Abstract: China has abundance of mineral resources such as magnesite, hematite and bauxite, which constitute a key component of its economy. The relatively low grade, and the widely varying and complex compositions of the raw extracts, however, pose difficult processing challenges including specialized equipment with excessive energy demands. The energy intensive furnaces together with widely uncertain features of the extracts form hybrid complexities of the system, where the existing modeling, optimization and control methods have met only limited success. Currently, the mineral processing plants generally employ manual control and are known to impose greater demands on the energy, while yielding unreasonable waste and poor operational efficiency. The recently developed Cyber-Physical System (CPS) provides a new key for us to address these challenges. The idea is to make the control system of energy intensive equipment into a CPS, which will lead to a CPS driven control system.

This talk presents the syntheses and implementation of a CPS driven control system for energy-intensive equipment under the framework of CPS. The proposed CPS driven control system consists of four main functions: (I) setpoint control; (II) tracking control; (III) self-optimized tuning; and (IV) remote and mobile monitoring for operating condition. The key in realizing the above functions is the integrated optimal operational control methods to implement setpoint control, tracking control and self-optimized tuning together seamlessly. This talk introduces the integrated optimal operational control methods we proposed.

Hardware and software platform of CPS driven control system for energy-intensive equipment is then briefly introduced, which adopts embedded control system, wireless network and industrial cloud. It not only realizes the functions of computer control system using DCS (PLS), optimization computer and computer for abnormal condition identification and self-optimized tuning, but also achieves the functions of mobile and remote monitoring for industrial process.

Then, using fused magnesium furnace as an example, a hybrid simulation system for CPS driven control system for energy-intensive equipment developed by our team is introduced. The results of simulation experiments show the effectiveness of the proposed method that integrates the setpoint control, tracking control, self-optimized tuning and remote and mobile monitoring for operating condition in the framework of CPS.

The industrial application of the proposed CPS driven control system is also discussed. It has been successfully applied to the largest magnesia production enterprise in China, resulting in great returns. Finally, future research on the CPS driven control system is outlined.

Keywords: Modeling and Identification, Optimization and Scheduling, Model-based Control
Abstract: Stiction in control valves is considered one of the most common causes of poor performance in control loops. Thus, non-invasive, reliable and efficient methods which can detect and quantify this malfunction are highly desirable in the process industry. Under the framework of Hammerstein model identification and nonlinear optimization, this paper proposes an approach to estimate stiction amount on the basis of a recently proposed smoothed model. One of the motivations of the work is to improve the performance of a stiction unaware model predictive controller which exhibits sustained oscillations in the presence of valve stiction. By augmenting process model with the identified valve dynamics, the controller is turn to a stiction embedding formulation which can actually remove fluctuations and then guarantees good set-point tracking. Applications to simulation case studies and industrial loops are used to demonstrate the validity of the proposed method. Results are compared with a standard grid-search approach and other identification techniques of the literature.

Keywords: Energy Processes and Control, Model-based Control
Abstract: This paper presents an analysis of the non-inertive-feedback thermofluidic engine (NIFTE) under cyclic steady-state conditions. The analysis is based on a nonlinear model of NIFTE that had previously been validated experimentally, and applies an optimization-based approach to detect the cyclic steady states (CSS). The stability of the CSS is furthermore determined by analyzing their monodromy matrix. It is found that NIFTE can exhibit multiple CSS for certain values of the design parameters, which may be either stable or unstable, a result that had not been reported before. Subsequently, a parametric study is conducted by varying key design parameters, revealing that higher efficiencies could be achieved by controlling the engine at different CSS, including unstable ones. Lastly, the paper investigates the trade-offs between efficiency and work output in NIFTE.

Keywords: Big Data Analytics and Monitoring, Process Applications, Model-based Control
Abstract: This paper addresses the fault detectability problem for PCA-based process monitoring methods under closed-loop control. Unlike previous research assuming the same process variation for both modeling and monitoring periods, the impact of controller on data variation is considered. The effective fault direction is given in order to describe the effect of process variation on fault detection index under different controllers. The minimum variance along the effective fault direction is developed by introducing the concept of block-lower-triangular interactor matrix and conditional minimum variance control. The sufficient condition for a fault to be detectable under any controllers is provided. The proposed method is demonstrated with the simulation examples.

Keywords: Modeling and Identification
Abstract: Reliable and timely diagnosis of system faults under uncertainties is imperative for safe, reliable, and profitable operation of technical systems. This paper presents an input design method for active fault diagnosis for nonlinear systems that are subject to probabilistic model uncertainty and stochastic disturbances, and are under operational constraints. A computationally efficient sample-based method is presented for joint propagation of model uncertainty and stochastic disturbances using non-intrusive generalized polynomial chaos and unscented transformation. A tractable sample-based distance measure, inspired by the k-nearest neighbors algorithm, is used for fault diagnosis, which seeks to discriminate between probabilistic predictions of the model hypotheses for normal and faulty operation. Simulation results on a benchmark bioreactor case study demonstrate the effectiveness of the proposed input design method for reliable fault diagnosis under uncertainty through online model discrimination.

