Data Dropouts Research Articles

Model Predictive Control Based on Modified Smith Predictor for Networked Control Systems

This study presents a predictive controller for networked control systems (NCSs), which is a model predictive control (MPC) combined with Smith predictor. The network delays and data dropouts are problems, which greatly weaken the controller performance. In the proposed controller, there are two internal loops. The first is the loop around the MPC with the model of the system, which predicted future outputs. The other is the loop around the plant to give the error between the model and actual plant. The proposed controller is designed for controlling a DC servo system via a network. The practical results based on Matlab/Simulink are established. The practical results show that the performance of the proposed controller is greatly improved over a wide range of networked time delay and data dropouts comparing to other controllers.

A Technical Overview of Recent Progresses on Stochastic Iterative Learning Control

This paper contributes to a technical overview of recent progresses on stochastic iterative learning control (ILC), where stochastic ILC implies the learning control for systems with various random signals and factors such as stochastic noises, random data dropouts and inherent random asynchronism. The fundamental principles of ILC are first briefed with emphasis on the system formulations and typical analysis methods. Then the recent progresses on stochastic ILC are reviewed in three parts: additive randomness case, multiplicative randomness case, and coupled randomness case, respectively. Three major approaches, i.e., expectation-based method, Kalman filtering-based method, and stochastic approximation-based method, are clarified. Promising research directions are also presented for further investigation.

Stabilisation to input‐to‐state stability for continuous‐time dynamical systems via event‐triggered impulsive control with three levels of events

This study investigates the issue of stabilisation to input-to-state stability (ISS) by event-triggered impulsive control (ETIC) for continuous-time dynamical systems (CDSs) with external inputs/disturbances. The CDSs are more general systems with possible multiple vector fields and dynamics. The ETIC scheme is based on three key indices: threshold value, control-free index, and check period, and it is executed by designing three levels of events from which ETIC is triggered. Both ETIC and ETIC with time-delays (delayed ETIC) are studied, respectively, and the criteria for ISS are derived for CDSs under ETIC. Thus the stabilisation to ISS is achieved for CDSs by the designed ETIC with or without time delays. The data dropout of ETIC is also investigated, where the maximal allowable dropout rates are estimated, respectively. It shows that the proposed ETIC scheme has advantages over the reported event-triggered control and has robustness with respect to the network-induced time-delays although larger time-delays might lead to a slower convergence speed for the stabilisation to ISS. Finally, two examples with numerical simulations are given to illustrate the obtained results.

Open-closed-loop iterative learning control for a class of nonlinear systems with random data dropouts

In this paper, an open-closed-loop iterative learning control (ILC) algorithm is constructed for a class of nonlinear systems subjecting to random data dropouts. The ILC algorithm is implemented by a networked control system (NCS), where only the off-line data is transmitted by network while the real-time data is delivered in the point-to-point way. Thus, there are two controllers rather than one in the control system, which makes better use of the saved and current information and thereby improves the performance achieved by open-loop control alone. During the transfer of off-line data between the nonlinear plant and the remote controller data dropout occurs randomly and the data dropout rate is modeled as a binary Bernoulli random variable. Both measurement and control data dropouts are taken into consideration simultaneously. The convergence criterion is derived based on rigorous analysis. Finally, the simulation results verify the effectiveness of the proposed method.

Iterative Learning Control for Network Data Dropout in Nonlinear System

The paper presents iterative learning control for data dropout in nonlinear system. The parallel distribution compensation method is used to determine the T-S nonlinear model and the nonlinear model is converted into local linear model. Assuming the probability of data loss is known. It is assumed that the probability of data loss is known, and the loss of data is described using a sequence that satisfies the Bernoulli distribution. The design of the learning control controller for linear discrete systems with data loss is studied. The iterative learning controller for data dropout is designed with the T-S model. The iterative learning controller designed has expected convergence characteristics and quadratic performance index. The simulation results show that the design method is effective.

Data-driven adaptive fractional order PI control for PMSM servo system with measurement noise and data dropouts

In this paper, data-driven adaptive fractional order proportional integral (AFOPI) control is presented for permanent magnet synchronous motor (PMSM) servo system perturbed by measurement noise and data dropouts. The proposed method directly exploits the closed-loop process data for the AFOPI controller design under unknown noise distribution and data missing probability. Firstly, the proposed method constructs the AFOPI controller tuning problem as a parameter identification problem using the modified lp norm virtual reference feedback tuning (VRFT). Then, iteratively reweighted least squares is integrated into the lp norm VRFT to give a consistent compensation solution for the AFOPI controller. The measurement noise and data dropouts are estimated and eliminated by feedback compensation periodically, so that the AFOPI controller is updated online to accommodate the time-varying operating conditions. Moreover, the convergence and stability are guaranteed by mathematical analysis. Finally, the effectiveness of the proposed method is demonstrated both on simulations and experiments implemented on a practical PMSM servo system.

