Iterative Identification Method Research Articles

Classification of Ice Particle Shapes Using Machine Learning on Forward Light Scattering Images

Abstract Machine learning (ML) has rapidly transitioned from a niche activity to a mainstream tool for environmental research applications including atmospheric science cloud microphysics studies. Two recently developed cloud particle probes measure the light scattered in the near forward direction and save digital images of the scattering light. Scattering pattern images collected by the Particle Phase Discriminator mark 2, Karlsruhe edition (PPD-2K), and the Small Ice Detector, version 3 (SID-3) provide valuable information for particle shape and size characterization. Since different particle shapes have distinctly different light scattering characteristics, the images are ideally suited for ML. Here, results of a ML project to characterize ice particle shapes sampled by the PPD-2K in ice fog and diamond dust during a 3-yr project in Fairbanks, Alaska. About 2.15 million light-scattering pattern images were collected during 3 years of measurements with the PPD-2K. Visual Geometry Group (VGG) convolutional neural network (CNN) was trained to categorize light-scattering patterns into eight categories. Initial training images (120 each category) were selected by human visual examination of data, and the training dataset was augmented using an automated iterative method for image identification of further images which were all visually inspected by a human. Results were well correlated to similar categories identified from previously developed classification algorithms. ML identifies characteristics not included in automated analysis such as sublimation. Of the 2.15 million images analyzed, 1.3% were categorized as spherical (liquid), 43.5% were categorized as having rough surfaces, 15.3% were pristine, 16.3% were categorized as sublimating, and the remaining 23.6% did not fit into any of those categories (irregular or saturated). Significance Statement The shapes and sizes of cloud particles can be extremely important for understanding the conditions that exist in the cloud. In this study, we show that more information about cloud particle characteristics can be identified by using machine learning than by traditional means. To demonstrate this, data are analyzed from a 3-yr study of ice fog and diamond dust events in Fairbanks, Alaska. The Particle Phase Discriminator instrument collected 2.15 million light-scattering pattern images of cloud particles during ground-based measurements. Neither traditional techniques nor machine learning were able to identify all categories, but a combination of both techniques led to a more complete view of ice particle shapes.

Decomposition‐based maximum likelihood gradient iterative algorithm for multivariate systems with colored noise

SummaryIn this paper, we use the maximum likelihood principle and the negative gradient search principle to study the identification issues of the multivariate equation‐error systems whose outputs are contaminated by an moving average noise process. The model decomposition technique is used to decompose the system into several regressive identification subsystems based on the number of the outputs. In order to improve the parameter estimation accuracy, a decomposition‐based multivariate maximum likelihood gradient iterative algorithm is proposed by means of the maximum likelihood principle and the iterative identification method. The numerical simulation example indicates that the proposed method has better parameter estimation results than the compared decomposition‐based multivariate maximum likelihood gradient algorithm.

Filtered generalized iterative parameter identification for equation‐error autoregressive models based on the filtering identification idea

SummaryBy using the collected batch data and the iterative search, and based on the filtering identification idea, this article investigates and proposes a filtered multi‐innovation generalized projection‐based iterative identification method, a filtered generalized gradient‐based iterative identification method, a filtered generalized least squares‐based iterative identification method, a filtered multi‐innovation generalized gradient‐based iterative identification method and a filtered multi‐innovation generalized least squares‐based iterative identification method for equation‐error autoregressive systems described by the equation‐error autoregressive models. These filtered generalized iterative identification methods can be extended to other linear and nonlinear scalar and multivariable stochastic systems with colored noises.

