Recursive Method Research Articles

Unbiased recursive least squares identification methods for a class of nonlinear systems with irregularly missing data

SummaryMissing data often occur in industrial processes. In order to solve this problem, an auxiliary model and a particle filter are adopted to estimate the missing outputs, and two unbiased parameter estimation methods are developed for a class of nonlinear systems (e.g., bilinear systems) with irregularly missing data. Firstly, an auxiliary model is constructed to estimate the unknown output, and an auxiliary model‐based multi‐innovation recursive least squares algorithm is presented by expanding the scalar innovation to an innovation vector. Secondly, according to the bias compensation principle, an auxiliary model‐based bias compensation multi‐innovation recursive least squares algorithm is proposed to compensate the bias caused by the colored noise. Thirdly, for further improving the parameter estimation accuracy, the unknown true output is estimated by a particle filter, and a particle filtering‐based bias compensation multi‐innovation recursive least squares algorithm is developed. Finally, a numerical example is selected to validate the effectiveness of the proposed algorithms. The simulation results indicate that the proposed algorithms have good performance in identifying bilinear systems with irregularly missing data.

A composite weighted human learning network and its application for modeling of the intermediate point temperature in USC

Ultra supercritical power plant (USC) is a complex system associated with nonlinearity, uncertainties and multivariable couplings. Generally, it is difficult to build an accurate model to approximate the dynamic behavior of USC. This paper presents a novel composite weighted human learning optimization network (CWHLO) to tackle the above-mentioned problem. Firstly, by fully using of the statistic characteristic of the history operating data, K-means clustering algorithm is applied to partition the raw date, which extremely reduces the operating nonlinearity. Then, an improved real-coded human learning optimization (HLO) is adopted to built linear models in local regions. Different from conventional methods, the advantage of the proposed CWHLO is that the nonlinear model of the object is effectively replaced by a real-time dynamic linear model, which is more suitable for other control methods. Finally, the CWHLO model is compared with the traditional recursive least square method (RLS), and four other meta-heuristic algorithms, to show the advantages in approximating the dynamic behavior of USC.

Joint optimization of mission abort and system structure considering dynamic tasks

Mission abort has recently attracted considerable attention to enhance the safety of critical systems during the primary mission (PM). Most of the existing research focuses on mission abort policies for systems performing a deterministic PM, i.e., operating for a fixed mission duration or completing a specified amount of work. However, in practice, systems are commonly required to perform dynamic tasks. This paper first makes advancements by jointly optimizing condition-based mission abort policies and system structure for the l-out-of-n: G warm standby system, where the dynamic arrival of tasks with a random amount of work is considered. In such systems, some components are initially in active mode, and the remaining warm standby components provide fault tolerance. Two types of mission success criteria are considered and corresponding mission abort policies are proposed based on different decision criteria. Mission reliability (MR) and system survivability (SS) are derived using recursive methods, considering the random switching of the active, idle, and warm standby modes under dynamic arrival of tasks. Mission abort policies and system structure are jointly optimized to balance MR and SS with the objective of minimizing the expected total cost. An example of a multiprocessor system is presented to illustrate the proposed model.

ISpliter: an intelligent and automatic surface mesh generator using neural networks and splitting lines

In this paper, we present a novel surface mesh generation approach that splits B-rep geometry models into isotropic triangular meshes based on neural networks and splitting lines. In the first stage, a recursive method is designed to generate plentiful data to train the neural network model offline. In the second stage, the implemented mesh generator, ISpliter, maps each surface patch into the parameter plane, and then the trained neural network model is applied to select the optimal splitting line to divide the patch into subdomains continuously until they are all triangles. In the third stage, ISpliter remaps the 2D mesh back to the physical space and further optimizes it. Several typical cases are evaluated to compare the mesh quality generated by ISpliter and two baselines, Gmsh and NNW-GridStar. The results show that ISpliter can generate isotropic triangular meshes with high average quality, and the generated meshes are comparable to those generated by the other two software under the same configuration.

