System Identification Performance Research Articles

Dynamic mode decomposition based on expectation–maximization algorithm for simultaneous system identification and denoising

The present study proposes a novel dynamic mode decomposition (DMD) that can simultaneously estimate the reduced-order model, the original signal, and the system/observation noise model only from the noisy data. An expectation–maximization (EM)-algorithm DMD (EMDMD) combines DMD and the parameter adjustment of the linear dynamical system (LDS) based on the EM algorithm. The initial parameters based on the linearity of the reduced-order data are set by using DMD. Subsequently, the log-likelihood of the complete data is maximized by adjusting the LDS parameters while separating the noise. The proposed algorithm is applied to the benchmark data of the short-fat and tall-skinny data matrices with different noise and the time-series velocity fields of the flow around a circular cylinder and the separated flow around an airfoil. The performance of EMDMD in terms of system identification and noise separation from the noisy data is evaluated, and the EMDMD shows the highest system identification and noise separation performance in all data.

Optimizing kernel width for new risk-sensitive loss: A generalized algorithmic approach

This paper outlines a novel approach to the development of adaptive filtering algorithms, which minimizes susceptibility to the effects of non-Gaussian measurement noise. Using the generalized Gaussian density (GGD) function as the kernel function, we developed a generalized kernel risk-sensitive loss (GKRSL) criterion by which to derive the generalized minimum kernel risk-sensitive loss (GMKRSL) algorithm. We also developed a variable kernel width algorithm for GMKRSL (VKW-GMKRSL) to address the trade-off between convergence rate and steady-state misalignment resulting from a constant kernel width. In simulations, GMKRSL and VKW-GMKRSL outperformed existing works in system identification performance.

Artificial Ecosystem Optimizer-Based System Identification and Its Performance Evaluation

This study delves into the realm of system identification, a crucial sub-field in control engineering, aimed at constructing mathematical models of systems based on input/output data. This work particularly proposes the application of artificial ecosystem algorithm (AEO) for solving system identification problems. Inspired by the energy flow of natural ecosystems, AEO has undergone specific modifications leading to derived versions. Additionally, five diverse meta-heuristic algorithms are employed to assess their applicability and performance in system identification using data from an air stream heater experiment kit. A comprehensive performance comparison is made, considering time bounds, maximum generations, early stopping, and function evaluation constraints, presenting their respective performances. Among the evaluated algorithms, the AEO algorithm enhanced with the sine and cosine strategy stands out with a determined R2 value of 0.951. This algorithm consistently outperforms others in Wilcoxon tests, showcasing its significant success. Our study affirms that meta-heuristic algorithms, particularly the proposed AEO algorithm, can be effectively applied to system identification problems, yielding successful calculations of transfer function parameters.

Pattern Recognition in Embedded Systems for Event Occurrences

Abstract: PR (pattern recognition) typically includes interaction with humans and other complicated processes in the real world, embedded systems are ideal candidates. A typical PR application often considered the more perceptual branch of AI, responds to external events that the system detects through physical sensors or input devices by activating actuators or displaying relevant information. To explore the embedded recognition system and apply the deep learning algorithm to face detection, the deep learning-based Convolutional Neural Network (CNN)suggests two deep face detection methods. These are presented to use the deep learning algorithm.This was done to make it possible for us to use the deep learning algorithm for face detection. Because of this, to analyze the built-in face recognition system and applied the deep learning algorithm to the process of identifying faces. In addition, to do both things simultaneously. OMTCNN's training accuracy is 85.14%, higher than the unimproved algorithm. Accuracy of the recognition and calculation acceleration modules boosts embedded system face detection and identification performance. Embedded deep learning recognition is helpful.