Gradient-based Iterative Identification Method for Non-uniformly Sampled Input-nonlinear Systems

Gradient-based Iterative Identification Method for Non-uniformly Sampled Input-nonlinear Systems

Highly‐computational hierarchical iterative identification methods for multiple‐input multiple‐output systems by using the auxiliary models

AbstractThe identification of multiple‐input multiple‐output (MIMO) systems is an important part of designing complex control systems. This article studies an auxiliary model least squares iterative (AM‐LSI) algorithm for MIMO systems. With the expansion of the system scale and limitations of the computer resources, there is an urgent need for an identification algorithm that provides higher computational efficiency. To address this issue, this article further derives a hierarchical identification model and proposes a new auxiliary model hierarchical least squares iterative (AM‐HLSI) algorithm for MIMO systems by applying the hierarchical identification principle. Through the analysis of the computational efficiency, the AM‐HLSI algorithm has higher computational efficiency than the AM‐LSI algorithm. Additionally, the feasibility of the AM‐LSI and AM‐HLSI algorithms is validated by a simulation example.

An iterative identification method for the dynamics and hysteresis of robots with elastic joints

An iterative identification method for the dynamics and hysteresis of robots with elastic joints

Multi‐innovation gradient‐based iterative identification methods for feedback nonlinear systems by using the decomposition technique

SummaryThis paper studies the parameter estimation problems of feedback nonlinear systems. Combining the multi‐innovation identification theory with the negative gradient search, we derive a multi‐innovation gradient‐based iterative algorithm. In order to reduce the computational burden and further improve the parameter estimation accuracy, a decomposition multi‐innovation gradient‐based iterative algorithm is proposed by using the decomposition technique. The key is to transform an original system into two subsystems and to estimate the parameters of each subsystem, respectively. A simulation example is provided to demonstrate the effectiveness of the proposed algorithms.

An iterative identification method of joint properties and its applications in an end-toothed connection structure

Accurate identification of mechanical connections plays a significant role in predicting the dynamic behaviors of assembled structures. Traditional identification methods based on frequency response function (FRF) decoupling may not be suitable for the joints with many interface degrees of freedom (DOFs), especially in the case of insufficient measured FRF data and large experimental noise. In this paper, a new iterative method is proposed to solve this problem. First, the incremental basic formulas on the measured DOFs and pseudo-measured DOFs are derived, respectively. The former avoids the measurements of rotational or interface DOFs whereas the latter is suitable for the identification of the joint with many interface DOFs. Then, the joint properties expressed in terms of mass, stiffness and damping matrices are updated from the solution of the linear equation system combined with the iterative basic identification equations at different frequencies. Thus, the original problem is transformed into estimating the parameters iteratively by minimizing the differences between predicted responses and measured ones, which effectively improves the identification accuracy and numerical stability. Moreover, the iterative method is extended to the identification of nonlinear joints based on describing function theory. The validity and superiority of the proposed method are verified by a simple lumped parameter system. Then, a numerical example of a structure connected with bolted joints, which has a large number of interface DOFs, is simulated to verify the convergence and the robustness of the method to noise. Finally, the joint dynamic properties of an end-toothed connection under different contact states are identified, and the assembly responses are also accurately predicted. The effectiveness and the applicability of the proposed iterative identification method to the real structures are well validated.

Filtered multi‐innovation‐based iterative identification methods for multivariate equation‐error ARMA systems

SummaryThis paper focuses on the parameter estimation issues of multivariate equation‐error autoregressive moving average systems. By applying the gradient search and the multi‐innovation theory, we derive a multi‐innovation gradient based iterative (MI‐GI) algorithm. In order to improve the computational efficiency and the parameter estimation accuracy, a filtering and decomposition based gradient iterative (F‐D‐GI) algorithm is presented by using the data filtering technique and the decomposition technique. The key is to choose an appropriate filter to filter the input‐output data and to transform an original system into several subsystems. Compared with the MI‐GI algorithm, the F‐D‐GI algorithm can generate more accurate parameter estimates. Finally, an illustrative example is provided to indicate the effectiveness of the proposed algorithms.

