Unleashing the future of wind energy systems using SiO 2 and TiO 2 based insulation: A deep dive into Artificial Intelligence and Response Surface Methodology
Restricted accessResearch articleFirst published online July, 2025
Unleashing the future of wind energy systems using SiO 2 and TiO 2 based insulation: A deep dive into Artificial Intelligence and Response Surface Methodology
This study presents the design and optimization of an induction generator for wind energy system applications. Artificial neural networks (ANNs) with feedforward backpropagation and response surface methodology (RSM) are widely used for modeling complex engineering processes. In this work, wt.%, stirring speed, and stirring time were optimized using ANN and RSM to achieve the desired response. The developed models demonstrated high predictive accuracy, with significant R2 values of 0.9982 for the RSM model and 0.99686 for the ANN model. For SiO2 samples, the optimal parameters were determined as wt.% of 2.15, stirring speed of 617, and stirring time of 16, while for TiO2 samples, the optimum values were wt.% of 6.96, stirring speed of 823, and stirring time of 20. This parameter optimization is crucial as it directly affects generator components’ properties, which affect the efficiency and reliability of power conversion. The improved material properties make induction generators more effective for wind energy applications because they reduce energy loss and enhance durability. These findings provide a systematic approach to improving material selection and processing parameters, contributing to enhanced performance and efficiency in induction generators for wind energy applications. The developed models serve as valuable tools for optimizing generator design, ultimately supporting more reliable and efficient renewable energy systems. The efficiency and dependability of induction generators are enhanced by the optimized parameters, which are essential for raising the effectiveness and sustainability of wind energy systems. These results suggest a methodical approach to material optimization to enhance electrical performance, which enriches the technology of renewable energy.
FernándezLMJuradoFSaenzJR.Aggregated dynamic model for wind farms with doubly fed induction generator wind turbines. Renew Energy2008; 33(1): 129–140.
2.
AtallahMMezouarAFernández-RamírezLM, et al. Supervisory control of reactive power in wind farms with doubly fed induction generator-based wind turbines for voltage regulation and power losses reduction. Elect Power Syst Res2024; 228: 110059.
3.
TapiaATapiaGOstolazaJX, et al. Reactive power control of a wind farm made up with doubly fed induction generators (I). In 2001 IEEE Porto Power Tech Proceedings (Cat. No. 01EX502) (Vol. 4, pp. 6–pp). IEEE.
4.
YeWLiuFYanY, et al. Application of response surface methodology and desirability approach to optimize the performance of an ultra-low temperature cascade refrigeration system. Appl Therm Eng2024; 239: 122130.
5.
GoudaSPShiDBasumataryS, et al. Sulfamic acid modified uio-66 metal-organic framework for biodiesel production: process optimization using response surface methodology, kinetics, and thermodynamic study. Chem Eng J2024; 480: 148154.
6.
KadrićDAganovicAKadrićE, et al. Applying the response surface methodology to predict the energy retrofit performance of the TABULA residential building stock. J Build Eng2022; 61: 105307.
7.
KavithaDSindhuTKNambiarTNP. Effect of nanoparticles on the dielectric strength and PD resistance of epoxy nanocomposites. In Proceedings of the International Conference on Soft Computing Systems: ICSCS 2015, Volume 1 (pp. 277–286). Springer India.
8.
GhouseSMVenkateshSRajeshR, et al. Effects of SiO 2 and TiO 2 nano fillers in enhancing the insulation breakdown strength of epoxy nano composite dielectric under divergent electric fields. Int J Electr Eng Inform2013; 5(4): 501.
9.
ChenYFDaiQWLinCW, et al. Characteristics and properties of SiO2-Al2O3/EP-PU composite. J Central South Univ2014; 21(11): 4076–4083.
10.
ChoiYMHwangboSALeeTG, et al. Effect of particle size on the mechanical properties of tio2–epoxy nanocomposites. Materials2021; 14: 2866.
11.
NetoPJDSBarrosTADSde PaulaMV, et al. Design of computational experiment for performance optimization of a switched reluctance generator in wind systems. IEEE Trans Energy Convers2018; 33(1): 406–419.
12.
RaheemATAzizARAZulkifliSA, et al. Optimisation of operating parameters on the performance characteristics of a free piston engine linear generator fuelled by CNG–H2 blends using the response surface methodology (RSM). Int J Hydrogen Energy2022; 47(3): 1996–2016.
13.
MostafaAMouradMMustafaA, et al. Influence of aluminum oxide nanoparticles addition with diesel fuel on emissions and performance of engine generator set using response surface methodology. Energy Convers Manag2023; 19: 100389.
14.
AsefPPerpinaRBBarzegaranMR, et al. Multiobjective design optimization using dual-level response surface methodology and Booth’s algorithm for permanent magnet synchronous generators. IEEE Trans Energy Convers2018; 33(2): 652–659.
15.
HernandezCLaraJArjonaMA, et al. Multi-objective electromagnetic design optimization of a power transformer using 3D finite element analysis, response surface methodology, and the third generation non-sorting genetic algorithm. Energies2023; 16(no): 5.
16.
ZhuJLiSSongD, et al. Multi-objective optimisation design of aircored axial flux PM generator. IET Electric Power Appl2018; 12(9): 1390–1395.
17.
JianLXuGSongJ, et al. Optimum design for improving modulating-effect of coaxial magnetic gear using response surface methodology and genetic algorithm. Prog Electromagn Res2011; 116: 297–312.
