Data-driven robust optimization for minimum nitrogen oxide emission under process uncertainty

Kim, Minsu; Cho, Sunghyun; Jang, Kyojin; Hong, Seokyoung; Na, Jonggeol; Moon, Il

doi:10.1016/j.cej.2021.130971

Cited by 19 publications

(9 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We assume that solving the multi-response model equations requires only modest computational effort. Researchers who are faced with computationally intensive models (e.g., CFD models) may opt to develop surrogate models (e.g., using polynomial chaos expansion) 39,40 to make the bootstrap calculations tractable. Steps required for implementing the extended algorithm using a modest multi-response model are shown in Table 3.…”

Section: Bootstrap Methods For Quantifying Parameter Uncertainties In...mentioning

confidence: 99%

Parameter estimation and prediction uncertainties for multi‐response kinetic models with uncertain inputs

2023

View full text Add to dashboard Cite

Error-in-variables model (EVM) methods are used for parameter estimation when independent variables are uncertain. During EVM parameter estimation, output measurement variances are required as weighting factors in the objective function. These variances can be estimated based on data from replicate experiments. However, conducting replicates is complicated when independent variables are uncertain. Instead, pseudo-replicate runs may be performed where the target values of inputs for repeated runs are the same, but the true input values may be different. Here, we propose a method to estimate output-measurement variances for use in multivariate EVM estimation problems, based on pseudo-replicate data. We also propose a bootstrap technique for quantifying uncertainties in resulting parameter estimates and model predictions. The methods are illustrated using a case study involving n-hexane hydroisomerization in a well-mixed reactor. Case-study results reveal that assumptions about input uncertainties can have important influences on parameter estimates, model predictions and their confidence intervals.

show abstract

Section: Bootstrap Methods For Quantifying Parameter Uncertainties In...mentioning

confidence: 99%

Parameter estimation and prediction uncertainties for multi‐response kinetic models with uncertain inputs

2023

View full text Add to dashboard Cite

show abstract

“…Polynomial chaos expansions are the most popular surrogate models used for prediction uncertainty quantification. [ 49–51,53,54 ]…”

Section: Propagating Input Uncertainties Into Prediction Uncertaintiesmentioning

confidence: 99%

“…System Types of uncertain inputs Al et al [52] Wastewater treatment • Surface flow rate of clarifier tank • Height of clarifier tank • Hydraulic retention time in aerobic tanks • Solid retention time in activated sludge • Ratio of anoxic tank volume to aerobic tank volume Sterr et al [53] Heat transfer in a microchannel heat sink • Surface roughness of a heat sink Kim et al [54] NOx emissions from waste materials incineration…”

Section: Authorsmentioning

confidence: 99%

Propagating input uncertainties into parameter uncertainties and model prediction uncertainties—A review

2023

View full text Add to dashboard Cite

A review of uncertainty quantification techniques is provided for a variety of situations involving uncertainties in model inputs (independent variables). The situations of interest are divided into three categories: (i) when model prediction uncertainties are quantified based on uncertainties in uncertain inputs, (ii) when parameter estimate uncertainties are calculated by propagation of uncertainties from measured inputs and outputs, and (iii) when model prediction uncertainties are quantified based on corresponding uncertainties in measured inputs and uncertain parameter estimates. For all three situations, linearization‐based and Monte Carlo‐based techniques are reviewed and details for their corresponding algorithms are presented. Recommendations are provided on which uncertainty quantification techniques are best for different types of chemical engineering models based on the amount of input uncertainty and nonlinearity over the range of plausible input and parameter values.

show abstract

“…For the construction of data-driven uncertainty sets, principal component analysis (PCA), kernel learning, support vector clustering, and Dirichlet process mixture model have been applied and achieved good results. − In addition to the above methods, Gumte et al recently proposed a new neural fuzzy clustering mechanism to cluster the uncertain space, to optimally identify the precise uncertain region and solve the uncertainty problem in the supply chain. Nowadays, DDRO methods have been extensively implemented in planning and scheduling, , supply chain design, , energy system optimization under unit efficiencies or demand uncertainty, , nitrogen oxide emission or flexible oxygen distribution, , and scheduling solutions in mutual energy networks and up- and downstream equipment …”

Section: Introductionmentioning

confidence: 99%

“…49−53 In addition to the above methods, Gumte et al 54 recently proposed a new neural fuzzy clustering mechanism to cluster the uncertain space, to optimally identify the precise uncertain region and solve the uncertainty problem in the supply chain. Nowadays, DDRO methods have been extensively implemented in planning and scheduling, 55,56 supply chain design, 57,58 energy system optimization under unit efficiencies or demand uncertainty, 59,60 nitrogen oxide emission or flexible oxygen distribution, 61,62 and scheduling solutions in mutual energy networks and up-and downstream equipment. 63 However, the actual refining process has many uncertain factors because of multiple technological processes, complex composition, and multiscale business layers; thus, only considering the uncertainty of a certain factor or the optimization of a single-tier business sometimes cannot make the optimized solutions fully feasible and immune to uncertainty.…”

mentioning

confidence: 99%

Modified Hybrid Strategy Integrating Online Adjustable Oil Property Characterization and Data-Driven Robust Optimization under Uncertainty: Application in Gasoline Blending

et al. 2022

View full text Add to dashboard Cite

With the increasing challenges in environmental protection, product quality, and market profit, the development of a feasible optimization approach coupling accurate oil property online prediction has become crucial to the intelligent production of refineries. However, traditional property modeling with near-infrared (NIR) spectroscopy cannot dynamically extract the synergistic effect between wavelengths and cause inaccurate property prediction. The success rate of the data-driven robust optimization (DDRO) of the refining process is influenced by process uncertainty. Thus, we proposed a modified hybrid strategy on online NIR prediction modeling with adjustable characteristic wavelength selection and DDRO under uncertainty for the refining process. An adjustable feature space variable extraction (AFSVE) method was first proposed to dynamically select the characteristic wavelengths for constructing accurate property prediction models. A data-driven robust optimization of gasoline blending model was rendered through dual transformation using the uncertainty set of oil property derived from principal component analysis combined with robust kernel density estimation. An industrial application showed the best prediction accuracy of property models with the AFSVE method, and a high blending success rate was obtained by the proposed modified hybrid strategy, which hedges against uncertainties, including measuring error and blending effect, and boosts environmentally friendly and intelligent process operation.

show abstract

Data-driven robust optimization for minimum nitrogen oxide emission under process uncertainty

Cited by 19 publications

References 53 publications

Parameter estimation and prediction uncertainties for multi‐response kinetic models with uncertain inputs

Parameter estimation and prediction uncertainties for multi‐response kinetic models with uncertain inputs

Propagating input uncertainties into parameter uncertainties and model prediction uncertainties—A review

Modified Hybrid Strategy Integrating Online Adjustable Oil Property Characterization and Data-Driven Robust Optimization under Uncertainty: Application in Gasoline Blending

Contact Info

Product

Resources

About