基于FCM的多支持向量机模型在高炉炉温预测中的应用

针对高炉炼铁智能控制专家系统中单一支持向量机(SVM)炉温预测模型的改进研究,提出一种基于模糊C均值聚类(FCM)的多支持向量机模型。首先运用模糊C均值聚类对模型训练集进行聚类划分,然后对每一类进行支持向量机的训练,建立相应的子模型,并对测试集中的同一样本点分别进行预测,以测试样本点的输入对应于每一类的隶属度为权值,进行加权求和,最终得到预测值。通过对在线采集的数据分析表明,基于FCM的多支持向量机模型比单一的支持向量机模型在多方面预测性能得到改善,连续预测100炉命中率达86%。     

ANN and Fuzzy Logic Models for Simulating Event-Based Rainfall-Runoff

This study presents the development of artificial neural network (ANN) and fuzzy logic (FL) models for predicting event-based rainfall runoff and tests these models against the kinematic wave approximation (KWA). A three-layer feed-forward ANN was developed using the sigmoid function and the backpropagation algorithm. The FL model was developed employing the triangular fuzzy membership functions for the input and output variables. The fuzzy rules were inferred from the measured data. The measured event based rainfall-runoff peak discharge data from laboratory flume and experimental plots were satisfactorily predicted by the ANN, FL, and KWA models. Similarly, all the three models satisfactorily simulated event-based rainfall-runoff hydrographs from experimental plots with comparable error measures. ANN and FL models also satisfactorily simulated a measured hydrograph from a small watershed 8.44?km2 in area. The results provide insights into the adequacy of ANN and FL methods as well as their competitiveness against the KWA for simulating event-based rainfall-runoff processes.     

Neural Networks for Profiling Stress History of Clays from PCPT Data

This paper evaluates the feasibility of using artificial neural network (ANN) models for estimating the overconsolidation ratio (OCR) of clays from piezocone penetration tests (PCPT). Three feed-forward, back-propagation ANN models are developed, and trained using actual PCPT records from test sites around the world. The soil deposits range from soft, normally consolidated intact clays to very stiff, heavily overconsolidated fissured clays. ANN model 1 is a general model applicable for both intact and fissured clays. ANN model 2 is suited for intact clays, and ANN model 3 is applicable to fissured clays only. The models are validated using new PCPT data (not used for training), and by comparing model predictions with reference OCR values obtained from oedometer tests. For intact clays, ANN model 2 gives better OCR estimates compared to ANN model 1. For fissured clays, ANN model 3 gives better estimates compared to ANN model 1. Some of the existing interpretation methods are reviewed. Compared to the existing methods, ANN models 2 and 3 give very good estimates of OCR.     

基于Kohonen网络的模糊诊断技术在钢管材质分检中的应用

介绍了钢管材质计算机在线分检系统的有关算法与应用 ,并用生产数据对统计分析和基于Kohonen神经网络聚类的模糊诊断的分选结果进行了分析与比较 ,指出了智能化分检方法涉及的问题和应用前景。介绍了钢管材质计算机在线分检系统的有关算法与应用 ,并用生产数据对统计分析和基于Kohonen神经网络聚类的模糊诊断的分选结果进行了分析与比较 ,指出了智能化分检方法涉及的问题和应用前景。     

Support Vector Machines Approach to HMA Stiffness Prediction

The application of artificial intelligence (AI) techniques to engineering has increased tremendously over the last decade. Support vector machine (SVM) is one efficient AI technique based on statistical learning theory. This paper explores the SVM approach to model the mechanical behavior of hot-mix asphalt (HMA) owing to high degree of complexity and uncertainty inherent in HMA modeling. The dynamic modulus (|E?|), among HMA mechanical property parameters, not only is important for HMA pavement design but also in determining HMA pavement performance associated with pavement response. Previously employed approaches for development of the predictive |E?| models concentrated on multivariate regression analysis of database. In this paper, SVM-based |E?| prediction models were developed using the latest comprehensive |E?| database containing 7,400 data points from 346 HMA mixtures. The developed SVM models were compared with the existing multivariate regression-based |E?| model as well as the artificial neural networks (ANN) based |E?| models developed recently by the writers. The prediction performance of SVM model is better than multivariate regression-based model and comparable to the ANN. Fewer constraints in SVM compared to ANN can make it a promising alternative considering the availability of limited and nonrepresentative data frequently encountered in construction materials characterization.     

