Performance Prediction of Carbon Fiber Protofilament Based on SAGA-SVR

The existing optimized performance prediction of carbon fiber protofilament process model is still unable to meet the production needs. A way of performance prediction on carbon fiber protofilament was presented based on support vector regression （SVR） which was optimized by an optimization algorithm combining simulated annealing algorithm and genetic algorithm （SAGA-SVR）. To verify the accuracy of the model, the carbon fiber protofilament production test data were analyzed and compared with BP neural network （BPNN）. The results show that SAGA-SVR can predict the performance parameters of the carbon fiber protofilament accurately.     

Processing Time Prediction Method Based on SVR in Semiconductor Manufacturing

The prediction problem of the actual value of the dynamic parameters in the simulation model in semiconductor manufacturing was discussed. Considering the fact that the default value of processing time of one certain equipment in the simulation model was not the same as its actual value. a general data driven prediction model of the processing time was built based on support vector regression （SVR） , with the utilization of manufacturing information in manufacturing execution system （MES）. The processing time of one certain equipment was highly related to the status of the equipment itself and the wafers being processed. To uncover the relationship of the processing time with the information of historical products. process flow. technical standard of silicon wafers and manual intervention. data were extracted from MES and used to build a prediction model. This model was employed on an ion implantation equipment as a case. and the effectiveness of the proposed method was shown by comparing with other approaches.     

Method of So.Sensor Modeling for Fermentation Process Based on Geometric Support Vector Regression

The soft-sensor modeling for fermentation process based on standard support vector regression(SVR) needs to solve the quadratic programming problem(QPP) which will often lead to large computational burdens, slow convergence rate, low solving efficiency, and etc. In order to overcome these problems, a method of soft-sensor modeling for fermentation process based on geometric SVR is presented. In the method, the problem of solving the SVR soft-sensor model is converted into the problem of finding the nearest points between two convex hulls (CHs) or reduced convex hulls (RCHs) in geometry. Then a geometric algorithm is adopted to generate soft-sensor models of fermentation process efficiently. Furthermore, a swarm energy conservation particle swarm optimization (SEC-PSO) algorithm is proposed to seek the optimal parameters of the augmented training sample sets, the RCH size, and the kernel function which are involved in geometric SVR modeling. The method is applied to the soft-sensor modeling for a penicillin fermentation process. The experimental results show that, compared with the method based on the standard SVR, the proposed method of soft-sensor modeling based on geometric SVR for fermentation process can generate accurate soft-sensor models and has much less amount of computation, faster convergence rate, and higher efficiency.     

Integration of a statistical emulator approach with the SCE-UA method for parameter optimization of a hydrological model

Parameter optimization of a hydrological model is an indispensable process within model development and application.The lack of knowledge regarding the efficient optimization of model parameters often results in a bottle-neck within the modeling process,resulting in the effective calibration and validation of distributed hydrological models being more difficult to achieve.The classical approaches to global parameter optimization are usually characterized by being time consuming,and having a high computation cost.For this reason,an integrated approach coupling a meta-modeling approach with the SCE-UA method was proposed,and applied within this study to optimize hydrological model parameter estimation.Meta-modeling was used to determine the optimization range for all parameters,following which the SCE-UA method was applied to achieve global parameter optimization.The multivariate regression adaptive splines method was used to construct the response surface as a surrogate model to a complex hydrological model.In this study,the daily distributed time-variant gain model(DTVGM) applied to the Huaihe River Basin,China,was chosen as a case study.The integrated objective function based on the water balance coefficient and the Nash-Sutcliffe coefficient was used to evaluate the model performance.The case study shows that the integrated method can efficiently complete the multi-parameter optimization process,and also demonstrates that the method is a powerful tool for efficient parameter optimization.     

