Chonghun Han

Real-time monitoring for a process with multiple operating modes

Abstract Real-time monitoring of a chemical process with multiple operation modes is a challenging problem. The frequent changes of operation modes require frequent updates of monitoring models, which lead to the frequent pauses of the real-time monitoring activities. This paper proposes a monitoring methodology for a process with multiple operation modes based on hierarchical clustering and a super PCA model. The case studies have shown that the super peA model has shown better performances than a single PCA model for all operation modes and local PCA models developed for each operation mode.

A nonlinear soft sensor based on multivariate smoothing procedure for quality estimation in distillation columns

Abstract An accurate on-line measurement of quality variables are essential for the successful monitoring and control tasks in chemical process operations. However, due to the measurement difficulties such as the large time delays, the soft sensor, an inferential model, for the target quality variable, has been widely used as an alternative for the physical sensors. Partial least-squares (PLS) was used to develop a soft sensor because it can handle the correlations among many variables. However, the successful applications of linear projection methods like PLS were limited to only the cases without strong nonlinearities. This paper proposes a design methodology to build a soft sensor for chemical processes that can handle the correlations among many process variables and nonlinearities based on smoothness concept. The method has been directly motivated by the locally weighted regression that estimates a regression surface through multivariate smoothing. The proposed method will be illustrated by comparisons with other familiar methods. The industrial case studies have shown that the proposed method gives a better or equal performances over other methods such as PLS, nonlinear PLS and artificial neural networks.

Estimation methods for fatigue properties of steels under axial and torsional loading

Uniaxial and torsional fatigue tests have been conducted on eight steels. The cyclic equivalent stress and strain amplitudes can be fitted by the Ramberg-Osgood relationship. Fatigue lives are found correlated with the equivalent strain amplitude. Seven methods for estimating uniaxial fatigue properties from tensile properties or hardness have been evaluated. The modified universal slopes method by Muralidharan and Manson, the uniform material law by Baumel and Seeger and the hardness method by Roessle and Fatemi predicted over 93% of test cases within the factor of 3 compared with observed lives. These methods are also found applicable to torsional fatigue with fatigue properties estimated from uniaxial fatigue properties based on the equivalent strain criterion.

A NOVEL MILP MODEL FOR PLANTWIDE MULTIPERIOD OPTIMIZATION OF BYPRODUCT GAS SUPPLY SYSTEM IN THE IRON- AND STEEL-MAKING PROCESS

A novel MILP model for plantwide multiperiod optimization of byproduct gas supply system in the iron- and steel-making process is proposed. Compared with the previous optimization model, proposed approach simultaneously optimizes the byproduct gasholder levels and gas distribution among conflicting objectives. Both integer and continuous variables are used in determining the optimal fuel load change according to the fuel types. Objectives include the minimization of the unfavorable byproduct gas emission or shortage, oil consumption, the number of turn on/off of the burner, maintaining the normal holder levels, and maximizing fuel usage efficiency. Case study results show that the proposed model finds the optimal solution in terms of total cost reduction and the different optimization model structure makes the solution more applicable than the previous approach.

Intelligent integrated plant operation system for Six Sigma

Abstract Six Sigma has been widely adopted in a variety of industries as a proven management innovation methodology to produce high-quality products and reduce the cost at all the levels of an enterprise. This paper proposes a plant operation system which can guide plant engineers and operators to pursue Six Sigma activities by providing supports for key elements of Six Sigma: measurement, analysis, improvement and control. Multivariate statistical process control (MSPC) techniques have been employed as key technologies for the system, along with the plant information systems. This paper also discusses the future research issues that should be addressed to implement the described system.

