All issues

This paper presents the results of experimental validation of some structural issues concerning the practical use of methods to overcome catastrophic forgetting of neural networks. A comparison of current effective methods like EWC (Elastic Weight Consolidation) and WVA (Weight Velocity Attenuation) is made and their advantages and disadvantages are considered. It is shown that EWC is better for tasks where full retention of learned skills is required on all the tasks in the training queue, while WVA is more suitable for sequential tasks with very limited computational resources, or when reuse of representations and acceleration of learning from task to task is required rather than exact retention of the skills. The attenuation of the WVA method must be applied to the optimization step, i. e. to the increments of neural network weights, rather than to the loss function gradient itself, and this is true for any gradient optimization method except the simplest stochastic gradient descent (SGD). The choice of the optimal weights attenuation function between the hyperbolic function and the exponent is considered. It is shown that hyperbolic attenuation is preferable because, despite comparable quality at optimal values of the hyperparameter of the WVA method, it is more robust to hyperparameter deviations from the optimal value (this hyperparameter in the WVA method provides a balance between preservation of old skills and learning a new skill). Empirical observations are presented that support the hypothesis that the optimal value of this hyperparameter does not depend on the number of tasks in the sequential learning queue. And, consequently, this hyperparameter can be picked up on a small number of tasks and used on longer sequences.

We present results of a study aimed at identification of a controlled object’s channels based on postprocessing of measurements with development of a model of a multiple-input controlled object and subsequent active modelling experiment. The controlled object model is developed using approximation of its behavior by a neural network model using trends obtained during a passive experiment in the mode of normal operation. Recurrent neural network containing feedback elements allows to simulate behavior of dynamic objects; input and feedback time delays allow to simulate behavior of inertial objects with pure delay. The model was taught using examples of the object’s operation with a control system and is presented by a dynamic neural network and a model of a regulator with a known regulation function. The neural network model simulates the system’s behavior and is used to conduct active computing experiments. Neural network model allows to obtain the controlled object’s response to an exploratory stimulus, including a periodic one. The obtained complex frequency response is used to evaluate parameters of the object’s transfer system using the least squares method. We present an example of identification of a channel of the simulated control system. The simulated object has two input ports and one output port and varying transport delays in transfer channels. One of the input ports serves as a controlling stimulus, the second is a controlled perturbation. The controlled output value changes as a result of control stimulus produced by the regulator operating according to the proportional-integral regulation law based on deviation of the controlled value from the task. The obtained parameters of the object’s channels’ transfer functions are close to the parameters of the input simulated object. The obtained normalized error of the reaction for a single step-wise stimulus of the control system model developed based on identification of the simulated control system doesn’t exceed 0.08. The considered objects pertain to the class of technological processes with continuous production. Such objects are characteristic of chemical, metallurgic, mine-mill, pulp and paper, and other industries.

Solution of Dynamic Light Scattering problem makes it possible to determine particle size distribution (PSD) from the spectrum of the intensity of scattered light. As a result of experiment, an intensity curve is obtained. The experimentally obtained spectrum of intensity is compared with the theoretically expected spectrum, which is the Lorentzian line. The main task is to determine on the basis of these data the relative concentrations of particles of each class presented in the solution. The article presents a method for constructing and using a neural network trained on synthetic data to determine PSD in a solution in the range of 1–500 nm. The neural network has a fully connected layer of 60 neurons with the RELU activation function at the output, a layer of 45 neurons and the same activation function, a dropout layer and 2 layers with 15 and 1 neurons (network output). The article describes how the network has been trained and tested on synthetic and experimental data. On the synthetic data, the standard deviation metric (rmse) gave a value of 1.3157 nm. Experimental data were obtained for particle sizes of 200 nm, 400 nm and a solution with representatives of both sizes. The results of the neural network and the classical linear methods are compared. The disadvantages of the classical methods are that it is difficult to determine the degree of regularization: too much regularization leads to the particle size distribution curves are much smoothed out, and weak regularization gives oscillating curves and low reliability of the results. The paper shows that the neural network gives a good prediction for particles with a large size. For small sizes, the prediction is worse, but the error quickly decreases as the particle size increases.

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.

