文化差分进化算法及其在化工过程建模中的应用

提出了一种新的文化差分进化算法,该算法将差分进化算法作为文化算法的种群空间,在文化算法的信念空间和影响函数设计中提出了基于多种知识源的设计方法,通过多种知识指导差分进化的变异操作和交叉操作,使知识的表达和指导种群进化的能力得到加强。函数测试结果表明,基于知识机制的引入使得文化差分进化算法在寻优性能上比差分进化算法有了较大的提高,而对参数的敏感性却相对较小。将文化差分进化算法用于训练补偿模糊神经网络,建立乙烯精馏塔产品质量软测量模型。通过训练与泛化能力的比较结果表明,基于文化差分进化算法的补偿模糊神经网络软测量模型在建模精度和泛化性能上均优于常规补偿模糊神经网络、模糊神经网络以及采用遗传算法优化的模型,具有更好的应用前景。     

基于差分进化粒子群混合优化算法的软测量建模

针对乙烯生产过程中,用传统方法难以直接完成对乙烯收率的在线测量的问题,提出了一种新型差分进化粒子群混合优化算法,建立了乙烯收率软测量建模。改进算法将优化过程分成两阶段,两分群分别采用粒子群算法和差分进化算法同时进行。迭代过程中引入进化速度因子进行算法局部收敛性判断,通过两个群体间的信息交流阻止算法陷入局部最优。对高维复杂函数寻优测试表明,算法的整体优化性能均强于基本粒子群算法和差分进化算法。应用结果表明,基于改进算法的软测量模型具有测量精度较高、泛化性能较好等优点。     

基于Alopex的进化算法及其在乙炔加氢反应器建模中的应用

分析进化优化算法的共性,参考Alopex优化思想提出一种新的进化优化算法.给出该算法的具体步骤,将该算法与标准粒子群算法、遗传算法和差分算法在性能上做了仿真时比,并将其应用于乙炔加氢反应器出口乙炔浓度软测量的建模中.这种新的优化算法具有全局收敛能力,并具有较快的收敛速度.对典型函数的测试和基于神经网络的软测量建模表明:新算法具有较强的全局搜索能力,特别是对易于陷入局部最优的多峰函数能够有效地避免早熟收敛问题.     

基于改进PSO-RBFNN的海洋蛋白酶发酵过程软测量

针对海洋蛋白酶（marine protease,MP）发酵过程中某些关键参量难以在线检测,离线测量存在大滞后、易染菌的问题,提出了一种基于改进的粒子群-径向基神经网络（PSO-RBFNN）的MP发酵过程软测量建模方法。首先采用指数下降惯性权重（exponential decreasing inertia weight,EDIW）策略对粒子群算法进行改进,克服了固定惯性权重和自适应惯性权重的粒子群算法易于陷入局部极小,进化后期收敛速度慢以及全局搜索能力弱的缺点;然后,采用改进后的粒子群算法对径向基神经网络连接权值进行在线优化,确定RBFNN拓扑结构;最后,根据MP发酵过程的输入/输出向量构建RBFNN软测量模型。实验仿真结果表明,EDIW策略改进的PSO-RBFNN软测量模型训练时间缩短了40%左右,模型预测精度提高了3%以上。     

基于改进PSO-RBFNN的海洋蛋白酶发酵过程软测量

针对海洋蛋白酶(marine protease,MP)发酵过程中某些关键参量难以在线检测,离线测量存在大滞后、易染菌的问题,提出了一种基于改进的粒子群-径向基神经网络(PSO-RBFNN)的MP发酵过程软测量建模方法。首先采用指数下降惯性权重(exponential decreasing inertia weight,EDIW)策略对粒子群算法进行改进,克服了固定惯性权重和自适应惯性权重的粒子群算法易于陷入局部极小,进化后期收敛速度慢以及全局搜索能力弱的缺点;然后,采用改进后的粒子群算法对径向基神经网络连接权值进行在线优化,确定RBFNN拓扑结构;最后,根据MP发酵过程的输入/输出向量构建RBFNN软测量模型。实验仿真结果表明,EDIW策略改进的PSO-RBFNN软测量模型训练时间缩短了40%左右,模型预测精度提高了3%以上。     