Keywords: Big Data Analytics and Monitoring, Process Applications, Optimization and Scheduling
Abstract: In this article, a novel change detection algorithm is proposed based on the Kantorovich distance concept. Incorporating the proposed change detection algorithm with the existing process monitoring tools may assist the operator in detecting dynamic changes in process plants and provide fewer unnecessary (false) alarms as well as fewer missed alarms. The proposed change detection method was tested through simulation data. It is applied to the benchmark Tennessee Eastman (TE) process in online mode. Results prove the efficacy of the proposed method to detect both the sharp and incipient changes.

Keywords: Modeling and Identification, Big Data Analytics and Monitoring, Process Applications
Abstract: Continuous glucose monitoring (CGM) sensors are a critical component of artificial pancreas (AP) systems that enable individuals with type 1 diabetes to achieve tighter blood glucose control. CGM sensor signals are often afflicted by a variety of anomalies, such as biases, drifts, random noises, and pressure-induced sensor attenuations. To improve the accuracy of CGM measurements, an on-line fault detection method is proposed based on sparse recursive kernel filtering algorithms to identify faults in glucose concentration values. The fault detection algorithm is designed to effectively handle the nonlinearity of the measurements and to differentiate the normal variability in the glycemic dynamics from sensor anomalies. The effectiveness of the proposed recursive kernel filtering algorithm for sensor error detection is demonstrated using simulation studies.

Keywords: Process and Control Monitoring, Process Applications
Abstract: To improve the performance of the abnormal condition identification, the multi-source information of the abnormal conditions in the electro-fused magnesia smelting process is analyzed in this paper. An intelligent abnormal condition identification method is proposed based on Bayesian network (BN). By analyzing three main abnormal conditions and the experience of the operators, the characteristics related with the abnormal conditions are extracted. The BNs are established to identify the abnormal conditions by fusing the multi-source information. The simulation results show that the proposed method can realize abnormal condition identification, distinguish the degree of the abnormal condition, and obtain better performance.

Keywords: Big Data Analytics and Monitoring, Modeling and Identification, Energy Processes and Control
Abstract: Large wide-area monitoring systems generate a large amounts of phasor measure- ment unit (PMU) data. Although single variable analysis methods are often applied to relative phase angle difference (RPAD) between two PMU locations for event detection, the possible locations of the events cannot be estimated by such methods. In this paper, we apply a Dynamic- inner canonical correlation analysis (DiCCA) based discrete Fourier transform method to detect the events in the PMU data and identify the possible location of the events. A case study on a real PMU dataset demonstrates the effectiveness of the proposed method.

Keywords: Optimization and Scheduling, Model-based Control
Abstract: This work contributes to the optimal design of closed-loop nonlinear systems with input saturation in the presence of unknown uncertainty. Stability conditions based on contractive constraints were developed for a general class of nonlinear systems under some Lipschitz assumptions. Closed-loop robust stability and robustly optimal performance can be guaranteed in the presence of input bounds, if the solution of the design problem, formulated as a nonlinear semi-infinite program (SIP) with differential equation constraints, can be guaranteed to be feasible. In this work, the SIP is solved by means of a local reduction approach, which requires a local representation of the so-called lower level problems associated with the SIP. The suggested design method is illustrated by means of chemical reactor control problem.

Keywords: Process Applications, Modeling and Identification
Abstract: Belousov-Zhabotinsky reaction generates self-organized oscillatory pattern which is common in biological systems, synergistic study of oscillatory patterns will assist understanding and modeling complex processes in biological sphere. However, the analytical solution of a self-oscillator is difficult because the system exhibits nonlinear dynamics. In this study, a frequency domain analysis of Hopf bifurcation based on a closed-loop representation is addressed. For better understanding of the closed-loop mechanism, a Laplace-Borel transform is implemented, which is proved to be effective in identifying the coefficients of the harmonics.

Keywords: Process Applications, Optimization and Scheduling
Abstract: This study is concerned with the development of an extremum seeking (ES) strategy based on recursive least square (RLS) for on-line estimation, and a regression model in the form of a Hammerstein-Wiener model. RLS usually provides a faster convergence than the classical bank of filter estimators, and the consideration of process dynamics allows to take account for the phase-shift and attenuation occurring when increasing the frequency of the dither signal. The resulting ES scheme achieves very significant improvement in convergence speed, as illustrated with a numerical example, and a more realistic application to micro-algae cultures in a photobioreactor in simulation.

