Nonlinear Support Research Articles

Generalization of linear and non-linear support vector machine in multiple fields: a review

Support vector machines (SVMs) are a set of related supervised learning methods used for classification and regression. They belong to a family of generalized linear classifiers. In other terms, SVM is a classification and regression prediction tool that uses machine learning theory to maximize predictive accuracy. In this article, the discussion about linear and non-linear SVM classifiers with their functions and parameters is investigated. Due to the equality type of constraints in the formulation, the solution follows from solving a set of linear equations. Besides this, if the under-consideration problem is in the form of a non-linear case, then the problem must convert into linear separable form with the help of kernel trick and solve it according to the methods. Some important algorithms related to sentimental work are also presented in this paper. Generalization of the formulation of linear and non-linear SVMs is also open in this article. In the final section of this paper, the different modified sections of SVM are discussed which are modified by different research for different purposes.

Generalization of linear and non-linear support vector machine in multiple fields: a review

Support vector machines (SVMs) are a set of related supervised learning methods used for classification and regression. They belong to a family of generalized linear classifiers. In other terms, SVM is a classification and regression prediction tool that uses machine learning theory to maximize predictive accuracy. In this article, the discussion about linear and non-linear SVM classifiers with their functions and parameters is investigated. Due to the equality type of constraints in the formulation, the solution follows from solving a set of linear equations. Besides this, if the under-consideration problem is in the form of a non-linear case, then the problem must convert into linear separable form with the help of kernel trick and solve it according to the methods. Some important algorithms related to sentimental work are also presented in this paper. Generalization of the formulation of linear and non-linear SVMs is also open in this article. In the final section of this paper, the different modified sections of SVM are discussed which are modified by different research for different purposes.

On the quality of stability and bifurcation sets in rotors with permanent shaft bow on nonlinear supports

The paper investigates the dynamics of simple and complex geometry rotors of various configurations with permanent bow, mounted on nonlinear bearings, in regards to nonlinear stability, quality of motion, and bifurcation set. Permanent shaft bow is a fault existing in most slender rotors in thermal machines, e.g. gas turbines, steam turbines, and turbopumps, and other applications of rotating machinery like transmission shafts and geared shafts. In this work, finite element models of multi-segment rotors are implemented, including permanent bow, and a generic rotor system of a uniform rotor with three masses is studied first on its stability characteristics for various bow values, and for different slenderness. A realistic rotor of complex geometry is used as the second application. The rotors are mounted on two plain short bearings to include the nonlinearity of impedance forces with analytical formulas; this is the unique source of nonlinearity in the systems.The rotor-bearing systems are found to produce the well-known oil whirl instability through a Hopf, or a secondary Hopf bifurcation, depending the unbalance grade, in speeds higher than the first critical speed, when bow is not included. Supercritical and subcritical Neimark-Sacker and period doubling bifurcations are produced in speeds lower than critical speed when bow is included above some threshold values. Stable and unstable limit cycles are generated both for unbalanced or balanced rotors with bow, and their periodicity is investigated. A bifurcation set of the rotor-bearing systems is evaluated utilizing the pseudo-arc length continuation method with embedded collocation method, for the evaluation of periodic limit cycles (where these appear). Neimark-Sacker bifurcations are found to collide and vanish as the bow magnitude changes from high to low values. Bow phase angle with respect to unbalance force is found not to influence the respective bifurcation sets.

Comparative Study on Seismic Response of Regular and Plan Irregular Base isolated structures

Earthquake is one of most destructing threat which cause severe damage to life and property. Recently, due to increasing demand and need for the present generations has made the designers and engineers to plan for structures having irregularities. It is impossible to avoid designing uneven buildings in reality. It is important to research how this irregular structure behaves in earthquake-prone areas, especially for structures with base isolation. When compared to regular structures with base isolation, structures with plan irregularity react differently. When a reinforced concrete building with varied plan configurations, such as an H, U, or L shape, is subjected to seismic loading, the impact of plan irregularity is explored. In the current study, a regular structure is also considered to compare the responses of the isolated irregular base structure. Building response spectrum analysis for seismic analysis was done in seismic zone III using IS 1893-2002 code [5]. The structural analysis software tool ETABS 18 was used to conduct the analysis. Among different types of isolators present, in the present study LRB isolator and friction isolators was used as a Nonlinear Support. The dynamic response of the structure such as base shear, time period and story displacement are used to compare the regular and irregular structures. The effectiveness of the base isolators in the structures in earthquake prone area is evaluated in the present study.

