Closed-loop Data Research Articles

UB-Net: Improved Seismic Inversion Based on Uncertainty Backpropagation

Seismic inversion is aimed at building a mapping from low-resolution seismic data to high-resolution impedance data. Most of the traditional methods have satisfactory interpretability, and most parameters tend to have specific physical definitions. On the other hand, deep learning-based methods present poor interpretability as their prediction performance is not always clearly explainable. One of the significant challenges of the deep learning-based methods is to quantify the uncertainty of the model. The uncertainty includes aleatoric uncertainty and epistemic uncertainty, and epistemic uncertainty can be used to evaluate the predicted accuracy of the trained model. In this paper, we propose a new deep learning model called uncertainty backpropagation network (UB-Net) to perform impedance inversion. The proposed UB-Net is based on a closed-loop framework and can predict the impedance and the epistemic uncertainty simultaneously. UB-Net has three closed-loop data flows, whereby the predicted uncertainty is utilized as the weight of loss functions to improve the inversion accuracy. Experimental analyses demonstrate that UB-Net presents advanced inversion accuracy on both synthetic and real examples. Specifically, the mean absolute error (MAE) on synthetic examples drops by 40%, and the Pearson correlation coefficient (PCC) on real examples increases by 2%. Besides, compared with existing approaches, UB-Net presents superior spatial continuity and preserves more geological structures such as little faults in real examples.

Design of an Optimized GMV Controller Based on Data-Driven Approach

This paper presents a data-driven scheme that can obtain the optimized Generalized Minimum Variance (GMV) control parameters by applying the Nelder—Mead (NM) method based on Proportional-Integral Derivative (PID) controller for linear system. An adjustable λ is included in the GMV-PID controller. In the existing GMV-PID controller, the PID parameters are calculated by simply changing λ manually. Therefore, it is hard to get desirable control performance. The NM method is introduced to improve the control performance. The application of NM method can optimize the calculation of λ. However, the objective function in NM method needs the output calculation. In other words, the model information is inevitably to obtain output. To achieve model-free design scheme, the estimation of closed-loop response method is introduced in database-driven approach. Using obtained closed-loop data can predict the output and then substitute it into the objective function without any model information of the process. The effectiveness is verified by experiment.

MPC closed-loop identification without excitation

This paper presents a method of closed-loop identification for multivariable systems without external excitation. The method is specially designed for model predictive control (MPC) systems. Without using external excitation (test signals), the method ensures the informativity of the closed-loop data and, at the same time, improve the control performance during the test period. The purpose of the study is to reduce the cost of identification test. The basic idea is to switch the input weighting matrix in the MPC controller which leads to the informativity of the data-set. A preliminary test is carried out in order to find a new input weighting matrix which improve the control performance; then a switching scheme is developed based on the two weighting matrices. Traditional simulation based model validation no longer works in closed-loop identification without excitation, and model error bounds on the frequency responses can be used instead. The effectiveness of the proposed method is demonstrated by a simulation study.

Closed-Loop Identification of Baroreflex Properties in the Frequency Domain.

The arterial baroreflex system plays a key role in maintaining the homeostasis of arterial pressure (AP). Changes in AP affect autonomic nervous activities through the baroreflex neural arc, whereas changes in the autonomic nervous activities, in turn, alter AP through the baroreflex peripheral arc. This closed-loop negative feedback operation makes it difficult to identify open-loop dynamic characteristics of the neural and peripheral arcs. Regarding sympathetic AP controls, we examined the applicability of a nonparametric frequency-domain closed-loop identification method to the carotid sinus baroreflex system in anesthetized rabbits. This article compares the results of an open-loop analysis applied to open-loop data, an open-loop analysis erroneously applied to closed-loop data, and a closed-loop analysis applied to closed-loop data. To facilitate the understanding of the analytical method, sample data files and sample analytical codes were provided. In the closed-loop identification, properties of the unknown central noise that modulated the sympathetic nerve activity and the unknown peripheral noise that fluctuated AP affected the accuracy of the estimation results. A priori knowledge about the open-loop dynamic characteristics of the arterial baroreflex system may be used to advance the assessment of baroreflex function under closed-loop conditions in the future.

