Just-in-time Learning Research Articles

Locally Weighted Kernel Principal Component Regression Model for Soft Sensing of Nonlinear Time-Variant Processes

The principal component regression (PCR) based soft sensor modeling technique has been widely used for process quality prediction in the last decades. While most industrial processes are characterized with nonlinearity and time variance, the global linear PCR model is no longer applicable. Thus, its nonlinear and adaptive forms should be adopted. In this paper, a just-in-time learning (JITL) based locally weighted kernel principal component regression (LWKPCR) is proposed to solve the nonlinear and time-variant problems of the process. Soft sensing performance of the proposed method is validated on an industrial debutanizer column and a simulated fermentation process. Compared to the JITL-based PCR, KPCR, and LWPCR soft sensing approaches, the root-mean-square errors (RMSE) of JITL-based LWKPCR are the smallest and the prediction results match the best with the actual outputs, which indicates that the proposed method is more effective for quality prediction in nonlinear time-variant processes.

Identification and Compensation of Force Ripple in PMSLM using a JITL Technique

AbstractThis paper presents an adaptive compensation scheme for reducing the force ripple effects in permanent magnet synchronous linear motor (PMSLM). Firstly, when the velocity of a linear motor is close to zero, based on a fast Fourier transform (FFT) analysis, the dominant displacement periodicity of the force ripple can be obtained for simplifying the ripple model. Next, to enhance the identification accuracy and guarantee the real‐time performance of the linear servo system simultaneously, an improved just‐in‐time learning (JITL) algorithm is proposed to estimate the ripple model parameters. Finally, based on the identification results, an adaptive feedforward controller is implemented in the linear servo system to compensate the force ripple. The main motivation of this paper is the extension of the identification methods to improve the servo control performance in industrial applications. Simulation and experimental results demonstrate the effectiveness of the proposed method, even when compared to a traditional compensation method based on a recursive least squares algorithm.

Batch process monitoring based on just-in-time learning and multiple-subspace principal component analysis

Batch or fed-batch process monitoring is a challenging task because of its characteristics such as batch-to-batch variations, inherent time-varying dynamics, and multiple operating phases. Thus, a new batch process monitoring method based on just-in-time learning (JITL) and multiple-subspace principal component analysis (MSPCA) is developed. Based on offline one batch normal data, the division algorithm of multiple subspace is proposed, in which mutual information (MI) and K-means are employed to derive the segmentation rule of variable subspace and then the variables are divided into several subspaces according to the segmentation rule of variable subspace. At online monitoring, the training data set for modeling is obtained by JITL and separated into each subspace according to the segmentation rule of variable subspace. Principal component analysis is employed to build the model in each subspace, and all components are retained to calculate T2 statistics. A unique probability index is obtained by Bayesian inference (BI) as the decision fusion strategy of T2 statistics of all subspaces. A simple numerical example is used to show the advantages of the proposed MSPCA method. The feasibility and effectiveness of JITL–MSPCA is demonstrated by fed-batch penicillin fermentation.

JITL-based concentration control for semi-batch pH-shift reactive crystallization of l-glutamic acid

Although concentration control (C-control) strategy has been shown to give effective and robust control performance for batch cooling and semi-batch antisolvent crystallizations in recent years, no research work was reported concerning the potential application of conventional C-control for the more challenging semi-batch pH-shift reactive crystallization that is common in the process industries. To this end, this paper presents detailed analysis to find out that it is not feasible to apply the C-control to semi-batch pH-shift reactive crystallization. To circumvent this problem, a variant of C-control strategy by incorporating the Just-in-Time Learning (JITL) method to cope with strong process nonlinearity inherent in the pH-shift reactive crystallization is developed in this paper. Simulation results are presented to illustrate the proposed design and a comparison with conventional optimal control is made.

Adaptive speed control based on just-in-time learning technique for permanent magnet synchronous linear motor

In this paper, an adaptive two degrees of freedom (2Dof) PI controller based on a just-in-time learning (JITL) method is proposed for predictive speed control of permanent magnet synchronous linear motor (PMSLM). Firstly, to guarantee the high identification accuracy and high real-time performance simultaneously, an improved JITL method is proposed to estimate the controlled model parameters of speed control system. Then, based on the dynamic controlled model, a simplified generalized predictive control (GPC) supplies a 2Dof proportional integral (PI) controller with suitable control parameters to follow a sinusoid-type speed command in operating conditions. The main motivation of this paper is the extension of the predictive controller to replace traditional PI controller in industrial applications. Finally, the efficacy and usefulness of the proposed controller are verified through the experimental results.

Enhanced batch process monitoring using just-in-time-learning based kernel partial least squares

Batch process has the characteristics of dynamic behavior, strong nonlinearity and multiplicity of operation phases. In this paper, an integrated framework consisting of just-in-time-learning (JITL) method and kernel partial least squares (KPLS) is described for the monitoring of the performance of batch processes. The training data set for modeling is defined by using JITL, and KPLS is employed to build the model for process monitoring purpose. The JITL model based on local neighborhoods of similar samples is very accurate and sensitive because it can well track the change of the process. Therefore, the proposed method is suited for multiphase and time-varying batch processes. Meanwhile, KPLS is a very efficient technique for tacking complex nonlinear data sets. Furthermore, the proposed monitoring strategy does not require the estimation of future values of the process variables during online application. The effectiveness of the proposed monitoring approach is demonstrated through two sets of benchmark data. The comparison of monitoring results shows that the proposed JITL–KPLS approach is superior to multiway KPLS and hierarchical KPLS, and it can achieve accurate detection of various types of process faults occurring in the batch operation.

