Product and Process Development via Sequential Pseudo-Uniform Design

Chang, Jyh-Shyong

doi:10.1021/ie034294j

Cited by 11 publications

(14 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their UD results in simulated and real fluorescence systems were satisfactory. Chang and Lin developed the sequential pseudo‐UD method to locate additional experiments in the experimental region. Their simulation results demonstrated that the product and process development based on the proposed method only required a reasonable number of experiments.…”

Section: Applications Of Uniform Designmentioning

confidence: 99%

Uniform experimental design in chemometrics

et al. 2018

Journal of Chemometrics

View full text Add to dashboard Cite

Experimental designs and modeling are very important in chemometrics and chemical engineering. There are many kinds of experimental designs, which include the fractional factorial design (including the orthogonal design), the optimal regression design, and the uniform design. The uniform experimental design can be regarded as a fractional factorial design with model uncertainty, a space filling design for computer experiments, a robust design against model specification, and a supersaturated design. This paper gives a brief introduction to the recent theoretical developments on uniform experimental design as well as its applications in chemometrics.

show abstract

Section: Applications Of Uniform Designmentioning

confidence: 99%

Uniform experimental design in chemometrics

et al. 2018

Journal of Chemometrics

View full text Add to dashboard Cite

show abstract

“…To determine the possible optima of an unknown process, an experimental design scheme that uses UD, SPUD, artificial neural networks (ANN), and random search was proposed in our previous studies. , The main objective of the proposed experimental design scheme is to build a reliable model and locate the optima of an unknown process simultaneously using only a reasonable number of experiments. Presently, the most commonly used ANN type is a multilayer artificial feed-forward neural network (FNN), which is trained by the back-propagation (BP) algorithm.…”

Section: Construction Of Data-driven Model For Process Optimizationmentioning

confidence: 99%

“…For an unknown complex process, one usually uses the experimental design to arrange effective experiments to obtain process information from a small-scale system. When the amount of experimental information is enough to construct a reliable model, the optimal recipe and operating conditions of the small-scale system could be obtained effectively on the basis of the constructed data-driven model. − ,, Similarly, if the data-driven model is built from the experimental data eventually taken from a scale-up system, the optimal recipe and operating conditions could be determined accordingly if the cost of the experimental resources is not an important issue. In general, the optimal recipe and operating conditions for a small-scale system may not be necessarily the same as those for the scale-up system.…”

Section: Introductionmentioning

confidence: 99%

Use of Data-Driven Model to Optimize Scale-up Production of Nanosized Silver Particles

Chang

Tang

2008

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

Optimization of the synthesis of nanosized silver particles by chemical reduction using formaldehyde in aqueous solution was carried out based on small-scale experimental data (with a 100 cc reaction mixture loaded in a 200 cc beaker and shaken in an isothermal shaker). Effects of the possible processing variables were achieved. In this study, the scale-up production (2 L or more of the reaction mixture) of nanosized silver particles is intended; however, for a scale-up system, the experiments generally require large and expensive resources. To obtain the optimal recipe and operating conditions using limited resources for a scale-up system, an algorithm for extending the identified model built from a small-scale system to a scale-up system was first developed through a numerical example. On the basis of the developed algorithm, 19 experiments were carried out in the scale-up system to achieve the objectives (minimum of the mean particle size of the silver colloid with or without the constraint of conversion of silver nitrate) of the synthesis of nanosized silver particles. With the modeling and optimization method proposed in this work, limited experiments were required, and most of the experiments can produce near on-specification products such that the off-specification products can be minimized.

show abstract

“…To provide an identified HNNRF model with rich information, the developed sequential pseudo-uniform design (SPUD) method [20] will be applied to locate limited but sufficient experiments for gathering the experimental data. The developed SPUD method is an extended version of the uniform design (UD) method [21].…”

Section: Neural Network Trainingmentioning

confidence: 99%

“…The developed SPUD method is an extended version of the uniform design (UD) method [21]. With respect to the detail of the SPUD method, one could make reference to our previous work [20]. In this work, the experiments for gathering the training data based on the UD method will be performed at first.…”

Section: Neural Network Trainingmentioning

confidence: 99%