Search

Mathematical models of dynamic processes are often required for assessment, diagnosis and improvement of control loop performances in process industries. Data samples collected in daily operations of feedback control loops may enclose data segments suitable for process identification to obtain the mathematical models. This paper proposes a new method to search such data segments. The searching criterion is that the reference and process output in a feedback control loop should experience significant magnitude changes. Hypothesis tests are exploited to find changing positions of data segments with different probability distributions and to verify whether the reference and process output make significant magnitude changes inside one data segment or between two adjacent segments. Simulation and industrial examples are provided to illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In view of the problem that the traditional augmented reality maintenance system can not effectively perceive and judge die state of maintenance, this paper proposed a method based on PNN for maintenance process identification. The strategy shifted the focus from the maintenance object itself to the nearby removed parts placement area, used the eurvelet transform to generate the edge image of the disassembled parts, then generated the Hu invariant moment characteristic value, which was used as the input of PNN neural network to classify and identify the parts of the process, finally got the current maintenance process. Experimental results show that the recognition accuracy of the proposed method is significantly higher than that of the image matching method and eurvelet invariant moment method, which provides a new idea and solution for the scene perception. [ABSTRACT FROM AUTHOR]

This paper is concerned with robust identification of processes with time-varying time delays. In reality, the delay values do not simply change randomly, but there is a correlation between consecutive delays. In this paper, the correlation of time delay is modeled by the transition probability of a Markov chain. Furthermore, the measured data are often contaminated by outliers, and therefore, t -distribution is adopted to model the measurement noise. The variational Bayesian (VB) approach is applied to estimate the model parameters along with time delays. Compared with the classical expectation-maximization algorithm, VB approach has the advantage of capturing the uncertainty of the estimated parameter and time delays by providing their full probabilities. The effectiveness of the proposed method is demonstrated by both a numerical example and a pilot-scale hybrid-tank experiment. [ABSTRACT FROM PUBLISHER]

• A digital twin-driven intelligent controller is designed for of flowmeter calibration station. • Combining first principle and data driven model to achieve online operation and real-time correction of digital twin model. • On the basis of identifying calibration process with the GT-ST method, MPC-BP intelligent controller is proposed for fast and accurate control. • The digital twin framework has been tested in an industrial application for intelligent calibrations, and the control effect was better than that of the skilled manual worker. With the rapid growth of the natural gas industry, the handover volume of natural gas measurement is increasing every year, thus making the calibration of natural gas flowmeters more and more critical. The flow calibration process includes adjusting pressure and flow rate and performing calibration operations, which are manually carried out by testing personnel. The increasing demand for expertise and proficiency in the testing personnel poses significant challenges to on-site safety production. In this paper, we propose a digital twin-driven intelligent control of natural gas flowmeter calibration station to improve the efficiency of flowmeter calibration. First, a virtual simulation platform combining first-principle models and data-driven models is established, which can simulate the hydraulic state in the calibration station to provide verification for the control algorithm. Then, a fully automated flow control system is developed, which performs predictive control based on a hybrid driven model. This process first identifies the current process and automatically switches to the best process based on the diameter of the calibrated meter. Then, it performs initial control using a model predictive controller, and finally, improves control accuracy by using a BP neural network. The results show that the digital twin-driven intelligent control of natural gas flowmeter calibration station can accurately obtain hydraulic state and control strategy, and significantly improve the calibration efficiency. [ABSTRACT FROM AUTHOR]

The quality of the air in underground mines is a challenging issue due to many factors, such as technological processes related to the work of miners (blasting, air conditioning, and ventilation), gas release by the rock mass and geometry of mine corridors. However, to allow miners to start their work, it is crucial to determine the quality of the air. One of the most critical parameters of the air quality is the carbon monoxide (CO) concentration. Thus, in this paper, we analyze the time series describing CO concentration. Firstly, the signal segmentation is proposed, then segmented data (daily patterns) is visualized and statistically analyzed. The method for blasting moment localization, with no prior knowledge, has been presented. It has been found that daily patterns differ and CO concentration values reach a safe level at a different time after blasting. The waiting time to achieve the safe level after blasting moment (with a certain probability) has been calculated based on the historical data. The knowledge about the nature of the CO variability and sources of a high CO concentration can be helpful in creating forecasting models, as well as while planning mining activities.

