A novel robust ensemble model integrated extreme learning machine with multi-activation functions for energy modeling and analysis: Application to petrochemical industry

Zhang, Xiaohan; Zhu, Qunxiong; He, Yan‐Lin; Xu, Yuan

doi:10.1016/j.energy.2018.08.069

Cited by 28 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this way, an alarm time sequence consisting of 0 and 1 is converted into a sequence of time nodes, which is a sequence composed of time node data. For example, the 0 and 1 sequences shown in Figure 2 can be converted into a new time node sequence (1,11,15,25,31,40) with this method.…”

Section: Correlation Analysis Of Alarm Time Sequencesmentioning

confidence: 99%

“…With the widespread application of computer control technologies such as distributed control system (DCS) and programmable logic controller (PLC) in industrial processes, it has become much more convenient to set up process alarm specifications. 1,2 However, because many process variables have strong correlations among each other, a large amount of sequential alarms usually take place which may lead to alarm flooding. In this sense, it is difficult for human operators to take timely and effective interventions on abnormal conditions.…”

Section: Introductionmentioning

confidence: 99%

“…It is acknowledged that the alarm management system plays an important role in ensuring safe and efficient operations of industrial productions. With the widespread application of computer control technologies such as distributed control system (DCS) and programmable logic controller (PLC) in industrial processes, it has become much more convenient to set up process alarm specifications. , However, because many process variables have strong correlations among each other, a large amount of sequential alarms usually take place which may lead to alarm flooding. In this sense, it is difficult for human operators to take timely and effective interventions on abnormal conditions. , The suitable method will have promising performance for quality-related fault diagnosis and control. , Therefore, it is of practical importance to suppress flood alarms so as to prevent the occurrence of alarm flooding.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Novel Correlation Analysis of Alarms Based on Block Matching Similarities

Yang

Chu

et al. 2019

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Correlation analysis has been widely applied for process detection and alarm analysis. However, traditional correlation analysis methods have been found inefficient due to the influence of process time delays on analysis results. In this paper, we propose an alarm correlation analysis method based on block matching similarities. By converting alarm data sequences to time node sequences, the block matching similarity (BMS) method is able to reduce the calculation burden of the correlation analysis. Considering time delays between the alarm variables, the improved maximal correlation coefficient method is presented to reduce the effect of time delays and to improve the accuracy of the correlation analysis. A numerical case and the Tennessee Eastman process are employed to demonstrate the effectiveness and efficiency of the proposed method.

show abstract

Section: Correlation Analysis Of Alarm Time Sequencesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Novel Correlation Analysis of Alarms Based on Block Matching Similarities

Yang

Chu

et al. 2019

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…5 Similarly, Zhang et al did a manual feature extraction based on their domain knowledge for developing energy analysis models for petrochemical industry. 6 In addition to the manual approaches, there are some model-based methods, such as Affinity Propagation (AP), Partial Least Squares (PLS), Principal Components Regression (PCR), and Principal Component Analysis (PCA), which have been successfully applied in similar ML workflows to do feature selection and handle dimensionality reduction in the context of regression problems. Another interesting recent trend in the modeling research in the field is to use modeling techniques such as Data Envelopment Analysis (DEA) and Artificial Neural Network (ANN) which have been proven to be applicable for developing predictive models and identifying energy saving potentials.…”

Section: Introductionmentioning

confidence: 99%

“…Geng et al selected five key process inputs based on experience for the modeling of ethylene manufacturing units 5 . Similarly, Zhang et al did a manual feature extraction based on their domain knowledge for developing energy analysis models for petrochemical industry 6 …”

Section: Introductionmentioning

confidence: 99%

Steam consumption prediction of a gas sweetening process with methyldiethanolamine solvent using machine learning approaches

2020

View full text Add to dashboard Cite

This study proposes a comprehensive data processing and modeling framework for building high-accuracy machine learning model to predict the steam consumption of a gas sweetening process. The data pipeline processes raw historical data of this application and identifies the minimum number of modeling variables required for this prediction in order to ease the applicability and practicality of such methods in the industrial units. On the modeling end, an empirical comparison of most of the state-of-the-arts regression algorithms was run in order to find the best fit to this specific case study. The ultimate goal is to leverage this model to identify the achievable energy conservation opportunity in such plants. The historical data for this modeling was collected from a gas treating plant at South Pars Gas Complex for 3 years from 2017 to 2019. This data gets passed through a multistage data processing scheme that conducts multicollinearity analysis and model-based feature selection. For model selection, a wide range of regression algorithms from different classes of regressor have been considered. Among all these methods, the Gradient Boosting Machines model outperformed the others and achieved the lowest crossvalidation error. The results show that this model can predict the steam consumption values with 98% R-squared accuracy on the holdout test set. Furthermore, the offline analysis demonstrates that there is a potential of 2% energy saving, equivalent to 24 000 metric tons of annual steam consumption reduction, which can be achieved by mapping the underperforming energy consumption states of the unit to the expected performances predicted by the model.

show abstract

A hierarchical assessment of resilience engineering indicators in petrochemical industries using AHP and TOPSIS

Shirali

Rashnoudi

Salehi

et al. 2022

Hum Ftrs & Erg Mfg Svc

View full text Add to dashboard Cite

Resilience engineering (RE) is a proactive approach that enables complex systems to deal with adverse events and improve safety management by enhancing structural and organizational capabilities. A methodological examination of RE-related studies showed that they had only focused on some major indicators so that subindicators have been mostly neglected. This study aims to present a hierarchical analysis to identify the importance degree of indicators and subindicators of RE using analytic hierarchy process (AHP) in a petrochemical plant. To accomplish this, a pairwise comparison matrix of the indicators and subindicators was used to collect the data required for AHP approach. To demonstrate the applicability of the AHP results, this study ranks the units of the petrochemical plant using the technique for the order of preference by similarity to an ideal solution (TOPSIS) approach based on the importance degree of RE indicators.A questionnaire was used to gather data related to RE indicators so we could use the TOPSIS method. The results of the AHP showed that management commitment, buffering capacity, and reporting culture were the most influential RE indicators. In addition, anticipation had the lowest impact on RE. The most important subindicators of the RE indicators were also identified using a hierarchical analysis through AHP. The results of TOPSIS provided a best-worst analysis of the units of the petrochemical plant. The findings of this study could help safety managers formulate better-targeted safety policies by investing in influential indicators and subindicators of RE.

show abstract

A novel robust ensemble model integrated extreme learning machine with multi-activation functions for energy modeling and analysis: Application to petrochemical industry

Cited by 28 publications

References 28 publications

Novel Correlation Analysis of Alarms Based on Block Matching Similarities

Novel Correlation Analysis of Alarms Based on Block Matching Similarities

Steam consumption prediction of a gas sweetening process with methyldiethanolamine solvent using machine learning approaches

A hierarchical assessment of resilience engineering indicators in petrochemical industries using AHP and TOPSIS

Contact Info

Product

Resources

About