All issues

The work is devoted to the construction of efficient and applicable to real tasks first-order methods of convex optimization, that is, using only values of the target function and its derivatives. Construction uses OGMG, fast gradient method which is optimal by complexity, but requires to know the Lipschitz constant for gradient and the strong convexity constant to determine the number of steps and step length. This requirement makes practical usage very hard. An adaptive on the constant for strong convexity algorithm ACGM is proposed, based on restarts of the OGM-G with update of the strong convexity constant estimate, and an adaptive on the Lipschitz constant for gradient ALGM, in which the use of OGM-G restarts is supplemented by the selection of the Lipschitz constant with verification of the smoothness conditions used in the universal gradient descent method. This eliminates the disadvantages of the original method associated with the need to know these constants, which makes practical usage possible. Optimality of estimates for the complexity of the constructed algorithms is proved. To verify the results obtained, experiments on model functions and real tasks from machine learning are carried out.

First-order optimization methods are workhorses in a wide range of modern applications in economics, physics, biology, machine learning, control, and other fields. Among other first-order methods accelerated and momentum ones obtain special attention because of their practical efficiency. The heavy-ball method (HB) is one of the first momentum methods. The method was proposed in 1964 and the first analysis was conducted for quadratic strongly convex functions. Since then a number of variations of HB have been proposed and analyzed. In particular, HB is known for its simplicity in implementation and its performance on nonconvex problems. However, as other momentum methods, it has nonmonotone behavior, and for optimal parameters, the method suffers from the so-called peak effect. To address this issue, in this paper, we consider an averaged version of the heavy-ball method (AHB). We show that for quadratic problems AHB has a smaller maximal deviation from the solution than HB. Moreover, for general convex and strongly convex functions, we prove non-accelerated rates of global convergence of AHB, its weighted version WAHB, and for AHB with restarts R-AHB. To the best of our knowledge, such guarantees for HB with averaging were not explicitly proven for strongly convex problems in the existing works. Finally, we conduct several numerical experiments on minimizing quadratic and nonquadratic functions to demonstrate the advantages of using averaging for HB. Moreover, we also tested one more modification of AHB called the tail-averaged heavy-ball method (TAHB). In the experiments, we observed that HB with a properly adjusted averaging scheme converges faster than HB without averaging and has smaller oscillations.

The paper provides a method for selecting the composition of a refrigerant with a given isobaric cooling curve using an artificial neural network (ANN). This method is based on the use of 1D layers of a convolutional neural network. To train the neural network, we applied a technological model of a simple heat exchanger in the UniSim design program, using the Peng – Robinson equation of state.We created synthetic database on isobaric boiling curves of refrigerants of different compositions using the technological model. To record the database, an algorithm was developed in the Python programming language, and information on isobaric boiling curves for 1 049 500 compositions was uploaded using the COM interface. The compositions have generated by Monte Carlo method. Designed architecture of ANN allows select composition of a mixed refrigerant by 101 points of boiling curve. ANN gives mole flows of mixed refrigerant by composition (methane, ethane, propane, nitrogen) on the output layer. For training ANN, we used method of cyclical learning rate. For results demonstration we selected MR composition by natural gas cooling curve with a minimum temperature drop of 3 К and a maximum temperature drop of no more than 10 К, which turn better than we predicted via UniSim SQP optimizer and better than predicted by $k$-nearest neighbors algorithm. A significant value of this article is the fact that an artificial neural network can be used to select the optimal composition of the refrigerant when analyzing the cooling curve of natural gas. This method can help engineers select the composition of the mixed refrigerant in real time, which will help reduce the energy consumption of natural gas liquefaction.