Testing Okun's Law Over the Business Cycle: The Case of South Africa

ABSTRACT: Understanding the relationship between output and unemployment (Okun's law) is important for policymakers, particularly during crises. One of the aspects of Okun's law that is commonly examined during crises is the structural stability and behavior of Okun's law over the business cycle. While this issue has been addressed empirically in developed countries, little is known about it in developing countries such as South Africa. Hence, this study examines the structural stability and behavior of Okun's law over the business cycle in South Africa. This study adopts both the difference and gap methods of formalizing Okun's law to implement its objectives. The recursive rolling window method was used to test the structural stability of Okun's law over time. The behavior of Okun's law over the business cycle is examined using the Markov-switching regression method. This study was conducted using the data from 1994Q1 to 2021Q1. The results of this study show that Okun's law varies over the business cycle because the response of unemployment to economic expansion is quite different from that of economic contraction. Okun's law holds during economic expansion but fails to hold during an economic recession. Evidence from this study also reveals that the coefficients of Okun's law vary over time. For instance, in the case of the difference method, Okun's coefficients range from 0.063 to 1.221 in absolute terms. For the gap method, especially when the H-P filtering method is applied, Okun's coefficients range from 0.509 to 9.412 in absolute terms. In summary, the findings from this study suggest that cyclical unemployment responds more significantly to cyclical output growth than the response of changes in unemployment to changes in output. In addition, this study proves that Okun's law is unstable over time in South Africa. These findings imply a high degree of variability between the labor and goods markets in the country. Hence, policies must be designed to harmonize the divergence between the two markets. This will require the implementation of macroeconomic policies such as fiscal and monetary policies. From a fiscal policy perspective, the government should reduce taxes or provide tax holidays to firms. The interest rate should be set low in the case of monetary policy.

Influences of spatial structure of plasma shock wave on electromagnetic wave propagation characteristics

Plasma shock waves represent high-speed, nonlinear motion states of plasma in which the physical parameters of fluid, such as density, temperature, and velocity, vary dramatically within a limited space. These variations make studying electromagnetic wave propagation in plasma shock waves difficult. In this study, we calculate the spatial distribution of the plasma frequency and collision frequency in the plasma shock layer based on the one-dimensional structure of the plasma shock wave. In the process from upstream to downstream of the plasma shock wave, the plasma frequency increases as a function of electron density; downstream, the plasma frequency increases as a function of the free stream Mach number, while the collision frequency decreases first and then increases. We also use the recursive convolution finite-difference time-domain method to calculate the propagation of the electromagnetic wave in the plasma shock layer. The absorption of the plasma shock layer to the electromagnetic wave decreases gradually as a function of the electromagnetic wave frequency. The absorption of the plasma shock layer downstream of the electromagnetic wave is greater than that upstream owing to the larger plasma frequency. In the case of low-Mach numbers, the wavelength of the electromagnetic wave downstream is larger than that upstream. In the case of high-Mach numbers, the thickness of the entire shock layer is much greater than that at low-Mach numbers. Due to the nonlinear effects of electromagnetic processes in plasmas, a modulation phenomenon occurs when electromagnetic waves propagate in the shock layer. In the case of low frequencies, the electromagnetic wave gave rise to a modulation phenomenon that resembled the beat phenomenon. When the electromagnetic wave frequency increased, a modulation phenomenon gradually appeared that resembled the oscillation superimposed on a sine wave. Further increases in the electromagnetic wave frequency led to the gradual disappearance of the modulation phenomenon.

Computation of Transmission Eigenvalues by the Regularized Schur Complement for the Boundary Integral Operators

This paper is devoted to the computation of transmission eigenvalues in the inverse acoustic scattering theory. This problem is first reformulated as a two by two boundary system of boundary integral equations. Next, utilizing the Schur complement technique, we develop a Schur complement operator with regularization to obtain a reduced system of boundary integral equations. The Nystrom discretization is then used to obtain an eigenvalue problem for a matrix. We employ the recursive integral method for the numerical computation of the matrix eigenvalue. Numerical results show that the proposed method is efficient and reduces computational costs.

Receding Horizon D-Optimal Input Design for Identification of Wireless Power Transfer Systems

This paper develops a method of input design for the identification of wireless power transfer (WPT) systems based on receding horizon D-optimization. The dynamic behavior of the system is characterized by a discrete-time, Single-Input, Single-Output (SISO) model, and the associated model parameters are estimated by a recursive least-squares method. To improve the quality of the parameter estimates, the excitation signal of the system is designed based on receding horizon D-optimization. To improve the convergence and efficiency of the input design process, the Sequential Quadratic Programming (SQP) method is employed to solve the receding horizon D-optimization. Numerical and experimental results are presented to demonstrate the effectiveness of the proposed method.

