Assessment of the uniaxial compressive strength of intact rocks: an extended comparison between machine and advanced machine learning models

Khatti, Jitendra; Grover, Kamaldeep Singh

doi:10.1007/s41939-024-00408-4

Cited by 10 publications

(2 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Microsoft excel 2019(Zhu et al 2022;Khatti and Grover 2024). Other than this Different statistical parameters like, root mean squared error (RMSE) (Al-Haddad and Mahdi 2024), mean squared error (MSE)(Ahmad et al 2024), mean absolute error (MAE)(Zhao et al 2024), mean absolute percentage error (MAPE)(Ding et al 2023), and mean bias error (MBE)(Huang 2022)were calculated for comparing the different prediction capabilities at various CHTCs.…”

mentioning

confidence: 99%

Framework for Accurate Estimation of Freezing Time and Convective Heat Transfer Coefficient for Freezing of Food product in Domestic Refrigerator: A Numerical and Simulation Modeling approach

Ray,

Minz,

Sinha

2024

Preprint

View full text Add to dashboard Cite

Freezing time estimation is essential for quality assurance, process optimization, and innovation in a variety of industries. The total freezing time is greatly influenced by the heat transfer parameters like convective heat transfer coefficient (CHTC), which in turn affects the product's quality characteristics. The correct calculation of CHTC, especially in domestic freezers, remains largely unexplored despite the development of several mathematical models for freezing time prediction. To fill this gap, this research presents a framework-driven analysis of CHTC for estimating the freezing time of frozen sweetened yoghurt in a domestic refrigerator. This research not only compares the performance of several freezing time prediction models, but also seeks to determine the optimal CHTC range for accurate freezing time forecasts, determining the most accurate freezing time prediction model and to replicate the freezing process by simulation accurately for this specific scenario. The results showed great accuracy for the determined CHTC range (RMSE = 27.24, CV = 0.16, MSE = 741.85, MAE = 22.4, MAPE = 12.24 and MBE = 8.14) and lowest average residual (3.95 min) for the fitted prediction for the freezing time calculation of frozen yoghurt. The simulation analysis further backed the results by showcasing marginal temperature difference between the simulated and actual temperature (0.232℃ to 0.684℃) of the frozen yoghurt by using the determined CHTC range. This showed the developed framework’s reliability in freezing time prediction and CHTC range determination, which will eventually help in modeling the freezing process of different food products with high accuracy.

show abstract

mentioning

confidence: 99%

Framework for Accurate Estimation of Freezing Time and Convective Heat Transfer Coefficient for Freezing of Food product in Domestic Refrigerator: A Numerical and Simulation Modeling approach

Ray,

Minz,

Sinha

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, machine learning techniques have found extensive applications in engineering. For instance, references [16][17][18] demonstrate modeling using the cosine amplitude method, and relevant algorithms were then employed to analyze and predict various engineering issues. Simultaneously, numerous machine learning studies to solve engineering problems have proposed several valuable and innovative designs [19][20][21][22] .…”

mentioning

confidence: 99%

Calibration of contact parameters of sandy soil for planting tiger nut based on non-linear tools

Qi,

Chen,

Yang

et al. 2024

Sci Rep

View full text Add to dashboard Cite

A methodology combining physical experiments with simulation was employed to acquire contact parameters of sandy soil precisely for planting tiger nuts in the desert area of Xinjiang. The stacking angle under different parameter combinations was applied as a response value. Through the Plackett–Burman test, several factors that have a significant influence were determined. The steepest ascent test was conducted to establish the finest scope of values for these parameters. The stacking angle was considered the response variable, and non-linear tools were used to optimize these parameters for simulation. The findings showed that applying response surface methodology (RSM) resulted in a relative error of 1.24%. In the case of BP-GA, the relative error compared to the physical test value was 0.34%, while for BP, it was 2.18%. After optimization using Wavelet Neural Network (WNN), the relative error was reduced to only 0.15%. Results suggest that WNN outperforms the RSM model, and the sandy soil model and parameters generated using WNN can be effectively utilized for discrete element simulation research.

show abstract

Predictive Genetic Programming Approaches for Swell-Shrink Soil Compaction

Jalal,

Bao,

Omar

2024

Earth Sci Inform

View full text Add to dashboard Cite

Assessment of the uniaxial compressive strength of intact rocks: an extended comparison between machine and advanced machine learning models

Cited by 10 publications

References 69 publications

Framework for Accurate Estimation of Freezing Time and Convective Heat Transfer Coefficient for Freezing of Food product in Domestic Refrigerator: A Numerical and Simulation Modeling approach

Framework for Accurate Estimation of Freezing Time and Convective Heat Transfer Coefficient for Freezing of Food product in Domestic Refrigerator: A Numerical and Simulation Modeling approach

Calibration of contact parameters of sandy soil for planting tiger nut based on non-linear tools

Predictive Genetic Programming Approaches for Swell-Shrink Soil Compaction

Contact Info

Product

Resources

About