When recognizing workers on images of a construction site obtained from surveillance cameras, a situation is typical in which the objects of detection have a very different spatial scale relative to each other and other objects. An increase in the accuracy of detection of small objects can be achieved by using the Feature-Fused modification of the SSD detector. Together with the use of overlapping image slicing on the inference, this model copes well with the detection of small objects. However, the practical use of this approach requires manual adjustment of the slicing parameters. This reduces the accuracy of object detection on scenes that differ from the scenes used in training, as well as large objects. In this paper, we propose an algorithm for automatic selection of image slicing parameters depending on the ratio of the characteristic geometric dimensions of objects in the image. We have developed a two-pass version of the Feature-Fused SSD detector for automatic determination of optimal image slicing parameters. On the first pass, a fast truncated version of the detector is used, which makes it possible to determine the characteristic sizes of objects of interest. On the second pass, the final detection of objects with slicing parameters selected after the first pass is performed. A dataset was collected with images of workers on a construction site. The dataset includes large, small and diverse images of workers. To compare the detection results for a one-pass algorithm without splitting the input image, a one-pass algorithm with uniform splitting, and a two-pass algorithm with the selection of the optimal splitting, we considered tests for the detection of separately large objects, very small objects, with a high density of objects both in the foreground and in the background, only in the background. In the range of cases we have considered, our approach is superior to the approaches taken in comparison, allows us to deal well with the problem of double detections and demonstrates a quality of 0.82–0.91 according to the mAP (mean Average Precision) metric.

The article is devoted to solving ill-posed problems of mathematical physics for elliptic and parabolic equations, such as the Cauchy problem for the Helmholtz equation and the retrospective Cauchy problem for the heat equation with constant coefficients. These problems are reduced to problems of convex optimization in Hilbert space. The gradients of the corresponding functionals are calculated approximately by solving two well-posed problems. A new method is proposed for solving the optimization problems under study, it is component-by-component descent in the basis of eigenfunctions of a self-adjoint operator associated with the problem. If it was possible to calculate the gradient exactly, this method would give an arbitrarily exact solution of the problem, depending on the number of considered elements of the basis. In real cases, the inaccuracy of calculations leads to a violation of monotonicity, which requires the use of restarts and limits the achievable quality. The paper presents the results of experiments confirming the effectiveness of the constructed method. It is determined that the new approach is superior to approaches based on the use of gradient optimization methods: it allows to achieve better quality of solution with significantly less computational resources. It is assumed that the constructed method can be generalized to other problems.

The paper presents the results of three-dimensional numerical simulation of thermal-hydraulic processes in the Intermediate Heat Exchanger of the advanced Sodium-Cooled Fast-Neutron (BN) Reactor considering a developed methodological approach.

The Intermediate Heat Exchanger (IHX) is located in the reactor vessel and intended to transfer heat from the primary sodium circulating on the shell side to the secondary sodium circulating on the tube side. In case of an integral layout of the primary equipment in the BN reactor, upstream the IHX inlet windows there is a temperature stratification of the coolant due to incomplete mixing of different temperature flows at the core outlet. Inside the IHX, in the area of the input and output windows, a complex longitudinal and transverse flow of the coolant also takes place resulting in an uneven distribution of the coolant flow rate on the tube side and, as a consequence, in an uneven temperature distribution and heat transfer efficiency along the height and radius of the tube bundle.

In order to confirm the thermal-hydraulic parameters of the IHX of the advanced BN reactor applied in the design, a methodological approach for three-dimensional numerical simulation of the heat exchanger located in the reactor vessel was developed, taking into account the three-dimensional sodium flow pattern at the IHX inlet and inside the IHX, as well as justifying the recommendations for simplifying the geometry of the computational model of the IHX.

Numerical simulation of thermal-hydraulic processes in the IHX of the advanced BN reactor was carried out using the FlowVision software package with the standard $k-\varepsilon$ turbulence model and the LMS turbulent heat transfer model.

To increase the representativeness of numerical simulation of the IHX tube bundle, verification calculations of singletube and multi-tube sodium-sodium heat exchangers were performed with the geometric characteristics corresponding to the IHX design.

To determine the input boundary conditions in the IHX model, an additional three-dimensional calculation was performed taking into account the uneven flow pattern in the upper mixing chamber of the reactor.

The IHX computational model was optimized by simplifying spacer belts and selecting a sector model.

As a result of numerical simulation of the IHX, the distributions of the primary sodium velocity and primary and secondary sodium temperature were obtained. Satisfactory agreement of the calculation results with the design data on integral parameters confirmed the adopted design thermal-hydraulic characteristics of the IHX of the advanced BN reactor.