Real-time classification of petroleum products using near-infrared spectra

Abstract This study describes a real-time classification methodology for petroleum products based upon the near-infrared (NIR) spectra. The proposed real-time classifier (RTC) is designed based on the combination of principal component analysis (PCA) and a Bayesian classifier. Principal component analysis is employed to extract essential features that are selected considering both classification power and easiness of implementation for classification of the spectra. Bayesian classifier minimizes classification error. The RTC based on NIR spectra offers the faster and more accurate identification capacity of products on-line than the conventional analyzers. The proposed RTC has been applied to classify six petroleum products: diesel, gasoline, kerosene, light gas oil, light straight-run, and naphtha. It has shown good classification power for industrial petroleum products.

Calibration transfer of near-infrared spectra based on compression of wavelet coefficients

Abstract Near-infrared (NIR) spectrometers are widely used as online analyzers. We propose a transfer algorithm of a calibration model for the near-infrared spectrometer. The method, which is based on direct standardization in the compressed wavelet domain, is called WTDS. Calibration transfer in the compressed wavelet domain permits a reduction in processing time for the analysis of large spectral data sets. A calibration transfer based on a compression of wavelet coefficients offers increased robustness against testing errors, with no significant loss of performance. We select the proper set of wavelet coefficients based on the hard thresholding approach. The WTDS shows an improvement in calibration transfer performance, although wavelet coefficients are compressed.

Application of neural network to the supervisory control of a reheating furnace in the steel industry

Abstract The efficient and reliable control of a reheating furnace is a challenging problem, due to: (a) the many different types of billets to be processed, (b) the strong intercorrelation among process variables, (c) the large dimension of the input and output space, (d) the strong interaction among process variables, (e) a large time delay, and (f) highly nonlinear behavior. Thus, conventional reheating furnace operation has been heavily dependent upon look-up tables which list the optimal set points. This paper describes a modified modular neural network for the supervisory control of a reheating furnace. Based on the divide-and-conquer concept, a modular network is capable of dividing a complex task into subtasks, and modeling each subtask with an expert network. To model such activities, a gating network is used for the classification and allocation of the input data to the corresponding expert network. To overcome the correlation effects among process variables and the problem of dimensionality, principal component analysis (PCA) has been employed to remove the correlation and reduce the problem dimension. From PCA analysis, it was possible to decide on the optimal dimension for the problem, to describe the dynamic behavior of the furnace. The proposed neural network has been trained and tested using operational data from the reheating furnace and has been implemented on the wire rod mill process of POSCO TM .

Hierarchical time-optimal control of a continuous copolymerization reactor during start-up or grade change operation using genetic algorithms

Abstract During start-up or grade-change operation of a continuous polymerization reactor, the transition time, required to reach the desired steady state from the initial state, decides the amount of off-specification polymer product which causes treatment problems. It is very important to calculate the minimum transition time and subsequently determine optimal control input trajectories. Due to complex reaction mechanism and highly nonlinear dynamics, however, this time-optimal control problem for a continuous copolymerization reaction during start-up or grade-change operation has been a challenging problem. We propose two level hierarchical time-optimal control methodology that is based on the discretization of transition time interval and application of genetic algorithms combined with heuristic constraints. The proposed methodology has shown better performance than IDP or SQP in terms of accuracy and efficiency. The methodology is illustrated by the application to the time-optimal control problem of continuous MMA-VA copolymerization to produce polymer with a desired molecular weight and composition of VA in a dead copolymer.

Plant-wide Optimal Byproduct Gas Distribution and Holder Level Control in the Iron and Steel Making Process

A new plant-wide multiperiod optimization approach is proposed for optimal byproduct gas distribution to prevent unfavorable byproduct gas emission and equipment trip and simultaneously to maximize the efficiency of energy resource usage in the iron and steel making process. Compared with the previous approach, the proposed approach finds the optimal trade off among conflicting objectives such as holder level control, minimization of oil consumption and number of burner switching, and the maximization of generating electricity. To consider the different fuel load change operation according to the fuel types, both integer and continuous variables are used. Case studies were performed to verify the usefulness of the proposed approach, and the results show good performance in terms of the reduced number of burner switching which leads to the reduction of total cost and producing operation-easy solutions.