Reaction network flux analysis: Optimization‐based evaluation of reaction pathways for biorenewables processing

Even though biomass is attracting increasing interest as a raw material in the chemical and the fuel industries, only few biobased production processes are yet established. At the same time a lot of new catalytic routes are proposed, but their potential in biorefinery applications is hard to predict. Reaction network flux analysis (RNFA) is introduced as a novel, rapid screening method which bridges the gap between chemo‐ or biocatalysis and process design by (1) systematically identifying and (2) subsequently analyzing and ranking the large number of alternative reaction pathways based on limited data. This optimization‐based method helps to detect promising production routes as well as bottlenecks in possible pathways. The potential and the application of the RNFA methodology will be demonstrated by means of a case study for the production of the potential biofuel 3‐methyl‐tetrahydrofuran (3‐MTHF) from the platform chemical itaconic acid (IA). © 2011 American Institute of Chemical Engineers AIChE J, 58: 1788–1801, 2012     

基于PCA-BP神经网络的精馏塔产品组成软测量模型

通过共轭梯度算法和BP神经网络对精馏过程进行软测量建模.依据工艺原理和经验知识,初选了精馏塔顶产品组成的神经网络输入变量,运用主元分析法对变量进行主元分解,降低了变量维数,并且消除了变量之间的相关性,最后对网络进行了训练与测试.仿真结果表明,该模型具有较快的收敛速度,较高的精度,可以满足大规模生产诊断的要求.     

基于PCA的反应网络简化策略

研究了复杂反应网络的模拟及简化方法,文中对于自由基模型的建立、模型中反应速率系数的计算以及模型的求解和简化均做了阐述。对于文献中已有的简化方法进行了拓展与创新,提出了基于主成分分析（PCA）的反应网络简化策略。以丙烷热裂解反应为例,针对自由基模型大规模刚性的微分方程组,利用Gear法进行求解,并利用所提出的策略缩小了模型规模,通过原丙烷裂解模型与简化后模型的计算结果的比较,展示了所提出的策略的有效性。     

煤种成浆性的人工神经网络预测模型研究

为更好地预测煤的成浆性,以大量煤种成浆浓度试验数据为基础,建立了3个输出因子的神经网络成浆浓度预测模型,模型采用L-M算法,对输入数据进行数据预处理,最后对比分析了神经网络预测模型与回归分析模型的预测结果。结果表明,以A_d、哈氏可磨性指数HGI和氧含量O为输入因子的模型预测结果平均绝对误差为0.63%,以M_(ad)、HGI和O为输入因子的模型预测结果平均绝对误差为0.60%,以M_(ad)、HGI和氧碳比O/C为输入因子的模型预测结果平均绝对误差为0.40%,3种组合的模型结果均小于回归分析模型的平均绝对误差1.15%。因此神经网络模型比回归分析模型有更好的预测能力,其中以M_(ad)、HGI和O/C为输入因子的神经网络模型预测结果最好。     

Robustness analysis of chemical process systems based on complex network non-linear load capacity model

In the complex network of chemical process systems, if a node fails, it may trigger cascading failures and affect normal operation. To enhance the ability of chemical process systems to maintain normal operation after the cascading failure, this paper presents cascading failure modelling and robustness analysis of chemical process systems based on the complex network non-linear load capacity model. First, based on complex network theory, a complex network model of the chemical process is constructed; then, three cascading failure models are constructed using a combination of linear and non-linear load capacity models and initial load and initial residual capacity redistribution strategies; and finally, the nodes with the maximum node degree are deliberately attacked to analyze the robustness of the chemical process system in response to cascading failure. The case study shows that the proposed models are valid and feasible, and the robustness of the chemical process system is enhanced as the load and capacity parameters are increased. By reasonably setting the initial load and adjusting the model parameters, the robustness can be effectively improved, providing a theoretical reference for improving the robustness of the actual chemical process system in response to cascading failure.     

Folding rate prediction based on neural network model

One of the most important challenges in biology is to understand the relationship between the folded structure of a protein and its primary amino acid sequence. A related and challenging task is to understand the relationship between sequences and folding rates of proteins. Previous studies found that one of contact order (CO), long-range order (LRO), and total contact distance (TCD) has a significant correlation with folding rate of protein. Although the predicted results from TCD can provide better results, the deviation is also large for some proteins. In this paper, we adopt back-propagation neural network to study the relationship between folding rate and protein structure. In our model, the input nodes are CO, LRO, and TCD, and the output node is folding rate. The number of nodes in the hidden layer is seven. Our results show that the relative errors for the predicted results are even lower than other methods in the literature. We also observe a best excellent correlation between the folding rate and contact parameters (including CO, LRO, and TCD), and find that the folding rate depends on CO, LRO and TCD simultaneously. This means that CO, LRO and TCD are similarly important in folding rate of protein. Some comparisons are made with other methods.     