This article proposes the sumo-atclib framework, which provides a convenient uniform interface for testing adaptive control algorithms with different limitations, for example, restrictions on phase durations, phase sequences, restrictions on the minimum time between control actions, which uses the open source microscopic transport modeling environment SUMO. The framework shares the functionality of controllers (class TrafficController) and a monitoring and detection system (class StateObserver), which repeats the architecture of real traffic light objects and adaptive control systems and simplifies the testing of new algorithms, since combinations of different controllers and vehicle detection systems can be freely varied. Also, unlike most existing solutions, the road class Road has been added, which combines a set of lanes, this allows, for example, to determine the adjacency of regulated intersections, in cases when the number of lanes changes on the way from one intersection to another, and therefore the road graph is divided into several edges. At the same time, the algorithms themselves use the same interface and are abstracted from the specific parameters of the detectors, network topologies, that is, it is assumed that this solution will allow the transport engineer to test ready-made algorithms for a new scenario, without the need to adapt them to new conditions, which speeds up the development process of the control system, and reduces design overhead. At the moment, the package contains examples of MaxPressure algorithms and the Q-learning reinforcement learning method, the database of examples is also being updated. The framework also includes a set of SUMO scripts for testing algorithms, which includes both synthetic maps and well-verified SUMO scripts such as Cologne and Ingolstadt. In addition, the framework provides a set of automatically calculated metrics, such as total travel time, delay time, average speed; the framework also provides a ready-made example for visualization of metrics.

GIS INTEGRO is the geo-information software system forming the basis for the integrated interpretation of geophysical data in researching a deep structure of Earth. GIS INTEGRO combines a variety of computational and analytical applications for the solution of geological and geophysical problems. It includes various interfaces that allow you to change the form of representation of data (raster, vector, regular and irregular network of observations), the conversion unit of map projections, application blocks, including block integrated data analysis and decision prognostic and diagnostic tasks.

The methodological approach is based on integration and integrated analysis of geophysical data on regional profiles, geophysical potential fields and additional geological information on the study area. Analytical support includes packages transformations, filtering, statistical processing, calculation, finding of lineaments, solving direct and inverse tasks, integration of geographic information.

Technology and software and analytical support was tested in solving problems tectonic zoning in scale 1:200000, 1:1000000 in Yakutia, Kazakhstan, Rostov region, studying the deep structure of regional profiles 1:S, 1-SC, 2-SAT, 3-SAT and 2-DV, oil and gas forecast in the regions of Eastern Siberia, Brazil.

The article describes two possible approaches of parallel calculations for data processing 2D or 3D nets in the field of geophysical research. As an example presented realization in the environment of GRID of the application software ZondGeoStat (statistical sensing), which create 3D net model on the basis of data 2d net. The experience has demonstrated the high efficiency of the use of environment of GRID during realization of calculations in field of geophysical researches.

In this paper we consider the controlled motion of a helical body with three blades in an ideal fluid, which is executed by rotating three internal rotors. We set the problem of selecting control actions, which ensure the motion of the body near the predetermined trajectory. To determine controls that guarantee motion near the given curve, we propose methods based on the application of hybrid genetic algorithms (genetic algorithms with real encoding and with additional learning of the leader of the population by a gradient method) and artificial neural networks. The correctness of the operation of the proposed numerical methods is estimated using previously obtained differential equations, which define the law of changing the control actions for the predetermined trajectory.

In the approach based on hybrid genetic algorithms, the initial problem of minimizing the integral functional reduces to minimizing the function of many variables. The given time interval is broken up into small elements, on each of which the control actions are approximated by Lagrangian polynomials of order 2 and 3. When appropriately adjusted, the hybrid genetic algorithms reproduce a solution close to exact. However, the cost of calculation of 1 second of the physical process is about 300 seconds of processor time.

To increase the speed of calculation of control actions, we propose an algorithm based on artificial neural networks. As the input signal the neural network takes the components of the required displacement vector. The node values of the Lagrangian polynomials which approximately describe the control actions return as output signals . The neural network is taught by the well-known back-propagation method. The learning sample is generated using the approach based on hybrid genetic algorithms. The calculation of 1 second of the physical process by means of the neural network requires about 0.004 seconds of processor time, that is, 6 orders faster than the hybrid genetic algorithm. The control calculated by means of the artificial neural network differs from exact control. However, in spite of this difference, it ensures that the predetermined trajectory is followed exactly.