A controllable spin flip and filter in zigzag graphene nanoribbons with triangular defect

Based on the recursive Green-function method together with Landauer-Büttiker formalism, we theoretically study the controllable spin-dependent electron transport properties by using zigzag graphene nanoribbons (ZGNRs) with triangular nanoconstrictions in the presence of the ferromagnetic strips and spatial modulation of Rashba spin-orbit coupling (RSOC). In this structure, the peaks of spin-flip and spin-conserved conductance can be modulated by the strength of RSOC and exchange magnetic field. The distribution of the local density of states (LDOS) around triangular nanoconstrictions is predicted. The spin filtering effect of ZGNRs with triangular and rectangular nanoconstrictions are compared by spin conductance from left (L) to right (R) lead and R to L lead, respectively. The results might be useful in designing a bidirectional spin converter and spin filter with the modulation of the multi-parameter.

Evolving Neuro-Fuzzy Systems-Based Design of Experiments in Process Identification

This paper presents a new design of experiment approach based on an evolving neuro-fuzzy model. The input of the process is proposed by a space-filling method that uses a sequence of step functions to maximise the coverage of the antecedent clusters in the input space of the evolving model, which is called the heuristic step sequence (HSS). When the system reaches a steady state, a new rule is added, existing rules are merged based on the similarity of the consequent transfer functions, and a new optimal excitation is calculated. The output error model structure was chosen because it assumes a noise distribution common to real processes and is suitable for identification, while the step function input signal is one of the most commonly used signals in identification and control. A filtered recursive least squares method is used to identify the consequent parameters of the output error models and the optimisation filter is adapted based on the the confidence interval of the local model. Evaluations were performed on a Hammerstein type model comparing the HSS method with a staircase excitation and on a real plate heat-exchanger pilot plant. The experiments show that the proposed HSS method outperforms the staircase excitation and can be used to identify nonlinear dynamical systems of Hammerstein-Wiener type.

Hyperspectral remote sensing image dimensionality reduction method based on adaptive filtering

With the rapid development of hyperspectral image technology, remote sensing technology has ushered in an innovation in theory and application, and hyperspectral remote sensing images have come into being. However, due to its high data dimensionality, it is difficult for statistical classifiers to work on it, making the technology face development difficulties. Therefore, how to effectively reduce the dimensionality of hyperspectral remote sensing images has gradually become a research hotspot in this field. The study clusters bands by K-means algorithm, and then combines the least mean square algorithm in adaptive filtering and recursive least squares method, and uses this as the basis for band selection. Finally, the dimension reduction effect is verified. The experimental results show that the improved band selection method achieves an overall accuracy of over 80% and 90% in the hyperspectral datasets of Pavia University and Idian Pine respectively, with the Kappa coefficient reaching 0.9. In the overall dimensionality reduction classification of the Indianan data, the accuracy also reaches 90% and can be maintained consistently, indicating that the method has high accuracy and can effectively reduce the dimensionality of hyperspectral remote sensing images.

Semi-analytical study of defective pile-beam systems under transient excitation: Implications for low-strain testing

This paper establishes a theoretical model of defective pile-beam system. The soil around the pile is modeled as a continuous three-dimensional model. The pile is modeled as a Rayleigh-Love rod model that considers its transversal inertia effect. The defect of the pile is simulated by a pile segment with a different radius from the normal pile segment. The impedance function recursion method, along with RSPT and AIFTM, is utilized to derive the impedance of the pile top. The beam is modeled as a Timoshenko beam and the transient excitation is applied at the connection between the pile and the beam. Analytical solution for the dynamic response of the system in the frequency domain is derived, and the discrete Fourier transform is used to obtain a semi-analytical solution for the time-domain response. The velocity at the connection between the pile and beam and the pile body can be measured to assess defects. The semi-analytical solution presented in this paper was validated through comparison with FEM results. Reflection from necked segment is more pronounced at pile-beam connection than at pile body. Both velocity curves at these two points should be measured when beam cross-section height is large to determine pile integrity. The peak value and arrival time of characteristic waves are influenced by burial depth, necking section parameters, beam cross-section height, cantilever length, and soil parameters. These findings can provide guidance for the application of low strain integrity testing in pile-beam systems.

