Mechanism Analysis and Neural Network Predictive Method for Cement Calcination Process

采用神经网络的方法建立水泥预分解窑煅烧工段的预测模型.选择合理的状态与控制变量,通过采集实际运行数据来训练神经网络.构建的基于BPNN神经网络的煅烧预测模型能够较好地拟合采样数据,具有较好的泛化能力.     

遗传算法优化BP神经网络的水泥窑温度预测模型研究

水泥窑内温度的控制效果直接决定水泥熟料的质量。由于回转窑煅烧具有大滞后、非线性、多变量等复杂特性,并且各个因素之间存在相互联系和不确定性,这给回转窑温度的控制带来了困难。本文利用BP神经网络建立回转窑温度预测模型,并利用遗传算法对模型进行优化,将优化前后的仿真结果对比后发现,使用遗传算法优化后的BP神经网络回转窑温度预测模型的预测精度高于单独使用BP神经网络搭建的回转窑温度预测模型精度,为回转窑温度的预测提供了一种新的方法。     

BP神经网络在膜分离处理废水中的应用

为能更好地预测工艺条件对膜分离过程的影响,运用BP神经网络技术建立输入变量为压差、流速、浓度和温度,输出变量为膜通量的预测模型。通过大量实验数据训练预测模型,得到的网络模型整体误差平方和仅为0.014 5;计算值与模拟值相比,10组不同条件的膜通量平均预测误差仅为1.1,证实了所建立的BP神经网络膜通量预测模型与实验值吻合程度较好,有较好地预测能力。在此基础上进一步考察了工艺参数对膜分离过程的影响。     

循环冷却水系统腐蚀速率的预测模型

通过对某石化公司循环冷却水系统生产运行数据的分析,选取了对腐蚀速率影响较大的水质参数,借助神经网络良好的非线性能力,基于BP神经网络建立了腐蚀速率的预测模型.利用该模型对循环冷却水系统一定周期腐蚀速率的预测结果较好.     

基于改进粒子群小波神经网络的瓦斯浓度预测研究

结合小波神经网络基本原理,提出一种粒子群优化小波神经网络的瓦斯浓度预测模型。对小波神经网络基本原理进行分析,然后,利用粒子群对小波神经网络参数进行优化,并构建预测模型;最后,以P₁～P₅监测点的煤矿瓦斯浓度数据为基础,将其输入预测模型中进行训练。结果表明：粒子群优化后的小波神经网络在瓦斯浓度预测方面,数值更接近真实值,同时迭代次数在110次左右即达到稳定。     

徐深气田CO_2腐蚀速率预测方法研究

通过室内CO2腐蚀模拟实验获得实验数据,利用灰色理论对CO2腐蚀参数进行分析确定CO2腐蚀的主要影响因素,建立BP神经网络腐蚀速率预测模型,利用主要影响因素进行网络训练。利用此模型预测徐深气田某井的腐蚀剖面,预测结果表明：BP神经网络预测结果与气井实验结果接近,体现了BP神经网络在处理非线性数据方面的优越性。灰色理论、神经网络预测模型的研究对于徐深气田CO2腐蚀研究有一定的指导意义。     

基于MIC-CFS-LSTM的SCR出口NOx浓度动态预测

针对燃煤机组选择性催化还原(SCR)系统出口氮氧化物(NO_x)预测模型精度不高的问题，提出一种基于最大信息系数(MIC)和长短期记忆(LSTM)神经网络的预测模型方法。首先采用MIC估计各变量的延迟时间，对数据进行时延重构；然后采用重构后数据的MIC值作为评价各输入变量和输出变量间相关性大小的指标，并结合基于关联性的特征选择算法(CFS)进行输入变量筛选；最后基于时延重构和变量筛选后的数据，采用LSTM神经网络建立了SCR出口氮氧化物浓度动态预测模型。该模型被用于广东某320 MW燃煤机组实际运行数据分析。结果表明，经时延重构和变量筛选后所建立的LSTM预测模型具有较高精度，优于深度神经网络(DNN)模型和径向基函数(RBF)神经网络模型，平均绝对百分比误差达2.58%,均方根误差达2.02,可满足现场运用要求。     

信息扩散及神经网络在建立注塑预测模型中的应用

注塑工艺参数和塑料件质量之间的关联模型常采用神经网络建立,然而训练样本不足将直接影响模型的质量。为了在适量训练样本条件下得到较好的预测模型,对原始样本数据进行信息扩散处理,然后使用神经网络建模。笔者以汽车音响面板为例建立预测模型,确定了最佳工艺参数,证明了该方法的有效性。     

工业循环冷却水腐蚀速率模型的研究

循环冷却水在工业生产中占有很大比重,而腐蚀是循环冷却水系统中常见的水质故障,严重影响工业生产中设备的运行。通过对某石化公司水质数据的分析,选取对腐蚀速率影响较大的水质参数,凭借神经网络良好的非线性能力,基于NARX神经网络建立了腐蚀速率预测模型。利用该模型对循环冷却水系统腐蚀速率进行预测,结果较好,说明该模型可行,且具有良好的应用前景。     

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。     

BP神经网络在乙烯裂解炉预测分析中的应用研究

利用MATLAB工具箱中的BP神经网络模型建立了乙烯裂解炉的三层神经网络模型,应用该模型分析和预测了裂解产物乙烯和丙烯的收率。预测结果与生产过程数据的比较表明,该模型能适合实际生产过程,可用于乙烯生产的预测分析和预测控制。     

