Modeling nanofluid viscosity: comparing models and optimizing feature selection—a novel approach
نویسندگان
چکیده
Abstract Background The accurate prediction of viscosity in nanofluids is essential for comprehending their flow behavior and enhancing effectiveness different industries. This research delves into modeling the assessing various models through cross-validation techniques. are compared based on root mean square error sets, which served as selection criteria. main body abstract Four feature algorithms namely minimum redundancy maximum relevance, F-test, RReliefF were evaluated to identify most influential features prediction. physical meaning was algorithm that yielded best results, outlined this study. methodology takes account relevance aspects nanofluid's viscosity. To assess predictive performance models, a process conducted, provided robust evaluation. squared validation sets used compare models. rigorous evaluation identified reliable model predicting nanofluid Results results showed novel outclassed established approaches single material nanofluid. proposed had 0.022 an r value 0.9941 set, while test 0.0146, 0.0157, 0.9924. Conclusions provides valuable insights offers guidance choosing suitable modeling. study also highlights importance using select performance. developed can be helpful optimizing use industrial processes.
منابع مشابه
Modeling of Effective Thermal Conductivity and Viscosity of Carbon Structured Nanofluid
This paper was aimed to address the modeling of effective thermal conductivity and viscosity of carbon structured nanofluids. Response surface methodology, D_optimal design (DOD) was employed to assess the main and interactive effects of temperature (T) and weight percentage (wt %) to model effective thermal conductivity and viscosity of multi wall and single wall carbon nanotube, CVD and RGO G...
متن کاملHysteresis Phenomenon on Water-Based Nanofluid Viscosity
Dynamic viscosity values for three particular water-based nanofluids have been experimental determined using a commercial ‘piston-type’ viscometer. Viscosity data, expressed as a function of temperature and particle volume fraction, are presented for 29nm particle-size CuO-water and Al2O3-water with 36nm and 47nm particle-sizes, temperatures up to 75°C and particle volume fraction up to 9%. In ...
متن کاملmodeling of effective thermal conductivity and viscosity of carbon structured nanofluid
this paper was aimed to address the modeling of effective thermal conductivity and viscosity of carbon structured nanofluids. response surface methodology, d_optimal design (dod) was employed to assess the main and interactive effects of temperature (t) and weight percentage (wt %) to model effective thermal conductivity and viscosity of multi wall and single wall carbon nanotube, cvd and rgo g...
متن کاملComparing Feature-Based Models of Harmony
Predictive processing is a fundamental process in music cognition. While there are a number of predictive models of melodic structure, fewer approaches exist for harmony/chord prediction. This paper compares the predictive performance of n-gram, HMM, autoregressive HMMs as well as feature-based (or multiple-viewpoint) n-gram and Dynamic Bayesian Network Models of harmony, which used a basic set...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Bulletin of the National Research Centre
سال: 2023
ISSN: ['2522-8307', '1110-0591']
DOI: https://doi.org/10.1186/s42269-023-01114-w