The paper offers identification algorithms enabling process parameters estimation under limited output samples and time-varying inputs. The algorithms are based on the associative search of analogs and wavelet-analysis of signals. The effectiveness of the techniques proposed is demonstrated with the example of product quality prediction in oil refining industry. [ABSTRACT FROM AUTHOR]

In this paper, we consider an important practical industrial process identification problem where the time delay can change at every sampling instant. We model the time-varying discrete time-delay mechanism by a Markov chain model and estimate the Markov chain parameters along with the time-delay sequence simultaneously. Besides time-varying delay, processes with both time-invariant and time-variant model parameters are also considered. The former is solved by an expectation-maximization (EM) algorithm, while the latter is solved by a recursive version of the EM algorithm. The advantages of the proposed identification methods are demonstrated by numerical simulation examples and an evaluation on pilot-scale experiments. [ABSTRACT FROM AUTHOR]

• A new Wiener-type process identification method is proposed. • The proposed method is developed based on a narrow pulse response. • It requires a relatively short process excitation than previous methods but estimates the process almost exactly. • The performance of the proposed method is verified by the simulation and experiment. The Wiener-type nonlinear process where a nonlinear static block follows a linear dynamic subsystem is widely used to describe the dynamics of chemical processes. The present study proposes a new Wiener-type process identification method based on a narrow pulse response to cope with the relatively long process excitation requirement in the previous methods. The proposed identification method is developed based on the property of the Wiener-type nonlinear process that outputs of a linear dynamic subsystem is the same at two points when process outputs are equivalent at those points. The proposed method requires a relatively short process excitation (a single pulse response) but almost exactly estimates the Wiener-type nonlinear process. The simulation and experimental studies show the outstanding performance of the proposed method. [ABSTRACT FROM AUTHOR]

A new model Gm(s), defined by the four measurable parameters ultimate gain, ultimate frequency, amplitude of ultimate oscillation and dead-time, is obtained from tangent to the Nyquist curve at the critical point. This model represents the generalization of the Ziegler-Nichols process dynamics characterization method, for a large class of stable, integrating and unstable processes Gp(s). In the present paper, by applying time and amplitude scaling technique to the model Gm(s), a new tuning formulae are developed for the PI controller. These tuning formulae guarantee almost optimal performance/robustness tradeoff for a large class of processes with integral action.

During the development of intelligent systems inspired by biological neural system, in the last two decades the researchers from various scientific fields have created neural networks for solving a series of problems from pattern recognition, prediction, diagnostic, software sensor, modelling and identification, control and optimization. In this paper a review of neural network application in the field of chemical engineering with emphasis on identification and process control is given.The neural networks have been proven usefull in the applications which include complex chemical and biochemical reactions. In such a processes use of standard methods of process modelling and control structure are frequently not suitable. The ability of neural network to model dynamics of nonlinear process makes them an important tool for implementation in model-based control. Due to intensively theory development and many practical applications, there are numerous neural network structures and algorithms.In this paper neural networks are categorized under three major control schemes: model-base predictive control, inverse model-based control, and adaptive control. The major applications are summarized. It reveals prospect of using neural networks in process identification and control.The future of neural network application lies not only in their explicite use, but in cross connecting to other advanted technnologies as well. Fusion of neural networks and fuzzy logic in the form of neural-fuzzy network is one of the possibilites. Other important field is hibrid modelling and identification methods which supplement simplified mechanistic models. Software sensors and their application, especially in controlling of bioprocesses, present a very promising field.

Both complex impedance and electric modulus formalisms are considered to study the contribution of dominant electrical conduction mechanisms in bulk p-type CuIn 3 Se 5 semiconductor. An equivalent electrical circuit has been proposed to explain the impedance results that show semicircle arcs at different temperatures. Two different conduction mechanisms were obtained by fitting the data to Cole-Cole equation. The values of calculated resistances for the “grain” and “grain-boundary” contributions were found to decrease with increasing temperature, in agreement with the Arrhenius law associated with activation energy of 179 meV and 199 meV for the “grain” and “grain-boundary” conduction, respectively. The activation energies for relaxation mechanisms were estimated and compared with those obtained from the analysis of electric modulus. [ABSTRACT FROM AUTHOR]

Spatial spread of infectious diseases among populations via the mobility of humans is highly stochastic and heterogeneous. Accurate forecast/mining of the spread process is often hard to be achieved by using statistical or mechanical models. Here we propose a new reverse problem, which aims to identify the stochastically spatial spread process itself from observable information regarding the arrival history of infectious cases in each subpopulation. We solved the problem by developing an efficient optimization algorithm based on dynamical programming, which comprises three procedures: 1) anatomizing the whole spread process among all subpopulations into disjoint componential patches; 2) inferring the most probable invasion pathways underlying each patch via maximum likelihood estimation; and 3) recovering the whole process by assembling the invasion pathways in each patch iteratively, without burdens in parameter calibrations and computer simulations. Based on the entropy theory, we introduced an identifiability measure to assess the difficulty level that an invasion pathway can be identified. Results on both artificial and empirical metapopulation networks show the robust performance in identifying actual invasion pathways driving pandemic spread. [ABSTRACT FROM PUBLISHER]