Lattice-Based Homomorphic Encryption For Privacy-Preserving Smart Meter Data Analytics

Abstract Privacy-preserving smart meter data collection and analysis are critical for optimizing smart grid environments without compromising privacy. Using homomorphic encryption techniques, smart meters can encrypt collected data to ensure confidentiality, and other untrusted nodes can further compute over the encrypted data without having to recover the underlying plaintext. As an illustrative example, this approach can be useful to compute the monthly electricity consumption without violating consumer privacy by collecting fine-granular data through small increments of time. Toward that end, we propose an architecture for privacy-preserving smart meter data collection, aggregation and analysis based on lattice-based homomorphic encryption. Furthermore, we compare the proposed method with the Paillier and Boneh–Goh–Nissim (BGN) cryptosystems, which are popular alternatives for homomorphic encryption in smart grids. We consider different services with different requirements in terms of multiplicative depth, e.g. billing, variance and nonlinear support vector machine classification. Accordingly, we measure and show the practical overhead of using the proposed homomorphic encryption method in terms of communication traffic (ciphertext size) and latency. Our results show that lattice-based homomorphic encryption is more efficient than Paillier and BGN for both multiplication and addition operations while offering more flexibility in terms of the computation that can be evaluated homomorphically.

A novel PAPR reduction scheme based on nonlinear real-valued support vector regression for CO-OFDM systems

Coherent optical orthogonal frequency division multiplexing (CO-OFDM) systems suffer from the high peak-to-average ratio (PAPR) of orthogonal frequency division multiplexing (OFDM) signals, which leads to the back-off of laser power and fiber nonlinear effects, thereby deteriorating the bit error rate (BER) performance and limiting the transmission distance. Iterative clipping and filtering (ICF) scheme, as a direct, simple and effective PAPR reduction scheme that only processes signals at the transmitter, but requires multiple fast Fourier transform (FFT)/inverse fast Fourier transform (IFFT), resulting in high computational complexity. We propose a novel scheme using the machine learning (ML) technique, which is based on nonlinear real-valued support vector regression (NRSVR) and trained on ICF scheme. The simulation results show that compared with ICF scheme, the proposed scheme has a similar PAPR0 for 10−4 complementary cumulative distribution function (CCDF), an additional 15 km of single mode fiber (SMF) transmission for 16 dB optical signal-to-noise ratio (OSNR) and 10−3BER, and a significant time complexity reduction. Compared with other ML-based schemes, the proposed scheme also achieves better performance and lower computational complexity.

Effect of Support Stiffness Nonlinearity on the Low-Frequency Vibro-Acoustic Characteristics for a Mechanical Equipment—Floating Raft—Underwater Cylindrical Shell Coupled System

The low-frequency vibro-acoustic characteristics of a mechanical equipment—floating raft—cylindrical shell—underwater acoustic field coupled system with nonlinear supports are studied in this paper. Firstly, the state space equations were established by a modal superposition theory for the coupled system, and a modal parameter identification method was deduced and verified for the cylindrical shell—underwater acoustic field coupled subsystem. On this basis, the formulas were derived for transmitted power flow in the coupled system, and the nonlinear stiffness constitutive relation of the vibration isolation supports was expressed by softening and hardening characteristics. Finally, dynamic simulations were carried out by the Runge—Kutta method to analyze the effect of nonlinear stiffness characteristic parameters on the low-frequency vibration modes and vibro-acoustic transfer characteristics in the coupled system. The research shows that a superharmonic phenomenon is common in the steady vibration mode of the coupled system with a nonlinear softening (or hardening) stiffness characteristic under harmonic excitation. The stronger the softening (or hardening) stiffness characteristic is, the more complex the vibration form is, and the smaller (or larger) the low-frequency vibro-acoustic transfer level in resonance regions is.