Modeling and state-preview height control of electronically controlled air suspension system based on experiment

In the control of electronically controlled air suspension system (ECAS), the precision and stability of height control are important targets to measure the efficiency. In the control process of an ECAS system, the phenomenon of “overinflation,”“overdeflation,” and frequent inflating and deflating is commonly existed, which is rooted in the hysteresis characteristic of the system. Based on the bench test, the system identification and quantitative performance analysis of solenoid valves and air springs were carried out, and a refined ECAS system model was built. A PID control strategy is optimized and a state-preview height control strategy is established. AMESim and Simulink are used to build the vehicle and controller models. The accuracy of the model is verified by the comparison between the co-simulation and the field test, and the conclusion is drawn that the state preview height control strategy can improve the smoothness of the height control.

Dendrite Net: A White-Box Module for Classification, Regression, and System Identification.

The simulation of biological dendrite computations is vital for the development of artificial intelligence (AI). This article presents a basic machine-learning (ML) algorithm, called Dendrite Net or DD, just like the support vector machine (SVM) or multilayer perceptron (MLP). DD's main concept is that the algorithm can recognize this class after learning, if the output's logical expression contains the corresponding class's logical relationship among inputs (and \ or \ not). Experiments and main results: DD, a white-box ML algorithm, showed excellent system identification performance for the black-box system. Second, it was verified by nine real-world applications that DD brought better generalization capability relative to the MLP architecture that imitated neurons' cell body (Cell body Net) for regression. Third, by MNIST and FASHION-MNIST datasets, it was verified that DD showed higher testing accuracy under greater training loss than the cell body net for classification. The number of modules can effectively adjust DD's logical expression capacity, which avoids overfitting and makes it easy to get a model with outstanding generalization capability. Finally, repeated experiments in MATLAB and PyTorch (Python) demonstrated that DD was faster than Cell body Net both in epoch and forwardpropagation. The main contribution of this article is the basic ML algorithm (DD) with a white-box attribute, controllable precision for better generalization capability, and lower computational complexity. Not only can DD be used for generalized engineering, but DD has vast development potential as a module for deep learning. DD code is available at https://github.com/liugang1234567/Gang-neuron.

Online System Identification and Optimal Control for Mission-Critical IoT Systems Over MIMO Fading Channels

With the rapid development of mobile computing, mission-critical Internet of Things (IoT) systems have become popular. Typical mission-critical IoT systems may contain complicated unknown and unstable elements and it is of particular importance to identify and stabilize them as unstable systems may experience catastrophic consequences. We consider the identification and optimal control for a mission-critical IoT system over multiple-input–multiple-output (MIMO) fading channels. First, we focus on the optimal control of the mission-critical IoT system, assuming that the system dynamics are known, and propose a novel stochastic-approximation-based algorithm to learn the optimal control solution for the IoT controller in an online manner. Second, we extend the optimal control framework to deal with the unknown mission-critical IoT system and propose a novel normalized-stochastic-gradient-descent-based algorithm to simultaneously identify and control the system in an online manner. Using the Lyapunov stability analysis, we theoretically show the asymptotic optimality of the proposed learning algorithms. Numerical results are analyzed for our proposed scheme and for several state-of-the-art learning schemes in terms of the computational complexity, convergence, and stability performance. Specifically, the proposed scheme can be implemented more than 50% faster than the state-of-the-art learning schemes. Moreover, the system identification performance of the proposed scheme can achieve a normalized system identification mean square error (MSE) of around 0.01 in 100 iterations. This is a substantial improvement compared to the baseline algorithms, where the normalized system identification MSE diverges.

An Enhanced Fractional Least Mean Square Filter Encountering the Specific Unknown System Vector

This brief proposes an enhanced fractional derivative that can prevent the tap weight coefficients from destroying the gradient information, solve the problem caused by the fractional extreme point, and improve the convergence speed with the help of error estimation information and Sign function. Based on this fractional derivative, an enhanced fractional least mean square (EFLMS) filter algorithm is proposed. We analyze the influence of unknown system vector on the convergence performance of the EFLMS algorithm. The computational complexity of the proposed algorithm is also analyzed. Simulation experiments show that the EFLMS algorithm achieves better performance in system identification than the classic least mean square (LMS) algorithm and the existing algorithms based on fractional calculus.