An Ensemble of Recurrent Neural Networks for Real Time Performance Modeling of Three-Spool Aero-Derivative Gas Turbine Engine

Abstract Gas turbine is a complex system operating in nonstationary operation conditions for which traditional model-based modeling approaches have poor generalization capabilities. To address this, an investigation of a novel data driven neural networks based model approach for a three-spool aero-derivative gas turbine engine (ADGTE) for power generation during its loading and unloading conditions is reported in this paper. For this purpose, a nonlinear autoregressive network with exogenous inputs (NARX) is used to develop this model in matlab environment using operational closed-loop data collected from Siemens (SGT-A65) ADGTE. Inspired by the way biological neural networks process information and by their structure which changes depending on their function, multiple-input single-output (MISO) NARX models with different configurations were used to represent each of the ADGTE output parameters with the same input parameters. First, data preprocessing and estimation of the order of these MISO models were performed. Next, a computer program code was developed to perform a comparative study and to select the best NARX model configuration, which can represent the system dynamics. Usage of a single neural network to represent each of the system output parameters may not be able to provide an accurate prediction for unseen data and as a consequence provides poor generalization. To overcome this problem, an ensemble of MISO NARX models is used to represent each output parameter. The major challenge of the ensemble generation is to decide how to combine results produced by the ensemble's components. In this paper, a novel hybrid dynamic weighting method (HDWM) is proposed. The verification of this method was performed by comparing its performance with three of the most popular basic methods for ensemble integration: basic ensemble method (BEM), median rule, and dynamic weighting method (DWM). Finally, the generated ensembles of MISO NARX models for each output parameter were evaluated using unseen data (testing data). The simulation results based on datasets consisting for experimental data as well as data provided by Siemens high fidelity thermodynamic transient simulation program show improvement in accuracy and robustness by using the proposed modeling approach.

Offset-free ARX-based adaptive model predictive control applied to a nonlinear process

An adaptive steady state constrained autoregressive with extra input (SSARX) model is developed and embedded into a model predictive control (MPC) framework to regulate a nonlinear process for setpoint tracking. The resulting linear MPC can minimize the deviation of output from the setpoint along the entire prediction horizon and finally achieve zero-offset control even with model mismatch. Our contributions lie in three aspects: First, we develop a steady state constrained ARX model identification method. Second, we design an information matrix based index for model re-identification using online closed-loop data. Third, when steady state input values are unknown, our algorithm can adaptively adjust parameters in the objective function and model to track the setpoint crossing a wide range of operating conditions. A fermenter control problem is studied to demonstrate the effectiveness of the proposed approach.

A probabilistic framework with concurrent analytics of Gaussian process regression and classification for multivariate control performance assessment

Control performance assessment (CPA) is vital to ensure the safety of control systems. However, most multivariate CPA methods are limited to the system with explicit knowledge. Recently, it has been recognized that high predictability of closed-loop outputs implies poor control performance. This paper proposes a probabilistic CPA (PCPA), which is compatible with the above promising idea. This paper constructs a modified Gaussian process regression (GPR) model to quantitatively​ estimate the prediction uncertainty of outputs with only routine closed-loop data as input and focuses on the prediction variance of interest. As a non-parametric and probabilistic method, the proposed framework can handle the nonlinearity and random uncertainties inherent in complex control systems. Combined with the varying window size strategy, a novel performance metric, called disruption resistance (DR) here, is designed to characterize different control performance. The evaluation confidence and uncertainty can be revealed with concurrent analytics of GPR and Gaussian process classification when performing the performance assessment. This gives rise to a reliable and pragmatic PCPA framework, which shows more accurate and comprehensive results in the application to both simulated and real industrial processes.

Modeling Method for the Coupling Relations of Microgrid Cyber-Physical Systems Driven by Hybrid Spatiotemporal Events

The essence of the microgrid cyber-physical system (CPS) lies in the cyclical conversion of information flow and energy flow. Most of the existing coupling models are modeled with static networks and interface structures, in which the closed-loop data flow characteristic is not fully considered. It is difficult for these models to accurately describe spatiotemporal deduction processes, such as microgrid CPS attack identification, risk propagation, safety assessment, defense control, and cascading failure. To address this problem, a modeling method for the coupling relations of microgrid CPS driven by hybrid spatiotemporal events is proposed in the present work. First, according to the topological correlation and coupling logic of the microgrid CPS, the cyclical conversion mechanism of information flow and energy flow is analyzed, and a microgrid CPS architecture with multi-agents as the core is constructed. Next, the spatiotemporal evolution characteristic of the CPS is described by hybrid automata, and the task coordination mechanism of the multi-agent CPS terminal is designed. On this basis, a discrete-continuous correlation and terminal structure characteristic representation method of the CPS based on heterogeneous multi-groups are then proposed. Finally, four spatiotemporal events, namely state perception, network communication, intelligent decision-making, and action control, are defined. Considering the constraints of the temporal conversion of information flow and energy flow, a microgrid CPS coupling model is established, the effectiveness of which is verified by simulating false data injection attack (FDIA) scenarios.

