rss_2.0Sciendo RSS Feed for International Journal of Applied Mathematics and Computer Sciencehttps://sciendo.com/journal/AMCShttps://www.sciendo.comhttps://sciendo-parsed.s3.eu-central-1.amazonaws.com/64709a5a71e4585e08aa1167/cover-image.jpghttps://sciendo.com/journal/AMCS140216https://sciendo.com/article/10.61822/amcs-2024-0028<abstract> <title style='display:none'>Abstract</title> <p>The article presents a comprehensive quantitative comparison of four analytical models that, in different ways, describe the flow process in transmission pipelines necessary in the task of detecting and isolating leaks. First, the analyzed models are briefly presented. Then, a novel model comparison framework is introduced along with a methodology for generating data and assessing diagnostic effectiveness. The study presents basic assumptions, experimental conditions and scenarios considered. Finally, the quality of the model-based diagnostic estimators is assessed, focusing on their bias, standard deviation, and computational complexity. Here, several optimality criteria are used as detailed indicators of the quality and performance of the estimators in a multi-criteria Pareto optimality assessment.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00282024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0024<abstract> <title style='display:none'>Abstract</title> <p>New approaches to transformations of linear continuous-time systems to their positive asymptotically stable canonical controllable (observable) forms are proposed. It is shown that, if the system matrix is nonsingular, then the desired transformation matrix can be chosen in block diagonal form. Procedures for the computation of the transformation matrices are proposed and illustrated with simple numerical examples.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00242024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0025<abstract> <title style='display:none'>Abstract</title> <p>An innovative control strategy addressing the complexities of discrete interconnected nonlinear Hammerstein subsystems is presented. The approach combines decentralized sliding mode control (DSMC) with an event-triggered mechanism (ETM) to efficiently manage complex systems characterized by discrete elements, nonlinear behavior, and interconnections. The event-triggered sliding mode control (ETSMC) framework offers a distributed control solution that utilizes the robustness and disturbance tolerance of sliding mode control while optimizing resource usage and network communication through an event-triggered mechanism. A comprehensive analysis of stability and robustness ensures that the proposed control strategy stabilizes the system and achieves its design objectives, even in the presence of uncertainties or disturbances. The effectiveness of the approach is demonstrated through two simulation examples.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00252024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0027<abstract> <title style='display:none'>Abstract</title> <p>The paper discusses the inventory management problem with a single product stored in two warehouses, where each has its unique suppliers with certain lead times. Moreover, one of the warehouses may act as a backup supplier for the other. In other words, product exchange between two different warehouses within one company is allowed. The first warehouse operates under an <italic>a priori</italic> known time-variant contractual demand and a bounded random one. Its secondary goal is to accumulate emergency stock that can be delivered to the second warehouse within one time period. For this warehouse we use a desired trajectory generator to shape the required stock level and then utilize a trajectory following control law. The demand in the second warehouse is unknown but bounded, and its suppliers have limited delivery capacity. The challenge is to fulfill the customers’ needs, although they might exceed the order limit. Therefore, occasional backup supplies from the first warehouse are necessary. For the control of the second warehouse, a simple sliding mode (SM) scheme is applied. The paper proves that, with appropriate compensation of the emergency deliveries in the first warehouse, our proposed control scheme ensures full demand satisfaction in both warehouses despite the second one’s control limit.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00272024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0035<abstract> <title style='display:none'>Abstract</title> <p>Nowadays, applying additive manufacturing (AM) technologies into a supply chain (SC) permits realization of the so-called “demand chains” and transformation of conventional production to mass customization. However, integration of AM technologies within an SC indicates the need to support managers’ decision about such an investment. Therefore, this work develops a Petri net-based decision support model that determines the changes in an SC by adopting AM and improving customer-perceived value (CPV), based on a case study regarding a real-life metal production process. The basis for building such a model is the supply chain operation reference model (SCOR), focusing on CPV, due to the need for redesigning the SC starting from the customer instead of the company. To achieve the research objective, this work introduces a novel verification methodology for a Petri net-based decision model. The research results show that applying the developed model, which is based on the selected characteristics of the production process and parameters describing the potential integration of AM within the SC, allows managers to perceive a scenario in the form of graphical models about positive or negative impacts of introducing AM into the SC. The managers find the Petri net-based decision support model presented in this paper a beneficial tool to support the implementation of changes in an SC and show the potential increase in customer satisfaction thanks to the integration of AM within an SC.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00352024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0026<abstract> <title style='display:none'>Abstract</title> <p>An observer-based hybrid event-triggered sliding mode fault-tolerant consistent control strategy is proposed for actuator faults in nonlinear second-order leader–follower multi-agent systems. A fault observer is designed to obtain the velocity and additive fault of the agents at the current moment. In order to save network resources and avoid the proliferation of actuator fault information, a hybrid event-triggered mechanism is given based on the actuator fault output from the fault observer. Then, a sliding mode fault-tolerant control strategy is investigated based on the speed and hybrid event-triggered mechanism of the fault observer output and combined with a linear sliding mode surface. As a result, the multi-agent system can still realize state consistency when there is an actuator fault. Conditions under which the consistent error of the multi-agent system is bounded are given. Finally, the effectiveness of the designed fault observer, sliding mode fault-tolerant controller, and hybrid event-triggered mechanism is verified by simulation in a leader–follower multi-agent system connected by a directed graph.