Optimization of back propagation algorithm and GAS-assisted ANN models for hot metal desulphurization

Adaptive neural net (ANN) model of hot metal desulphurization is first optimized by various search methods including the golden section search and Davies-Swann-Campey methods. Logarithmic preprocessing of input data leads to a further improvement in generalization ability of the net. Genetic adaptive search (GAS) method is used to optimize the mathematical model for desulphurization and when the input data are preprocessed with this optimized model and fed into an artificial neural net, the generalization ability of the net becomes even better. Best results are obtained when using GAS to optimize the interconnection weights during the training phase, while training data are preprocessed through a mathematical model already optimized by GAS. For every process several options presented by a combination of ANN and GAS must be systematically investigated before choosing the ultimate model for predictions on shop floor.     

Subtractive Clustering-Based K-means Technique for Determining Optimum Time-of-Day Breakpoints

In many countries, the most widely used method for timing plan selection and implementation is the time-of-day (TOD) method. In TOD mode, a few traffic patterns that exist in the historical volume data are recognized and used to find the signal timing plans needed to achieve optimum performance of the intersections during the day. Traffic engineers usually determine TOD breakpoints by analyzing 1 or 2?days worth of traffic data and relying on their engineering judgment. The current statistical methods, such as hierarchical and K-means clustering methods, determine TOD breakpoints but introduce a large number of transitions. This paper proposes adopting the Z-score of the traffic flow and time variable in the K-means clustering to reduce the number of transitions. The numbers of optimum breakpoints are chosen based on a microscopic simulation model considering a set of performance measures. By using simulation and the K-means algorithm, it was found that five clusters are the optimum for a major arterial in Al-Khobar, Saudi Arabia. As an alternative to the simulation-based approach, a subtractive algorithm-based K-means technique is introduced to determine the optimum number of TODs. Through simulation, it was found that both approaches results in almost the same values of measure of effectiveness (MOE). The proposed two approaches seem promising for similar studies in other regions, and both of them can be extended for different types of roads. The paper also suggests a procedure for considering the cyclic nature of the daily traffic in the clustering effort.     

Machinability Investigations on High Chrome White Cast Iron Using Multi Coated Hard Carbide Tools

This study investigated the performance of multilayer hard coated carbide tool and multi-response optimization of the turning process for an optimal parametric combination to yield the minimum cutting forces and machining power with a maximum material removal rate (MRR) using Taguchi and artificial neural network (ANN) methods. In recent times, high chrome white cast iron finds increasing applications in aerospace, mining, mineral process industries. Its machinability using carbide insert (TiC/TiCN/Al₂O₃) cutting tool has been studied. The influences of cutting parameters on the cutting forces, MRR and machining power of the process have been analyzed using analysis of variance and the results are correlated using ANN. Linear regression method was used to establish the relation between the cutting parameters and the process responses. The confirmation test reveals that, the accuracy of prediction of ANN is better than that of the regression analysis. In view of the good performance of the carbide tools (at optimum conditions), it can replace the cosly CBN, with improved economic benefits.     

Application of Statistical and Machine Learning Techniques for Laboratory-Scale Pressure Filtration: Modeling and Analysis of Cake Moisture

This study deals with the modeling and analysis of the pressure filtration process using statistical and machine learning techniques. The effects of externally controllable process-influencing factors such as pressure, pH, temperature, solids concentration, filtration time, air-blow time, and cake thickness on filtration performance, measured in terms of cake moisture, were modeled. A 9-factor regression model based on an exhaustive search algorithm and a 7-6-1 artificial neural network (ANN) model based on a resilient backpropagation algorithm were developed and gave R² values of 0.84 and 0.94, respectively. Relative importance of input variables was analyzed using novel methods such as added-variable plots based on the regression model and Olden’s method based on the ANN model. Results from both methods established a negative correlation for pressure, solids concentration, filtration time, temperature, and air-blow time and a positive correlation for cake thickness and pH. Analysis from regression and ANN models indicated pH to be the most significant process-influencing factor. Even though both models served as good interpretable models, the ANN model outperformed the regression model in terms of predictive capability, with an R² value of 0.965 compared with the regression model’s 0.750 for the test dataset.     

Prediction of carbonisation time in byproduct recovery coke ovens through gooseneck temperature measurement

Precise prediction of the end of carbonisation possesses intangible benefits in the coke making process. The coke ovens in Tata Steel measure the raw gas temperature (at the gooseneck arrangement in the oven top) to identify the end of coking. Based on the gooseneck temperature profile, the carbonisation time is divided into active carbonisation time (ACT) and soaking time. As the soaking time is varied between 45?min and 1.5?h as per the need, the current study focusses on developing a mathematical model to predict the ACT given the coal blend properties and the operating conditions of the oven. Different statistical methods ranging from linear regression to artificial neural network (ANN) have been used to arrive at a robust model. Piece-wise linear regression and ANN have been found to out-perform the other statistical techniques. However, the ANN model is preferred in terms of the predictability of unseen data.