The application of a multivariable predictive controller to an activated sludge process is discussed in this work. Emphasis is given to the model identification and the long term assessment of the controller efficiency in terms of economical and environmental performances. A recurrent neural network model is developed for the identification problem and the dynamic matrix control is chosen as suitable predictive control algorithm for controlling the nitrogen compounds in the bioreactor. Using the Benchmark Simulation Model No. 1 as virtual platform, different predictive controller configurations are tested and further improvements are achieved by controlling the suspended solids at the end of the bioreactor. Based on the simulation results, this work shows the potentiality of the dynamic matrix control that together with a careful identification of the process, is able to decrease the energy consumption costs and, at the same time, reduce the ammonia peaks and nitrate concentration in the effluent. [ABSTRACT FROM AUTHOR]

Original Equipment Manufacturers have recognized that offering After Sales Services enables high revenues. Above this, the offer of After Sales Services combined with primary products is an appropriate instrument to improve customer satisfaction and loyalty. But due to the noticeable increasing competitive pressure in the After Sales Service, a possible solution is the implementation of Lean principles to optimize the customer service-processes. These processes all pursue the target to satisfy the customer's needs. But due to the variety of these processes and difficult, changeable conditions in customer service (varying workload, required material and tools, changing working environment …), there is no valid framework concerning the general conditions and objectives of customer service-processes or no holistic approach for process structuration in these processes. But the framework identification of general conditions as well as a holistic process structuration in customer service is the necessary basis for the implementation of Lean principles. Within this paper (based on a research project funded by the German Research Foundation), the development of a methodical approach to identify general conditions, customer service-objectives and to create a holistic process structuration will be described, which should help the OEMs offering customer services to identify and structure these processes. By the use of this methodical approach, OEMs will be enabled to create a structured process landscape. In the future, this process landscape provides an initial point to implement Lean principles, such as Continuous Improvement or Waste Reduction, in customer service. [ABSTRACT FROM AUTHOR]

This work initially focuses on developing a computational fluid dynamics (CFD) model of an industrial-scale steam methane reforming reactor (reforming tube) used to produce hydrogen. Subsequently, we design and evaluate three different feedback control schemes to drive the area-weighted average hydrogen mole fraction measured at the reforming tube outlet ( x ¯ H 2 outlet ) to a desired set-point value ( x ¯ H 2 set ) under the influence of a tube-side feed disturbance. Specifically, a CFD model of an industrial-scale reforming tube is developed in ANSYS Fluent with realistic geometry characteristics to simulate the transport and chemical reaction phenomena with approximate representation of the catalyst packing. Then, to realize the real-time regulation of the hydrogen production, the manipulated input and controlled output are chosen to be the outer reforming tube wall temperature profile and x ¯ H 2 outlet respectively. On the problem of feedback control, a proportional (P) control scheme, a proportional-integral (PI) control scheme and a control scheme combining dynamic optimization and integral feedback control to generate the outer reforming tube wall temperature profile based on x ¯ H 2 set are designed and integrated into real-time CFD simulation of the reforming tube to track x ¯ H 2 set . The CFD simulation results demonstrated that feedback control schemes can drive the value of x ¯ H 2 outlet toward x ¯ H 2 set in the presence of a tube-side feed disturbance and can significantly improve the process dynamics compared to the dynamics under open-loop control. [ABSTRACT FROM AUTHOR]

This paper considers the problem of estimating the frequency response of an unknown plant from impulse or step response measurements, possibly with non-ideal step signals. The proposed method is direct (no model identification of the plant is needed) and it can be applied using a single set of data generated by the plant, with no need for specific experiments nor iterations. The frequency response of the plant, at a desired frequency, is estimated by a virtual experiment generating a filtered version of the data set. Numerical simulations show the effectiveness of the method. [ABSTRACT FROM AUTHOR]

In this study, the application of adaptive neuro-fuzzy inference system (ANFIS) architecture to build prediction models that represent the pH neutralization process is proposed. The dataset used to identify the process was obtained experimentally in a bench scale plant. The prediction model attained was validated offline and online and demonstrated as able to precisely predict the one step-ahead value of effluent pH leaving the neutralization reactor. The input variables were the current and one past value of the acid and base flow rates and the current value of the output variable. Variance accounted for (VAF) indices greater than 99% were achieved by the model in experiments in which the disturbances in the acid and basic solutions flow rates were applied separately. For tests with simultaneous disturbances, conditions never seen in the training and suffering from reactor level oscillations, the prediction model VAF index was still approximately 96%. The validations demonstrated the capability of ANFIS to build precise fuzzy models from input–output datasets.R2values achieved were always larger than 0.96. [ABSTRACT FROM AUTHOR]