A Deep Residual Recurrent Neural Network Model-Augmented Attention With Physical Characteristics: Application to Turntable Servo System

Turntable servo systems are affected by working status changing in actual work, thus, bringing some uncertain factors. Having models that have accurate predictive capabilities can be valuable for control and decision-making purposes. This article proposes a deep residual recurrent neural network (DRRNN) model-augmented attention with physical characteristics, which is used to build an accurate speed prediction model for a turntable servo system. Integrating the known physical features into the design of the residual network model can increase the interpretability of the overall model. Meanwhile, the recurrent neural network-augmented attention model is used to represent the unknown nonlinear characteristics. The addition of the attention mechanism can effectively improve the accuracy of the nonlinear model. Finally, this article proposes an iterative identification method based on the linear and nonlinear model error with alternative compensation. Through alternating error compensation, the linear model’s accuracy can be continuously improved, then the overall and the nonlinear part of the model can be established accurately by retraining the whole model. Experimental results show that the DRRNN model designed in this article can predict the rotational speed in single-step or multistep well. Compared with the traditional mechanism modeling method, the modeling accuracy can be greatly improved.

A New Adaptive Switching Control Method for the Complex Operating Condition of the Grid-Connected Power System

Due to the increase in wind power accessing power system in recent years, the low-frequency oscillation of power system is more complicated. Therefore, a new adaptive switching control method combined with a nonlinear system is proposed for the complex operating conditions of the grid-connected power system. Firstly, this method decomposes the low-frequency oscillation control problem of the power system into two parts, which are the linear system control and the nonlinear system control. Then, in the linear system control part, the iterative identification method is used to control this part. In the nonlinear system control part, the nonlinear term is decomposed into a combination of the previous beating measurable and the nonlinear estimation increment. Secondly, the nonlinear system is divided into three different operating conditions by the nonlinear estimation increment. Thus, four different adaptive controllers are obtained. Then, the designed controller is switched according to the controlled object with four controllers by the switching function, which not only ensures the stability of the closed-loop system, but also improves the control performance of the controller under different operating conditions. Finally, the feasibility and effectiveness of this method are verified, and the controller can fit the control efficiency of the real power system well.

A Subspace Identification Technique for Real-Time Stability Assessment of Droop Based Microgrids

Real-time detection of the microgrid stability is crucial for determining and adjusting the power-sharing droop control's gains to maintain a sufficient stability margin. Maintaining minimum relative stability should be ensured at different operating conditions to accommodate sudden system changes. This article develops a novel subspace-based identification technique to assess microgrid stability in real-time without relying on offline analytical small or large-signal models. Unlike conventional system identification techniques that require the introduction of external excitation (signal probing, load switching, or contingency event), the proposed method employs a simple internal routine through small and short-duration perturbations in the active power droop gain of inverter-based distributed generation (IBDG). The use of subspace identification does not require pre-defining the system's order, avoids the exhaustive computations associated with iterative identification methods and can handle a change of microgrid's network configuration. The proposed stability assessment method has been tested on an IBDG microgrid considering different operating conditions in MATLAB/Simulink. The accuracy of the proposed real-time stability assessment tool is determined by comparing the results to the analytically-derived small-signal model as well as the Matrix pencil and Prony methods.

Particle Filtering-based Iterative Identification Methods for a Class of Nonlinear Systems with Interval-varying Measurements

Particle Filtering-based Iterative Identification Methods for a Class of Nonlinear Systems with Interval-varying Measurements

Vibration load identification in the time-domain of high arch dam under discharge excitation based on hybrid LSQR algorithm

The identification of vibration load is of great importance in studying the vibration induced by discharge of high arch dam. Considering the ill-posedness in time-domain identification of vibration load, a hybrid least squares QR (LSQR) iterative identification method is proposed. The system response is expressed as the convolution of the unit impulse response function and the excitation load. It is discretized into a set of linear equations, and the mathematical model of the inverse problem of load identification is established. On the basis of vibration signal filtering and noise reduction, Tikhonov regularization method is used to pre optimize LSQR iterative algorithm. This process is performed because the LSQR algorithm is prone to nonconvergence when the error of observation data is large. Thus, a hybrid LSQR algorithm is obtained to improve the ill-posedness of the inverse problem. Numerical examples show that the proposed method can recognize multiple vibration loads effectively and stably under different noise levels, and the recognition accuracy is better than the conventional regularization method. The proposed method is applied to an engineering example. Results show that the proposed method is feasible and effective for the time-domain identification of vibration load of a high arch dam.