基于粒子群算法的软仪表技术

在软测量建模中,最常见的非机理建模方式就是利用神经网络进行建模,而近年来兴起的粒子群算法目前已应用于神经网络的训练。在对粒子群算法提出改进方案后,提出了基于改进的粒子群算法的前馈神经网络训练方案。然后再将神经网络应用到焦化装置分流塔柴油95％点软仪表模型参数估计中,得到了满意的结果,可以满足工业过程中的实际需要。     

一种基于Alopex的参数自适应进化算法及其在软测量建模中的应用

提出了一种基于Alopex的参数自适应进化算法(SaAEA)。SaAEA算法将进化分为两个层面,即种群个体利用AEA算法进化,算法参数利用粒子群算法进化,实现参数的自适应调整。并将差分算法中使用的交叉操作引入到AEA算法以改善种群多样性。SaAEA算法在14个典型测试函数上进行了测试,测试结果表明,与基本的AEA算法相比,SaAEA算法寻优性能有了较大的提高,获得的解的质量和收敛速度均有明显提高。最后,将SaAEA算法应用于乙烯裂解深度神经网络软测量建模,得到的模型有较好的泛化能力。     

微粒群神经网络在常压塔汽油干点软测量建模中的应用

首先将微粒群优化算法用于神经网络连接权值和阈值的训练,构造微粒群神经网络,然后将微粒群神经网络用于常压塔汽油干点软测量建模。通过与实际值的对比,结果表明基于微粒群神经网络的软测量模型具有良好的性能和极好的应用前景。     

自适应粒子群优化算法在聚丙烯熔融指数预报上的应用

针对丙烯聚合生产控制中聚丙烯熔融指数在线测量的控制要求,以及过程变量间相关性高的特点,提出一种基于自适应粒子群优化算法和径向基函数神经网络的聚丙烯熔融指数预报新方法。该方法采用变参数的自适应粒子群优化算法提高优化算法的效率和收敛性,并且融合了主成分分析、统计建模以及智能优化方法,从而降低了预报模型的复杂度。提出了一种基于径向基函数神经网络的统计预报模型的参数优化和结构优化方法。使用该统计模型对工厂实际生产过程进行预报,并与国内外相关研究报道相比较,表明了本文所提出的预报方法的有效性和更高的准确性。     

多频率系统动态插值神经网络软测量建模

针对某些化工过程关键变量难以在线测量的问题,提出了一种基于多采样率系统的时间序列神经网络的软测量建模方法,建立了动态插值神经网络模型,并利用增强粒子群算法实现了网络参数的优化。将此方法用于实验室模拟建模,实现了变量的在线预估,并对网络的训练效果和泛化性能进行了分析,表明其建模效果明显优于普通静态神经网络。     

A novel 3-layer mixed cultural evolutionary optimization framework for optimal operation of syngas production in a Texaco coal-water slurry gasifier☆

Optimizing operational parameters for syngas production of Texaco coal-water slurry gasifier studied in this paper is a complicated nonlinear constrained problem concerning 3 BP (Error Back Propagation) neural networks. To solve this model, a new 3-layer cultural evolving algorithm framework which has a population space, a medium space and a belief space is firstly conceived. Standard differential evolution algorithm (DE), genetic algorithm (GA), and parti-cle swarm optimization algorithm (PSO) are embedded in this framework to build 3-layer mixed cultural DE/GA/PSO (3LM-CDE, 3LM-CGA, and 3LM-CPSO) algorithms. The accuracy and efficiency of the proposed hybrid algo-rithms are firstly tested in 20 benchmark nonlinear constrained functions. Then, the operational optimization model for syngas production in a Texaco coal-water slurry gasifier of a real-world chemical plant is solved effective-ly. The simulation results are encouraging that the 3-layer cultural algorithm evolving framework suggests ways in which the performance of DE, GA, PSO and other population-based evolutionary algorithms (EAs) can be improved, and the optimal operational parameters based on 3LM-CDE algorithm of the syngas production in the Texaco coal-water slurry gasifier shows outstanding computing results than actual industry use and other algorithms.     

