Search

This paper presents the research results related to the development of an intelligent system for monitoring and assessing the state of natural systems (PROSTOR), which was tested in the area from the city of Velikiy Ustyug to the city of Kotlas on the Northern Dvina River. It is one of the most vulnerable places in Russia to spring snow-melt and ice-jams induced floods.The proposed automated flood forecasting technology is based on the concept of a multi-model description of complex natural objects implementing a mechanism of the selection and adaptation of parameters of the most adequate model for each specific situation. The computational core of PROSTOR is the two-dimensional hydrodynamic model STREAM_2D and its newer version STREAM_2D_CUDA based on the numerical solution of the shallow water equations with discontinuous bottom. Additional hydraulic resistance due to the ice roughness and decrease in the flow cross-section due to ice-caused congestion were taken into account for modeling the ice-jams water levels. The forecasting capabilities of the system are secured by the prediction of water levels at the gauging stations located upstream from Velikiy Ustyug basing either on neural networks, or by means of linking with the runoff formation model ECOMAG and using prognostic meteorological information.The system was built with the use of a service-oriented architecture, that provides flexible interaction between software modules, implementing hydrodynamic and hydrological models; modules of collecting and processing of heterogeneous data, including data from gauging stations and remote sensing data; control modules, etc. All system components are realized as web services and can be geographically distributed and localized in various organizations, cities and countries. All results of the system implementation, including the results of flooded zones calculations, flow parameters there, as well as satellite images are available via the geoportal.Models parameters were justified on the base of numerical experiments and simulations of the floods of 1980-2016 period, including more than 18 significant cases of ice-jamming. Grouping of model parameters according to the height of the ice-jam induced water levels suggested for the implementation of the hydrodynamic model incorporated into intelligent information system of river floods monitoring. Operational flood forecasting mode of the system was tested during 2017 – 2019 years under support of Russian Science Foundation project № 17-11-01254.

The paper proposes a method and a multi-model complex for managing computations, which make it possible to increase the efficiency of production processes at existing and prospective industrial enterprises due to the optimal (rational) functioning of their information systems. The features of computational processes and architectures of information systems of modern enterprises based on the concept of Industry 4.0 are considered. A brief description of the software package is given, in which the task of structural-functional synthesis of the structure of an enterprise information system, as well as the task of building an operational schedule for its work, are simultaneously solved in an automated (automatic) mode.

An important factor in the recognition of magnetic resonance images is not only the accuracy, but also the speed of the segmentation procedure. In some cases, the speed of the procedure is more important than the accuracy and the choice is made in favor of a less accurate, but faster procedure. This means that the segmentation method must be fully adaptive to different image models, that reduces its accuracy. These requirements are satisfied by developed hyper-heuristical particle swarm method for image segmentation. The main idea of the proposed hyper-heuristical method is the application of several heuristics, each of which has its strengths and weaknesses, and then their use depending on the current state of the solution. Hyper-heuristical particle swarm segmentation method is a management system, in the subordination of which there are three bioinspired heuristics: PSO-K-means, Modified Exponential PSO, Elitist Exponential PSO. Developed hyper-heuristical method was tested using the Ossirix benchmark with magnetic-resonance images (MRI) with various nature and different quality. The results of method’s work and a comparison with competing segmentation methods are presented in the form of an accuracy chart and a time table of segmentation methods.

The paper presents the state-of-the-art in flood forecasting and simulation applied to a river flood analysis and risk prediction. Different water flow forecasting and river simulation models and systems are analysed. An advanced river flood monitoring, modelling and forecasting approach is introduced. It extends the traditional approach based on modelling river physical processes by integration of different types of models and technologies such as input data clustering and filtering, digital maps of a relief and river terrain, data crowdsourcing, heterogeneous data processing, hydrological models for time scale modelling water flows and geo-simulation, inundation visualisation and duly warning on flooding. A case study on river flow forecasting and simulation for river flood risk analysis and management is given.

The chapter presents a holistic system that implements an advanced river flood modeling and forecasting approach. This approach extends traditional methods based on separate satellite monitoring or river physical processes modeling, by integration of different technologies such as satellite and in situ data processing, input data clustering and filtering, digital mapping of river valleys relief, data crowdsourcing, hydrodynamic modeling, inundation visualization, and also duly warning of stakeholders.

Elements of the methodological basis of the theory of monitoring automated (computer-aided) systems structure-dynamics control are proposed in the paper. This theory can be widely used in practice. It has an interdisciplinary basis provided by the classic control theory, operations research, artificial intelligence, systems theory and systems analysis. The proposed approaches were implemented in software prototypes. The software prototypes were developed to simulate control processes for space-facilities control system (SF CS). The unified description of various control processes allows synthesizing both technical and functional structures of SF CS simultaneously. The presented multiple-model complex, as compared with known analogues, has several advantages. It simplifies decision-making in SF CS structure dynamics management, for it allows seeking for alternatives in finite dimensional spaces rather than in discrete ones. The complex permits to reduce dimensionality of SF CS structure-functional synthesis problems in a real-time operation mode.

In this paper a new approach to the creation of short-term forecasting systems of river flooding is being further developed. It provides highly accurate forecasting results due to operative obtaining and integrated processing of the remote sensing and ground-based water flow data in real time. Forecasting of flood areas and depths is performed on a time interval of 12–48 h to be able to perform the necessary steps to alert and evacuate the population. Forecast results are available as web services. The proposed system extends the traditional separate methods based on satellite monitoring or modeling of a river’s physical processes, by using an interdisciplinary approach, integration of different models and technologies, and through intelligent choice of the most suitable models for a flood forecasting.