Optimal Energy Consumption of the Distillation Process and Its Product Purity Analysis Using Ultraviolet Spectroscopy.

This paper addresses the energy consumption of distillation process via an actuator, which is a challenging problem in process industries. Precise control action would enhance energy consumption and improve the productivity. This paper is an experimental validation of EPC-PI control algorithm and analysis of distillate purity of a lab-scale distillation column. The PI control scheme uses closed-loop data of extended predictive controller (EPC) that has been performed through off-line simulation. The performance of control method is compared with different schemes such as Hägglund’s one-third rule and Skogestad’s overshoot method. The issue of integral windup in the multivariable process is addressed in the aspect of optimal energy consumption. The energy consumption calculations are made with respect to power utility of actuators throughout the process. The distillate product of post-controller implementation is processed to qualitative analysis using UV spectroscopy. Performance index is carried out via integral time absolute error (ITAE) by perturbing plant parameters up to 30% uncertainty.

Mountain Top-Based Atmospheric Radio Occultation Observations with Open/Closed Loop Tracking: Experiment and Validation

Through Global Navigation Satellite System (GNSS) occultation measurement, the global ionosphere and atmosphere can be observed. When the navigation satellites’ signal passes through the lower atmosphere, the rapid change of the atmospheric refractive index gradient will cause serious multipath phenomena in radio wave propagation. Atmospheric doppler frequency shift and amplitude signal fluctuations increase drastically. Due to the attenuation of signal amplitude and the rapid change of the Doppler frequency, the general phase locked loop (PLL) cannot work properly. Hence, a more stable tracking technology is needed to track the occultation signal passing through the lower atmosphere. In this paper, a mountain-top based radio occultation experiment is performed, where we employ an open-loop receiver and remove the navigation bits by the internal demodulation. In the process of the experiment, we adopt the open-loop tracking technique and there is no feedback between the observed signal and the control model. Specifically, taking the pseudo-range and doppler information from models as input, three key parameters, i.e., accurate code phase, carrier doppler and code doppler, can be obtained, and furthermore, the accurate accumulation is determined by them. For the full open-loop occultation data, a closed-loop observation assisted strategy is presented to compare the tracking results between open-loop and closed-loop occultation data. Through the compared results, we can determine whether the initial phase has been reversed or not, and obtain the high consistency corrected open-loop data that can be directly used for subsequent atmospheric parameters inversion. To verify the effect of open-loop tracking and open-loop inversion, we used the company’s self-developed occult receiver system for verification. The company’s self-developed occult receiver system supports Global Position System (GPS)/Beidou satellites constellation (BD, the 2nd and 3rd generations) dual systems. We have verified GPS and BD open-loop tracking and inversion, carried out in a three-week mountain-based experiment. We used closed-loop and open-loop strategies to track and capture the same navigation star to detect its acquisition effect. Finally, we counted the results of a week (we only listed the GPS data; BD’s effect is similar). The experimental results show that the open-loop has expanded the signal-cut-off angle by nearly 20% under the condition of counting all angles, while the open-loop has increased the signal-cut-off angle value by nearly 89% when only calculating the negative angle. Finally, the atmosphere profiles retrieved from observations in open-loop tracking mode are evaluated with the local observations of temperature, humidity and pressure provided by the Beijing Meteorological Bureau, and it is concluded that the error of open-loop tracking method is within ~4% in MSER (mean square error of relative error), which meets the accuracy of its applications (<5%, in MSER).

A Comparison Study of Subspace Identification of Blast Furnace Ironmaking Process

Blast Furnace (BF) iron making process is an extremely complex industrial process. The molten iron silicon content is considered as an important indicator of the thermal status of the blast furnace. The stabilization control of blast furnace depends on the molten iron silicon content. Three classic subspace identification methods including MOESP (Multivariable Output-Error State sPace), CVA (Canonical Variate Analysis) and SSARX (Subspace identification method ARX) are considered to establish the state space model of blast furnace ironmaking process. The inputs to the model are the most responsible and easily measured variables for the fluctuation of thermal state in blast furnace while the output to the model is the molten iron silicon content. The identified state space models are then tested on datasets obtained from No.1 BF in LiuGang Iron and Steel Group Co. of China. Experiment results show that the blast furnace ironmaking process can be reliably modeled by these subspace identification methods. Further, the SSARX method outperforms over the other two subspace identification methods with closed loop data.