Keywords: Model-based Control, Process Applications
Abstract: A robust adaptive controller is developed for a chemical process using a generalized Polynomial Chaos (gPC) expansion-based Markov decision model, which can account for time-invariant probabilistic uncertainty and overcome computational challenge for building Markov models. To calculate the transition probability, a gPC model is used to iteratively predict probability density functions (PDFs) of system’s states including controlled and manipulated variables. For controller tuning, these PDFs and controller parameters are discretized to a finite number of discrete states for building a Markov model. The key idea is to predict the transition probability of controlled and manipulated variables over a finite future control horizon, which can be further used to calculate an optimal sequence of control actions. This approach can be used to optimally tune a controller for set point tracking within a finite future control horizon. The proposed method is illustrated by a continuous stirred tank reactor (CSTR) system with stochastic perturbations in the inlet concentration. The efficiency of the proposed algorithm is quantified in terms of control performance and transient decay.

Keywords: Model-based Control, Process Applications, Optimization and Scheduling
Abstract: This paper is concerned with optimal feedback control synthesis for periodic processes with economic control objectives. The focus is on tube-based methods which optimize over robust forward invariant tubes (RFITs) in order to determine the nonlinear feedback law. The main contribution is an approach to conservatively approximating this set-based periodic feedback control optimization problem by a tractable optimal control problem, which can be solved with existing optimal control solvers. The approach is applied to an uncertain periodic biochemical production process, where the objective is to maximize the profit subject to robust safety constraints.

Keywords: Model-based Control, Optimization and Scheduling, Process Applications
Abstract: This paper concerns the input-output model identification and zone model predictive control of an agro-hydrological system modeled by a partial differential equation. The primary control objective is to maintain the soil moisture within a desired range which is suitable for grass grow. There is also a secondary control objective which is to reduce the total irrigation amount. First, a linear parameter varying (LPV) model is identified for controller design purpose using a maximum likelihood gradient-based iterative estimation method. Then, based on the LPV model, a zone model predictive control (MPC) is designed which uses an output disturbance and state observer to reduce model-plant mismatch and an asymmetric target zone to reduce irrigation amount under weather uncertainties while maintaining the soil moisture within the target range. Simulation studies show that the LPV model is a good approximation of the original nonlinear model and effectively reduces the online computational load of the MPC, and that the proposed zone MPC can lead to significant water conservation.

Keywords: Batch Process Modeling and Control, Model-based Control, Process Applications
Abstract: The control strategies for batch processes in the past can be categorized into two levels. The higher level is economic optimization running at low frequency and the lower one tracks the given reference using MPC or PID at the higher level. The lower level regards all the disturbances as something to be rejected according to a quadratic based optimization objective. However, not all the disturbances are unfavorable to batch processes; some of them are helpful. In this paper, an economic batch-to-batch optimization method for batch processes is directly applied at the lower level. It replaces the former tracking strategy. With the information of disturbances collected from the previous batches, the iterative learning control strategy (ILC) can find out better operation profiles. ILC has the advantage of continuously improving the economic performance of the current batch with enriched information of disturbances from batch to batch. To demonstrate the potential applications of the proposed design method, a typical fed-batch bioreactor is applied.

Keywords: Model-based Control, Batch Process Modeling and Control, Modeling and Identification
Abstract: This paper presents a scheme for dual robust control of batch processes under parametric uncertainty. Some recently proposed approaches can be used to tackle this problem, however, this will be done at the price of conservativeness or significant computational burden. In order to increase computational efficiency, we propose a scheme that uses parametrized conditions of optimality in the adaptive predictive-control fashion. The dual features of the controller, i.e., balance between the control moves that excite the system to improve accuracy of the parameter estimation and between the moves that optimize process performance, is realized through scenario-based (multi-stage) approach, which allows for modeling of the adaptive robust decision problem and for projecting this decision into predictions of the controller. The proposed approach is illustrated on a case study from batch membrane filtration.

Keywords: Batch Process Modeling and Control, Model-based Control, Process Applications
Abstract: Butylated urea formaldehyde (BUF) is a key intermediate for manufacturing paint and coating. The quality of BUF resins can be measured in terms of the concentration of free formaldehyde in the BUF resins and the extent of butylation. We in this work present an optimal control study to obtain minimum free formaldehyde concentration and minimum butanol concentration at the end of the batch operation. Reactor temperature is used as the manipulated variable and optimum temporal reactor temperature profiles are obtained using control vector parameterization approach. The two aforementioned criteria are observed to be mutually conflicting and hence the multi-objective optimal control problem is solved in this work yielding the pareto optimal curve showing the trade-off solutions of the MOO problem. Such pareto optimal curve helps the operator to choose an operating condition for a desired operation.