An Iterative Closed-Loop Identification Method for Continuous-Time FOPDT Model Using Equality Constraints

An Iterative Closed-Loop Identification Method for Continuous-Time FOPDT Model Using Equality Constraints

Parameter estimation of MIMO two-dimensional ARMAX model based on IGLS method

Abstract This paper presents an iterative method for the unbiased identification of linear Multiple-Input Multiple-Output (MIMO) discrete two-dimensional (2D) systems. The system discussed here has Auto-Regressive Moving-Average model with exogenous inputs (ARMAX model). The proposed algorithm functions on the basis of the traditional Iterative Generalized Least Squares (IGLS) method. In summary, this paper proposes a two-dimensional Multiple-Input Multiple-Output Iterative Generalized Least Squares (2DMIGLS) algorithm to estimate the unknown parameters of the ARMAX model. Finally, simulation results show the efficiency and accuracy of the presented algorithm in estimating the unknown parameters of the model in the presence of colored noise.

Iterative identification methods for a class of bilinear systems by using the particle filtering technique

SummaryThis article mainly studies the iterative parameter estimation problems of a class of nonlinear systems. Based on the auxiliary model identification idea, this article utilizes the estimated parameters to construct an auxiliary model, and uses its outputs to replace the unknown noise‐free process outputs, and develops an auxiliary model least squares‐based iterative (AM‐LSI) identification algorithm. For further improving the parameter estimation accuracy, we use a particle filter to estimate the unknown noise‐free process outputs, and derive a particle filtering least squares‐based iterative (PF‐LSI) identification algorithm. During each iteration, the AM‐LSI and PF‐LSI algorithms can make full use of the measured input–output data. The simulation results indicate that the proposed algorithms are effective for identifying the nonlinear systems, and can generate more accurate parameter estimates than the auxiliary model‐based recursive least squares algorithm.

A novel iterative identification based on the optimised topology for common state monitoring in wireless sensor networks

Power consumption and data redundancy of wireless sensor networks (WSN) are widely considered for a distributed state monitoring network. For reducing the energy consumption and data amount, we propose a topology optimisation and an iterative parameter identification method for estimating the common model factors in WSN. The former method optimises the decentralised topology such that all the leaf nodes in a community connect to the head node directly. A circle topology is built to enable the remote leaf nodes to link to the head node through two adjoining relay nodes to reduce the whole communication distance and power consumption. Based on the optimised topology, an iterative identification method is proposed to minimise the information capacity by transmitting the processed results instead of raw data to reduce the data amount for calculation and storage. Then, we prove the consensus and convergence of the proposed identification method. Finally, two simulations verify the effectiveness of the proposed method and the comparative results present the data reduction for the on-board calculation, communication, and storage in the practical use of WSN.

Recursive–iterative identification method for power converters

Control of power electronic converters is an application area where system identification is necessary due to the implementation of suitable controllers. Thanks to the identification process, it is possible to obtain a suitable mathematical model of the converter, which can be used for the design of its control. This article deals with the design of a new recursive–iterative identification method applicable in practice. The method has the task of finding a parametric model. The converter plant linear time-invariant model is found and described mathematically by its transfer function. The method is verified by simulation and also on a real sample of a synchronous buck converter.

Model-plant mismatch detection for cross-directional processes

This paper presents a two-component framework to detect model-plant mismatch (MPM) in cross-directional (CD) processes on paper machines under model-predictive control. First, routine operating data is used for system identification in closed loop; second, a one-class support vector machine (SVM) is trained to predict MPM. The iterative identification method alternates between identifying the finite impulse response coefficients of the spatial and temporal models. It converges, and the parameter estimates are asymptotically consistent. Coefficient estimates drawn from normal operation are used to train a one-class SVM, which then detects model-plant mismatch in subsequent routine operation. This approach applies to routine operating data without requiring external excitations. It can also distinguish mismatches in the process model from changes in the noise model. Examples of CD processes on paper machines are provided to verify the effectiveness of both components.