Application of artificial neural networks for the optimal design of sheet molding compound (SMC) compression molding

A sequential design optimization scheme based on artificial neural networks (ANN) is proposed. It is a combination of an ANN model and a nonlinear programming algorithm. The proposed scheme is implemented with network training, optimization, and sheet molding compound (SMC) process simulation in a closed loop. A “cyclic coordinate search” technique is employed to initiate the optimization process, to collect training data for the neural network model, and to perform a preliminary design sensitivity analysis. Emphasis is placed on the development of an integrated, automatic optimization-simulation design tool that does not rely on the designer's experience and interpretation. Testing results based on the design of heating channels in an SMC compression molding tool show that the optimal design can be achieved with fewer data points than other methods, such as factorial design. The efficiency of the ANN method would be greater as the number of design variables grows.     

基于改进的GSO算法和BP神经网络的氨合成塔出口氨含量软测量模型(英文)

The ammonia synthesis reactor is the core unit in the whole ammonia synthesis production. The ammonia concentration at the ammonia converter outlet is a significant process variable, which reflects directly the production efficiency. However, it is hard to be measured reliably online in real applications. In this paper, a soft sensor based on BP neural network (BPNN) is applied to estimate the ammonia concentration. A modified group search optimization with nearest neighborhood (GSO-NH) is proposed to optimize the weights and thresholds of BPNN. GSO-NH is integrated with BPNN to build a soft sensor model. Finally, the soft sensor model based on BPNN and GSO-NH (GSO-NH-NN) is used to infer the outlet ammonia concentration in a real-world application. Three other modeling methods are applied for comparison with GSO-NH-NN. The results show that the soft sensor based on GSO-NH-NN has a good prediction performance with high accuracy. Moreover, the GSO-NH-NN also provides good generalization ability to other modeling problems in ammonia synthesis production.     

Multi-model predictive control of SCR flue gas denitrification system in coal-fired power plant based on kernel fuzzy c-means clustering and integrated model

A multi-model predictive control strategy based on kernel fuzzy c-means (KFCM) clustering and integrated model is proposed for the complex problem of rapid and accurate control of ammonia injection in selective catalytic reduction (SCR) denitrification systems of coal-fired power plants under a wide range of variable load conditions. First, the SCR data samples are clustered using the KFCM clustering algorithm, and the number of clusters is determined by introducing the Xie-Beni index. Second, the prediction model of the SCR denitrification system is established by an integrated modelling approach, and the sub-learners of the integrated model are the genetic algorithm optimized back propagation (GA-BP) neural network model and the least squares support vector machine (LSSVM) model. Third, a multi-model prediction controller based on the particle swarm optimization (PSO) algorithm and the integrated model is designed and developed. To ensure the stability of the system, a model-switching strategy based on the minimum Euclidean distance is proposed. Finally, simulation verification and industrial field application verification are fulfilled by comparing with proportion integral differential (PID) control and single model predictive control (MPC). The results show that the multi-model predictive control method proposed in this paper can obtain higher control accuracy and better control stability and meet the control requirements for the long-term operation of the SCR denitrification system.     

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MLP网络的智能控制及应用

讨论了一种基于ＭＬＰ网络的智能控制系统。它利用一个多层感知器（ＭＬＰ）作为被控对象的在线辨识器，而用另一线性神经网络结构构成ＰＩＤ控制器。采用变步长ＢＰ加速算法求得多层感知器的各联接权重与阀值的学习规律与控制器参数的自整定算法。通过化肥厂变换炉控制的仿真研究与一列管换热器温度控制系统的仿真试验，说明该算法是有效的。     

Evaluation of neural network models and quality forecasting based on process time-series data