Support vector machine based on chaos particle swarm optimization for fault diagnosis of rotating machine

The performance of the support vector machine models depends on a proper setting of its parameters to a great extent. A novel method of searching the optimal parameters of support vector machine based on chaos particle swarm optimization is proposed. A multifault classification model based on SVM optimized by chaos particle swarm optimization is established and applied to the fault diagnosis of rotating machines. The results show that the proposed fault classification model outperforms the neural network trained by chaos particle swarm optimization and least squares support vector machine, and the precision and reliability of the fault classification results can meet the requirement of practical application. It indicates that chaos particle swarm optimization is a suitable method for searching the optimal parameters of support vector machine.     

Application of support vector machine in the prediction of mechanical property of steel materials

The investigation of the influences of important parameters including steel chemical composition and hot rolling parameters on the mechanical properties of steel is a key for the systems that are used to predict mechanical properties. To improve the prediction accuracy, support vector machine was used to predict the mechanical properties of hot-rolled plain carbon steel Q235B. Support vector machine is a novel machine learning method, which is a powerful tool used to solve the problem characterized by small sample, nonlinearity, and high dimension with a good generalization performance. On the basis of the data collected from the supervisor of hot-rolling process, the support vector regression algorithm was used to build prediction models, and the off-line simulation indicates that predicted and measured results are in good agreement.     

Artificial neural network prediction of mechanical properties of hot rolled low carbon steel strip

Conventionally, direct tensile tests are employed to measure mechanical properties of industrially pro- duced products. In mass production, the cost of sampling and labor is high, which leads to an increase of total pro- duction cost and a decrease of production efficiency. The main purpose of this paper is to develop an intelligent pro- gram based on artificial neural network （ANN） to predict the mechanical properties of a commercial grade hot rolled low carbon steel strip, SPHC. A neural network model was developed by using 7 x 5 x 1 back-propagation （BP） neural network structure to determine the multiple relationships among chemical composition, product pro- cess and mechanical properties. Industrial on-line application of the model indicated that prediction results were in good agreement with measured values. It showed that 99.2 % of the products＇ tensile strength was accurately pre- dicted within an error margin of ~ 10 %, compared to measured values. Based on the model, the effects of chemical composition and hot rolling process on mechanical properties were derived and the relative importance of each in- put parameter was evaluated by sensitivity analysis. All the results demonstrate that the developed ANN models are capable of accurate predictions under real-time industrial conditions. The developed model can be used to sub- stitute mechanical property measurement and therefore reduce cost of production. It can also be used to control and optimize mechanical properties of the investigated steel.     

Yarn Quality Prediction and Diagnosis Based on Rough Set and Knowledge-Based Artificial Neural Network

In the spinning process, some key process parameters（ i. e.,raw material index inputs） have very strong relationship with the quality of finished products. The abnormal changes of these process parameters could result in various categories of faulty products. In this paper, a hybrid learning-based model was developed for on-line intelligent monitoring and diagnosis of the spinning process. In the proposed model, a knowledge-based artificial neural network（ KBANN） was developed for monitoring the spinning process and recognizing faulty quality categories of yarn. In addition,a rough set（ RS）-based rule extraction approach named RSRule was developed to discover the causal relationship between textile parameters and yarn quality. These extracted rules were applied in diagnosis of the spinning process, provided guidelines on improving yarn quality,and were used to construct KBANN. Experiments show that the proposed model significantly improve the learning efficiency, and its prediction precision is improved by about 5. 4% compared with the BP neural network model.     

An optimized short-term wind power interval prediction method considering NWP accuracy

In recent years, the accuracy of the wind power prediction has been urgently studied and improved to satisfy the requirements of power system operation. In this paper, the relevance vector machine(RVM)-based models are established to predict the wind power and its interval for a given confidence level. An NWP improvement module is presented considering the characteristic of NWP error. Moreover, two parameter optimization algorithms are applied to further improve the prediction model and to compare each performance. To take three wind farms in China as examples, the performance of two RVM-based models optimized, respectively, by genetic algorithm(GA)and particle swarm optimization(PSO) are compared with predictions based on a genetic algorithm–artificial neural network(GA–ANN) and support vector machine. Results show that the proposed models have better prediction accuracy with GA–RVM model and more efficient calculation with PSO–RVM.     