基于SVR误差补偿技术的神经网络城市污水厂水质预测

城市污水处理是一个复杂的生化处理过程,现在的神经网络技术无法做到对此过程精准建模。为了解决该问题,提出了一种基于SVR误差补偿技术的神经网络水质预测算法。该算法先利用BP神经网络对水质处理过程进行映射,再利用SVR误差补偿模型获得BP网络的预测补偿,进行预测数据校正。为了验证补偿模型的性能,还组织了马尔科夫补偿模型的对比试验。试验结果表明,SVR误差补偿模型可有效提高模型预测的精度,且模型性能优于马尔科夫补偿模型。     

Neural‐Networks‐Based Feedback Linearization versus Model Predictive Control of Continuous Alcoholic Fermentation Process

In this work advanced nonlinear neural networks based control system design algorithms are adopted to control a mechanistic model for an ethanol fermentation process. The process model equations for such systems are highly nonlinear. A neural network strategy has been implemented in this work for capturing the dynamics of the mechanistic model for the fermentation process. The neural network achieved has been validated against the mechanistic model. Two neural network based nonlinear control strategies have also been adopted using the model identified. The performance of the feedback linearization technique was compared to neural network model predictive control in terms of stability and set point tracking capabilities. Under servo conditions, the feedback linearization algorithm gave comparable tracking and stability. The feedback linearization controller achieved the control target faster than the model predictive one but with vigorous and sudden controller moves.     

Machine-learning-based construction of barrier functions and models for safe model predictive control

In this paper, we propose a control Lyapunov-barrier function-based model predictive control method utilizing a feed-forward neural network specified control barrier function (CBF) and a recurrent neural network (RNN) predictive model to stabilize nonlinear processes with input constraints, and to guarantee that safety requirements are met for all times. The nonlinear system is first modeled using RNN techniques, and a CBF is characterized by constructing a feed-forward neural network (FNN) model with unique structures and properties. The FNN model for the CBF is trained based on data samples collected from safe and unsafe operating regions, and the resulting FNN model is verified to demonstrate that the safety properties of the CBF are satisfied. Given sufficiently small bounded modeling errors for both the FNN and the RNN models, the proposed control system is able to guarantee closed-loop stability while preventing the closed-loop states from entering unsafe regions in state-space under sample-and-hold control action implementation. We provide the theoretical analysis for bounded unsafe sets in state-space, and demonstrate the effectiveness of the proposed control strategy using a nonlinear chemical process example with a bounded unsafe region.     

反向传播网络在乙烯裂解炉预测分析中的应用研究

利用MATLAB工具箱中的BP神经网络模型建立了乙烯裂解炉的三层神经网络模型,分析和预测了裂解产物乙烯和丙烯的收率,将预测的结果和生产过程数据作比较,结果表明,该模型的预测值和实际生产数据值吻合很好,可用于乙烯生产的预测分析和预测控制。     

多级库存集成模型优化预测研究

以供应链多级库存为研究背景,建立以生产为中心的供应、生产和销售3个环节的多级库存集成化动态模型,并对模型进行仿真优化。结合训练神经网络的混合算法GA-BP算法,提出了基于遗传算法与人工神经网络相结合的优化预测模型。最后给出实例说明GA-BP算法优化预测模型的求解过程,验证了模型的可行性。     

Neural network aided design of Pt-Co-Ce/Al2O3 catalyst for selective CO oxidation in hydrogen-rich streams

In this study, the design of Pt-Co-Ce/Al₂O₃ catalyst for the low temperature CO oxidation in hydrogen streams was modeled using artificial neural networks. The effects of five design parameters, namely Pt wt.%, Co wt.%, Ce wt.%, calcination temperature and calcination time, on CO conversion were investigated by modeling the experimental data obtained in our laboratory for 30 catalysts. Although 30 points data set can be considered as small for the neural network modeling, the results were quite satisfactory apparently due to the fact that the experimental data generated with response surface method were well balanced over the experimental region and it was very suitable for neural network modeling. The success of neural network modeling was more apparent when the number of data points was increased to 120 by using the time on stream as another input parameter. It was then concluded that the neural network modeling can be very helpful to improve the experimental works in catalyst design and it may be combined with the statistical experimental design techniques so that the successful models can be constructed using relatively small number of data points.     

The design of experiments,training and implementation of nonlinear controllers based on neural networks

In the area of nonlinear predictive control, several control schemes using artificial neural networks have been proposed. In this work, the issues relating to the information contents of the data used to train the neural network components of these nonlinear predictive control schemes are considered. This raises questions about the design of experiments. A class of feedback-feedforward nonlinear controller based on the model predictive structure (also known as Internal Model Control, IMC, structure) is investigated. The implementation and performance of these neural network based controllers, together with comparisons to other nonlinear and linear controllers, are illustrated on two nonlinear continuous-stirred-tank-reactor simulations.     

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。     

基于Hammerstein-Wiener模型的连续搅拌反应釜神经网络预测控制

针对化工过程中广泛使用的连续搅拌反应釜(CSTR),提出一种基于神经网络的模型预测控制策略,采用分段最小二乘支持向量机辨识Hammerstein-Wiener模型系数的方法,在此基础上建立线性自回归模式〖DK〗(ARX)结构和高斯径向基神经网络串联的非线性预测控制器。利用BP神经网络训练预测控制输入序列和拟牛顿算法求解非线性预测控制律,从而实现一种基于支持向量机Hammerstein-Wiener辨识模型的非线性神经网络预测控制算法。对CSTR的仿真结果表明,该方法能够更有效地跟踪控制反应物浓度。