Online performance monitoring and diagnosis of multivariate systems

Online performance monitoring can be used to improve the performance of control systems in industry. The purpose of this article is to detect a performance deterioration and determine its cause in a system. In this article, two indices are used for online performance monitoring of a nonlinear multivariate system with optimally tuned proportional integral controllers. The first index is defined based on a squared distance measurement between the closed-loop system outputs and chosen set-points. The second index is a statistical index that uses all the information in the covariance matrices of the closed-loop system output data. Both indices are used and compared for performance monitoring of a quadruple-tank system. Moreover, hypothesis testing method has been used to determine the cause of the performance deterioration, so that appropriate solutions according to the cause can be applied to the system to improve the performance.

Server-Edge dualized closed-loop data analytics system for cyber-physical system application

• We define six essential components for the data analytics system for CPS application. • We propose an architecture of dualized closed-loop data analytics with server and edge-computing devices. • A PMML-based data analysis model information parsing structure is proposed to deal with the automation issue of the data analytics application. • A case study that applies the proposed server-edge dualized closed-loop data analytics system for CPS application to the die-casting factory in Korea is introduced. In this research, we propose a system architecture of the server-edge dualized closed-loop data analytics system for cyber-physical system (CPS) application. We define six essential components for the data analytics system for CPS application: (1) the cyber model, (2) the data analytics module, (3) the data analysis model execution module, (4) the decision making module, (5) the system control module, and (6) the visualization module. We then propose an architecture of dualized closed-loop data analytics with server and edge-computing devices. The proposed dualized architecture of the data analytics system has advantages in handling the three issues of applying data analytics systems to the manufacturing context: (1) the system overload issue of the data analytics module due to large volumes of data, (2) the automation issue in the sequences of data analysis model generation, data analysis module execution, and system control, and (3) the real-time issue of data analysis model execution. In particular, a PMML-based data analysis model information parsing structure is proposed to deal with the automation issue. A case study that applies the proposed server-edge dualized closed-loop data analytics system for CPS application to the die-casting factory in Korea is introduced.

Integrating dynamic neural network models with principal component analysis for adaptive model predictive control

This work addresses one aspect of the overparameterization problem in using artificial/recurrent neural networks (ANN/RNN) based dynamic models for model predictive control (MPC) implementations. The manuscript presents an approach to handle situations where the training data may not be sufficiently rich, and in particular, for handling historical data with correlated inputs. Two approaches are proposed. The key idea in the first method is to perform principal component analysis (PCA) on input space and then utilize the scores to build a PCA-RNN model. Next, a PCA-RNN-based MPC is designed to compute the optimal values of scores and subsequently determined the manipulated inputs. An alternative solution is proposed in the second approach by proposing a new constraint on squared prediction error (SPE) statistic in the RNN-based MPC to make prescribed inputs follow the PCA model constructed for training input data. Finally, an approach is presented that allows to break the correlation in the MPC implementation while maintaining model validity. This is done by first generating richer closed-loop data by implementing the SPE based MPC with slightly relaxed constraints (thus compromising only slightly on the closed loop performance). Then the new data is utilized to re-identify the model, and for use in the MPC. The efficacy of the proposed approaches to handle the problem of set-point tracking is evaluated using a chemical reactor example. The results are compared with a nominal MPC design, and the superior performance under the proposed formulations demonstrated.

Comparison between Plant Impulse Responses Estimated in Time and Frequency Domains Using Closed-loop Data

Comparison between Plant Impulse Responses Estimated in Time and Frequency Domains Using Closed-loop Data

MPC closed-loop identification without excitation

This paper presents a method of closed-loop identification for multivariable systems without external excitation. The method is specially designed for model predictive control (MPC) systems. Without using external excitation (test signals), the method ensures the informativity of the closed-loop data and, at the same time, improve the control performance during the test period. The purpose of the study is to reduce the cost of identification test. The basic idea is to switch the input weighting matrix in the MPC controller which leads to the informativity of the data-set. A preliminary test is carried out in order to find a new input weighting matrix which improve the control performance; then a switching scheme is developed based on the two weighting matrixes. Traditional simulation based model validation no longer works in closed-loop identification without excitation, and model error bounds on the frequency responses can be used instead. The effectiveness of the proposed method is demonstrated by a simulation study.