Stabilization of Discrete-Time Dynamical Systems Under Event-Triggered Impulsive Control with and Without Time-Delays

This paper investigates the issue of stabilization for discrete-time dynamical systems (DDS) by event-triggered impulsive control (ETIC). Based on some relatively simple threshold constants, three levels of event conditions are set and thus the ETIC scheme is designed. Three cases for ETIC with and without time-delays and data dropouts are studied respectively, and the criteria on exponential stability are derived for the controlled DDS. The stabilization in the form of exponential stability is achieved for DDS under the designed ETIC with or without time-delays. And in the case of the ETIC data dropouts, the conditions of exponential stabilization are derived for DDS and the maximal allowable dropout rates for ETIC are estimated. Finally, one example with numerical simulations is worked out for illustration.

An Enhanced PIRATA Dataset for Tropical Atlantic Ocean–Atmosphere Research

AbstractThe Prediction and Research Moored Array in the Tropical Atlantic (PIRATA) provides measurements of the upper ocean and near-surface atmosphere at 18 locations. Time series from many moorings are nearly 20 years in length. However, instrumental biases, data dropouts, and the coarse vertical resolutions of the oceanic measurements complicate their use for research. Here an enhanced PIRATA dataset (ePIRATA) is presented for the 17 PIRATA moorings with record lengths of at least seven years. Data in ePIRATA are corrected for instrumental biases, temporal gaps are filled using supplementary datasets, and the subsurface temperature and salinity time series are mapped to a uniform 5-m vertical grid. All original PIRATA data that pass quality control and that do not require bias correction are retained without modification, and detailed error estimates are provided. The terms in the mixed-layer heat and temperature budgets are calculated and included, with error bars. As an example of ePIRATA’s application, the vertical exchange of heat at the base of the mixed layer (Q−h) is calculated at each PIRATA location as the difference between the heat storage rate and the sum of the net surface heat flux and horizontal advection. Off-equatorial locations are found to have annual mean cooling rates of 20–60 W m−2, while cooling at equatorial locations reaches 85–110 W m−2 between 10° and 35°W and decreases to 40 W m−2 at 0°. At most off-equatorial locations, the strongest seasonal cooling from Q−h occurs when winds are weak. Possible explanations are discussed, including the importance of seasonal modulations of mixed-layer depth and the diurnal cycle.

Evaluation of linear autoregressive models for estimation of energy spectra in gappy turbulent velocity data

AbstractThis paper deals with the problem of data interpolation in velocity time series measured by acoustic Doppler velocimeters and acoustic Doppler current profilers; the gap‐filled data are often used to determine turbulent kinetic energy (TKE) dissipation using Kolmogorov's inertial subrange scaling. For the latter to estimate dissipation accurately, it is important that the interpolation scheme preserves the attributes, both spectral slope and component magnitudes, of the true energy spectrum. We show that this goal can be achieved using the simple zeroth‐order sample and hold interpolation in situations where isolated data gaps, having durations shorter than the integral time scale of flow, occur. Its success is explained using the framework of stochastic autoregressive (AR) processes, which we also compare to the Langevin equation for the Lagrangian velocity of a turbulent flow field. We also demonstrate that linear interpolation is not appropriate because it can be interpreted as a nonstationary second order AR process, leading to erroneous conclusions for spectral slope and magnitude. When data dropouts occur in clusters, i.e., of durations longer than the integral time scale, we propose to use the first order AR process, of which sample and hold is its limiting case, for interpolation. The effectiveness of our proposal is tested and demonstrated with synthetic time series having a range of spectral slopes, from −7/6 to −8/3, and with experimental data measured in a turbulent channel flow. A comparison is also made with the more sophisticated proper orthogonal decomposition‐based interpolation. The paper ends with a step‐by‐step procedure on using the proposed method in applications.

