Study on the Impact of Catalyst Factors on Ethanol Conversion and C4 Olefin Selectivity Based on Random Forest Regression

Jiandong Wei

doi:10.54097/b3q9x602

Authors

Jiandong Wei

DOI:

https://doi.org/10.54097/b3q9x602

Keywords:

Random Forest Regression, Catalyst Portfolio, Ethanol Conversion, C4 Olefin.

Abstract

Optimizing catalyst combinations for efficient Ethanol Conversion and selective generation of C4 Olefins is important for resource conservation and environmental protection. Specifically, this study started from the temperature data in the catalysts. Firstly, the data were preprocessed to unify the temperature intervals using Hermite interpolation, and then Pearson correlation coefficients were used to further determine the linear relationship between the Ethanol Conversion rate and the C4 Olefin selectivity, respectively, and the temperature. On this basis, a random forest regression model was established to explore the importance of each catalytic factor on ethanol conversion rate and C4 olefin selectivity. Finally, multifactor ANOVA was used in this paper to explore the degree of influence of each catalytic factor on ethanol conversion and C4 olefin selectivity. In this study, it was found that for ethanol conversion rate, the F-value of ethanol concentration was 594.66, which was the largest compared to the other catalyst factors, indicating that it had the greatest influence on ethanol conversion rate.

Downloads

Download data is not yet available.

References

[1] Zheng Bin, SUN Hongxia, WANG Weimin. Octane prediction of gasoline research method based on random forest regression [J]. Petroleum Refining and Chemical Industry, 2020, 51 (12): 69 - 75.

[2] REN Xiaotian, CHU Xiaoli, TIAN Songbai. Prediction of thiophene sulfide composition distribution in decompression distillate based on random forest algorithm [J]. Journal of Petroleum (Petroleum Processing), 2020, 36 (05): 995 - 1002.

[3] JIN Li-Ping, DENG Jin-Gen. Oil formation fouling prediction method based on differential evolutionary optimization random forest model[J]. Energy and Chemical Industry, 2022, 43 (06): 28 - 32.

[4] QIN Qiang, CHEN Huazhou, LI Linghui, et al. Multidimensional scaling and random forest applied to near-infrared quantitative analysis ofgasoline [J]. Science Technology and Engineering, 2014, 14 (36): 137 - 140.

[5] Yi Siqi, Wei Kai. Random forest stuck drill risk assessment method based on SMOTE under sampling [J]. Petroleum Geology and Engineering,2023, 37 (04): 100 - 103.

[6] Liu Lirui, Li Peiqiang, Chen Jiayu, et al. Hybrid energy storage capacity allocation based on Pearson correlation analysis and T-tFD under VMD decomposition [J]. China Electric Power, 2024, 57 (07): 82 - 97.

[7] Sun Shaocong, Xu Yang, Cao Bin. Aluminum output prediction from BiLSTM electrolyzer based on random forest feature selection[J]. Light Metals, 2023, (10): 30 - 36.

[8] HU Chunyan, HU Liangping. How to use ANOVA correctly--covariance chi-square test and SAS implementation [J]. Sichuan Mental Health, 2022, 35 (01): 11 - 15.

[9] Wang R. Research on work order scheduling based on multifactor ANOVA method [J]. Telecommunications Engineering Technology and Standardization, 2023, 36 (08): 65 - 68.

[10] Qiu Nianwei, Liu Qian, Liu Hui. Multiple comparison method in statistical analysis of experimental data in biology [J]. Journal of Biomathematics, 2015, 30 (03): 535 - 541.