A family of affine projection-type least lncosh algorithms and their step-size optimization

The Least lncosh (Llncosh) algorithm is a promising adaptive algorithm due to its robustness in impulsive interference and has a faster convergence rate than the classical sign algorithm (SA). However, when the input signal is correlated, it may still suffer from a slow convergence rate. To address this problem, this paper incorporates the method of data-reusing into the lncosh cost function to develop a family of affine projection-type Llncosh (AP-Llncosh) algorithms. To promote its performance for sparse system identification, a low-order norm constraint is also considered. Moreover, to address the problem of tradeoff between fast convergence rate and small steady-state misalignment, we optimize its step-size by minimizing the mean-square deviation (MSD). Simulation results are provided to verify the superior performance of the proposed algorithms.

An evaluation of the influence of Eigensystem Realization Algorithm settings on multiple input multiple output system identification

One characteristic of the Eigensystem Realization Algorithm method for system identification concerns about the difficulty of finding more appropriate parameters to run the algorithm. One of this work’s purposes is to tackle the cumbersome task of achieving the ideal algorithm settings, providing additional knowledge about algorithm parameters’ influence, and searching to improve results with quicker settings of the algorithm’s parameters, especially for Multiple Input Multiple Output (MIMO) systems. The application of a Fit Rate indicator to evaluate the identified system arises as a novelty in the Eigensystem Realization Algorithm applications, aiming to assess the system identification performance and drive the algorithm to better adjustments. Another objective of this article regards the application of a Pseudo Random Binary Sequence as excitation signals, which has not been used until now with the Eigensystem Realization Algorithm, despite being successfully applied in the system identification process. The proposed approach is verified and analyzed with numerical simulations for a mass–spring–damper model of 5 degrees of freedom. The results reported in time response analysis and frequency response analysis allow us to realize the effect of settings accordingly for the system identification improvement. The results analysis was extended to simulate and compare the Pseudo Random Binary Sequence with Gaussian white noise excitation, and the system was also submitted to the presence of measurement noise.

An improved model predictive control approach for fuel efficiency optimization of vessel propulsion systems

This paper studies the fuel efficiency improvement issues for the vessel propulsion systems (VPSs). Specifically, the fuel efficiency is optimized by a novel model predictive control (MPC) approach. Our study consists of two parts. In the first part, using only the measurable data, the predictive model is constructed with the aid of the sparse regression method (SRM). The second part is devoted to improving the fuel efficiency of the VPS based on the suboptimality estimate of the MPC scheme. A complete design method of the model predictive controller for the VPSs is given in this paper, which consists of the data-driven system identification, controller formulation and on-line closed-loop performance improvement. A case study on a very large ore carrier (VLOC) propulsion system is given in the end to show that the proposed approach achieves better fuel efficiency than the traditional MPC scheme.

Blind Identification of Sparse Systems Using Symbolic Dynamics Encoding

The unique properties of chaotic signals have led to their application in improving blind system identification performance. However, the role of chaos in blind identification of a sparse system has not been investigated. In this letter, we apply symbolic dynamics to encode a random signal to reap the benefits of chaos in improving blind identification of a sparse Moving Average (MA) system. We derive an estimation technique using the encoded signal by training a machine learning model that mimics a chaotic map. The novelty of our work is to exploit the merits of chaos in improving blind estimation performance of sparse systems at low signal-to-noise (SNR) ratio. The estimation error of our method is close to the minimum mean square error of the nonblind method for sparse system estimation and works well for a short data sequence.