Keywords: Batch Process Modeling and Control, Model-based Control, Optimization and Scheduling
Abstract: Optimization-based iterative learning control (OILC) has been widely applied to batch processes due to its fast convergence, good control performance and ability to handle constraints. However, how to guarantee constraint satisfaction and convergence of tracking error in the presence of unknown system nonlinearity remains open in the framework of OILC. It is important to address this issue since unknown nonlinearity is common in practice and detrimental to good control performance. In this paper, we propose a tube feedback OILC to investigate the applicability of linear-model based control strategy on batch processes with an unknown nonlinear term. First, a state feedback control law is designed to stabilize the system. The stabilized system is then decomposed into two subsystems: repeatable and unrepeatable subsystems; Second, an invariant set of states corresponding to the unrepeatable subsystem is computed, based on which an OILC is further developed for the repeatable subsystem to improve the control performance. Meanwhile, the feedback controller steers the states within a tube around the trajectory of OILC. In this way, convergence and constraint satisfaction are ensured simultaneously. Compared with the currently existing methods, the proposed method has the following advantages: (1) generality covering both stable and unstable systems; (2) low computation complexity; and (3) rigorous stability. The simulation results on injection molding velocity control demonstrate that the proposed method has superior performance.

Keywords: Batch Process Modeling and Control, Optimization and Scheduling
Abstract: In this paper, a novel dual optimization method based on process transfer model is proposed for the quality optimization control for new batch process, which combines within batch optimization method and batch-to-batch modifier-adapt strategy. Firstly, a process transfer model, JY-PLS, is applied to solve the problem that the number of new batch process data is not sufficient to build reliable latent variable process model, which transfer the data information from a based and similar batch process to the aimed new process. However, there are always difference between similar batch processes, which lead to serious plant-model mismatch. In order to cope with this problem, MCC control method is utilized to determine the optimization points and optimal setting compensation method is used to eliminate the gap between suboptimal and optimal, especially for solving the within-batch mismatch. In addition, batch-to-batch modifier adaptation is used to further overcome the batch-to-batch disturbance and solve the plant-model mismatch. Finally, the proposed approach is illustrated on the cobalt oxalate synthesis process.

Keywords: Batch Process Modeling and Control, Optimization and Scheduling, Modeling and Identification
Abstract: Model-plant mismatch commonly arises from simplifications and assumptions during the development of first-principles models. Hence, when employing such models in iterative optimization schemes, structural mismatch may lead to inaccurate prediction of the necessary conditions of optimality. This results in convergence to a predicted optimum which does not coincide with the actual process optimum. The method of simultaneous identification and optimization aims to correct for errors in the predicted gradients of the cost and constraints by adapting the model parameters. In a former implementation of this approach, the gradients have been corrected only locally at the current operating point. To achieve a better prediction of the cost function over a wider range of input conditions, we propose to consider cost measurements from previous batch experiments combined with an optimal experimental design of future experiments. Using this approach, it is possible to achieve a better prediction, especially around the optimum, and to make the gradient correction step less susceptible to uncertainty in local gradient measurements. The improvements are illustrated using a simulated run-to-run optimization study of a cell-culture process.

Keywords: Optimization and Scheduling, Process Applications
Abstract: With ever increased needs for an improved product quality, production efficiency, and cost in today’s globalized world market, advanced process control should not only realize the accuracy of each control loops, but also has the ability to achieve an optimization control of production indices that are closely related to the improved product quality, enhanced production efficiency and reduced consumption. As a result, the control and optimization of complex industrial process has attracted an increased attention of various process industries. The whole production line of industrial processes is usually composed of multiple subprocesses. The automation system of each subprocess consists of the process control part and the operational optimization system. The challenging issue is how the automation systems can be integrated to realize optimal control of the global production indices (i.e., the product quality, yield, cost and profit, etc.). This talk provides the problem description of integrated optimization for the whole plant and the proposed integrated optimization strategy for the automation systems. The proposed strategy consists of three layers because the global production indices, the production indices, and the operational indices are of multiple time scale. Optimization and adaptation, prediction and adjusting are adopted to realize the hierarchical optimization structure of different time scales. Our group has focus on the problem for ten years and the recent progress will be presented in this talk.

Keywords: Optimization and Scheduling, Process Applications
Abstract: To enhance competitiveness of industry, the manufacturing should focus on high-quality and value-added products. Quality optimization and control based on high-resolution models is still a challenging task in both academia and industry. An international research center for quality-targeted process optimization & control has been founded at Zhejiang University. Recent progress in polymerization processes with embedded molecular weight distributions will be presented in this talk. Unlike their counterparts for the manufacture of commodity chemicals, polymer processes are characterized by more complex nonlinear behavior due to nonideal thermodynamics as well complex kinetics among individual polymer chains. In particular, it is essential to capture quality indices through more detailed molecular descriptions, e.g., molecular weight distribution (MWD) models based on population balances of polymer chains. The introduction of MWD also brings forth a number of challenges. First, fundamental model identification and parameter estimation of polymerization systems are essential to establish predictive and accurate reactor models that can be solved in a time critical manner. Second, polymerization processes need to synthesized and designed to exploit chemical and physical phenomena to yield superior production, quality and resource utilization. Third, these processes need to be operated in an efficient and stable manner so that they can realize the potential predicted by an optimal design. Incorporating all of these features within large-scale systems with MWD leads to challenging research in process modeling, process design, and process operation. Novel methods with industrial applications are presented. Perspectives on future work will also be discussed.

