PNIPU: Neurodin Software Modeling Complex

Main articles: Neural networks (neural networks)

2022: Announcement of the software-modeling complex "Neurodin"

Perm Polytechnic scientists have optimized the operation of power plants using neural networks. The university announced this on June 22, 2022.

EnergyDesc

As reported, gas turbine power plants are often created on the basis of converted aircraft engines. To optimize the quality of electricity generation, specialists simulate power supply systems using classic models. But they do not act quickly enough, and therefore it is inconvenient and long to test various control algorithms with their help. Scientists of the Perm Polytechnic proposed to use pre-trained artificial neural networks for this. The development was carried out within the framework of the Priority-2030 Strategic Academic Leadership Program.

Researchers have created a software modeling complex "Neurodin," which allows you to obtain neural network mathematical models of gas turbine power plants. Its use to model complex technological systems will help replace foreign analogues, for example, MATLAB-Simulink, scientists say.

EnergyDesc

They presented the results of the study in the journal IOP Conference Series: Earth and Environmental Science.

The automatic control system makes it possible to optimize the quality of electricity generation that consumers receive. Using models, it is possible to create algorithms for controlling gas turbine power plants and conduct their computer tests. The most critical modes of operation that may arise during the use of the power plant are difficult or impossible to reproduce in real conditions or on a test bench. To do this, a semi-horizontal stand with a mathematical model is used, which "recreates" the behavior of the electrical system. explained by one of the developers Grigory Kilin, senior lecturer at the Department of Electrical Engineering and Electromechanics of the Perm Polytechnic

According to scientists, it is quite difficult to build a model of an electrical system, since it includes a large number of interacting elements. Because of this, it is impossible to conduct computer tests quickly. Perm researchers have found a way to simplify the model to speed up its operation and optimize the quality of electricity generation.

We proposed to create a number of high-speed neural network models for the main characteristic modes of operation of a gas turbine power plant, reproducing them with the required degree of adequacy. With their help, you can quickly test the operation of automatic control algorithms. We developed the models using a specially designed artificial neural network. At the same time, to create it, we have developed an original methodology that allows you to choose a rational architecture and network hyperparameters aimed at the problem being solved. told associate professor Boris Kavalerov, scientific director of the developer, head of the Department of Electrical Engineering and Electromechanics of the Perm Polytechnic Institute, leading researcher at the Center for Additive Technologies of the Center for Collective Use, Doctor of Technical Sciences

Perm Polytechnic scientists have developed a matmodel that includes various electricity consumers. According to the researchers, it is they who determine the nature of the impact of the electrical system on the automatic control system of a gas turbine plant. It is also necessary to take into account important parameters in the load nodes of the power supply network.

The study showed that simplified models based on artificial neural networks work correctly. They can function quickly as part of semi-natural stands in real time. Pre-trained neural networks allow you to optimize the accuracy of models and the quality of electricity for the consumer.