Variational algorithms for identifying power source of impulse point pollution
UDC
51.37Abstract
The environmental situation in the Azov sea requires the establishment of reliable systems of environmental monitoring that can effectively evaluate its condition in areas exposed to technogenic impact. To obtain such information on the state of the object is the efficient application of modern computing technology. The decision of problems of different pollutions propagation in the sea is possible on the basis of methods of mathematical modeling and methods of solving inverse problems, when due to the assimilation of measurement data is required to identify certain parameters of the transport model. Currently, variational methods of assimilation and the method of adjoint equations are actively developing and are used to solve Oceanographic problems. Algorithms for data assimilation of measurements based, as a rule, on the minimization of a quadratic cost function, characterizing the deviation of model solutions from measurement data. Transport model of a passive impurity acts as constraints on variation of input parameters. In the present paper the variational methods of identifying power instantaneous point source pollution based on the solution of adjoint tasks and variational task. In addition to search for the desired value applied valuation method and the variational method of filtering of linear systems of equations. Numerical experiments were carried out using the hydrodynamic model of the Azov sea. The received fields of the currents used in the passive admixtures transport modeling. Numerical experiments have shown that using variational algorithms for identifying power instantaneous point source required a single iteration. Due to identification of the variables in the space and time parameters of the pollution source additional iterations are required. When implementing variational algorithms for identification in case of noisy measurement data is filtered. When implementing the valuation method and solving the corresponding system of equations necessary additional filtering. In General, the numerical experiments showed a reliable performance of the considered algorithms identify the power source of the contamination, in relation of transport model of passive tracer in the Azov sea.
Keywords:
variational algorithm, identification of input parameters, passive admixture, transport model, Azov Sea, transport and diffusion of pollutions, assimilation of data measurementsAcknowledgement
References
- Malanotte-Rizzoli P., Holland W.R. Data constraints applied to models of the ocean general circulation. Part II: The Transient. Eddy-Resolving Case. Journal of Physical Oceanography, 1988, vol. 18, iss. 8, pp. 1093-1107.
- Yu. L., O'Brien J.J. Variational estimation of the wind stress drag coefficient and the oceanic eddy viscosity profile. J. Phys. Oceanogr., 1991, vol. 21, pp. 709-719.
- Marchuk G.I. Osnovnyye i sopryazhennyye uravneniya dinamiki atmosfery i okeana [Basic and conjugate equations of dynamics of atmosphere and ocean]. Meteorologiya i gidrologiya [Meteorology and hydrology], 1974, no. 2, pp. 17-34. (In Russian)
- Penenko V.V. Metody chislennogo modelirovaniya atmosfernykh protsessov [Methods for numerical modeling of atmospheric processes]. Leningrad, Gidrometeoizdat Pub., 1981, 350 p.
- Kochergin V.S., Kochergin S.V. Identifikaciya parametrov mgnovennogo tochechnogo istochnika zagryazneniya v Azovskom more na osnove metoda sopryazhennyh uravnenij [Identification of parameters of an instantaneous point source of pollution in the Azov sea on the basis of conjugate equations]. Morskoj gidrofizicheskij zhurnal [Marine hydrophysical journal], 2017, no. 1. P. 66-71. (In Russian)
- Strahov V.N. Metod fil'tracii sistem linejnyh algebraicheskih uravnenij - osnova dlya resheniya linejnyh zadach gravimetrii i magnitometrii [Filtering method systems of linear algebraic equations is the basis for solving linear problems of gravimetry and magnetometry]. Doklady AN SSSR [Rep. of the USSR Academy of Sciences], 1991, vol. 320, no. 3, pp. 595-599. (In Russian)
- Marchuk G.I. Matematicheskoye modelirovaniye v probleme okruzhayushchey sredy [Mathematical modeling in the environmental problem]. Moscow, Nauka Pub., 1982, 320 p. (In Russian)
- Eremeyev V.N., Kochergin V.P., Kochergin S.V., Sklyar S.N. Matematicheskoye modelirovaniye gidrodinamiki glubokovdnykh basseynov [Mathematical modeling of hydrodynamics of deep-water basins]. Sevastopol, EKOSI-Gidrofizika Pub., 2002, 238 p. (In Russian)
- Ivanov V.A., Fomin V.V. Matematicheskoye modelirovaniye dinamicheskikh protsessov v zone more - susha [Mathematical modeling of dynamic processes in the zone of the sea - land]. Sevastopol: EKOSI-gidrofizika. 2008. 363 p. (In Russian)
- Kochergin V.S., Kochergin S.V., Identifikaciya moshchnosti istochnika zagryazneniya v Kazantipskom zalive na osnove primeneniya variacionnogo algoritma [Identification of power source of pollution in the Kazantipsky Gulf through the application of a variational algorithm]. Morskoj gidrofizicheskij zhurnal [Marine hydrophysical journal], 2015, no. 2, pp. 79-88. (In Russian)
- Kochergin V.S., Kochergin S.V. Ispol'zovanie variacionnyh principov i resheniya sopryazhennoj zadachi pri identifikacii vhodnyh parametrov modeli perenosa passivnoj primesi [The use of variational principles and the solution of the adjoint problem, identification of input parameters for models of transport of passive tracer]. In Ehkologicheskaya bezopasnost' pribrezhnoj i shel'fovoj zon i kompleksnoe ispol'zovanie resursov shel'fa [Ecological safety of coastal and shelf zones and complex use of shelf resources], iss. 22. Sevastopol', MGI NANU, 2010, pp. 240-244. (In Russian)
Downloads
Submitted
Published
How to Cite
Copyright (c) 2017 Kochergin V.S., Kochergin S.V.
This work is licensed under a Creative Commons Attribution 4.0 International License.
