Treffer: РОЗРОБКА ТА ІНТЕГРАЦІЯ МОДЕЛІ ТЕПЛОВОГО НАСОСА НА ОСНОВІ НЕЙРОННОЇ МЕРЕЖІ В СЕРЕДОВИЩЕ SIMULINK ДЛЯ МОДЕЛЮВАННЯ СИСТЕМИ ОПАЛЕННЯ БУДИНКУ.

The article describes in detail the process of developing and integrating a model of a heat pump unit based on the LSTM neural network into the Simulink environment. This model makes it possible to simulate the operation of a building's heat supply system, taking into account changing external conditions. Due to its high accuracy and flexibility, the model can be used as a basis for synthesizing and studying various control algorithms for the heat supply system. This brings new opportunities for optimizing energy consumption and improving the efficiency of heat pump systems. The modeling results can be used to develop intelligent control systems that adapt to changing operating conditions and improve the energy efficiency of a building. The model development process involved several key steps. First, a large data set was prepared to train the model, based on an analytical model of the heat pump system. After that, a neural network was developed and optimized using the Adam algorithm and the mean square error (MSE) loss function. An important part of the work was the conversion of the developed model into C code using the keras2c library, which made it possible to integrate the model into the Simulink environment through the S-function block. Tests have shown that the integrated Simulink model not only kept the high accuracy of predictions, but also significantly outperformed the original Python model in terms of computing speed. This is achieved through code optimization and efficient use of computing resources. The obtained results demonstrate the prospects of the proposed approach for dynamic modeling of thermal power plants, synthesis of control systems for such plants, and solving problems of improving their energy efficiency. [ABSTRACT FROM AUTHOR]

Copyright of Automation of Technological & Business Processes / Avtomatizaciâ Tehnologiceskih i Biznes-Processov is the property of Odesa National University of Technology and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)