基于人工神经网络的轮胎侧偏特性模型

采用多层前馈神经网络系统建立轮胎侧偏特性模型,用Ｌ－Ｍ算法作为学习算法,利用人工神经网络反映轮胎输入和输出特性之间的非线性映射关系,试验结果表明,人工网络可以产生出具有很同的准确性和计算效率的轮胎侧偏特性模型。     

基于WOA-BP算法的持液率预测模型研究

湿气管道在运行过程中,不可避免地会在管道低洼处出现积液.积液的存在会诱发很多安全问题,严重时甚至引发事故.因此,对湿气管道持液率进行预测就显得至关重要.文中基于灰色理论,对影响水平管道持液率的6个影响因素进行灰色关联分析,选取影响较大的因素作为影响变量;基于鲸鱼算法,建立鲸鱼算法优化BP神经网络的持液率预测模型,并与传...     

基于转运模型的功交换网络综合

功交换网络作为能量回收系统的重要组成部分,其设计水平的高低对过程系统的能耗将有着重要的影响,但是目前对功网络综合的研究仍处于理论研究的起步阶段.本文根据功级联分析,首次提出了一种基于转运模型进行功交换网络综合的新方法.该方法以公用工程用量最小为目标函数,建立了适用于等温过程的LP数学模型.通过提出构造低压流股压力中间值的策略,解决直接式功交换匹配过程中压力约束的限制,在每个压力间隔内寻求可行流股匹配,从而得到功交换的初始网络结构;再根据所提的合并相邻压力间隔的策略,进一步减少公用工程用量,进而达到优化网络结构的目的.最后通过实例计算,验证了本文方法的可行性及有效性.     

A macroscopic model for shear-thinning flow in packed beds based on network modeling

The flow of non-Newtonian fluids in packed beds and other porous media is important in several applications such as polymer processing, filtration, and enhanced oil recovery. Expressions for flowrate versus pressure gradient are desirable for a-priori prediction and for substitution into continuum models. In this work, physically representative network models are used to model the flow of shear-thinning fluids, including power-law and Ellis fluids. The networks are used to investigate the effects of fluid rheology and bed morphology on flow.A simple macroscopic model is developed for the flow of power-law and Ellis fluids in packed beds using results from the network model. The model has the same general functionality as those developed using the popular bundle-of-tubes approach. The constant β, which appears in these models, is often directly derived from the tortuosity and a simple representation of the porous media. It is shown here that this can lead to incorrect and ambiguous values of the constant. Furthermore, the constant is a weak function of the shear-thinning index, indicating that no single bundle-of-tubes could ever properly model flow for a wide variety of shear-thinning fluids.The macroscopic model is compared to experimental data for shear-thinning fluids available in the literature. The model fits the data well when β is treated as an experimental parameter. The best-fit values of β vary, which is expected because even the constant C in the Blake-Kozeny equation varies depending on the source consulted. Additionally, physical effects, such as adsorption and filtration, as well as rheological effects such as viscoelasticity may affect the value of β. We believe that in the absence of these effects, β equals approximately 1.46 for packed beds of uniform spheres at relatively moderate values of the shear-thinning index (>0.3).     

A method of model validation for chaotic chemical reaction systems based on neural network

A chaotic system with measurable state variables fewer than the degrees of freedom of the system is identified with the Artificial Neural Network (ANN) method combined with dynamic training. Instead of using the usual method of Sum of Square Errors (SSE), the identified models are validated with the return maps (embedded trajectories), the largest Lyapunov exponent, and the correlation dimension when there is no exogenous input, and bifurcation diagram when there is an exogenous input. This method is demonstrated for nonisothermal, irreversible, first-order, series reaction A→ B → C in a CSTR.     

基于流量振荡的窄矩形通道内临界热通量机理模型

设备最大运行功率受临界热通量(CHF)限制,而流量振荡会导致沸腾危机早发,此时的临界热通量称为PM-CHF。为了研究流量振荡条件下窄矩形通道内的临界热通量,进行单侧加热窄矩形通道内竖直向上流动条件下沸腾危机可视化实验,实验工质为去离子水,质量流速范围为350~2000 kg/(m²·s),窄缝宽度范围为1~5 mm,系统压力范围为1~4 MPa。结果显示,在窄矩形通道中CHF随质量流速的增加而线性增加。当流速较小时会发生流量振荡,振荡周期约为0.1 s。流量振荡继而导致沸腾危机早发,其流型表现为弹状流-搅混流。此外,针对本实验观察到的流量振荡和窄矩形通道内气泡动力学特性,从流量振荡的角度进行理论分析与推导,建立窄矩形通道内由于流动失稳引起的PM-CHF机理模型,预测误差在30%以内。     