Keywords: Model-based Control, Process Applications
Abstract: As the founding institution of petrochemical automation, East China University of Science and Technology has been carrying out fundamental research and key technology development on intelligent control and optimization for industrial processes since 1958, focusing on the demands for high efficient utilization of energy and resources in petrochemical processes. For important and typical petrochemical processes such as Ethylene, Purified Terephthalic Acid (PTA) and Oil Refining, technologies on plant-wide modelling, advanced control and operation optimization have been developed by innovatively integrating material conversion mechanisms and real-time operation information. These technologies have been successfully applied to dozens of large-scale petrochemical plants ensuring long-term stable operation with high efficiency and low consumption. This report will emphasize on the introduction of three representative achievements: (1) operation optimization technology for ethylene plant for the maximization of high value-added products, (2) plant-wide optimization technology for PTA process, (3) integrated optimization of large-scale refinery process. By analyzing the characteristics of practical problems of each process, comprehensive solutions including process mechanism characterization, nonlinear process control and systematic multivariable optimization, etc. will be discussed, followed by demonstrations of their effectiveness in typical industrial applications. The research outcomes have 4 National Second-Prizes for Progress in Science and Technology, 10 first prizes of Ministerial and Provincial Science and Technology Advancement Award, and over 20 provincial and ministerial-level Science and Technology awards. 40 national invention patents have been authorized, in which 2 are China patent outstanding awards and 2 are First-Prize of Shanghai Invention Patent Award. 70 pieces of national computer software copyright are registered. 3 monographs and over 300 papers are published. The research outcomes was selected as the Top 10 cases of the university-industry cooperation.

Keywords: Model-based Control, Process Applications
Abstract: Nonferrous metallurgical industry is a cornerstone of a nation’s economy. In this talk, the Chinese researchers’ activities, which are aimed to the high-efficiency and green production of nonferrous metallurgical processes, are briefly summarized. For demonstration purpose, some typical nonferrous metallurgical process are selected to introduce the key automation technologies, e.g., process modeling, instrumentation for the online detection of key process variables, optimization and control of nonferrous metallurgical process, as well as their industry application results. Finally, China’s future developments in nonferrous metallurgical process control are analyzed to conclude the presentation.

Keywords: Big Data Analytics and Monitoring, Process Applications
Abstract: In this work, a generic process visualization method is introduced using parametric t-SNE and used to visualize real-time process information and correlations among variables on a 2D map. A deep network is used to learn the Kullback-Leibler divergence between the original high-dimensional space and the latent space. In practice, it is observed that a model trained with historical data is not robust enough to visualize shifts into unknown states. Due to the effect of greedy learning, the response of the model is biased toward the most-contributing inputs. To relieve this effect, combinatorial variation creation is applied in the training stage to allow the model to respond to each input more evenly. The proposed method is tested on the Tennessee Eastman Process (TEP) data for four types of faults. The result indicates that the proposed method outperforms conventional methods such as PCA and Isomap, and is able to provide clear visual indication of process changes.

Keywords: Big Data Analytics and Monitoring, Process Applications
Abstract: When data are collected sequentially from a chemical process, consecutive observations are correlated resulting in serial dependence. Such dependence (or autocorrelation) would violate the assumption of sample independence when carrying out most statistical process control schemes, such as Shewhart charts for individual measurements. In this paper, a model-free Shewhart individuals control chart for autocorrelated data is proposed to reduce/eliminate the effect of autocorrelation on chart performance. The modified Shewhart chart, based on the true mean and variance, is used as the benchmark chart for comparison. A single skipping chart (SSC) and a combined skipping chart (CSC) are established in the proposed Shewhart control scheme. The control performances of CSC, modified Shewhart chart and conventional Shewhart chart are compared in terms of their mean shift detection ability using an AR (1) process. The advantages of CSC are illustrated as a model-free approach and having a performance consistent with that of the benchmark, modified Shewhart chart. A practical application of CSC is illustrated using data from an industrial chemical process.

Keywords: Process Applications, Big Data Analytics and Monitoring
Abstract: This paper proposes a novel time-frequency method for plant-wide oscillation analysis based on the multivariate extension of standard empirical mode decomposition (EMD). The raised fast multivariate empirical mode decomposition (FMEMD) is generalized from EMD by solving an overdetermined system of linear equations. Due to its capability to analyze multiple channels data, FMEMD is especially suitable for characterizing plant-wide control loop oscillations. Unlike traditional methods, both the regularity of oscillations (in frequency domain) and evolution of local characteristics (in time scale) can be well captured via FMEMD. Validity of the raised approach is demonstrated on simulations as well as an industrial case.