Dynamic force identification in milling based on IRLS using acceleration signals

Measurement of cutting force is important for in-process monitoring, control and optimization. However, it is very difficult to measure the cutting force directly using invasive sensor in industrial scene. This paper develops an infinite-time recursive identification method of cutting force under finite buffer length condition, called the improved recursive least squares (IRLS) method. A cutting force identification framework based on time-domain convolutional theory is introduced, where the cutting force is identified using the acceleration signal easily obtained from industrial scene. Then, the decay characteristic of the impulse response function and the recursive characteristic of the transfer matrix are used to improve the recursive least squares method. The IRLS method overcomes the data saturation problem of the recursive least squares method and can identify the cutting force in infinite time. To reduce the error amplification of the proposed cutting force identification method, the averaged transfer function is used to determine the length of the buffer. In addition, the compensated values of the Kistler 9129AA dynamometer measurement results are used as a reference. Finally, several sets of milling experiments are performed to verify the effectiveness and rationality of the IRLS method.

A multivariate time series graph neural network for district heat load forecasting

Heat load prediction is essential for energy efficiency and carbon reduction in district heating systems. However, heat load is influenced by many factors, such as building characteristics, consumption behavior, and climate, making its prediction challenging. Traditional methods based on physical models are complex and insufficiently accurate, whereas most data-driven statistical methods ignore customer energy consumption behaviors and their correlation, and do not account for the temporal inertia of consumption. This paper proposes a graph ambient intelligence (GAIN) method for heat load prediction, which classifies customers based on their load profiles and uses collaborative attention on temporal graphs to capture their associations and the weather impact on heat loads. GAIN also incorporates recursive and autoregressive methods to model the temporal inertia of consumption. The proposed method is evaluated on four metrics and compared with fifteen baseline methods. The results show that GAIN achieves the lowest daily forecasting errors in terms of RMSE, MAE, and CV-RMSE, with values of 6.972, 4.442, and 0.191, respectively. Besides, the proposed method achieves a maximum reduction of 25%, 29%, and 25% in RMSE, MAE, and CV-RMSE, respectively, compared to other methods when taking meteorological factors into account.

An error covariance correction-adaptive extended Kalman filter based on piecewise forgetting factor recursive least squares method for the state-of-charge estimation of lithium-ion batteries

Accurate state-of-charge (SOC) estimation is essential for fully utilizing the battery performance of electric vehicles. Considering the demand for algorithms with the advantages of simplicity, fewer calculations, good stability, and high accuracy in practical applications, this paper proposes a novel error covariance correction-adaptive extended Kalman filter (ECC-AEKF) for accurate and robust SOC estimation. In this paper, the maximum likelihood function of the probability density function of the error series (conditional on the priori covariance) is calculated by mathematical derivation to obtain a new priori error covariance, which is used to obtain a more appropriate Kalman gain. The ECC-AEKF can minimize the estimation error and reduces the effect of process noise characteristics and inappropriate error covariances on priori estimates. Meanwhile, a piecewise forgetting factor recursive least square (PFFRLS) is presented for model parameter identification. The PFFRLS using error feedback for real-time adaptively adjusts the forgetting degree of data based on the principle of integral separation. Furthermore, a comparative analysis of SOC estimation in PFFRLS-ECC-AEKF and commonly used methods is presented for validation of the performance of the proposed method under different temperatures and operating conditions. The results prove that the PFFRLS-ECC-AEKF achieves higher accuracy with less computation time than other methods.

All plus four point (A)dS graviton function using generalized on-shell recursion relation

This paper presents a calculation of the four gravitons amplitude in (Anti)-de Sitter space, focusing specifically on external gravitons with positive helicity. To achieve this, we employ a generalized recursion method that involves complexifying all external momentum of the graviton function, which results in the factorization of AdS graviton amplitudes and eliminates the need for Feynman-Witten diagrams. Our calculations were conducted in three boundary dimensions, with a particular emphasis on exploring cosmology and aiding the cosmological bootstrap program. To compute the expression, we utilized the three-dimensional spinor helicity formalism. The final expression was obtained by summing over residues of physical poles, and we present both symbolic and numerical results. Additionally, we discuss the advantages and limitations of this approach, and highlight potential opportunities for future research.

Apakah Program Keluarga Harapan Mampu Mengurangi Pekerja Anak di Masa Pandemi COVID-19?

The COVID-19 pandemic can be an economic shock for low-income households. Several studies have shown that child labor is a strategy used by households to cope with economic shocks. This shock will exacerbate the phenomenon of child labor. Social protection is a form of mitigating the impact of the COVID-19 pandemic which is considered able to reduce the number of child labor. Program Keluarga Harapan (PKH) is one of the social protection programs in Indonesia. This study investigates the impact of Program Keluarga Harapan (PKH) on the phenomenon of child labor during the COVID-19 pandemic. To overcome the problem of self-selection, this study applies the variable of interest as an endogenous variable. By using the recursive bivariate probit method, this study concludes that PKH has no impact on reducing the likelihood of a child becoming child labor during the COVID-19 pandemic. To enrich the research, we also investigated the influence of PKH on the phenomenon of child labor in the period before the COVID-19 pandemic and found that PKH had an impact on reducing the likelihood of a child becoming child labor in the period before the COVID-19 pandemic. Several factors that influence the phenomenon of child labor, such as the characteristics of children, the characteristics of the head of the household, household characteristics, and environmental characteristics, have been shown to influence a child's tendency to become child labor, both in both periods (before the pandemic or during the COVID-19 pandemic) and in only one period.

