The aim of the project is to study mathematical models in the form of complex (infinite-dimensional nonlinear) dynamic systems that adequately describe modern technological processes, in particular, in nuclear power plant coolers and bioreactors for protein cultivation. The tasks are to generalize the multimodal method, which will allow to obtain modal systems (mathematical models) for new types of containers and their motions, to develop machine learning methods that allow to include in mathematical models a number of physical factors such as viscous damping, development of analysis methods of complex dynamic systems that can be applied to modal systems that describe sloshing. The strategic organizational goal is to create a "crystallization point" for young scientists who study practically important complex dynamical systems, as well as machine learning methods in the context of real applications.