Nonlinear superharmonic resonance and chaotic motion of a moving web under an intermediate nonlinear support

Roll-to-roll manufacturing is the primary means of flexible electronics to facilitate scale-up production, and the web as the printed substrate is guided by the intermediate roller in this process. This paper aims to investigate the nonlinear superharmonic resonance and chaotic motion of a moving web under an intermediate nonlinear support. This support is modeled as a nonlinear elastic spring with linear and nonlinear stiffness, and the equation of motion of the web attached with nonlinear elastic support is derived from the D’Alembert principle and von Karman theory. The resulting equation is reduced to the two-degree ordinary differential equations via the Galerkin truncation, the superharmonic resonance responses of the web system are obtained by the multi-scale method, and the bifurcation analysis and stability are analyzed by the Runge-Kutta numerical method. The results indicate that the linear and nonlinear stiffnesses have a significant effect on amplitude-frequency responses and chaotic motion. This study provides an exploration of vibration behaviors of the web in flexible manufacturing, thereby laying the foundation for the improvement of fabrication productivity.

The Influence of Speed Ratio on the Nonlinear Dynamics of a Magnetic Suspended Dual-Rotor System with a Fixed-Point Rubbing

Magnetic suspended dual-rotor systems (MSDS) provide the potential to significantly improve the performance of aero-engines by eliminating the wear and lubrication system, and solve vibration control issues effectively. However, the nonlinear dynamics of MSDS with rubbing is rarely investigated. In this work, the nonlinear support characteristics of active magnetic bearings (AMBs) are described by the equivalent magnetic circuit method, the impact force is characterized by the Lankarani–Nikravesh model, and the nonlinear dynamic model is established using the finite element method. On this basis, the influence of speed ratio on the nonlinear dynamics is investigated. Simulation results show that the fundamental sub-synchronous vibration of period n is the dominant motion of MSDS, where n is determined by the speed ratio. The frequency components of sub-synchronous vibrations of period k are integer multiples of the minimum dimensionless frequency component 1/k, where k is a positive integral multiple of n. Quasi-periodic and chaotic vibrations are more likely to occur near critical speeds, and their main frequency components can be expressed as a variety of combined frequency components of the rotating frequency difference and its fractional frequency. To reduce the severity of fluctuating stresses stemming from complicated non-synchronous vibrations, speed ratios, corresponding to smaller n and AMB control parameters attenuating vibration amplitude or avoiding critical speeds, are suggested.

Chemical space navigation by machine learning models for discovering selective MAO-B enzyme inhibitors for Parkinson’s disease

Monoamine Oxidase-B (MAO-B) is a key neuroprotective target that breaks neurotransmitters such as dopamine and releases highly reactive free radicals as the by-product. Its over-expression in the brain observed due to ageing and neurodegenerative diseases contributes to worsening neuronal degeneration. Being the primary enzyme for dopamine metabolism in the substantia nigra of the brain and due to the lack of efficient drug candidates, MAO-B selective, reversible inhibition is hot topic of research in Parkinson’s disease (PD). This study developed machine learning (ML) models that predict the activity of experimentally tested indole and indazole derivatives against MAO-B using linear genetic function approximation (GFA) and two non-linear support vector machine (SVM) and artificial neural network (ANN) techniques. ANN model with an R2 of 0.9704 for the training dataset, q2of 0.9436 for cross-validation and r2of 0.9025 for the test dataset were identified as the best-performing ML model with the seven significant molecular descriptors CATS2D_04_DA, CATS2D_05_DA, CATS3D_06_LL, Mor04u, Mor25m, P_VSA_v_2 and nO. The robust ML model was then employed to design novel MAO-B inhibitors with similar core scaffolds and their biological activity prediction. ANN model was further employed in the virtual screening of 4356 molecules from the ChEMBL database. Applicability domain analysis and pharmacokinetic and toxicity profiles predicted three newly designed molecules (22 N, 23 N and 24 N) and two virtually screened best ChEMBL molecules as potential drug candidates using the ANN ML model. Molecular docking studies of the best-identified compounds were performed to understand the molecular mechanism of interactions having high binding energy and selectivity with the MAO-B enzyme. The current study shortlisted 5 potential lead compounds as potent and selective MAO-B inhibitors, which could further be carried forward for in vitro and in vivo studies to discover small molecules against neurodegenerative disease.