基于流量振荡的窄矩形通道内临界热通量机理模型

设备最大运行功率受临界热通量(CHF)限制,而流量振荡会导致沸腾危机早发,此时的临界热通量称为PM-CHF。为了研究流量振荡条件下窄矩形通道内的临界热通量,进行单侧加热窄矩形通道内竖直向上流动条件下沸腾危机可视化实验,实验工质为去离子水,质量流速范围为350~2000 kg/(m²·s),窄缝宽度范围为1~5 mm,系统压力范围为1~4 MPa。结果显示,在窄矩形通道中CHF随质量流速的增加而线性增加。当流速较小时会发生流量振荡,振荡周期约为0.1 s。流量振荡继而导致沸腾危机早发,其流型表现为弹状流-搅混流。此外,针对本实验观察到的流量振荡和窄矩形通道内气泡动力学特性,从流量振荡的角度进行理论分析与推导,建立窄矩形通道内由于流动失稳引起的PM-CHF机理模型,预测误差在30%以内。     

重组枯草芽孢杆菌生产青霉素G酰化酶发酵条件的研究

对产青霉素G酰化酶的温度诱导型重组枯草芽孢杆菌发酵条件进行了研究。结果表明,培养基的最佳氮源和碳源组成为:35gL-1牛肉膏、3gL-1葡萄糖和6gL-1淀粉;最佳诱导时机是细胞对数生长的中后期,诱导前应将pH调节至中性;最佳诱导条件为升高温度至50℃并维持4min;随后将温度降低到34℃进行重组蛋白的表达。在上述条件下,PGA的表达水平可以达到5.8UmL-1。SDS-PAGE电泳分析表明该重组菌能够将绝大部分表达的PGA分泌到细胞外,而且表达的PGA蛋白占有高比例(90%)。     

Distributed model predictive control-oriented network decomposition based on full dynamic response

This article addresses network decomposition for distributed model predictive control (DMPC), which includes two improvements. First, in the weighted input–output bipartite graph construction of a process network, a new measure called frequency affinity is proposed to characterize the input–output interaction considering the full dynamic response and structural information of a process. Then, in community detection, which is used to decompose the process network, the gap metric is added to quantify stability and the loss of control performance of each subsystem. Through the proposed decomposition, the obtained subsystems can be dynamically well-decoupled since both transient and steady-state responses are measured by the frequency affinity. As structural information is considered, the decomposition is consistent with the process physical topology. Furthermore, the utilization of gap metric can facilitate controller design for DMPC. Case studies on a reactor separator process and an air separation process demonstrate the effectiveness of the proposed decomposition method.     

基于RBF网络的胶磷矿浮选精矿指标预测模型

本文基于RBF神经网络构造了云南某胶磷矿浮选多因素输入和浮选精矿品位、回收率之间的浮选模型,并在Matlab环境下进行了计算机仿真试验,结果表明,模型预测精度较高,验证了非参数建模的合理性,具有一定的实用价值,为浮选过程的控制奠定了基础.     

基于径向基函数神经网络的挤出温度预测模型

简要介绍了软测量方法建立挤出温度预测模型的方法,分析其优缺点后进而提出应用基于RBF神经网络建立渝度预测模型,试验后对比了实测值和预测值,结果表明该方法能达到较好的预测精度,同时具有使用简洁、快速等优点,具有较好的应用推广意义。     

基于BP网络的保护渣性能预测模型的研究

结合BP网络计算机实验研究 ,建立了保护渣化学组成与性能的预测模型 ,并利用化学组成与性能的关系 ,对网络的实用性进行了检验 ,结果符合保护渣化学组成与性能的关系。保护渣粘度随着保护渣碱度的增大而减小 ,而半球点温度随碱度增大而增大 ;保护渣的半球点温度和粘度都随着渣中CaF2 含量的增加而减小。用BP网络的误差反向传播算法建立的保护渣的化学组成与性能的预测模型 ,得出的预测值与实际值的误差小 ,对保护渣的设计与应用都有一定的指导作用。     

基于Hopfield网络的时滞分析故障诊断策略

振荡是化工过程中常见的对全流程运行性能有显著影响的故障类型,仅基于数据幅值域知识的故障诊断方法对这一类故障诊断性能不佳。时滞分析基于数据信号时域知识,根据波形相关性分析变量之间因果关系,通过得到的因果模型确定故障完整传播路径,可进一步识别出扰动发生的根本原因。将Hopfield网络与时滞分析相结合,解决了时滞分析当变量数众多时,从变量对的因果关系难以得到故障传播路径的问题,并同时讨论了时滞分析数据窗选取、对称时滞确立等的原则,提升了故障传播路径建立的准确度,建立了基于时滞分析的完备的故障诊断策略,最后通过TE模型验证了方法的优越性。