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00262024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0031<abstract> <title style='display:none'>Abstract</title> <p>Besides clustering and classification, detection of atypical elements (outliers, rare elements) is one of the most fundamental problems in contemporary data analysis. However, contrary to clustering and classification, an atypical element detection task does not possess any natural quality (performance) index. The subject of the research presented here is the creation of one. It will enable not only evaluation of the results of a procedure for atypical element detection, but also optimization of its parameters or other quantities. The investigated quality index works particularly well with frequency types of such procedures, especially in the presence of substantial noise. Using a nonparametric approach in the design of this index practically frees the proposed method from the distribution in the dataset under examination. It may also be successfully applied to multimodal and multidimensional cases.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00312024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0030<abstract> <title style='display:none'>Abstract</title> <p>Recent studies show that deep neural networks (DNNs) are extremely vulnerable to elaborately designed adversarial examples. Adversarial training, which uses adversarial examples as training data, has been proven to be one of the most effective methods of defense against adversarial attacks. However, most existing adversarial training methods use adversarial examples relying on first-order gradients, which perform poorly against second-order adversarial attacks and make it difficult to further improve the robustness of the model. In contrast to first-order gradients, second-order gradients provide a more accurate approximation of the loss landscape relative to natural examples. Therefore, our work focuses on constructing second-order adversarial examples and utilizing them for training DNNs. However, second-order optimization involves computing the Hessian inverse, which typically consumes considerable time. To address this issue, we propose an approximation method that transforms the problem into optimization within the Krylov subspace. Compared with the Euclidean space, the Krylov subspace method typically does not require storing the entire matrix. It only needs to store vectors and intermediate results, avoiding explicitly calculating the complete Hessian matrix. We approximate the adversarial direction by a linear combination of Hessian-vector products in the Krylov subspace to reduce the computation cost. Because of the non-symmetrical Hessian matrix, we use the generalized minimum residual to search for an approximate polynomial solution of the matrix. Our method efficiently reduces computational complexity and accelerates the training process. Extensive experiments conducted on the MNIST, CIFAR-10, and ImageNet-100 datasets demonstrate that our adversarial learning using second-order adversarial samples outperforms other first-order methods, leading to improved model robustness against various attacks.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00302024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0034<abstract> <title style='display:none'>Abstract</title> <p>An electrocardiogram (ECG) is an essential medical tool for analyzing the functioning of the heart. An arrhythmia is a deviation in the shape of the ECG signal from the normal sinus rhythm. Long-term arrhythmias are the primary sources of cardiac disorders. Shockable arrhythmias, a type of life-threatening arrhythmia in cardiac patients, are characterized by disorganized or chaotic electrical activity in the heart’s lower chambers (ventricles), disrupting blood flow throughout the body. This condition may lead to sudden cardiac arrest in most patients. Therefore, detecting and classifying shockable arrhythmias is crucial for prompt defibrillation. In this work, various machine and deep learning algorithms from the literature are analyzed and summarized, which is helpful in automatic classification of shockable arrhythmias. Additionally, the advantages of these methods are compared with existing traditional unsupervised methods. The importance of digital signal processing techniques based on feature extraction, feature selection, and optimization is also discussed at various stages. Finally, available databases, the performance of automated algorithms, limitations, and the scope for future research are analyzed. This review encourages researchers’ interest in this challenging topic and provides a broad overview of its latest developments.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00342024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0032<abstract> <title style='display:none'>Abstract</title> <p>Visual question answering (VQA) is a pivotal topic at the intersection of computer vision and natural language processing. This paper addresses the challenges of linguistic bias and bias fusion within invalid regions encountered in existing VQA models due to insufficient representation of multi-modal features. To overcome those issues, we propose a multi-feature enhancement scheme. This scheme involves the fusion of one or more features with the original ones, incorporating discrete cosine transform (DCT) features into the counterfactual reasoning framework. This approach harnesses finegrained information and spatial relationships within images and questions, enabling a more refined understanding of the indirect relationship between images and questions. Consequently, it effectively mitigates linguistic bias and bias fusion within invalid regions in the model. Extensive experiments are conducted on multiple datasets, including VQA2 and VQA-CP2, employing various baseline models and fusion techniques, resulting in promising and robust performance.