Iterative Kalman filter for biological tissue identification

Dynamic soft tissue identification plays an important role in robotic-assisted minimally invasive surgery to achieve realistic force feedback for precise and safe surgical operations. This article studies a dynamic soft tissue identification method by combination of the Hunt Crossley contact model with an iterated Kalman filter. The dynamic system equation of tool-tissue interaction is constructed by combining mechanical tool's kinematic dynamics with the Hunt Crossley contact dynamics. Upon this, an iterative Kalman filter is developed for online soft tissue identification by integrating the maximum a posteriori principle into the Kalman filtering framework to optimize the posterior state estimate to account for the strong nonlinearity of the Hunt Crossley contact model. Results and comparison analysis demonstrate that the proposed method can effectively identify the Hunt-Crossley model parameters, resulting in improved accuracy compared with the conventional recursive least square method and extended Kalman filter.

Optimization Design and Performance Analysis of a Bionic Knee Joint Based on the Geared Five-Bar Mechanism.

Animal joint motion is a combination of rotation and translational motion, which brings high stability, high energy utilization, and other advantages. At present, the hinge joint is widely used in the legged robot. The simple motion characteristic of the hinge joint rotating around the fixed axis limits the improvement of the robot's motion performance. In this paper, by imitating the knee joint of a kangaroo, we propose a new bionic geared five-bar knee joint mechanism to improve the energy utilization rate of the legged robot and reduce the required driving power. Firstly, based on image processing technology, the trajectory curve of the instantaneous center of rotation (ICR) of the kangaroo knee joint was quickly obtained. Then, the bionic knee joint was designed by the single-degree-of-freedom geared five-bar mechanism and the parameters for each part of the mechanism were optimized. Finally, based on the inverted pendulum model and the Newton-Euler recursive method, the dynamics model of the single leg of the robot in the landing stage was established, and the influence of the designed bionic knee joint and hinge joint on the robot's motion performance was compared and analyzed. The proposed bionic geared five-bar knee joint mechanism can more closely track the given trajectory of the total center of mass motion, has abundant motion characteristics, and can effectively reduce the power demand and energy consumption of the robot knee actuators under the high-speed running and jumping gait.

Machine learning models for prediction of invasion Klebsiella pneumoniae liver abscess syndrome in diabetes mellitus: a singled centered retrospective study

ObjectiveThis study aimed to develop and validate a machine learning algorithm-based model for predicting invasive Klebsiella pneumoniae liver abscess syndrome(IKPLAS) in diabetes mellitus and compare the performance of different models.MethodsThe clinical signs and data on the admission of 213 diabetic patients with Klebsiella pneumoniae liver abscesses were collected as variables. The optimal feature variables were screened out, and then Artificial Neural Network, Support Vector Machine, Logistic Regression, Random Forest, K-Nearest Neighbor, Decision Tree, and XGBoost models were established. Finally, the model's prediction performance was evaluated by the ROC curve, sensitivity (recall), specificity, accuracy, precision, F1-score, Average Precision, calibration curve, and DCA curve.ResultsFour features of hemoglobin, platelet, D-dimer, and SOFA score were screened by the recursive elimination method, and seven prediction models were established based on these variables. The AUC (0.969), F1-Score(0.737), Sensitivity(0.875) and AP(0.890) of the SVM model were the highest among the seven models. The KNN model showed the highest specificity (1.000). Except that the XGB and DT models over-estimates the occurrence of IKPLAS risk, the other models' calibration curves are a good fit with the actual observed results. Decision Curve Analysis showed that when the risk threshold was between 0.4 and 0.8, the net rate of intervention of the SVM model was significantly higher than that of other models. In the feature importance ranking, the SOFA score impacted the model significantly.ConclusionAn effective prediction model of invasion Klebsiella pneumoniae liver abscess syndrome in diabetes mellitus could be established by a machine learning algorithm, which had potential application value.