Just-in-Time Kernel Learning with Adaptive Parameter Selection for Soft Sensor Modeling of Batch Processes

An efficient nonlinear just-in-time learning (JITL) soft sensor method for online modeling of batch processes with uneven operating durations is proposed. A recursive least-squares support vector regression (RLSSVR) approach is combined with the JITL manner to model the nonlinearity of batch processes. The similarity between the query sample and the most relevant samples, including the weight of similarity and the size of the relevant set, can be chosen using a presented cumulative similarity factor. Then, the kernel parameters of the developed JITL-RLSSVR model structure can be determined adaptively using an efficient cross-validation strategy with low computational load. The soft sensor implement algorithm for batch processes is also developed. Both the batch-to-batch similarity and variation characteristics are taken into consideration to make the modeling procedure more practical. The superiority of the proposed soft sensor approach is demonstrated by predicting the concentrations of the active biomass and recombinant protein in the streptokinase fed-batch fermentation process, compared with other existing JITL-based and global soft sensors.

Development of Interval Soft Sensors Using Enhanced Just-in-Time Learning and Inductive Confidence Predictor

In the development of soft sensors for chemical processes, outliers of input variables and the time-varying nature of the process are difficult to address, thereby often resulting in unavailable prediction. Motivated by these issues, new just-in-time (JIT) learning is derived to track the normal changes of processes regardless of abrupt noises. Such an approach adapts a proposed robust nearest correlation (RNC) algorithm with multimodel ensemble learning to enhance conventional JIT learning. Furthermore, to gauge the quality of the given prediction, we integrate such JIT learning with the inductive confidence predictor (ICP) to yield a new soft sensor called the “interval soft sensor” which generates not only prediction values but also associated confidence values that represent the credibility of a soft sensor’s output. These ideas were applied to a wastewater treatment process. The proposed interval soft sensor was seen to be effective for prediction in the absence and presence of outliers in the proces...

적시학습 지원을 위한 모바일러닝이 학습만족도 및 학습역량에 미치는 효과

본 연구는 일터학습을 하는 직장인을 대상으로 적시학습 지원을 위한 모바일러닝이 학습만족도와 학습역량에 미치는 효과를 분석하기 위해 수행되었다. 이를 위해 2010년 공공기관에서 모바일러닝으로 학습한 597명을 대상으로 온라인 설문조사를 실시하였다. 연구결과, 모바일러닝의 학습만족도와 학습역량에 유의미한 예측변인은 학습콘텐츠인 것으로 나타났다. 학습역량의 하위변인인 자기주도 학습능력과 문제해결 학습능력에 대한 모바일러닝의 학습콘텐츠와 사용편의성의 예측정도를 분석한 결과, 자기주도 학습능력은 학습콘텐츠가 유의미한 예측변인인 반면, 문제해결 학습능력은 학습콘텐츠와 사용편의성 모두 예측변인으로 나타났다. 연구결과를 바탕으로 학습만족도 및 학습역량강화를 위한 모바일러닝의 실제적 전략을 통합적으로 제시하였다.

A comparative study of just-in-time-learning based methods for online soft sensor modeling

Most traditional soft sensors are built offline and only to be used online. In modern industrial processes, the operation condition is changed frequently. For these time-varying processes, online soft sensor modeling is required, since the prediction result is highly related to other components of the process control system. In the present paper, a comparative study of three different just-in-time-learning (JITL) methods for online soft sensor modeling is carried out, which are based on partial least squares (PLS), support vector regression (SVR) and least squares support vector regression (LSSVR). Different from traditional soft sensors which model the process through a global and offline manner, the JITL-based method exhibits an online local model structure, depending on which the change of the process can be well tracked. Besides, the process nonlinearity can also be addressed under this modeling framework. As a further contribution of this paper, a real-time performance improvement strategy is proposed to enhance the online modeling efficiency of the JITL-based soft sensor. For performance evaluation, two industrial case studies are provided.

Adaptive IMC controller design using linear multiple models

In this paper, an adaptive IMC controller using just-in-time learning (JITL) technique for nonlinear process control is proposed. Based on a set of linear models obtained on-line by the JITL, not only the parameters of IMC controller are updated, but also IMC filter parameter is adjusted on-line by an updating algorithm derived based on the Lyapunov method to guarantee the convergence of JITL's predicted tracking error. Simulation results are presented to demonstrate the advantage of the proposed adaptive IMC design over its conventional counterparts.