Keywords: Big Data Analytics and Monitoring, Modeling and Identification, Process Applications
Abstract: Quality of the control system significantly contributes to the overall process technological and financial results. Plant throughput, environmental footprint and energy consumption push plants towards their technological limitations requiring better operation closer to constraints. Any improvement initiative should be predated with the estimation of the potential benefits associated with the rehabilitation project. The same applies to the control improvements. The assessment is always based on the performance indicators. Classical the same limit method is based on the Gaussian approach. However, investigation of industrial data frequently is not compliant with normal assumption about the properties of the variables. This paper extends the Gaussian approach with the use of robust (Huber) statistics and asymmetric Pearson type IV probability density function.

Keywords: Modeling and Identification, Big Data Analytics and Monitoring, Batch Process Modeling and Control
Abstract: A passivity-based input observer is proposed. The problem is motivated by reaction rate and heat estimation in control of chemical reaction systems. The input observer assumes measurement of the output, and its first order time derivative. The observer gives asymptotically converging estimation when both are accurately available, the so-called ideal case. In the nonideal case, where the derivative is not available, differentiator can be used to reconstruct the derivative with some error. Simulation results show performance results using a deadbeat differentiator for derivative reconstruction.

Keywords: Batch Process Modeling and Control, Energy Processes and Control, Process Applications
Abstract: Multiple phases with transitions from phase to phase are important characteristics of many batch processes. The linear characteristics of batch processes are usually taken into consideration in the traditional algorithms while the nonlinearity is neglected. However, to monitor batch processes more accurately and efficiently, such process features are needed to be considered carefully. In this paper, a new similarity index based on KECA (kernel entropy component analysis) is defined for batch processes with nonlinear characteristics. A new phase division and monitoring method based on the proposed similarity index is brought forward simultaneously. First, nonlinear characteristics can be extracted in feature space via performing KECA on each preprocessed time-slice data matrix. Then phase division is achieved with the similarity change of the extracted feature information. By establishing a series of KECA models for transitions and steady phases, it reflects the diversity of transitional characteristics objectively and can preferably solve the stage-transition monitoring problem in multistage batch processes. Finally, in order to overcome the problem that the traditional contribution plot cannot be applied to the kernel mapping space, a nonlinear contribution plot diagnosis algorithm is proposed. Both results of simulation study and industrial application clearly demonstrate the effectiveness and feasibility of the proposed method.

Keywords: Model-based Control, Batch Process Modeling and Control
Abstract: Optimal operating conditions for a process plant are typically obtained via model-based optimization. However, due to modeling errors, the operating conditions found are often sub-optimal or, worse, they can violate critical process constraints. Hence, model corrections become a necessity and are done by exploiting measured process data. To this end, either model parameters are adapted and/or correction terms are added to the model-based optimization problem. The modifier-adaptation methodology does the latter by adding bias and gradient correction terms that are called modifiers. The role of modifiers and model parameters are often seen as competing, and which one of the two is better suited to track the optimality conditions is an open problem. This paper attempts to shed light on finding a synergy between the model parameters and the modifiers in the case when tracking constraints is sufficient for near-optimal performance. We demonstrate through the simulation study of a batch-to-batch optimization problem that a set of model parameters can be selected that mirror the role of modifiers. The modifiers are then added only when there is insufficient number of mirror parameters for independent constraint tracking.

Keywords: Model-based Control
Abstract: The earlier the plant measurements are available for a given plant input, the quicker iterative real-time optimization by modifier adaptation (MA) can steer the plant to its optimum. In practical applications, in addition to the time required for a plant to reach its steady state and to the time a sensor needs to perform the measurement, further delays can occur. For example, due to the time required for the sample to reach the location of the sensor, caused by a remote positioning of the measurement device. We propose a modifier adaptation strategy where additional plant perturbations are performed around an estimate of the solution to the adapted optimization problem during the waiting period, which is caused by the measurement delay. The strategy is tested on the benchmark Otto-Williams reactor case study and its performance is studied.

Keywords: Optimization and Scheduling, Model-based Control, Energy Processes and Control
Abstract: Integration of combined cycle (CC) with chemical-looping combustion power plants (CLC) is studied for its potential for high power plant efficiency and low-cost CO2 capture. Dynamic modeling of the integrated process is used as a tool to analyze the extrema of CLC-CC power plant efficiency. Specifically, this work proposes control architectures for the CLC reactor and the integrated power plant. The time-averaged optimal CLC-CC power plant efficiency is estimated at 51.84% with CO2 capture efficiency at 96%. The main factor that limits the CLC-CC power plant efficiency is the reactor temperature, which is constrained by the oxygen carrier material. Plant-level sensitivity analysis shows that the inlet air temperature at the heat removal stage and gas compressor/gas turbine pressure ratio are the most important operating variables and if properly tuned the CLC-CC power plant can reach efficiencies sufficiently high for its economic deployment as a carbon neutral power generation option.