Effect of linear and nonlinear stiffness support on the dynamic response of high-speed railway bridges

In this paper, the effect of linear and nonlinear support vertical stiffness of elastic support on the dynamic response of the railway bridge is investigated. The bridge is modeled by a Euler-Bernoulli beam supported by linear and nonlinear supports stiffness. A constant moving load model HSLM-A is utilized to simulate the moving real train, and based on Hamilton’s principle, the governing partial differential equation of motion of the studied system is derived. A method to deal with the nonlinear boundary conditions is proposed, the fourth-order Runge-Kutta method in combination with the Galerkin method is used to obtain the dynamic response of the bridge. The finite difference method (FDM) is used to validate the proposed solution. The obtained results have shown that the dynamics of the system are governed essentially by a Duffing-like oscillator and proved that the elastic bearings and the cubic nonlinearities of supports have a significant effect on the global behavior of the railway bridge particularly at resonance.

Deep Learning Radiomics Nomogram Based on Multiphase Computed Tomography for Predicting Axillary Lymph Node Metastasis in Breast Cancer.

This study aims to develop and validate a deep learning radiomics nomogram (DLRN) for prediction of axillary lymph node metastasis (ALNM) in breast cancer patients. We retrospectively enrolled 196 patients with non-specific invasive breast cancer confirmed by pathology, radiomics and deep learning features were extracted from unenhanced and biphasic (arterial and venous phase) contrast-enhanced CT, and the non-linear support vector machine was used to construct the radiomics signature and the deep learning signature, respectively. Next, a DLRN was developed with independent predictors and evaluated the performance of models in terms of discrimination and clinical utility. Multivariate logistic regression analysis showed that the radiomics signature, deep learning signature, and clinical n stage were independent predictors. The DLRN accurately predicted ALNM and yielded an area under the receiver operator characteristic curve of 0.893 (95% confidence interval, 0.814-0.972) in the validation set, with good calibration. Decision curve analysis confirmed that the DLRN had higher clinical utility than other predictors. The DLRN had good predictive value for ALNM in breast cancer patients and provide valuable information for individual treatment.

A novel hybrid method for feature selection based on gender analysis for early Parkinson's disease diagnosis using speech analysis

Parkinson's disease (PD) is one of the neurologic disorders for older people that, because of similar symptoms to other neurologic diseases, is difficult to diagnose, mainly in the early stages. Recently, neurologists are trying non-medical methods based on speech processing for early diagnosis of PD. Since some speech characteristics are different between women and men, we analyze them separately in this research and propose a hybrid method to classify the PD and healthy samples. The proposed method's main structure relies on the features' scores, which are based on a projection on a two-dimensional plane. In the first step, several features inherently different between women and men (healthy and PD) are removed from the features. Then, statistical-Based Feature Score (SBFS) and Classification-Based Feature Score (CBFS) are introduced in the proposed method and used as the dimensions of a two-dimensional hyperplane to rank the features. To increase the stability of the model in speech feature selection, resampling of the dataset is considered in the methodology. Different classifiers (linear and nonlinear support vector machine (LSVM and NSVM respectively), K-nearest neighborhood (KNN), naïve Bayesian (NB), and random forest (RF)) are applied for sample classification. Finally, two sets of features are separately introduced for both groups of men and women. We have achieved 86% and 84% accuracy rates for men and women groups with 14 and 12 features, respectively. The results show that the number of selected features is less than the previous work that introduced and used this dataset. The results also present that although the chosen features are different for men and women, most of the selected features are in similar categories. Unlike the previous similar study where the validation method was based on Leave-one-out, in this study, the results are validated based on an independent set of the used dataset to analyze the stability of the selected features.

Emergency shelter materials under a complex non-linear diophantine fuzzy decision support system

AbstractThe distribution of emergency shelter materials in emergency cases around the world is a hard task, the goal of this research is to offer a Complex Non-linear Diophantine Fuzzy (C-NLDF) decision-making model for earthquake shelter construction. Essentially, the article is divided into three sections to acquire acceptable and precise measures in emergency decision-making situations. First, we present the Complex Non-Linear Diophantine Fuzzy Set (CN-LDFS), a new generalization of the complex linear Diophantine fuzzy set (CLDFS) and q-linear Diophantine fuzzy set (q-LDFS), as well as explore its key aspects. Furthermore, aggregation operators are useful for aggregating uncertainty in decision-making issues. As a result, algebraic norms for CN-LDFSs are produced based on certain operational laws. In the second section of the work, we offer a series of averaging and geometric aggregation operators under CN-LDFS that are based on defined operating laws. In the final section of the work, under complex Non-linear Diophantine fuzzy information, the ranking algorithms based on suggested aggregation operators are present to address the case study regarding emergency situation of earthquakes. In comparison section, results of existing and proposed operators explore the effectiveness of proposed methodologies and provide accurate emergency measures to address the global uncertainty about the construction of emergency shelters in earthquakes.