Robust Time-Varying Parameter Proportionate Affine-Projection-Like Algorithm for Sparse System Identification

Due to its low computational burden, the affine-projection-like (APL) adaptive filtering algorithm has been extensively studied for colored signal input. Recently, a robust APL algorithm was designed by adopting the M-estimate cost function in impulsive noise environment; however, its convergence rate is very slow for sparse system identification. This paper proposed a proportionate APL M-estimate (PAPLM) algorithm, which is derived by using the proportionate matrix to heighten the convergence rate. To maintain good steady-state performance of the PAPLM algorithm, a time-varying parameter PAPLM (TV-PAPLM) algorithm is proposed, which uses a modified exponential function to adjust the time-varying parameter according to the ratio of the mean square score function to the system noise variance. Moreover, the steady-state excess mean-square error performance of PAPLM algorithm is analyzed and obtained in detail. Simulation results reveal that the proposed PAPLM and TV-PAPLM algorithms achieve fast convergence rate and good steady-state performance in sparse system identification.

System Identification and Seismic Performance Assessment of Representative RC Buildings in Kathmandu Valley

To identify dynamic characteristics of representative reinforced concrete frame buildings with brick infills, ambient vibration measurements were taken in two four-storied buildings—one situated in soft soil and the another in stiff soil. Non-parametric as well as parametric system identification (SID) algorithms were used to estimate vibration frequencies and damping of the two buildings. The numerical models of the buildings were created using the finite element method. The modal frequencies and damping ratios obtained from ambient measurements were used to calibrate and tune the finite element models. The comparison between measured vibration frequencies and those obtained from finite element model highlights the need for accuracy in modeling assumptions, in particular, consideration of the stiffness of infill walls and the flexibility of foundation soil. The finite element models calibrated with SID results were used to estimate the response of the two buildings when subjected to strong ground motion recorded at different places in the Kathmandu Valley during the 2015 Gorkha earthquake. The results show that not considering flexibility of foundation and stiffness of infill walls, as is commonly done in engineering practice, can lead to inaccurate estimates of seismic demand.

Deep Learning-Based Model Predictive Control for Continuous Stirred-Tank Reactor System.

A continuous stirred-tank reactor (CSTR) system is widely applied in wastewater treatment processes. Its control is a challenging industrial-process-control problem due to great difficulty to achieve accurate system identification. This work proposes a deep learning-based model predictive control (DeepMPC) to model and control the CSTR system. The proposed DeepMPC consists of a growing deep belief network (GDBN) and an optimal controller. First, GDBN can automatically determine its size with transfer learning to achieve high performance in system identification, and it serves just as a predictive model of a controlled system. The model can accurately approximate the dynamics of the controlled system with a uniformly ultimately bounded error. Second, quadratic optimization is conducted to obtain an optimal controller. This work analyzes the convergence and stability of DeepMPC. Finally, the DeepMPC is used to model and control a second-order CSTR system. In the experiments, DeepMPC shows a better performance in modeling, tracking, and antidisturbance than the other state-of-the-art methods.

Bayesian step least mean squares algorithm for Gaussian signals

Selecting the step of the least mean squares (LMS) algorithm is an old problem. This study uses a new approach to address this problem resulting in a new algorithm with excellent system identification performance. The LMS algorithm, with time-varying step, size can be shown to be equivalent to the Kalman filter in some conditions. This is as long as the state noise of the Kalman filter and the step size of the LMS algorithm are chosen carefully. The Kalman filter is the optimum linear estimator (Bayesian) given the state and the measurement noise covariance matrices, but these matrices are not always known. This work considers the case where these matrices are not known, in the special cases that the Kalman filter reduces to the LMS. This results in an algorithm to select the step-size of the LMS algorithm with few priors. The optimum step size can be calculated using estimates of the probability density function (PDF) of the coefficient estimation error variance (qw ) and measurement noise variance (qv ). The PDFs can be estimated from the data using Bayes’ rule and assuming Gaussian reference and measurement noise signals. The resulting algorithm to determine qw and qv is a second small Kalman filter, and the outputs of this filter (means and covariances) are used to determine the expected value of the step.