A Novel Markov Chain Based ILC Analysis for Linear Stochastic Systems Under General Data Dropouts Environments

This technical note contributes to the convergence analysis for iterative learning control (ILC) for linear stochastic systems under general data dropout environments, i.e., data dropouts occur randomly at both the measurement and actuator sides. Data updating in the memory array is arranged in such a way that data at every time instance is updated independently, which allows successive data dropouts both in time and iteration axes. The update mechanisms for both the computed input and real input are proposed and then the update process of both inputs is shown to be a Markov chain. By virtue of Markov modeling, a new analysis method is developed to prove the convergence in both mean square and almost sure senses. An illustrative example verifies the theoretical results.

Iterative Learning Control for Nonlinear Systems with Data Dropouts at Both Measurement and Actuator Sides

AbstractThis paper discusses the iterative learning control (ILC) for nonlinear systems under a general networked control structure, in which random data dropouts occur independently at both measurement and actuator sides. Both updating algorithms are proposed for the computed input signal at the learning controller and the real input signal at the plant, respectively. The system output is strictly proved to converge to the desired reference with probability one as the iteration number goes to infinity. A numerical simulation is provided to verify the effectiveness of the proposed mechanism and algorithms.

A framework of iterative learning control under random data dropouts: Mean square and almost sure convergence

SummaryThis paper addresses the iterative learning control problem under random data dropout environments. The recent progress on iterative learning control in the presence of data dropouts is first reviewed from 3 aspects, namely, data dropout model, data dropout position, and convergence meaning. A general framework is then proposed for the convergence analysis of all 3 kinds of data dropout models, namely, the stochastic sequence model, the Bernoulli variable model, and the Markov chain model. Both mean square and almost sure convergence of the input sequence to the desired input are strictly established for noise‐free systems and stochastic systems, respectively, where the measurement output suffers from random data dropouts. Illustrative simulations are provided to verify the theoretical results.

Automatic target recognition in missing data cases

Automatic target recognition (ATR) is the process where computer algorithms are used to detect and classify objects or regions of interest in sensor data [1]. ATR algorithms have been developed for a broad range of sensors, including electro-optic (EO), infrared (IR), and microwave sensors (radar). A key advantage of microwave radar is its utility for all weather and time of day compared with other types of sensors. One limitation of many radar systems' ATR performance is the poor spatial resolution in the range and cross-range dimensions. While the range resolution of the microwave radar system can be straightforwardly improved by increasing the system's radio frequency bandwidth, superior cross-range (or azimuth) resolution requires an increase in antenna size. In order to achieve resolution similar to EO or IR systems, an enormous antenna must be used for transmission and reception, hindering microwave radar system's general applicability to ATR problems. However, by operating a synthetic aperture radar (SAR), the large antenna requirement can be overcome, and very fine cross-range resolution can be obtained with a modest real aperture antenna. Indeed, SAR has been used to generate high-resolution 2-D or 3-D object images at a variety of different microwave frequencies. In particular, wideangle SAR images can contain important features of an object from a diverse set of observation angles collected by a radar system orbiting around a target or scene. These feature-rich wide-angle SAR images are useful for ATR due to their high resolution and comprehensive coverage of the target. Interference, jamming, or data dropouts, however, are commonplace in practical SAR environments, and result in an incomplete data set. These missing and corrupted data cause substantial degradation in the generated SAR imagery, hampering their utility for subsequent ATR processing especially when data-independent image formation algorithms are used. A prime example of these difficulties is operating a SAR system in the very high frequency (VHF) or ultrahigh frequency (UHF) bands, where the spectrum is frequently crowded [2].

A distributed secondary scheme with terminal sliding mode controller for energy storages in an islanded microgrid

Different levels of the stored energy is the main challenge in control of the distributed energy storages (DESs) in an islanded microgrid. The conventional power-droop and the secondary distributed controllers of distributed generators (DGs) does not consider the long time-span dynamics of state of charge (SoC) of the DESs. Subsequently, the DESs with lower initial SoCs are discharged before other DESs. Besides, the SoC-droop primary control results in further deviation of the frequency of the microgrid. In this paper, the secondary control scheme is employed to share the power mismatch, match the SoCs of the DESs, and regulate the frequency and voltage of the microgrid. The proposed scheme has distributed cooperative architecture, and employs distributed terminal sliding mode controller (DTSMC) for the state regulators, and proportional controller for distributed power-sharing and SoC-matching. The proposed scheme organizes the controllable DGs, the DESs with limited SoCs, and the uncooperative DGs with unknown generation, through communicating between the neighbor cooperative DGs and DESs. Performance of the designed DTSMC is verified for the changes of communication topology, time-delays, data drop-outs, load variations, and external disturbances. DTSMC provides finite-time convergence, fast transients, and improved robustness.