Effect of Material Non-linearities on the Transient Dynamic Behavior of the Beni-Bahdel Dam in the Presence of the Dam-foundation Interaction

The current paper's goal is to investigate how dynamic interaction phenomena and material nonlinearities affect the dynamic behavior of a concrete gravity multi-arch dam in terms of maximum displacements, natural stresses, principal stresses, and Von Mises stresses when subjected to seismic excitation. Utilizing ANSYS APDL software, linear and nonlinear transient analysis using the finite element method was employed to analyze the interaction between the dam and its rock foundation. As a case study for this work, the multi-arch Beni-Bahdel dam was selected, and the seismic excitation used data from a simulated earthquake. The direct method is employed to model the interaction between the multi arch dam and the rock foundation using two approaches; the fixed base approach and the mass foundation approach. Six finite element models were performed using ANSYS code, “linear dam-fixed support”, “nonlinear dam-fixed support”, “linear dam-linear rock foundation” “nonlinear dam-linear rock foundation”, “nonlinear dam-nonlinear rock foundation” and “linear dam-nonlinear rock foundation”. The bilinear kinematic hardening model is employed to represent the nonlinearity of both dam body and rock foundation. The results obtained are compared to understand the effect of both material nonlinearities and interaction phenomenon on the dynamic behavior of the studied dam. In contrast to the nonlinearity of the rock foundation material, it is concluded that the nonlinearity of the concrete dam material has a significant impact on the behavior of the system.

Effect of internal excitation induced by non-uniform contact of roller bearings on nonlinear vibration and stability of rotor system under combined loads

The misalignment of roller bearing rings will results in moment load, together with the unbalance and external loads, forming the combined dynamic loads. The complex combined load can cause non-uniform contact of rollers and rings, which reduces the bearing’s life and significantly affect the vibration of rotor systems. This paper mainly studies the roller bearing internal contact performances under combined loads and the influence of internal excitation on the nonlinear vibration of rotor systems. First, the “slices” method is used to establish the mechanical model of roller bearings, which considers the radial clearance and nonlinear Hertz contact. The obtained time-varying bearing nonlinear support reaction is introduced into the rotor concentrated mass model, and a roller bearing–rotor system dynamic model is proposed. Further, the system’s complex nonlinear vibrations, such as bifurcation and chaos, are analyzed by spectrum, orbits, bifurcation, and Poincare map. The roller bearing contact performances and stability under the influence of ring tilt angle, rotational speed, and unbalance, are analyzed by contact load statistical features. Finally, based on a rolling bearing–rotor test rig, the vibration experiment considering the combined load was performed, and the experiment results are compared with simulation results to verify the proposed model and conclusions.

