Regional Information Centre
for Scientific and Technological Cooperation with EU

FP7 Partner offer profile

Federal State Unitary Enterprise “TURBONASOS” - FoF.NMP.2010-2: Supply chain approaches for small series industrial production, NMP.2010.3.1-1: New industrial models for a sustainable and efficient production - Simulation and Optimization of Work Processes in Components, Units and Systems of Petroleum Production and Transportation on the Basis of Artificial Neural Networks and Structure Modulating Methods of Nonlinear Programming

Go back
Date of filling in 15.09.2009
Organization name Federal State Unitary Enterprise “TURBONASOS”
Organization adress 394052, Russia, Voronezh 107 Ostrogozhskaya Str.
Department/Unit -
Contact person Gennady Tsukanov, Director for Development and Marketing, PhD, MBA; Oleg Butuzov, Head of Marketing Department
Phone +7-4732-727608; 727605
Fax +7-4732-727619
E-Mail market@turbonasos.ru
Web-site http://www.turbonasos.ru
Organization type
  • SME
Special Programme
Nanosciences, nanotechnologies, materials and new production technologies (NMP)
Call identifier
Topic number FoF.NMP.2010-2: Supply chain approaches for small series industrial production, NMP.2010.3.1-1: New industrial models for a sustainable and efficient production - Simulation and Optimization of Work Processes in Components, Units and Systems of Petroleum Production and Transportation on the Basis of Artificial Neural Networks and Structure Modulating Methods of Nonlinear Programming
Call Deadline 03.11.2009
Short description of the organization (max 12 lines): Federal State Unitary Enterprise “TURBONASOS” is a modern, steadily developing, research-and-production enterprise. It includes design, production and experiment units jointed by common production and technical cycle. At present Russian Federal Space Agency has assigned to “TURBONASOS” designing, producing, testing and servicing pumps, turbines and power systems for space-system engineering and its basic industry branches. Nowadays “TURBONASOS” is engineering, testing pre-production models and producing on repetition work base the following equipment:
turbo-pump units for liquid-fuel rockets; special pumps for water-craft and warships; hydraulic, steam and gas turbine; turbo-pump units for pumping different kinds of medium; gas ejectors; level and vertical electric pump units for pumping aggressive, abrasive, neutral medium, oil and oil products and other liquids; automated pumping stations and systems based on multiphase pumps, chemical pumps, slurry pumps, est.; automated modular turbo-expander gas turbine electric power stations; ozone plants, water and gas treatment systems; locking and regulating equipment.
Quantum of output for Russian Federal Space Agency and Ministry of Defence comes to 10%. Quantum of output for ferrous and non-ferrous metallurgy plants, enterprises of fuel and energy complex, producers of chemical fertilizers, est. comes to 90%.
Expertise offered In compliance with the petroleum production and transportation systems optimization research, design and modernization strategies aimed at the systems effectiveness increasing on all phases of their life cycle, data-processing technology for optimizing research work on the base of neuronet response surface has been elaborated and tested. The technology is designed to scrutinize and make comparative and automated analysis of testing, to engineer automatically, to search for the ways increasing effectiveness of functioning, to enhance reliability and to predict development of petroleum production and transportation systems at the stages of design, finish and modernization. The technology includes:
  • complex of neuronet simulators for different level on-line processors in pumps and turbines of petroleum production and transportation systems grids;
  • neuronet data bases for thermalphysic and hydraulic gas dynamics processes, carburetion and combustion processes based on multidimensional approximation of results of the computing and physical experiments with built-in intellectual algorithm of specification fundamental systems of transport equation;
  • wide range of original techniques, methods and computation algorithms for solving problems of componentry and aggregation optimization in petroleum production and transportation systems;
  • application program package embodying methods of neuronet modeling and optimization including method of indirect optimization based on self-organization and method of parameter space investigation (all these methods allow solving problems of nonlinear optimization with various number of independent variables (up to 100) and limits, with complicated topology (non-differentiable, multiextremal, stochastic), with high labour-intensiveness of optimization criterion value definition.
Availability of high-quality calculation methods enables applying practically theoretical researches in processes of oil and gas transportation. There are some traditional sources of errors in numerical results obtained while using widespread finite difference or finite element methods of simulation. They are digitization errors, insufficient approximation capacity of decision functions, low accuracy of calculation for initial conditions and boundaries. The errors partly constrain the use of mathematical simulating of work processes at micro level. At present, modern \"heavy\" finite element packages are widely exploited in scientific research. But engineers make some resistance to apply in practice the results received with their help and to introduce them in designing processes. It is due to the lack of trust to numerical algorithms and the impossibility to control the results obtaining.
As a universal numerical algorithm of work processes simulation for pipeline transport it is proposed an integral conjugate numerical method of solving mathematical physics equations by weighted residuals on the base of neuronet trial decisions.
This method is selection of global neuronet trial decision parameters for minimizing total residual at solving equations in random reference points. Application of the method allows eliminating differential equations decision errors caused by digitization of derivatives and low accuracy of borders representation. That raises adequacy of simulation process. The method serves for adjusting parameters of neuronet models when there is a lack of available experimental information on the investigated phenomena but the physical laws described by the corresponding equations are known. That is a basis for creation of information databases of physical processes with built-in neuronet algorithm for decisions search using the method. The data bases are opened for specification and constant identification on the basis of new experimental knowledge.
Application of the method allows simulating unrestricted physical processes in uniform neuronet logic basis, at the same time numerical algorithms are remarkable for their logic simplicity, clearness and a transparency of numerical results reception.
The developed technology is an application basis for robastic methods of design involving stochastic algorithm simulation and optimization. That allows including information-statistics operated factors into mathematical model and, accordingly, operating probability characteristics of engineering criteria. It is possible to notice that now the methodology of stochastic simulation actively develops. The program of the European Commission NODESIM-CFD (Non-Deterministic Simulation http://www.nodesim.eu) can be given as an example.
Scientific keywords petroleum production and transportation systems simulation and optimization of work processes methods of nonlinear programming neuronet models total residual
Publications on the topic (other references) -
Description of previous and present experience in International Cooperation -
Previous participation in EU’s Framework Programme projects