Newswise —
Researchers from China propose that the use of artificial intelligence can enhance the safety and efficacy of fluidic catalytic cracking, a crucial process in converting heavy crude oil into gasoline and other products. This process was commercially introduced in 1915 and has undergone several improvements since then.
On April 07, the researchers published their findings on how artificial intelligence can enhance the catalytic cracking process, which entails breaking down the complex molecules present in heavy crude oil and separating them into the desired products. Their research was published in the Big Data Mining and Analytics journal.
According to Fan Yang, the first author of the research paper and a researcher at the College of Computer Science at Sichuan University, as well as being affiliated with New Hope Liuhe’s Algorithm and Big Data Center, the core objectives of petroleum processing are safety, efficiency, and environmental protection. The researchers focused on improving safety, efficiency, and environmental protection by investigating abnormal operating conditions, early warning systems, product yield analysis and optimization, and optimizing flue gas desulfurization analysis. These areas are considered hotspots for research in their efforts to enhance safety, efficiency, and environmental protection in the catalytic cracking process.
Mao Xu, a researcher at the New Hope Liuhe’s Data Intelligence Lab, and Fan Yang, the first author of the research paper, believe that progress in data acquisition and the application of artificial intelligence in data analysis can pave the way forward. They state that these advancements can aid in understanding the data, and thus improve the safety, efficiency, and environmental protection aspects of the fluidic catalytic cracking process.
"The enhancement of industrial data gathering technology empowers us to acquire more data for scrutiny," Xu expressed. "The advancement of artificial intelligence empowers us to scrutinize these data with more precision."
In order to enhance the scrutiny and fine-tune the process of catalytic cracking, the team delved into neural networks. This variety of artificial intelligence imitates the data processing mechanism of the human brain, utilizing interlinked nodes to rapidly scrutinize massive volumes of data. It gains knowledge through processing, recognizing how seemingly incongruent data can indicate a broader problem or a potential opening when merged. Take into account temperature, vapor production and the yield of the desired product. Neural networks, despite being modeled on human brains, are not constrained by human limitations. They can aggregate a plethora of data points associated with those three variables and observe how they might influence one another to generate distinct outcomes.
"According to Xu, this form of machine learning is data-oriented and capable of automatically and proficiently resolving high-dimensional issues. By integrating machine learning with a mechanism model - or an algorithm that comprehends how a mechanism produces a behavior - we can further diminish uncertainties and enhance prediction capabilities."
Xu pointed out that alternative machine learning models can be adapted to neural networks, charting the non-linear correlation of the chemical process in catalytic cracking. Through this methodology, the researchers can also cherry-pick specific characteristics or decrease the dimensions that the artificial intelligence considers, in order to explore more intricate connections.
Yang remarked, "The paper presents a thorough evaluation of the analysis of fluidic catalytic cracking process, primarily introducing approaches grounded in conventional mathematical mechanisms and artificial intelligence. The neural network approach showcases notable benefits since it can proficiently tackle the non-linear and high-dimensional attributes of catalytic cracking processes, attaining superior outcomes in the research of process analysis and optimization."
The researchers said they plan to eventually test their neural networks, which perform well in simulations, in the actual production process.
Yang added, "In forthcoming research, hybrid models that integrate mechanism models and artificial intelligence algorithms are projected to become formidable instruments for conducting more all-inclusive and precise analyses of chemical processes, and predicting production outcomes. These techniques will assume a significant role in the future progress of the chemical industry and hold immense value."
Other authors include Wenqiang Lei and Jiancheng Lv, both with the College of Computer Science at Sichuan University.
This work was backed by the State Key Program of the National Science Foundation of China, the National Natural Science Fund for Distinguished Young Scholars, the National Natural Science Foundation of China, and the Key Research and Development Project of Sichuan
##
About Big Data Mining and Analytics
Big Data Mining and Analytics (Published by Tsinghua University Press) discovers hidden patterns, correlations, insights and knowledge through mining and analyzing large amounts of data obtained from various applications. It addresses the most innovative developments, research issues and solutions in big data research and their applications. Big Data Mining and Analytics is indexed and abstracted in ESCI, EI, Scopus, DBLP Computer Science, Google Scholar, INSPEC, CSCD, DOAJ, CNKI, etc.
About Tsinghua University Press
Established in 1980, belonging to Tsinghua University, Tsinghua University Press (TUP) is a leading comprehensive higher education and professional publisher in China. Committed to building a top-level global cultural brand, after 41 years of development, TUP has established an outstanding managerial system and enterprise structure, and delivered multimedia and multi-dimensional publications covering books, audio, video, electronic products, journals and digital publications. In addition, TUP actively carries out its strategic transformation from educational publishing to content development and service for teaching & learning and was named First-class National Publisher for achieving remarkable results.
RSS