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00322024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0029<abstract> <title style='display:none'>Abstract</title> <p>The paper deals with the estimation of sensor faults for dynamic systems as well as the assessment of the uncertainty of the resulting estimates. For that purpose, it is assumed that the external disturbances are bounded within an ellipsoidal domain. This allows considering both stochastic and deterministic process and measurement uncertainties. Under such an assumption, a fault diagnosis scheme is developed with a prescribed convergence rate and accuracy. To achieve fault estimation, a conversion into an equivalent descriptor system is utilized. The paper provides a full stability and convergence analysis of the estimator including observability analysis. As a result, a set of complementary fault uncertainty intervals is obtained, which are minimized in such a way as to obtain a minimum detectable sensor fault. The final part of the paper exhibits a numerical example concerning fault estimation of a multi-tank system. The obtained results clearly confirm the performance of the proposed estimator expressed in the minimum detectable fault intervals.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00292024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0033<abstract> <title style='display:none'>Abstract</title> <p>The paper proposes a non-iterative training algorithm for a power efficient SNN classifier for applications in self-learning systems. The approach uses mechanisms of preprocessing of signals from sensory neurons typical of a thalamus in a diencephalon. The algorithm concept is based on a cusp catastrophe model and on training by routing. The algorithm guarantees a zero dispersion of connection weight values across the entire network, which is particularly important in the case of hardware implementation based on programmable logic devices. Due to non-iterative mechanisms inspired by training methods for associative memories, the approach makes it possible to estimate the capacity of the network and required hardware resources. The trained network shows resistance to the phenomenon of catastrophic forgetting. Low complexity of the algorithm makes in-situ hardware training possible without using power-hungry accelerators. The paper compares the complexities of hardware implementations of the algorithm with the classic STDP and conversion procedures. The basic application of the algorithm is an autonomous agent equipped with a vision system and based on a classic FPGA device.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00332024-10-01T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0023<abstract> <title style='display:none'>Abstract</title> <p>The imbalance and complexity of network traffic data are hot issues in the field of intrusion detection. To improve the detection rate of minority class attacks in network traffic, this paper presents a method for intrusion detection based on the recombination generative adversarial network (RGAN). In this study, dual-stage game learning is used to optimize the discriminator for efficient identification of attack samples. In the first stage, the proposed model trains a deep convolutional generative adversarial network (DCGAN) integrated with the self-attention (SA) mechanism, and simultaneously trains an independent convolutional neural network (CNN) classifier integrated with the gated recurrent unit (GRU). This stage allows the generator to generate minority class attack samples that closely resemble real samples, while the independent classifier possesses the basic classification ability. In the second stage, the generator and the independent classifier of the DCGAN together constitute the second layer of the model—the generative adversarial network. Through dual-stage game learning, the classifier’s discrimination ability for the minority samples is optimized, and it serves as the final output of the discriminator. In addition, the introduction of reconstruction loss helps prevent the detection rate of false positive samples. Experimental results on the CSE-IDS-2018 dataset demonstrate that our model performs well compared with various other intrusion detection techniques in terms of detection accuracy, recall, and F1-score for minority class attacks.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00232024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0014<abstract> <title style='display:none'>Abstract</title> <p>This work proposes a convex cooperative control scheme for a multiagent system of differential mobile robots in a leader–follower formation. First, the kinematic model of the differential robots is obtained in a linear parameter varying representation. Next, a reference model approach is considered to track the desired trajectory. The paper’s contribution is then to derive conditions to guarantee the convergence of the convex controller, which is achieved using a non-quadratic Lyapunov function. Subsequently, this control law is integrated into the agent that leads a distributed control protocol based on graph theory designed to reach the consensus of the followers. Simulations of five mobile robots are performed to illustrate the effectiveness of the proposed method.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00142024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0021<abstract> <title style='display:none'>Abstract</title> <p>In practical applications of machine learning, the class distribution of the collected training set is usually imbalanced, i.e., there is a large difference among the sizes of different classes. The class imbalance problem often hinders the achievable generalization performance of most classifier learning algorithms to a large extent. To ameliorate the learning performance, some effective approaches have been proposed in the literature, where the recently presented GAN-based oversampling methods are very representative. However, their generated minority class examples have the risk of high similarity and duplication degree. To further ameliorate the quality of the generated minority class examples, i.e., to make the generated examples effectively expand the minority class region, a novel oversampling approach named the GWGAN-GP is proposed, which is based on the Gaussian distribution label within the framework of aWasserstein generative adversarial network with gradient penalty (WGAN-GP). Our GWGAN-GP approach incorporates the Gaussian distribution as an input label, thereby making the generated examples more diverse and dispersive. The examples are then combined with the original dataset to form a balanced dataset, which is subsequently utilized to evaluate the classification performance of three selected classification algorithms. Experimental results on 16 imbalanced datasets demonstrate that the GWGAN-GP not only generates examples that better conform to the distribution of the original dataset, but also achieves superior classification performance. Specifically, when combined with the KNN classifier, the GWGAN-GP significantly outperforms other oversampling approaches considered in the study.