Yarn Quality Prediction for Small Samples Based on AdaBoost Algorithm

In order to solve the problems of weak prediction stability and generalization ability of a neural network algorithm model in the yarn quality prediction research for small samples, a prediction model based on an AdaBoost algorithm(AdaBoost model) was established. A prediction model based on a linear regression algorithm(LR model) and a prediction model based on a multi-layer perceptron neural network algorithm(MLP model) were established for comparison. The prediction experiments of the yarn ev...     

Multi-objective quality control method for cold-rolled products oriented to customized requirements

To deal with the increasing demand for low-volume customization of the mechanical properties of cold-rolled products, a two-way control method based on mechanical property prediction and process parameter optimization(PPO) has become an effective solution. Aiming at the multi-objective quality control problem of a company's cold-rolled products, based on industrial production data, we proposed a process parameter design and optimization method that combined multi-objective quality prediction and PPO. This method used the multi-output support vector regression(MSVR) method to simultaneously predict multiple quality indices. The MSVR prediction model was used as the effect verification model of the PPO results. It performed multi-process parameter collaborative design and realized the optimization of production process parameters for customized multi-objective quality requirements. The experimental results showed that, compared with the traditional single-objective quality prediction model based on support vector regression(SVR), the multi-objective prediction model could better take into account the coupling effect between process parameters and quality index, the MSVR model prediction accuracy was higher than that of the SVR, and the optimized process parameters were more capable and reflected the influence of metallurgical mechanism on the quality index,which were more in line with actual production process requirements.     

基于多输出支持向量机的聚乙烯相对分子质量分布建模与优化

相对分子质量分布(MWD)是聚合物产品重要的质量指标,目前实时检测相对分子质量分布仍缺乏有效的方法.在聚合物研究中,高精度的相对分子质量分布实时检测方法是当前的研究热点.不同于其他文献中描述的方法,通过结合聚合反应机理和过程信息建立相对分子质量分布的混合模型来解决预测精度和实时性问题.首先利用催化剂各活性中心的分布函数加权叠加拟合相对分子质量分布,分布函数参数和工艺条件之间的关系可通过多输出支持向量机回归(MSVR)算法来描述;其次应用无约束非线性优化方法优化上述方法建立的混合模型的工艺条件;最后将建立的混合模型应用于乙烯聚合过程,验证了上述方法的可行性.     

基于多目标遗传算法的热定型工艺参数优化设计

为了解决染整后整理中热定型工艺参数难以定量设计的关键技术难题,将工艺参数优化设计问题视为以成品门幅、克重与客户要求的相应值的绝对误差最小为目标函数,温度、车速、超喂率和上机门幅为优化变量,以根据实际情况中各优化变量的取值范围为约束条件的多目标优化问题。建立多目标优化模型,并基于该模型采用多目标遗传算法,实现了热定型参数精确定量设计。用该方法得到的工艺参数加工弹力布,生产成品的克重、门幅与用户要求指标的偏差小于1%,完全可以满足实际生产要求。     

基于偏最小二乘回归模型的带钢热镀锌质量监控方法

提出了基于偏最小二乘回归模型的带钢热镀锌质量监控方法. 以带钢热镀锌生产中带钢力学性能和锌层质量的质量监控为研究对象,用偏最小二乘方法建立了生产过程参数与质量结果之间的回归模型,对生产过程控制能力进行了分析,并给出了产品质量的预测方法. 用鞍钢股份有限公司带钢热镀锌的实际生产数据进行验证. 结果表明,偏最小二乘法比传统的多元线性回归方法具有更好的预测精度,基于偏最小二乘回归的锌层质量预测模型,其相对预测误差可达到5.93%.     