Variance reduction in Covariance Based Realization Algorithm with application to closed-loop data

The Covariance Based Realization Algorithm (CoBRA), one branch of subspace methods, enables the estimation of multivariable models using a large number of data points due to the use of finite size covariance matrices. In addition, the covariance pre-processing allows CoBRA to ignore any (high-order) noise dynamics and focus on the estimation of the low-order deterministic model. However, subspace methods and CoBRA in particular are not maximum-likelihood methods. In this paper, an in-depth study on the statistical behavior of the noise effects is conducted. An approach is provided to reduce the variance of an estimate obtained by CoBRA via the choice of optimal row and column weighting matrices. For closed-loop implementation of CoBRA, a two stage procedure is proposed with the estimate on an intermediate instrument. At the first stage, a least-length perturbation of a scalar accuracy function is used to obtain an analytic solution for the optimal weighting matrices. The resulting instruments produced by the first stage are used for the identification at the second stage to extract the plant model. Simulation examples are given to illustrate the efficiency of the proposed two stage CoBRA method on the basis of closed-loop data and compared with other subspace methods.

Real-Time Estimation of Bare-Airframe Frequency Responses from Closed-Loop Data and Multisine Inputs

A method is presented for estimating frequency responses of multiple-input multiple-output bare-airframe dynamics from flight test data containing feedback control and/or mixing of control effectors. Orthogonal phase-optimized multisines are used to simultaneously excite each input with unique harmonic frequencies, at which frequency responses are computed as ratios of output-to-input Fourier transform data. The confounding effects of feedback and mixing are resolved by interpolation of the frequency responses. The analysis can be run in batch for postflight analysis, or in real time as the aircraft is flying. The effectiveness of the method was verified using simulations of the subscale T-2 generic transport airplane with rate feedback. The method was also demonstrated using flight test data from the X-56A MUTT aeroelastic airplane, which was flown with both feedback control and mixing.

Stochastic modeling and prediction of IMEP for closed loop knock control performance assessment

Knock control systems aim to maximize power output while also protecting the engine from knock-induced damage. The extent to which this goal is achieved is rarely evaluated, both because the random nature of knock events means that the results of any single experiment are not repeatable, and because control loop performance is generally assessed with respect to the resultant spark angle (the manipulated variable), rather than some measure of output performance. In this work, a new stochastic model for the empirical IMEP distribution is developed and fitted to open-loop experimental data. Stochastic methods are then applied to predict both the steady state and transient closed loop distribution of IMEP based on the open-loop model and the control law that is applied. The results are validated against closed loop data gathered from the engine and show good agreement with the data.

Embedded Model Predictive Control for a Wearable Artificial Pancreas.

While artificial pancreas (AP) systems are expected to improve the quality of life among people with type 1 diabetes mellitus (T1DM), the design of convenient systems that optimize the user experience, especially for those with active lifestyles, such as children and adolescents, still remains an open research question. In this work, we introduce an embeddable design and implementation of model predictive control (MPC) of AP systems for people with T1DM that significantly reduces the weight and on-body footprint of the AP system. The embeddable controller is based on a zone MPC that has been evaluated in multiple clinical studies. The proposed embedded zone MPC features a simpler design of the periodic safe zone in the cost function and the utilization of state-of-the-art alternating minimization algorithms for solving the convex programming problems inherent to MPC with linear models subject to convex constraints. Off-line closed-loop data generated by the FDA-accepted UVA/Padova simulator is used to select an optimization algorithm and corresponding tuning parameters. Through hardware-in-the-loop in silico results on a limited-resource Arduino Zero (Feather M0) platform, we demonstrate the potential of the proposed embedded MPC. In spite of resource limitations, our embedded zone MPC manages to achieve comparable performance of that of the full-version zone MPC implemented in a 64-bit desktop for scenarios with/without meal-disturbance compensations. Metrics for performance comparison included median percent time in the euglycemic ([70, 180] mg/dL range) of 84.3% vs. 83.1% for announced meals, with an equivalence test yielding p = 0.0013 and 66.2% vs. 66.0% for unannounced meals with p = 0.0028.