The objective of this study is to determine the best formations for the joint movement of a group of small robots in an aquatic environment. Estimation of drag of the flow is a traditional and well-known area of research, but it is not always valid to extend the conclusions made for a single robot to a group of similar devices due to the physical effects that appear during joint movement, such as a wave shadow. For these reasons, it is necessary to study the hydrodynamic characteristics of certain robot formations as a stable structure. The hydrodynamic parameters of systems with two main types of propulsion were studied: locomotive (fishtails) and propellers. Formations similar in structure to schools of fish were mainly considered, and then their applicability for robots of different types was assessed. The relationship between the speed of movement of the group and the drag of each of its participants was also studied. Mathematical modeling of the flow around a group of robots was performed using the finite volume method using two software packages (FlowVision and OpenFoam). Robots with a screw propeller interfere with each other when packed into tight formations, and for the locomotive case, being in the disturbance zone, on the contrary, is preferable. Also, with poorly streamlined bodies, flows separating from the surface can turn into narrow turbulent jets that greatly interfere with the rear robots. It has been established that wake effect reduces energy costs only at low speeds of movement — about 5 cm/s; at high speeds, movement in columns becomes difficult for the rear robots. No large difference in frontal resistance was found between a single robot and a group for a fish-like tail. The studies made it possible to develop and substantiate recommendations for optimizing robot designs for group movement.

Currently, a significant development has been observed in the direction of distributed computing theory, where computational tasks are solved collectively by resource-constrained devices. In practice, this scenario is implemented when processing data in Internet of Things systems, with the aim of reducing system latency and network infrastructure load, as data is processed on edge network computing devices. However, the rapid growth and widespread adoption of IoT systems raise questions about the need to develop methods for reducing the resource intensity of computations. The resource constraints of computing devices pose the following issues regarding the distribution of computational resources: firstly, the necessity to account for the transit cost between different devices solving various tasks; secondly, the necessity to consider the resource cost associated directly with the process of distributing computational resources, which is particularly relevant for groups of autonomous devices such as drones or robots. An analysis of modern publications available in open access demonstrated the absence of proposed models or methods for distributing computational resources that would simultaneously take into account all these factors, making the creation of a new mathematical model for organizing distributed computing in IoT systems and its solution methods topical. This article proposes a novel mathematical model for distributing computational resources along with heuristic optimization methods, providing an integrated approach to implementing distributed computing in IoT systems. A scenario is considered where there exists a leader device within a group that makes decisions concerning the allocation of computational resources, including its own, for distributed task resolution involving information exchanges. It is also assumed that no prior knowledge exists regarding which device will assume the role of leader or the migration paths of computational tasks across devices. Experimental results have shown the effectiveness of using the proposed models and heuristics: achieving up to a 52% reduction in resource costs for solving computational problems while accounting for data transit costs, saving up to 73% of resources through supplementary criteria optimizing task distribution based on minimizing fragment migrations and distances, and decreasing the resource cost of resolving the computational resource distribution problem by up to 28 times with reductions in distribution quality up to 10%.

This paper proposes an architectural solution for organizing a convolutional neural network (CNN) oriented towards hardware implementation on edge devices under limited resources. To this goal, an approach to compressing spectrograms to a given size (28 × 28) is proposed using discretization, monoconversion, windowed Fourier transform, and two-dimensional interpolation. A balanced convolution procedure is developed based on compact convolutional filters, the size of which provides the balance between computational complexity and accuracy required for edge devices. An algorithm that enables convolution operations and calculation of the error function gradient in the convolutional layer in a single cycle ensuring increased performance in both inference and training modes of the CNN is proposed. The tradeoff between network trainability and its resistance to overfitting is optimized by applying the Dropout regularization method with a dropout coefficient of 0.5 for the fully connected layer.

The effectiveness of the proposed solution was demonstrated using the example of recognizing audio spectrograms of car and airplane engine sounds. The CNN was trained on a balanced dataset consisting of 7160 audio recordings. The trained network demonstrated high recognition accuracy (95%), low loss values (< 0.2), and balanced precision/recall/F-metric, demonstrating the effectiveness of the developed CNN model.

This work considers the problem of optimal control galvanic process in multianode bath. The nonstationary mathematical model of galvanic process, which considers change concentrations of electrolyte components, is developed. Demonstrated rationale for the choice of the form to extremal control on example chrome galvanic process in the standard electrolyte.