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00212024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0019<abstract> <title style='display:none'>Abstract</title> <p>In the paper a problem of assignment of tasks to machines is formulated and solved, where a criterion of data replication is used and a large size of data imposes additional constraints. This problem is met in practice when dealing with large genomic files or other types of vast data. The necessity of comparing all pairs of files within a big set of DNA sequencing results, which we collected, maintained, and analyzed within a national genomic project, brought us to the proposed results. This problem resembles that of generating a particular Steiner system, and a mechanism observed there is employed in one of our algorithms. Based on the problem complexity, we propose two heuristic algorithms, which work very well even for instances with tight constraints and a heterogeneous environment defined. In addition, we propose a simplified method, nevertheless capable of finding very good solutions and surpassing the algorithms in some special cases. The methods are validated in tests on a wide set of instances, where values of parameters reflect our real-world application and where their usefulness is proven.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00192024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0016<abstract> <title style='display:none'>Abstract</title> <p>This study develops a new conjugate gradient (CG) search direction that incorporates a well defined spectral parameter while the step size is required to satisfy the famous strong Wolfe line search (SWP) strategy. The proposed spectral direction is derived based on a recent method available in the literature, and satisfies the sufficient descent condition irrespective of the line search strategy and without imposing any restrictions or conditions. The global convergence results of the new formula are established using the assumption that the gradient of the defined smooth function is Lipschitz continuous. To illustrate the computational efficiency of the new direction, the study presents two sets of experiments on a number of benchmark functions. The first experiment is performed by setting uniform SWP parameter values for all the algorithms considered for comparison. For the second experiment, the study evaluates the performance of all the algorithms by considering the exact SWP parameter values used for the numerical experiments as reported in each work. The idea of these experiments is to study the influence of parameters in the computational efficiency of various CG algorithms. The results obtained demonstrate the effect of the parameter value on the robustness of the algorithms.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00162024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0018<abstract> <title style='display:none'>Abstract</title> <p>In this study, we investigate several online and semi-online scheduling problems related to two hierarchical machines with a common due date to maximize the total early work. For the pure online case, we design an optimal online algorithm with a competitive ratio of <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="graphic/j_amcs-2024-0018_ieq_001.png"/> . Additionally, for the cases when the largest processing time is known, we give optimal algorithms with a competitive ratio of 6/5 if the largest job is a lower-hierarchy one, and of <inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="graphic/j_amcs-2024-0018_ieq_002.png"/> − 1 if the largest job is a higher-hierarchy one.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00182024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0015<abstract> <title style='display:none'>Abstract</title> <p>This paper addresses the challenge of managing state constraints in vehicle platoons, including maintaining safe distances and aligning velocities, which are key factors that contribute to performance degradation in platoon control. Traditional platoon control strategies, which rely on a constant time-headway policy, often lead to deteriorated performance and even instability, primarily during dynamic traffic conditions involving vehicle acceleration and deceleration. The underlying issue is the inadequacy of these methods to adapt to variable time-delays and to accurately modulate the spacing and speed among vehicles. To address these challenges, we propose a dynamic adjustment neural network (DANN) based cooperative control scheme. The proposed strategy employs neural networks to continuously learn and adjust to time varying conditions, thus enabling precise control of each vehicle’s state within the platoon. By integrating a DANN into the platoon control system, we ensure that both velocity and inter-vehicular spacing adapt in response to real-time traffic dynamics. The efficacy of our proposed control approach is validated using both Lyapunov stability theory and numeric simulation, which confirms substantial gains in stability and velocity tracking of the vehicle platoon.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00152024-06-25T00:00:00.000+00:00https://sciendo.com/article/10.61822/amcs-2024-0013<abstract> <title style='display:none'>Abstract</title> <p>We introduce an algebraically active disturbance rejection-based control solution for the trajectory tracking problem of an uncertain second-order flat system with unknown external disturbances. To this end, we first algebraically identify the system’s unknown dynamics and the external disturbances with a linear set of time-varying integral expressions for the output and the control signal. We use the identified dynamics on an online feedback cancellation scheme to linearize the second-order system and cancel the uncertainties. With a proportional-integral controller we stabilize the linearized system without the need to estimate the velocity and have feedback from it. We carry out the stability analysis using linear systems theory. Finally, we evaluate the effectiveness of the proposed controller in a partially known 2-DOF manipulator.</p> </abstract>ARTICLEtruehttps://sciendo.com/article/10.61822/amcs-2024-00132024-06-25T00:00:00.000+00:00en-us-1