Adaptive generalized predictive control based on JITL technique

In this paper, an adaptive generalized predictive control (GPC) strategy based on the just-in-time learning (JITL) technique is developed. In the proposed controller design, process nonlinearities are accounted for by the set of local models obtained on-line by the JITL technique and the optimal control actions are obtained by solving the quadratic optimization problem formulated in the GPC design framework. Simulation results are presented to illustrate the advantage of the proposed GPC design and a comparison with its conventional counterparts is made.

Robust PID controller design for nonlinear processes using JITL technique

A robust PID controller design methodology for nonlinear processes is proposed based on the just-in-time learning (JITL) technique. To do so, a composite model consisting of a nominal ARX model and the JITL, where the former is used to capture linear process dynamics and the latter to approximate the inevitable modeling error caused by the process nonlinearity, is employed to model the process dynamics in the operating space of interest. The state space realizations of this composite model and PID controller are then reformulated as an uncertain closed-loop system, by which the corresponding robust stability condition is developed. Literature examples are employed to illustrate the proposed methodology and a comparison with the previous result is made.

Online monitoring of nonlinear multiple mode processes based on adaptive local model approach

A new adaptive local model based monitoring approach is proposed for online monitoring of nonlinear multiple mode processes with non-Gaussian information. To solve the multiple mode problem, just-in-time-learning (JITL) strategy is introduced. The local least squares support vector regression (LSSVR) model is built on the relevant dataset for prediction. To satisfy the online modeling demand, the real-time problem is considered. Then a two-step independent component analysis–principal component analysis (ICA–PCA) information extraction strategy is introduced to analyze residuals between the real output and the predicted one. Two case studies show that the new proposed method gives better performance compared to conventional methods.

Self-tuning PID controllers based on the Lyapunov approach

In this paper, a self-tuning PID controller using just-in-time learning (JITL) technique is proposed. A set of linear models obtained by the JITL provides the information required to adjust the parameters of PID controller. The self-tuning algorithm for the PID parameters is derived by a rigorous analysis based on the Lyapunov method such that the JITL's predicted tracking error converges asymptotically. Simulation results are presented to illustrate the proposed self-tuning PID controller and a comparison with its conventional counterparts is made.

Adaptive Single-Neuron Controller Design for Nonlinear Process Control

In this paper, a new adaptive single-neuron (ASN) controller is proposed based on the just-in-time learning (JITL) technology for nonlinear process control. To mimic the traditional PID controller, a single neuron is employed in the proposed controller design strategy. Incorporated with the neural network’s learning ability, the proposed controller can control the process adaptively through the updating of its parameters by the adaptive learning algorithm developed and the information provided from the JITL. Compared with the neural network based PID controller designs previously developed, ASN controller is more amenable to on-line implementation. Simulation results are presented to illustrate the proposed method and a comparison with its conventional counterparts is made.

Partitioned model-based IMC design using JITL modeling technique

A partitioned model-based internal model control (PM-IMC) design strategy is proposed for a class of nonlinear systems that can be described by just-in-time learning (JITL) modeling technique. The PM-IMC scheme consists of a conventional IMC controller augmented by an auxiliary loop to account for nonlinearities in the system. Two alternative implementations of the JITL are discussed and compared via simulation studies of an industrial polymerization reactor and an isothermal reactor exhibiting inverse response. It is shown that PM-IMC using the database-updating JITL is more desirable owing to the relative ease in collecting the process data required to construct its initial database, while achieving comparable control performance as that obtained by PM-IMC using the JITL with fixed-database, which requires process data collected over the entire operating region to construct its database. In addition, a comparison is made between PM-IMC and its linear counterpart.

Adaptive IMC Controller Design for Nonlinear Process Control

In this paper, a new adaptive internal model control (IMC) controller is proposed based on just-in-time learning (JITL) technique for nonlinear process control. The parameters of the proposed IMC controller are updated not only based on the information provided by the JITL, but also its filter parameter is adjusted online by an adaptive learning algorithm. Compared with the previous nonlinear IMC controller design methods, it is straightforward for the proposed method to obtain the model inversion based on the JITL modelling technique. Simulation results are presented to demonstrate the advantage of the proposed adaptive IMC controller design over its conventional counterpart.

Modelling and Control of Multivariable Processes Using Generalized Hammerstein Model

In this paper, a generalized Hammerstein model consisting of a static polynomial function in series with time-varying linear model is proposed in order to model the Hammerstein-like multivariable processes whose linear dynamics vary over the operating space. An iteration procedure is proposed to identify the generalized Hammerstein model by using the just-in-time learning (JITL) technique. Unlike the conventional Hammerstein model, only the static nonlinear part of generalized Hammerstein model, thus identified is retained in its subsequent application in controller design. Consequently, the on-line use of proposed model requires the identification of linear model by using the JITL technique and current process data. An adaptive decentralized PID control strategy based on the proposed model is also developed. The controller parameters are adjusted on-line by using the gradient descent algorithm and information provided by the JITL. Simulation results show that the proposed methods have better prediction accuracy and control performance than those achieved by the conventional Hammerstein model and its associated adaptive decentralized PID controller design, respectively.

Commentary: Just in time learning with radio frequency identification devices (RFIDs)

Commentary: Just in time learning with radio frequency identification devices (RFIDs)