Peter the Great St. Petersburg Polytechnic University

Laboratory "Industrial systems of streaming data processing", SPbPU

IT Data analysis Machine learning

Head of Laboratory

Bolsunovskaya, Marina Vladimirovna

PhD in Physics and Mathematics, Associate Professor

CoLab Scopus

Publications

Citations

129

h-index

Publications

Citations

197

h-index

Authorization required.

Lab website

The Industrial Data Streaming Systems Laboratory is engaged in high-tech research and commercial developments in the field of algorithms and systems for data collection and analysis. The laboratory specializes in creating data collection and processing systems for solving non-standard tasks with the absence of standard solutions, the need to develop new software or hardware components and the requirement to adapt to the existing infrastructure.

The research results are used to create analytical information systems for companies interested in digitalizing business and seeking to use advanced technologies in industry, agriculture, construction, medicine, IT, housing and communal services, etc.

The laboratory's experts work with large amounts of data from diverse sources, such as digital images, laser scanning point clouds, sensor data, and audio data. When creating systems, methods of image mining, machine learning, predictive analytics, mathematical modeling, etc. are used.

Digital modeling
Simulation modeling
Optimization algorithms
Artificial intelligence
Machine learning

Marina Bolsunovskaya

Head of Laboratory

Alexey Boykov

Research Engineer

Research directions

Sensor system for environmental and sanitary-epidemiological monitoring

A small-sized autonomous vehicle with a sensor system measures the microclimate automatically, with the required frequency, without restrictions on the length of the working day and can independently move through various areas of the premises, including those that are difficult to access. It allows you to obtain a higher level of detail of data on environmental monitoring of production areas compared to manual monitoring or fixed sensors installed in certain places

Production planning system in mechanical engineering

The system will allow you to redistribute orders for the production of individual parts between enterprises of the same holding company so that the location of parts on the cutting sheet during each operation is optimal for each of them. The developed software increases the material utilization factor (CMM) to 6-8%, which is a high indicator of production optimization in this area. Co-executors: Laboratory of CMIS SPbPU, LLC "2050 Integrator", LLC "Tetrakub".

Hardware and software complex for automatic detection and classification of fabric defects in the textile industry

The PACK is developed on the basis of technical vision and artificial intelligence technologies. The software implementation uses a deep convolutional neural network for semantic image segmentation. During the trial operation of the complex at the Teikovsky KHBK, the complex revealed 6.28 times more defects in harsh fabric than during manual disassembly in the quality department of the enterprise. It is implemented with the financial support of the Innovation Promotion Fund. Co-executors: IvSPU, Higher School of Economics, Visiumtex LLC.

Software for digital modeling of the regional innovation system of the Russian Federation as a driver of sustainable development

A software solution for integrated digital modeling, analysis and forecasting of sustainable regional development. The developed platform visualizes data on the dynamics of innovation indicators in 85 regions of the Russian Federation. The data is accessed via an interactive map of the regions of the Russian Federation. The project was implemented jointly with the Higher School of Engineering and Economics of SPbPU and the Laboratory of Digital Modeling of Industrial Systems" PISH SPbPU at the expense of a grant from the Russian Science Foundation (project No. 20-78-10123).

Virtual stands for laboratory work on the discipline "Life safety"

A set of digital models of equipment for measuring noise levels, microclimate, lighting, and concentration of pollutants at an industrial facility. The virtual space makes it possible to analyze in detail the student's actions during laboratory work, which significantly increases the effectiveness of the educational process. Together with the Higher School of Technosphere Safety of the SPbPU Institute of Civil Engineering as part of the Program to Improve the Quality of Education and Training for the 2021-2022 academic year of Gazprom PJSC/Peter the Great St. Petersburg Polytechnic University. Event No. 13 "Modernization of the laboratory base of the HSE ISI SPbPU.

Virtual Compressor Laboratory

Digital counterparts of screw, reciprocating and centrifugal compressors with interactive controls for practical work in a remote format. As part of the project to create a Center for Collective Use "Technologies and Gas Transport" of Gazprom PJSC in SPbPU in order to improve the educational process in the areas of training relevant to Gazprom PJSC at the Higher School of Energy Engineering of the Institute of Energy of SPbPU (program for improving the quality of education and training for the 2020-2021 academic year of Gazprom PJSC/FSAOU IN St. Petersburg Polytechnic University of Peter the Great).

A virtual model of a hydraulic drive for laboratory work

Development of a digital model of a hydraulic drive stand and scenarios of virtual laboratory work for students studying hydraulic drive technology to modernize the laboratory base of the Higher School of Transport of the Institute of Mechanical Engineering, Materials and Transport of SPbPU under the Program for Improving the Quality of Education and Training for 2021-2022. Gazprom PJSC/Peter the Great St. Petersburg Polytechnic University.

Mathematical modeling and forecasting of COVID-19 spread in Russia

Development of a new methodology and software tools for mathematical modeling and dynamic forecasting of the spread of COVID-19 in the territory of the Russian Federation. Within the framework of the project, a family of digital COVID-19 distribution models has been developed, taking into account various sets of significant factors affecting the behavior of the system, as well as software implementing them. The developed models allow us to estimate in advance the scale, speed and consequences of the spread of COVID-19 in the regions of the Russian Federation, taking into account various scenarios for countering the spread of COVID-19.