Though it becomes a routine to fit impedance data to an equivalent electric circuit model (EECM) using complex nonlinear least square (CNLS) to extract physical parameters from impedance data, two formidable challenges still remain: to build a physically meaningful EECM and to find good initial estimates for model parameters. In this study, combining graphical analysis of impedance data in both Bode and Nyquist plots, a two-step procedure is proposed to address the challenges: (1) a frequency derivative phase angle method is developed in Bode plot to identify the number of time constants (or electrochemical processes); (2) graphical analysis of impedance data in Nyquist plot is used sequentially for initial parameter determination. Major graphical analysis methods are compared in terms of frequency resolution, accuracy and complexity using synthetic data. The superiority of the proposed procedure is illustrated using the experimental data of a three-electrode lithium-ion cell. [ABSTRACT FROM AUTHOR]

The article presents the results of analysis, the use of continuous flow of logic at the stage of production planning and control of the company producing furniture. The concept of continuous flow tends to regulate the flow of materials in a manner that provides the shortest flow path without unnecessary activities (Muda is a Japanese word meaning waste), a constant takt and defined throughput at constant resource requirements for the so-called transfer of material through the whole process. In the study Glenday'd sieve method was used to identify the correct area, which requires the value stream mapping, and areas called excessive complexity, which do not provide added value. The use of Glenday's sieve method made it possible to identify areas in which it must be improve production capacity. [ABSTRACT FROM AUTHOR]

A method is proposed to detect if there is no coupling between an input and an output in systems operating in open-loop, that is, without a supervisory controller. The proposed technique is applicable to multiple input multiple output (MIMO) systems, whose intent is to detect no-model input/output (IO) combinations in a transfer matrix. Traditional approaches for selecting IO pairs are usually performed after the plant model is identified. The presented approach is applied during the pre-identification stage and is based on IO cross-correlation, signal filtering and fuzzy logic analysis. A case study involving the identification of a 7×6 simulated Fluid Catalytic Cracking (FCC) is discussed, as well as an influence analysis of detecting no-model IO pairs in the identification process and in the performance index of a Model Predictive Controller (MPC) applied to a 2×2 simulated distillation column. Finally, the method is tested with a real dataset obtained from an FCC unit of a petrol refinery. [ABSTRACT FROM AUTHOR]

A new frequency response identification method is presented to estimate frequency responses of discrete-time linear time-invariant processes. It provides multiple exact frequency response data set for any desired frequencies. The proposed method is applicable to all the process data types of both initial/final steady state, initial steady state/final cyclic steady state, initial cyclic steady state/final steady state and both initial/final cyclic steady state. Also, it completely removes the effect of static disturbances and shows acceptable robustness to measurement noises. [ABSTRACT FROM AUTHOR]

For a multivariable process with strong input-output cross-interactions, the control system design is difficult to do using only monovariable controllers, one for each input-output (I-O) process channel. The purpose of this paper is to present a practical solution for controlling a heat transfer multivariable 2X2 process using two PID monovariable classical controllers and a 2X2 decoupler connected in front of the process, in order to provide a decoupled process. [ABSTRACT FROM AUTHOR]

Most of the real systems are multivariable, having in addition to the input-output direct interactions, unwanted input-output cross interactions. The objective of this paper is to find a simple and feasible mathematical model of a heat transfer multivariable 2X2 process using the results of the process dynamic investigation. A simple and accurate model is useful in the design of a 2X2 structure decoupler connected in series with the process, which aims to cancel the multivariable process natural cross interactions in order to obtain a process having only direct interactions, which can easily be controlled. [ABSTRACT FROM AUTHOR]

Highlights: [•] PID auto-tuning using new model reduction method and PID tuning rule based on fractional order plus time delay model. [•] The method successfully incorporates various dynamics such as non-minimum phase, fractional order and under- or over-damped dynamics. [•] The explicit PID tuning rule for the fractional order plus time delay model. [ABSTRACT FROM AUTHOR]

Abstract: Model Predictive Control (MPC) Relevant Identification (MRI) methods are a good option for identification, if there is model structure mismatch. Herein a new MRI method, named Enhanced Multistep Prediction Error Method (EMPEM), is proposed. EMPEM combines the best characteristics of others MRI methods in a single algorithm. It was developed to identify either closed-loop or open-loop systems; its convergence and stability make it perform better than the other presented methods. To show the advantages of EMPEM, a comparison is made against two other methods (one MRI and one PEM). The statistical analysis indicates that in the cases studied, the performance and the robustness of the new method is equal or better than the other ones. [Copyright &y& Elsevier]