Keywords: Optimization and Scheduling, Energy Processes and Control, Model-based Control
Abstract: Optimized scheduling of compressor operation in compressed natural gas stations can achieve significant reduction in the cost of electricity in time-of-use electricity tariff environments. A model predictive control strategy for the scheduling of compressor activity is presented in this paper. The strategy ensures a robust responsiveness to meeting potential changes in gas demand patterns while at the same time minimizing electricity cost by a margin of up to 53.87%.

Keywords: Energy Processes and Control, Model-based Control, Optimization and Scheduling
Abstract: This paper presents model predictive control (MPC) strategies with a shrinking horizon approach to track local control systems when subject to supervisory trajectories. The supervisory trajectories are generated using economic receding horizon optimization based on energy management in energy-intensive industries (e.g., chlor-alkali process) with a hybrid renewable energy system (HRES), including solar, wind, and fuel cell sub-systems to provide sustainable power supply. A planer solid oxide fuel cell system is adopted in this study, and its power output is regulated using a constrained shrinking horizon MPC controller. The feasibility of MPC control algorithm in regulating energy sub-systems within a supervisory MPC framework will be studied and evaluated at different parameters. The main contribution of this paper is to provide practical control options when addressing technical viability concerns of hybrid energy system implementation.

Keywords: Optimization and Scheduling, Energy Processes and Control, Model-based Control
Abstract: This paper studies the optimal control strategies of hybrid renewable energy systems, focusing on offshore wind farms with energy storage systems (ESS), considering challenges of economic costs, operational reliability, and environmental impacts. Wind energy is widely exploited as a promising renewable energy source worldwide. The development of offshore wind farms (OWF) is emerging to utilize wind energy on a large scale, but face more complicated operational conditions and higher costs. A systematic methodology is proposed to optimize the costs of an OWF with ESS based on a framework that extends from the supervisory level dispatch strategies to individual pitch control for load reduction and fatigue mitigation. This holistic approach is able to improve the efficiency and economic performance of a wind farm through overall system optimization, while explicitly operating each wind turbine using a formally designed control framework leading to an extension of their service lifespan. This work provides novel solutions of wind energy and storage components deployment and has strong application potential.

Keywords: Modeling and Identification, Process Applications, Model-based Control
Abstract: Useful simulation of complex systems for real-time or otherwise computationally constrained applications often requires the creation of simplified system models from existing analytical models and data. In this work, the creation of candidate models through machine learning and heuristic-informed model simplification methods are explored, and the resulting candidate models are evaluated and compared through an optimal experiment design process. The dynamic system of interest is a counterflow air-to-water heat exchanger.

Keywords: Modeling and Identification, Process Applications, Model-based Control
Abstract: This paper presents a modeling approach for a class of multiphase chemical systems, based on non-equilibrium thermodynamics, specialized to an open flash-drum system. A compartmental model is considered to establish the dynamics of the gas and liquid phases, while a model of interface transport yields to constraints in the model. The overall system is thus written as a Differential-Algebraic system of Equations (DAE). The derived model is shown to be of index one, for which a stability analysis, based on Lyapunov first method, is briefly developed. An example is presented to illustrate the proposed the modeling and stability analysis approach, together with numerical simulations.

Keywords: Modeling and Identification, Process Applications, Model-based Control
Abstract: The aim of this paper is to introduce a simple model for wastewater treatment plants that can be used for evaluating control strategies on the basin of the Seine River under different scenarios. The model represents a bioreactor without oxygen limitation. The construction of the model was based on the ASM1 with only one microbial population. The performance of the model was tested on daily data for a four year period over the three main wastewater treatment plants of the Seine River in Paris (Seine Aval, Seine Centre, and Seine Grésillons). Results demonstrated that the model was effective for predicting the concentrations of ammonia, nitrites, nitrates, biological oxygen demand, and total suspended solids. The developed model was found to be parsimonious and it can thus provide a useful tool in optimizing river quality once coupled to control strategies.

Keywords: Big Data Analytics and Monitoring, Modeling and Identification
Abstract: Steam Assisted Gravity Drainage (SAGD) is an efficient and widely used technology to extract heavy oil from a reservoir. The accurate prediction of subcool plays a critical role in determining the economic performance of SAGD operations since it influences oil production and operational safety. This work focuses on developing a subcool model based on industrial datasets using deep learning and several other widely-used machine learning methods. Furthermore, this work compares and discusses the out-of-sample performance of different machine learning algorithms using industrial datasets. In addition, we also show that care has to be taken when using machine learning algorithms to solve engineering problems. Data quality and a priori process knowledge play a role in their performance.