基于机器学习和遗传算法的高炉参数预测与优化

针对目前高炉炼铁模型精度不高问题,提出建立高炉生产过程中精确的多目标优化模型.首先对高炉的海量数据进行了数据预处理,其次采用支持向量机、随机森林、梯度提升树、XGBoost、LightGBM、人工神经网络6种机器学习算法对高炉焦比、K值进行了预测,并采用特征工程和超参调优对机器学习预测进行了优化,最后采用新的集成学习方法进行预测.预测结果不仅精准度高而且具有很好的鲁棒性.在机器学习的基础之上,采用NSGA-Ⅱ遗传算法对高炉参数进行了多目标优化分析,得到了Pareto最优解,高炉操作者可以根据该多目标优化结果针对不同的需求选择相应的控制参数.     

基于人工蜂群优化支持向量机回归的隧道塌方风险预测

为预测隧道塌方风险等级,减少隧道塌方引起的灾害事故,建立基于人工蜂群(artificial bee colony, ABC)优化支持向量机回归(support vector machine regression, SVR)隧道塌方风险预测模型。首先,从工程地质、水文气象、设计因素、施工因素4个方面综合考虑,遴选13个主要影响因素,建立隧道塌方风险指标体系;其次,引入人工蜂群算法优化SVR的核参数C和惩罚参数g,解决传统SVR稳定性低的缺陷,提高模型的精确度,为验证模型性能采用相关系数(R²)、均方误差(mean squared error, MSE)、均方根误差(root mean squared error, RMSE)评价参数对比分析;最后,以新疆北部某供水工程为研究对象,对隧道塌方风险测试样本进行预测,分别将ABC-SVR、PSO-SVR、GA-SVR及SVR模型对比分析。研究结果表明：ABC-SVR预测结果为100%,PSO-SVR预测结果为83.3%,GA-SVR和SVR均为66.67%,ABC-SVR的预测结果与实际工程结果一致性更高,可为隧道塌方风险...     

基于PSO-SVM的多阶段产品质量预测控制方法研究

本文以钢铁产品为例,在分析多工序多阶段产品质量预测控制特点的基础上,建立了多控制点递阶SVM预测控制模型,在模型的求解过程中,提出了基于粗集理论和主成分分析法的数据预处理与模型简化,并利用带约束的PSO算法分别优化了SVM的核超参数和相关影响因素的决策范围,实现了多阶段产品质量预测和相关过程参数的全局优化,为生产过程的质量改进提供了科学的决策依据。     

基于替代模型的油藏注采参数多目标优化设计

针对水驱油藏生产过程中合适的注采参数选取难的问题,提出了以净现值和累产油量为目标函数的多目标优化注采参数设计方法。采用基于粒子群算法的最小二乘支持向量机作为替代模型代替数值模拟,并用带精英策略的非支配排序多目标优化遗传算法对注采参数进行优化。以某区块两注两采模型为例,选取生产井井底压力和注水井注入量为优化变量,通过粒子群算法优化的最小二乘支持向量机构建替代模型,在优化过程中代替数模,再利用非支配排序遗传算法对注采参数进行优化。对比分析替代模型和数值模拟优化设计的结果,其误差在3%以内,并在注采参数优化时间上得到了明显提升。     

基于多重多元回归的焦炭质量预测模型

 焦炭质量预测是焦化企业进行焦炭质量控制的重要方法。在诸多生产因素已固定的条件下,焦炭质量主要取决于原料煤性质。用煤质指标预测焦炭质量是焦炭质量预测的重要方法。考虑配合煤煤质指标和焦炭质量指标的多元性和相关性,采用多重多元回归分析技术,将焦炭质量的各指标作为一个整体,建立配合煤煤质指标对焦炭质量的预测模型。基于偏最小二乘回归思想,采用预测误差平方和（PRESS）和预测的方差验证回归模型的预测能力。根据实际焦炭生产建立的焦炭质量预测多重多元回归模型的显著性检验表明,该模型具有较高的预测能力。应用多重多元回归技术建立的焦炭质量预测模型对指导焦化企业的焦化生产,优化配煤和加强焦炭质量控制具有重要的现实意义。