18.
GaoZYXieJWangSM.Optimization design of transverse flux permanent generator based on response surface method. Adv Mater Res2014; 936: 2135–2139.
19.
ChoSJChoYWLeeMG, et al. Variable impact analysis of linear generator by using response surface method. Int J Precis Eng Manuf2016; 17(9): 1223–1228.
20.
MohammadiMDadvarMDabirB.Application of response surface methodology for optimization of the stability of asphaltene particles in crude oil by TiO2/SiO2 nanofluids under static and dynamic conditions. J Dispers Sci Technol2018; 39(3): 431–442.
21.
GelisKNaci CelikAOzbekK, et al. Experimental investigation into efficiency of SiO2/water-based nanofluids in photovoltaic thermal systems using response surface methodology. Sol Energy2022; 235: 229–241.
22.
ThangarajHDavidPWBalachandranGB, et al. Experimental study of bifacial photovoltaic module with waste polyvinyl chloride flex and acrylonitrile butadiene styrene road side safety sticker as an alternative reflector: optimization using response surface methodology. Environ Sci Pollut Res2023; 30(35): 83873–83887.
23.
GhafoorianFWanHCheginiS.A systematic analysis of a small-scale HAWT configuration and aerodynamic performance optimization through kriging, factorial, and RSM methods. J Appl Comput Mech2024.
24.
AlmasiSGhoraniMMHaghighiMHS, et al. Optimization of a vacuum cleaner fan suction and shaft power using Kriging surrogate model and MIGA. Proc IMechE, Part A: J Power and Energy2022; 236(3): 519–537.
25.
GhafoorianFMirmotahariSRBakhtiariF, et al. Exploring optimal configurations for a wind farm with clusters of Darrieus VAWT, using CFD methodology. J Comput Appl Mech2023; 54(4): 533–551.
26.
MirzabeGHKeshtkarAR.Application of response surface methodology for thorium adsorption on PVA/Fe3O4/SiO2/APTES nanohybrid adsorbent. J Ind Eng Chem2015; 26: 277–285.
27.
GhasemzadehBMatinAAHabibiB, et al. Cotton/Fe3O4@SiO2@H3PW12O40 a magnetic heterogeneous catalyst for biodiesel production: process optimization through response surface methodology. Ind Crops Prod2022; 181: 114806.
28.
ZulfiqarMSamsudinMFRTaqviSAA, et al. Modelling and optimization of photocatalytic degradation of phenol via TiO2 nanoparticles: an insight into response surface methodology and artificial neural network. J Photochem Photobiol A Chem2019; 384: 112649.
29.
SaberianMHAshenai GhasemiFGhasemiI, et al. Morphology, mechanical behavior, and prediction of A-glass/SiO2/epoxy nanocomposite using response surface methodology. J Elastomers Plast2019; 51(7–8): 669–683.
30.
LashariNGanatTOtchereD, et al. Navigating viscosity of GO-SiO2/HPAM composite using response surface methodology and supervised machine learning models. J Pet Sci Eng2021; 205: 108800.
31.
LiuHLChiouYR.Optimal decolorization efficiency of reactive red 239 by UV/TiO2 photocatalytic process coupled with response surface methodology. Chem Eng J2005; 112(1–3): 173–179.
32.
EremiaSChevalier-LuciaDRaduG, et al. Optimization of hydroxyl radical formation using TiO2 as photocatalyst by response surface methodology. Talanta2008; 77(2): 858–862.
33.
JiangWJoensJADionysiouDD, et al. Optimization of photocatalytic performance of TiO2 coated glass microspheres using response surface methodology and the application for degradation of dimethyl phthalate. J Photochem Photobiol A Chem2013; 262: 7–13.
34.
RenZLiuYYuanL, et al. Optimizing the content of nano-SiO2, nano-TiO2 and nano-CaCO3 in Portland cement paste by response surface methodology. J Build Eng2021; 35: 102073.
35.
LiYDongWYangQ, et al. An automatic impedance matching method based on the feedforward-backpropagation neural network for a WPT system. IRE Trans Ind Electron2019; 66(5): 3963–3972.
36.
ShankarRBalasubramanianKRSivapirakasamSP, et al. Ann and RSM models approach for optimization of HVOF coating. Mater Today Proc2019; 46: 9201–9206.
37.
MouloodiSRahmanpanahHGoheryS, et al. Feedforward backpropagation artificial neural networks for predicting mechanical responses in complex nonlinear structures: a study on a long bone. J Mech Behav Biomed Mater2022; 128: 105079.
38.
FayazMShahHAseereAM, et al. A framework for prediction of household energy consumption using feed forward back propagation neural network. Technologies2019; 7(2): 2.
39.
ThekkudenDTMouradAHI. Investigation of feed-forward back propagation ANN using voltage signals for the early prediction of the welding defect. SN Appl Sci2019; 1(12): 1615. DOI: 10.1007/s42452-019−1660-4
40.
UsluS.Optimization of diesel engine operating parameters fueled with palm oil-diesel blend: comparative evaluation between response surface methodology (RSM) and artificial neural network (ANN). Fuel2020; 276: 117990.
41.
Kamyabi-GolAZebarjadSMSajjadiSA.Fabrication of NiO/SiO2 nanocomposites using sol–gel method and optimization of gelation time using Taguchi robust design method. Colloids Surf A Physicochem Eng Asp2009; 336(1–3): 69–74.