#6225 MICROPLASTICS: FIRST EXTENDED PROTEOMIC ANALYSIS ON KIDNEY TUBULAR CELLS

Abstract Background and Aims Several studies have shown the ubiquitous presence of microplastics (plastic fragments between 5 mm and 1 μm in diameter) and nanoplastics (<1 μm) in the environment and their toxicity. Furthermore, microplastics (MPs) absorb a lot of environmental pollutants, such as bisphenol A (BPA), and release them into tissues increasing their toxicity; this is the so called “Trojan Horse” effect. Recent studies have also proved their presence in human blood and human tissues of healthy people, such as placenta and lung, and in cirrhotic liver. Most frequent biological effects of MPs are inflammation, oxidative stress and alteration of metabolic pathways. In this study proteomic analysis was performed to evaluate the toxicity of polyethylene (PE) and bisphenol-A (BPA) MPs on renal tubular cells (HK2) in vitro. This is the first extended proteomic study on human cells. Method HK-2 cultures were exposed for 5-24-48 hours to BPA, PE Microspheres (PE-MP) and MP combined with BPA. Then it was performed a proteomic analysis by mass spectrometry (MS). MS data were obtain by Orbitrap Fusion Tribrid mass spectrometer (ThermoScientific). Analysis of data were performed using unsupervised hierarchical clustering using multidimensional scaling, non-linear support vector machine (SVM) learning, and partial least squares discriminant analysis. In SVM learning, a fourfold cross-validation approach was applied to estimate the prediction and classification accuracy. Results The proteomic analysis showed a clear differentiation of the HK2 proteome based on conditioning and identified a “core” of proteins, significant at ANOVA and above the 95th percentile for “fold increase” and significant at T-test compared with controls, highly discriminatory between groups. Finally, among these, a final set of 6 proteins was selected to be validated for distinguishing features: Nephronectin, GDF15, Vasorin, IGFBP7, Midkine, Tissue factor-F3. Nephronectin is a structural membrane's protein involved in cellular adhesion. GDF15, tissue factor F3 and midkine are markers of stress conditions, including inflammation and oxidative stress. IGFBP7 is a biomarker of acute kidney damage. Vasorin is a transmembrane glycoprotein that protects against apoptosis and fibrosis. Conclusion MP and BPA significantly modify the protein expression in renal tubular cells. These findings highlight the urgent need for additional research into the toxic effects of plastic debris on human kidneys and the eventual link to kidney diseases.

Design & Analysis of Beam-Column Structures with Nonlinear Supports &Computer Program

Design & Analysis of Beam-Column Structures with Nonlinear Supports &Computer Program

A novel hybrid feature extraction and ensemble C3D classification for anomaly detection in surveillance videos

Anomaly detection in several deep learning frameworks are recently presented on real-time video databases as a challenging task. However, these frameworks have high false positive rate (FPR) and error rate due to various backgrounds, motion appearance and semantic high-level and low-level features for anomaly detection through action classification. Also, extraction of features and classification are the major problems in traditional convolution neural network (CNN) on real-time video databases. The proposed work is a novel action classification framework which is designed and implemented on large video databases with high true positive rate (TPR) and error rate. In this framework, Kalman based incremental principal component analysis (IPCA) feature extraction method; C3D and non-linear support vector machine (SVM) classifier are used to improve the action prediction (anomaly detection) on the large real-time video databases. The proposed frame work shown new results of high computation performance than the traditional deep learning frameworks for action classification.

Development and validation of explainable machine-learning models for carotid atherosclerosis early screening

BackgroundCarotid atherosclerosis (CAS), an important factor in the development of stroke, is a major public health concern. The aim of this study was to establish and validate machine learning (ML) models for early screening of CAS using routine health check-up indicators in northeast China.MethodsA total of 69,601 health check-up records from the health examination center of the First Hospital of China Medical University (Shenyang, China) were collected between 2018 and 2019. For the 2019 records, 80% were assigned to the training set and 20% to the testing set. The 2018 records were used as the external validation dataset. Ten ML algorithms, including decision tree (DT), K-nearest neighbors (KNN), logistic regression (LR), naive Bayes (NB), random forest (RF), multiplayer perceptron (MLP), extreme gradient boosting machine (XGB), gradient boosting decision tree (GBDT), linear support vector machine (SVM-linear), and non-linear support vector machine (SVM-nonlinear), were used to construct CAS screening models. The area under the receiver operating characteristic curve (auROC) and precision-recall curve (auPR) were used as measures of model performance. The SHapley Additive exPlanations (SHAP) method was used to demonstrate the interpretability of the optimal model.ResultsA total of 6315 records of patients undergoing carotid ultrasonography were collected; of these, 1632, 407, and 1141 patients were diagnosed with CAS in the training, internal validation, and external validation datasets, respectively. The GBDT model achieved the highest performance metrics with auROC of 0.860 (95% CI 0.839–0.880) in the internal validation dataset and 0.851 (95% CI 0.837–0.863) in the external validation dataset. Individuals with diabetes or those over 65 years of age showed low negative predictive value. In the interpretability analysis, age was the most important factor influencing the performance of the GBDT model, followed by sex and non-high-density lipoprotein cholesterol.ConclusionsThe ML models developed could provide good performance for CAS identification using routine health check-up indicators and could hopefully be applied in scenarios without ethnic and geographic heterogeneity for CAS prevention.