Electrical Output Stabilization of Nanofluid Spectral Splitting Photovoltaic/Thermal System by Regulating Parameters Based on Machine Learning Algorithms

Published: Jan. 1, 2025

The nanofluid spectral splitting photovoltaic/thermal (NSS-PV/T) system offers flexible regulation compared to traditional PV/T systems, enabling more thorough thermal and electrical output regulation. However, there is currently a lack of research on regulating the NSS-PV/T achieve stable output. Therefore, this paper provided two methods parameter stepping optimization that were applicable in both single multi-parameter application conditions. Firstly, energy equilibrium model was used calculate multiple sets parameters, simulating actual operating conditions serving as training data train subsequent prediction models. Based typical daily radiation data, lowest during study period determined standard system. Channel thickness, inlet temperature, concentration considered. Two machine learning algorithms, including support vector particle swarm optimization-back propagation neural network, employed design Subsequently, performed using pattern search algorithms. results illustrate relative errors do not exceed 0.070%, 0.099%, 0.120%, respectively. are than 0.369%. models convenient simple, but still face challenges parameters with weak influences deviate significantly from standard. method can lower deviations, variations smaller optimization, which further reduce fluctuations improve speed.

Language: Английский

Intelligent prediction of lost circulation based on improved k-nearest neighbor and self-attention mechanism-convolutional neural network

Geoenergy Science and Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 213712 - 213712

Published: Jan. 1, 2025

Language: Английский