A Variable Step Size for Maximum Correntropy Criterion Algorithm with Improved Variable Kernel Width

The maximum correntropy criterion (MCC) algorithm has popularly been used in suppressing impulsive noise. It mainly relies on the choice of step size and kernel width. With the study of step size, this paper proposes a novel algorithm that updates the step size based on constructing a variable step size function that uses the moving weighted average algorithm to keep performance stable and sigmoid function to speed the convergence rate. The proposed algorithm is robust against impulsive noises and generates a large step size to accelerate the convergence in iteration beginning, and system mutation makes the system identification performance more stable when the error is small. Based on the improved variable kernel width maximum entropy criterion (IVKW‐MCC) algorithm, which overcomes the shortcoming of how to choose a reliable kernel width in the MCC algorithm, simulation results are compared with other MCC algorithms under impulsive noise interference. Simulations in the system identification scenarios show that the proposed algorithm has better performance in improving the convergence speed over other algorithms. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

A Class of Diffusion Zero Attracting Stochastic Gradient Algorithms With Exponentiated Error Cost Functions

In this paper, a class of diffusion zero-attracting stochastic gradient algorithms with exponentiated error cost functions is put forward due to its good performance for sparse system identification. Distributed estimation algorithms based on the popular mean-square error criterion have poor behavior for sparse system identification with color noise. To overcome this drawback, a class of stochastic gradient least exponentiated (LE) algorithms with exponentiated error cost functions were proposed, which achieved a low steady-state compared with the least mean square (LMS) algorithm. However, those LE algorithms may suffer from performance deterioration in the spare system. For sparse system identification in the adaptive network, a polynomial variable scaling factor improved diffusion least sum of exponentials (PZA-VSIDLSE) algorithm and an l p -norm constraint diffusion least exponentiated square (LP-DLE2) algorithm are proposed in this work. Instead of using the l 1 -norm penalty, an l p -norm penalty and a polynomial zero-attractor are employed as a substitution in the cost functions of the LE algorithms. Then, we perform mean behavior model and mean square behavior modal of the LP-DLE2 algorithm with several common assumptions. Moreover, simulations in the context of distributed network sparse system identification show that the proposed algorithms have a low steady-state compared with the existing algorithms.

Block-Sparsity Log-Sum-Induced Adaptive Filter for Cluster Sparse System Identification

In this work, an effective adaptive block-sparsity log-sum least mean square (BSLS-LMS) algorithm is proposed to improve the convergence performance of cluster sparse system identification. The main idea of the proposed scheme is to add a new block-sparsity induced term into the cost function of the LMS algorithm. We utilize the ${l}_{{1}}$ norm of adaptive tap weights and log-sum as a mixed constraint, via optimizing the cost function through the gradient descent method, the proposed adaptive filtering method can iteratively move the identified signals towards the optimal solutions, and finally identify the unknown system accurately. The cluster-sparse system response, with block length and arbitrary average sparsity, is generated by a Markov-Gaussian (M-G) model. For the white Gaussian input data, the theoretical formulas of the steady-state mis-adjustment and convergence behaviors of the BSLS-LMS are derived in a general sparse system and a block sparse system, respectively. Numerical experiments demonstrate that the proposed adaptive BSLS-LMS algorithm achieves much better convergence behavior than conventional sparse signal recovery solutions. The experimental study also verifies the consistency between the simulation results and the theoretical analysis.

Nearest Kronecker Product Decomposition Based Generalized Maximum Correntropy and Generalized Hyperbolic Secant Robust Adaptive Filters

Robust adaptive signal processing algorithms based on a generalized maximum correntropy criterion (GMCC) suffers from high steady state misalignment. In an endeavour to achieve lower steady state misalignment, in this letter we propose a generalized hyperbolic secant function (GHSF) as a robust norm and derive the generalized hyperbolic secant adaptive filter (GHSAF). The new algorithm is seen to offer robust system identification performance over the conventional GMCC algorithm. To further improve the convergence performance under non-Gaussian noise environments, we propose the nearest Kronecker product decomposition based GMCC and GHSAF algorithms. Extensive simulation study show the improved convergence performance provided by the proposed algorithms for system identification.