Complex industrial process modelling is critically important within the context of industrial intelligence. In recent years, soft sensor techniques based on neural networks have become increasingly popular for modelling nonlinear industrial processes. This paper proposes an integrated framework of neural network modelling and evaluation for nonlinear dynamic processes. This framework achieves an integrated solution for modelling, prediction, evaluation, and network structure parameter selection. It can be applied to noisy sensors and dense data in the time domain. The framework's proposed evaluation mechanism employs two novel evaluation metrics, the variational auto-encoder (VAE)-based Kullback–Leibler (KL) divergence metric and the maximum likelihood estimation-based J metric, which both evaluate the model by mining the statistical properties of the residuals. The framework models the dynamic process with a model order based-gated recurrent units (MOb-GRU) neural network and a modified transformer model. Numerical experiments demonstrate that the evaluation mechanism functions properly in scenarios with multiple signal-to-noise ratios and multiple noise statistical properties and that the framework produces accurate modelling results.     

高纯度精馏过程的集成控制与在线优化策略

For high-purity distillation processes, it is difficult to achieve a good direct product quality control using traditional pro-portional-integral-differential (PID) control or multivariable predictive control technique due to some difficulties, such as long re-sponse time, many un-measurable disturbances, and the reliability and precision issues of product quality soft-sensors. In this paper, based on the first principle analysis and dynamic simulation of a distillation process, a new predictive control scheme is proposed by using the split ratio of distillate flow rate to that of bottoms as an essential controlled variable. Correspondingly, a new strategy with integrated control and on-line optimization is developed, which consists of model predictive control of the split ratio, surrogate model based on radial basis function neural network for optimization, and modified differential evolution optimization algorithm. With the strategy, the process achieves its steady state quickly, so more profit can be obtained. The proposed strategy has been successfully applied to a gas separation plant for more than three years, which shows that the strategy is feasible and effective.     

Optimization of ODHE membrane reactor based on mixed ionic electronic conductor using soft computing techniques

This work presents the optimization of the operating conditions of a membrane reactor for the oxidative dehydrogenation of ethane. The catalytic membrane reactor is based on a mixed ionic–electronic conducting material, i.e. Ba_0.5Sr_0.5Co_0.8Fe_0.2O_δ−3, which presents high oxygen flux above 750 °C under sufficient chemical potential gradient. Specifically, diluted ethane is fed into the reactor chamber and air (or diluted air) is flushed to the other side of the membrane. A framework based on Soft Computing techniques has been used to maximize the ethylene yield by simultaneously varying five operation variables: nominal reactor temperature (Temp); gas flow in the reaction compartment (QHC); gas flow in the oxygen-rich compartment (QAir); ethane concentration in the reaction compartment (%C₂H₆); and oxygen concentration in oxygen-rich compartment (%O₂). The optimization tool combines a genetic algorithm guided by a neural network model. This shows how the neural network model for this particular problem is obtained and the analysis of its behavior along the optimization process. The optimization process is analyzed in terms of: (1) catalytic figures of merit, i.e., evolution of yield and selectivity towards different products and (2) framework behavior and variable significance. The two experimental areas maximizing the ethylene yield are explored and analyzed. The highest yield reached in the optimization process exceeded 87%.     

A Surrogate Integrated Production Modeling Approach to Long-Term Gas-Lift Allocation Optimization

The conventional static gas-lift allocation optimization approaches are not appropriate for long-term gas-lift projects. A good choice for long-term optimization should predict gas-lift performance dynamically as a function of production time and other variables. A good solution approach for problem is a hybrid of surrogate integrated production modeling and genetic algorithm (GA). Hybrid GAs have received significant interest in recent years and are being increasingly used to solve real-world problems. GA incorporates other techniques within its framework to produce a hybrid that reaps the best from the combination. This study discusses a new method known as surrogate integrated production modeling that uses an artificial neural network to predict gas-lift performance based on a database of oil production. Then, a hybrid of the neural network and GA is used for long-term gas-lift allocation optimization in a group of wells under real constraints.