1 January 2022 in Energija Ekonomija Ekologija
A new approach to efficient, faster, and intelligent hydropower plant (HPP) control, where constituent equipment is described with highly non-linear mathematical models based on the recommendation from the working group of IEEE on prime movers, is represented in this paper. HPP stability and high efficiency are important factors dependent on the dynamic changes in the energy system demands and the starting time of the plant because the obtained energy is very flexible to those changes in the energy system. This paper is shown and analysed the implementation of the artificial neural network-based controller with PID as an auxiliary controller which helped achieve better behaviour, faster plant stabilization, and operation. The benefits of new technologies and possibilities led to improvements in HPP control and faster system operation. This is achieved by using MATLAB® – Deep Learning Toolbox whereas the simulations are prepared in Simulink. Artificial Neural Networks (ANN) as a technique used in the HPP control systems have advantages in getting a stable and faster response but the complexity of the structure behind the neural networks (NN), meaning algorithms, number of hidden layers, training function, activation function can complicate and destabilize the process. In this paper, the focus is put on the mechanical power responses improvement and the advantages of implementing new technologies contrary to the problems that can occur by using them such as plant destabilization by implementing minor changes, fitting parameters, learning, and training processes, number of hidden layers/neurons, number of epochs, etc.
31 March 2025 in Precision Mechanics & Digital Fabrication
The accurate determination of the conduit water starting time constant ($T_w$) is critical for optimizing hydro turbine performance and dynamic control in hydropower plants. Instead of relying on conventional calculation methods, machine learning (ML) techniques, specifically long short-term memory (LSTM) networks and multilayer perceptron (MLP) models, have been employed to identify $T_w$. The dataset used for model training and validation comprises real operational data collected from two hydropower plants. The effectiveness of both algorithms in $T_w$ identification has been evaluated through simulation, with Python serving as the primary programming environment. The findings indicate that, despite its more complex architecture, LSTM does not necessarily yield superior results. In contrast, MLP, as a relatively simpler model, demonstrates greater accuracy in estimating $T_w$, suggesting that intricate network structures are not always required for precise identification. Additionally, an optimization function ($F_\text{opt}$) has been utilized to assess the reliability of the identified $T_w$ values by comparing them with actual hydro turbine responses. The results underscore the practicality of MLP in hydropower system modeling, providing a computationally efficient alternative for conduit water starting time constant identification. These insights contribute to improving real-time turbine control and enhancing the efficiency of hydropower generation.