Learning control for linear systems under general data dropouts at both measurement and actuator sides: A Markov chain approach

This paper contributes to the convergence analysis of iterative learning control for linear systems under general data dropouts at both measurement and actuator sides. By using a simple compensation mechanism for the dropped data, the sample path behavior along the iteration axis is analyzed and formulated as a Markov chain first. Based on the Markov chain, the recursion of the input error is reformulated as a switching system, and then a novel convergence proof is established in the almost sure sense under mild design conditions. Illustrative examples are provided to verify the theoretical results.

Data-Driven Learning Control for Stochastic Nonlinear Systems: Multiple Communication Constraints and Limited Storage.

This paper proposes a data-driven learning control method for stochastic nonlinear systems under random communication conditions, including data dropouts, communication delays, and packet transmission disordering. A renewal mechanism is added to the buffer to regulate the arrived packets, and a recognition mechanism is introduced to the controller for the selection of suitable update packets. Both intermittent and successive update schemes are proposed based on the conventional P-type iterative learning control algorithm, and are shown to converge to the desired input with probability one. The convergence and effectiveness of the proposed algorithms are verified by means of illustrative simulations.

Data-Driven Control With Input Design-Based Data Dropout Compensation for Networked Nonlinear Systems

This brief investigates the regulation problem for a class of networked nonlinear systems with measurement noise, where random data dropouts in both the feedback and forward channels are considered. To actively compensate for the two-channel data dropouts, a data-driven networked compensation control method is proposed, which consists of two aspects: 1) to calculate a control increment based on the measured output error in the controller and 2) to design a data dropout compensation strategy based on the latest control increment available in the actuator. The proposed method merely depends on the input and output data of the controlled plant, without using explicit or implicit information of its mathematical model. Moreover, only one control command needs to be transmitted in the forward channel at each time instant. A sufficient condition is derived to guarantee the closed-loop stability and output error convergence. Both numerical simulations and experimental tests are conducted to demonstrate the effectiveness of the proposed method.

Intermittent and Successive ILC for Stochastic Nonlinear Systems with Random Data Dropouts

AbstractThe iterative learning control (ILC) problem is addressed in this paper for stochastic nonlinear systems with random data dropouts. The data dropout is modeled by the conventional Bernoulli random variable to describe the successful transmission or loss. Both intermittent and successive ILC are considered, where the former stops updating if no information is received, while the latter keeps updating based on the latest available data. It is strictly proved the almost sure convergence of both algorithms. The simulations on a mechanical model are provided to show the comparisons and effectiveness of the proposed algorithms.

Bayesian Inference for Single-cell Clustering and Imputing

Single-cell RNA-seq gives access to gene expression measurements for thousands of cells, allowing discovery and characterization of cell types. However, the data is noise-prone due to experimental errors and cell type-specific biases. Current computational approaches for analyzing single-cell data involve a global normalization step which introduces incorrect biases and spurious noise and does not resolve missing data (dropouts). This can lead to misleading conclusions in downstream analyses. Moreover, a single normalization removes important cell type-specific information. We propose a data-driven model, BISCUIT, that iteratively normalizes and clusters cells, thereby separating noise from interesting biological signals. BISCUIT is a Bayesian probabilistic model that learns cell-specific parameters to intelligently drive normalization. This approach displays superior performance to global normalization followed by clustering in both synthetic and real single-cell data compared with previous methods, and allows easy interpretation and recovery of the underlying structure and cell types.

Smart Augmentation Learning an Optimal Data Augmentation Strategy

A recurring problem faced when training neural networks is that there is typically not enough data to maximize the generalization capability of deep neural networks(DNN). There are many techniques to address this, including data augmentation, dropout, and transfer learning. In this paper, we introduce an additional method which we call Smart Augmentation and we show how to use it to increase the accuracy and reduce overfitting on a target network. Smart Augmentation works by creating a network that learns how to generate augmented data during the training process of a target network in a way that reduces that networks loss. This allows us to learn augmentations that minimize the error of that network. Smart Augmentation has shown the potential to increase accuracy by demonstrably significant measures on all datasets tested. In addition, it has shown potential to achieve similar or improved performance levels with significantly smaller network sizes in a number of tested cases.