an optimal temperature and pH control policy for a batch process of simultaneous saccharification and co-fermentation (SSCF) for ethanol production from lignocellulose, using the enzyme and recombinant strain Zymomonas mobilis ZM (pZB5). To determine an optimal temperature and pH control policy, we applied the Arrhenius relationship to each rate constant to express the temperature and pH effects in the kinetic model for both saccharification and fermentation. The goal of the optimal design was to determine the optimal temperature, pH value, initial lignocellulosic concentration, and fermentation time for maximizing the ethanol productivity under the constraints of the follow-up separation specifications. The interactive crisp and fuzzy optimization methods were applied to solve the trade-off optimization problems for obtaining a compromised design. The fuzzy goal attainment approach obtained a compromised design more flexibly than did the crisp optimization. We also compared the performances for batch and fed-batch SSCF, and used various composition proportions for the batch SSCF to determine a series of optimal designs for the fuzzy goal attainment problem. Batch SSCF was slightly more effective than fed-batch fermentation, and spruce exhibited the maximum productivity because of its higher cellulose and lower hemicellulose contents compared with those of other sources.     

Expected energy analysis for industrial process planning problem with fuzzy time parameters

Industrial process planning is to make an optimal decision in terms of resource allocation. The planning objective can be to minimize the time required to complete a task, maximize customer satisfaction by completing orders in a timely fashion and minimize the cost required to complete a task. Based on time and energy consumption in an industrial process planning problem, a novel energy analysis method is proposed to solve it. According to different constraints and credibility theory, typical expected value models of energy for it are presented. In addition, a hybrid intelligent optimization algorithm integrating fuzzy simulation, neural network and genetic algorithm is provided for solving the proposed expected value models. Some numerical examples are also given to illustrate the proposed concepts and the effectiveness of the used algorithm.     

Application of genetic algorithm to self-organizing fuzzy controller in fed-batch culture ofScutellaria baicalensis G. plant cell

A self-organizing fuzzy controller is constructed for control of substrate concentration in fed-batch operation of a cell culture process. A genetic algorithm is used to generate fuzzy rules of the self-organizing fuzzy controller and to modify the universe of discourse automatically. The fuzzy controller is designed for the application ofScutellaria baicalensis G. plant cell culture process as a model system. A substrate feeding strategy of a two stage culture method to maximize flavone glycoside production in fed-batch culture ofScutellaria baicalensis G. plant cell is proposed based on structured model of growth and product synthesis. As a two stage culture, the feeding strategy consists of the first period at 22 g/L of glucose concentration to promote cell growth and the second period at 10 g/L of glucose concentration to promote flavone glycoside synthesis. The designed self-organizing fuzzy controller is applied to regulate the glucose concentration at a given set-point to increase flavone glycoside synthesis. The simulation results show that the proposed feeding strategy in a fed-batch culture enhances flavone glycoside production and the self-organizing fuzzy logic controller generated by genetic algorithm improves controller performance.     

基于自适应模糊神经网络的非线性系统模型预测控制

针对非线性动态系统的控制问题,提出了一种基于自适应模糊神经网络(adaptive fuzzy neural network, AFNN)的模型预测控制(model predictive control, MPC)方法。首先,在离线建模阶段,AFNN采用规则自分裂技术产生初始模糊规则,采用改进的自适应LM学习算法优化网络参数;然后,在实时控制过程,AFNN根据系统输出和预测输出之间的误差调整网络参数,从而为MPC提供一个精确的预测模型;进一步,AFNN-MPC利用带有自适应学习率的梯度下降寻优算法求解优化问题,在线获取非线性控制量,并将其作用到动态系统实施控制。此外,给出了AFNN-MPC的收敛性和稳定性证明,以保证其在实际工程中的成功应用。最后,利用数值仿真和双CSTR过程进行实验验证。结果表明,AFNN-MPC能够取得优越的控制性能。     

基于混合遗传算法的人工神经网络模型及其对有机化合物熔点的预测

为了避免BP神经网络在训练过程中收敛于局部极小的缺陷,采用自适应交叉变异、最优保存的混合遗传算法对BP网络的权值和阈值进行优化,从而提出一种新的基于混合遗传算法的神经网络模型.该算法首先对一给定的网络结构,采用混合自适应交叉变异和最优保存策略,取各自的长处,用尽可能少的搜索代数找到问题的最优解,从而既防止算法陷入局部最优,又保证算法有较好的平均适应值和最佳的适应值个体.采用上述优化策略的人工神经网络可明显改善收敛的稳定性和收敛速度,并确保网络收敛于全局极小点.人工神经网络运用于物性数据的预测是一个具有潜力和有待开发的领域.运用该模型,根据有机化合物的分子量、临界密度、正常沸点和偶极矩,对其熔点进行预测.预测结果表明：提出的混合遗传算法神经网络优于其他算法神经网络,而且预测结果优于文献上已有的Joback方程和许氏方程的计算值.     

Modeling and Optimization of Anode‐Supported Solid Oxide Fuel Cells on Cell Parameters via Artificial Neural Network and Genetic Algorithm

An artificial neural network (ANN) and a genetic algorithm (GA) are employed to model and optimize cell parameters to improve the performance of singular, intermediate‐temperature, solid oxide fuel cells (IT‐SOFCs). The ANN model uses a feed‐forward neural network with an error back‐propagation algorithm. The ANN is trained using experimental data as a black‐box without using physical models. The developed model is able to predict the performance of the SOFC. An optimization algorithm is utilized to select the optimal SOFC parameters. The optimal values of four cell parameters (anode support thickness, anode support porosity, electrolyte thickness, and functional layer cathode thickness) are determined by using the GA under different conditions. The results show that these optimum cell parameters deliver the highest maximum power density under different constraints on the anode support thickness, porosity, and electrolyte thickness.     

基于混合神经网络与UG/KF方法的浇口位置优化设计

在分析遗传算法特点的基础上,针对遗传算法优化搜索过程中收敛速度慢和不成熟收敛的缺点,提出一种改进的实数编码混合遗传算法,借助于知识熔接技术即KF语言进行几何建模,建立基于混合神经网络与UG/KF方法的浇口位置优化设计,利用CAE软件进行模拟,获取训练样本,用Matlab语言编制应用程序,运用参数优化系统对浇口的位置进行优化计算,结合UG/KF,实现知识驱动.     

航空发动机磨损趋势预测模型研究

文章讨论了神经网络的BP算法和遗传算法,提出用遗传算法来优化BP神经网络,应用遗传算法训练神经网络权重,实现网络结构的优化,用优化后的BP人工神经网络建立了航空发动机磨损故障趋势预测模型,利用发动机的光谱监测数据作为预测磨损趋势的特征参数,进行了模型的训练和预测试验,并将该模型预测结果与BP算法和多元线性回归法的预测结果进行了比较,证明了基于遗传算法的人工神经网络是航空发动机磨损故障趋势预测的一种理想方法。     

Optimal control of batch processes using particle swam optimisation with stacked neural network models

An optimal control strategy for batch processes using particle swam optimisation (PSO) and stacked neural networks is presented in this paper. Stacked neural network models are developed form historical process operation data. Stacked neural networks are used to improve model generalisation capability, as well as provide model prediction confidence bounds. In order to improve the reliability of the calculated optimal control policy, an additional term is introduced in the optimisation objective function to penalize wide model prediction confidence bounds. The optimisation problem is solved using PSO, which can cope with multiple local minima and could generally find the global minimum. Application to a simulated fed-batch process demonstrates that the proposed technique is very effective.