Keywords: Modeling and Identification
Abstract: Stratification of patients into different risk subcategories for disease development plays an important role in medical treatments. It sets the basis for physicians to decide upon personalized interventions. This patient-specific therapy design increasingly becomes supported by mathematical models that describe the underlying disease processes on a detailed molecular level. However, the mathematical description of disease development is challenging. Often the underlying processes act on different time scales. Furthermore, the biomedical data and measurements have different quantities, qualities and uncertainties. New methods are required to address this heterogeneity in the data landscape and to integrate measurements on different time scales in order to extract meaningful information over the disease process. We devise an approach for integrating biological signals for short and long-term molecular processes into a coherent framework. To this end, we combine set-based estimation methods for short- term molecular pathways with classification approaches of long-term disease development. The developed framework is demonstrated by means of IL-6-induced Jak-STAT3 and MAPK trans- signaling.

Keywords: Process Applications, Model-based Control
Abstract: With the deep integration of information, communication and operation technologies, industrial wireless networks (IWNs) are playing key roles to drive the coming industrial revolution. This presentation provides a comprehensive study on two typical IWNs called WIA-PA and WIA-FA which are established as the twin international standards for process automation and factory automation, respectively. We ?rst introduce their technology roadmaps,and then present their system architectures including device types, network topologies and system managements. Furthermore, we compare their protocol stacks, key techniques and security mechanisms. Finally, we summarize the existing IWNs and envisage on future IWNs, wherein challenges and research issues are discussed.

Keywords: Model-based Control, Process Applications
Abstract: There is a class of complex plant-wide systems which are composed of many physically or geographically divided subsystems. Each subsystem interacts with some so called neighboring subsystems by their states and inputs. The technical target is to achieve a specific global performance of the entire system. The distributed (or decentralized) framework, where each subsystem is controlled by an independent controller, has the advantages of error-tolerance, less computational effort, and being flexible to system structure. Thus the distributed control framework is usually adopted in this class of system, in spite of the fact that the dynamic performance of centralized framework is better than it. Thus, how to improve global performance under distributed control framework is a valuable problem. This talk systematically will introduce the different distributed predictive control for the plant-wide system, including the system decomposition, classification of distributed predictive control, unconstraint distributed predictive control and the stabilized distributed predictive control with different coordinating strategies for different purposes, as well as the implementation examples of distributed predictive control. The major new contribution of this book is to show how the distributed MPCs can be coordinated efficiently for different control requirements, namely the network connectivity, error tolerance, performance of entire closed-loop system, calculation speed, etc., and how to design distributed MPC.

Keywords: Energy Processes and Control, Big Data Analytics and Monitoring
Abstract: The process control groups in Dalian University of Technology (DLUT) have explored some noteworthy research results in the fields of real-time process analytical technology (PAT), modeling complex process dynamics, data-driven process monitoring and scheduling, robust process control and batch optimization in the past decade. The main achievements include (i) Infrared spectra & image based in-situ measurement and monitoring methods for industrial crystallization and fermentation processes; (ii) Identification and robust control of industrial systems with long time delay; (iii) Robust control & optimization of industrial batch processes with time-varying uncertainties; (iv) Multi-scales modeling of multi-energy flows for energy systems; (v) Data-driven optimization of collaborative energy-material flows for manufacturing systems. Experimental results on pharmaceutical crystallizers and practical applications to some steel plants in China are given to demonstrate the effectiveness and merits of these developments. In addition, some challenges and open issues related to the above research developments will be discussed to call for more attentions.

Keywords: Process Applications, Big Data Analytics and Monitoring
Abstract: Cyber-Physical Systems are integrations of the computational processes and the physical processes via communication mediums to control, monitor, and report entities by using sensors and actuators in the physical environments. Industrial control systems (ICSs) are a typical CPS. In recent years, industrial control system safety & security incidents occurred more frequently such as 2010 stuxnet event, 2013 Haifa highway control system attack event, and 2014 havex event. This presentation will start with the basic concept of CPS and their safety and security, introduce the relationship of ICS safety and security with examples, and review the achievements we made in the field in China, in collaboration with Huazhong University of Science and Technology, including the interaction between safety and security, the inherent safe architecture of an ICS, intrusion detection, online risk assessment, functional safety protection, and fault tolerant control. The integrated design of safety and security for ICS is made at the end. Various case studies have been presented to illustrate the efficiency of the safety and security approaches.

Keywords: Model-based Control
Abstract: In this work, we consider the problem of controlling a two-stage cooling mixed suspension mixed product removal (MSMPR) crystallizer. For this process, the temperature in the first crystallizer is manipulated for controlling the average crystal dimension (d₄₃), while it is desired to maintain the temperature of the second crystallizer at the minimum allowed value to guarantee maximum yield. Due to system nonlinearities and process delays, the performance of traditional PID controllers are poor. An advanced control scheme is proposed to improve the closed loop performance and achieve desired control objectives. The control scheme is based on the coupling of a PI controller and a model-based nonlinear prediction block that serves as delay and disturbance compensator. The proposed scheme has been tested on the system in case of disturbances in the feed concentration, with and without measurement errors. It is observed that the proposed scheme outperforms the PI controller by reducing the output response settling time and overshoot.