Minimizing the complexity of surrogate models for optimization

Straus, Julian; Skogestad, Sigurd

doi:10.1016/b978-0-444-63428-3.50053-9

Cited by 6 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most important part to be optimized is the distillation arrangement since it is responsible for the largest fraction of a vast amount of energy consumed. There are several process simulations packages available and mostly used for example Aspen Plus, SimSci PRO/II, and UniSim (Strausa and Skogestada, 2016). These simulation packages use sequential-modular approach for solving the flowsheet, in which each unit operation is considered as a separate block and calculated sequentially (Biegler et al, 1997).…”

Section: Methodsmentioning

confidence: 99%

Energy Analysis and Remixing Effect of Thermal Coupling Petlyuk Column for Natural Gas Liquid (NGL) Fractionation Train

Zin

Abidin

Shahruddin

2021

Int. J. Renew. Energy Dev.

View full text Add to dashboard Cite

In this work, a non-conventional distillation sequence with thermal coupling (Petlyuk Column) was presented as a technique to perform the separation of the NGL consist of ethane, propane, butane or other higher alkanes. The improvements were investigated through the energy analysis and remixing effect. From the result obtained, it was found that the Petlyuk arrangement consumes less amount of energy and able to reduce the remixing effects as compared to the conventional column sequencing. The Petlyuk arrangement saved about 44.49% and 12.83% in terms of cooling and heating duty, respectively. The overall annual energy saving shown by this arrangement is 39.22%. This arrangement proved to be able to prevent the remixing effect occurrence that contributes to thermal and separation inefficiency. The desired separation efficiency also obtained by this arrangement as all the product specifications are met. The ability in avoiding remixing effect by the Petlyuk column permits a significant reduction in CO2 emission with an average of 29.43 % of each equipment involved. Hence, it can be concluded that the Petlyuk arrangement model is a better alternative to be implemented in the NGL fractionation train.

show abstract

Section: Methodsmentioning

confidence: 99%

Energy Analysis and Remixing Effect of Thermal Coupling Petlyuk Column for Natural Gas Liquid (NGL) Fractionation Train

Zin

Abidin

Shahruddin

2021

Int. J. Renew. Energy Dev.

View full text Add to dashboard Cite

show abstract

“…The integration of simulation tools with rigorous mathematical programming techniques is a challenging process due to the inherent complexity of calculations with a process simulator [1,2]. To overcome this limitation, a surrogate model for the process is derived based on the simulation results and included within an optimization framework.…”

Section: Introductionmentioning

confidence: 99%

Impact of Sampling Technique on the Performance of Surrogate Models Generated with Artificial Neural Network (ANN): A Case Study for a Natural Gas Stabilization Unit

et al. 2019

View full text Add to dashboard Cite

Data-driven models are essential tools for the development of surrogate models that can be used for the design, operation, and optimization of industrial processes. One approach of developing surrogate models is through the use of input–output data obtained from a process simulator. To enhance the model robustness, proper sampling techniques are required to cover the entire domain of the process variables uniformly. In the present work, Monte Carlo with pseudo-random samples as well as Latin hypercube samples and quasi-Monte Carlo samples with Hammersley Sequence Sampling (HSS) are generated. The sampled data obtained from the process simulator are fitted to neural networks for generating a surrogate model. An illustrative case study is solved to predict the gas stabilization unit performance. From the developed surrogate models to predict process data, it can be concluded that of the different sampling methods, Latin hypercube sampling and HSS have better performance than the pseudo-random sampling method for designing the surrogate model. This argument is based on the maximum absolute value, standard deviation, and the confidence interval for the relative average error as obtained from different sampling techniques.

show abstract

“…Surrogate models can be useful in the optimization of integrated flowsheets (Straus and Skogestad, 2016). Se-45 quential modular simulators often have problems with convergence due to recycles, whereas equation-oriented solvers simulators are difficult to initialize.…”

Section: Introductionmentioning

confidence: 99%

“…A better alternative, investigated in this paper, may be to use the concepts of self-optimizing control (SOC) (Skogestad, 2000) to identify new independent variables (Straus and Skogestad, 2016). Self-optimizing control is a philosophy from control theory.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Surrogate model generation using self-optimizing variables

Straus

Skogestad

2018

Computers & Chemical Engineering

Self Cite

View full text Add to dashboard Cite

This paper presents the application of self-optimizing concepts for more efficient generation of steady-state surrogate models. Surrogate model generation generally has problems with a large number of independent variables resulting in a large sampling space. If the surrogate model is to be used for optimization, utilizing self-optimizing variables allows to map a close-to-optimal response surface, which reduces the model complexity. In particular, the mapped surface becomes much "flatter", allowing for a simpler representation, for example, a linear map or neglecting the dependency of certain variables completely. The proposed method is studied using an ammonia reactor which for some disturbances shows limit-cycle behaviour and/or reactor extinction. Using self-optimizing variables, it is possible to reduce the number of manipulated variables by three and map a response surface close to the optimal response surface. With the original variables, the response surface would include also regions in which the reactor is extinct.

show abstract

Minimizing the complexity of surrogate models for optimization

Cited by 6 publications

References 9 publications

Energy Analysis and Remixing Effect of Thermal Coupling Petlyuk Column for Natural Gas Liquid (NGL) Fractionation Train

Energy Analysis and Remixing Effect of Thermal Coupling Petlyuk Column for Natural Gas Liquid (NGL) Fractionation Train

Impact of Sampling Technique on the Performance of Surrogate Models Generated with Artificial Neural Network (ANN): A Case Study for a Natural Gas Stabilization Unit

Surrogate model generation using self-optimizing variables

Contact Info

Product

Resources

About