We consider one of the problems of transport modelling — searching the equilibrium distribution of traffic flows in the network. We use the classic Beckman’s model to describe time costs and flow distribution in the network represented by directed graph. Meanwhile agents’ behavior is not completely rational, what is described by the introduction of Markov logit dynamics: any driver selects a route randomly according to the Gibbs’ distribution taking into account current time costs on the edges of the graph. Thus, the problem is reduced to searching of the stationary distribution for this dynamics which is a stochastic Nash – Wardrope equilibrium in the corresponding population congestion game in the transport network. Since the game is potential, this problem is equivalent to the problem of minimization of some functional over flows distribution. The stochasticity is reflected in the appearance of the entropy regularization, in contrast to non-stochastic case. The dual problem is constructed to obtain a solution of the optimization problem. The universal primal-dual gradient method is applied. A major specificity of this method lies in an adaptive adjustment to the local smoothness of the problem, what is most important in case of the complex structure of the objective function and an inability to obtain a prior smoothness bound with acceptable accuracy. Such a situation occurs in the considered problem since the properties of the function strongly depend on the transport graph, on which we do not impose strong restrictions. The article describes the algorithm including the numerical differentiation for calculation of the objective function value and gradient. In addition, the paper represents a theoretical estimate of time complexity of the algorithm and the results of numerical experiments conducted on a small American town.

The article considers a mathematical model of decomposition of a complex product into assembly units. This is an important engineering problem, which affects the organization of discrete production and its operational management. A review of modern approaches to mathematical modeling and automated computer-aided of decompositions is given. In them, graphs, networks, matrices, etc. serve as mathematical models of structures of technical systems. These models describe the mechanical structure as a binary relation on a set of system elements. The geometrical coordination and integrity of machines and mechanical devices during the manufacturing process is achieved by means of basing. In general, basing can be performed on several elements simultaneously. Therefore, it represents a variable arity relation, which can not be correctly described in terms of binary mathematical structures. A new hypergraph model of mechanical structure of technical system is described. This model allows to give an adequate formalization of assembly operations and processes. Assembly operations which are carried out by two working bodies and consist in realization of mechanical connections are considered. Such operations are called coherent and sequential. This is the prevailing type of operations in modern industrial practice. It is shown that the mathematical description of such operation is normal contraction of an edge of the hypergraph. A sequence of contractions transforming the hypergraph into a point is a mathematical model of the assembly process. Two important theorems on the properties of contractible hypergraphs and their subgraphs proved by the author are presented. The concept of $s$-hypergraphs is introduced. $S$-hypergraphs are the correct mathematical models of mechanical structures of any assembled technical systems. Decomposition of a product into assembly units is defined as cutting of an $s$-hypergraph into $s$-subgraphs. The cutting problem is described in terms of discrete mathematical programming. Mathematical models of structural, topological and technological constraints are obtained. The objective functions are proposed that formalize the optimal choice of design solutions in various situations. The developed mathematical model of product decomposition is flexible and open. It allows for extensions that take into account the characteristics of the product and its production.

This article presents a classification of cities, taking into account their spatial planning and possible transport solutions for cities of various types. It also discusses examples of various strategies for the development of urban public transport in Russia and the European Union with a comparison of their efficiency. The article gives examples of the impact of urban planning on mobility of citizens. To implement complex strategic decisions, it is necessary to use micro and macro models which allow a comparison of situations “as is” and “as to be” to predict consequences. In addition, the authors propose a methodology to improve public transport route network and road network, which includes determining population needs in working and educational correspondences, identifying bottlenecks in the road network, developing simulation models and developing recommendations based on the simulation results, as well as the calculation of efficiency, including the calculation of a positive social effect, economic efficiency, environmental friendliness and sustainability of the urban transport system. To prove the suggested methodology, the macro and micro models of the city under study were built taking into account the spatial planning and other specifics of the city. Thus, the case study of the city of Naberezhnye Chelny shows that the use of our methodology can help to improve the situation on the roads by optimizing the bus route network and the road infrastructure. The results showed that by implementing the proposed solutions one can decrease the amount of transport load on the bottlenecks, the number of overlapping bus routes and the traffic density.