Warning: sizeof(): Parameter must be an array or an object that implements Countable in /homepages/26/d94734260/htdocs/clickandbuilds/PSECommunity/wp-content/plugins/wpor/includes/class-wpor.php on line 4338
Records with Keyword: Numerical Methods
Rethinking Computing and Statistics Instruction with Vocareum and Gradescope
Alexander Dowling
November 22, 2019 (v2)
Subject: Education
Keywords: Active Learning, Classroom Technology, Education, Jupyter Notebooks, Multivariate Statistics, Numerical Methods, Python
I will share ongoing efforts to retool CBE 20258 Numerical and Statistical Analysis (required) to provide a scaffolding for all chemical engineering undergraduates to develop core competencies in computing, applied statistics, and mathematical modeling. Key aspects of the course redesign include i) modernizing content including the adoption of the Python programming language and Jupyter notebooks, ii) moving initial exposure to outside of the classroom, and iii) incorporating active learning in all class sessions. I will share how classroom technologies Vocareum and Gradescope have been critical to the success of the redesign by reducing grading time, giving students fast feedback, and enabling regular accountability.
Training All Chemical Engineers in Computing and Data Science
Alexander Dowling
November 11, 2019 (v3)
Subject: Education
Keywords: Active Learning, Multivariate Statistics, Numerical Methods, Python, Undergraduate Education
In this contribution, I will discuss ongoing efforts to retool the sophomore-level “Numerical and Statistical Analysis” course (required) to provide a scaffolding for all students to develop core competencies in computing, applied statistics, and mathematical modeling throughout their undergraduate experience and profession careers. Beginning in Spring 2019, we are transitioning from MATLAB to Python for several reasons including consistency with “Chemical Process Control” (junior, required) and college-wide electives in data science and statistical computing that already use Python. I will also share experiences using Jupyter notebooks and cloud-based computing platforms such as Colaboratory to incorporate active learning into lectures and tutorials and to remove technical barriers for students. Content and assignments have been reorganized to emphasize mastery of foundational skills in preference over content breadth. For example, students are now required to submit hand-written pseu... [more]
Toward Integrating Python Throughout the Chemical Engineering Curriculum: Using Google Colaboratory in the Classroom
Alexander Dowling
July 21, 2019 (v2)
Subject: Education
Keywords: Active Learning, Cloud Computing, Data Analysis, Numerical Methods, Python, Statistics, Undergraduate
Computing and data science skills are without doubt extremely valuable for modern (chemical) engineers. Big data, machine learning, predictive modeling, decision science and similar terms are ever-present in job posting, scientific literature, funding announcements, and popular news. Yet, many chemical engineers lack a background in the fundamentals of computer programming, applied statistics, and mathematical modeling for problem solving. Often, student excitement in data-centric topics manifest through self-study with tutorials, extracurricular projects, and online classes whereby students assemble a toolbox of skills but do not learn the fundamentals that transcend each technique.

In this contribution, I will discuss our ongoing efforts at the University of Notre Dame to create a coherent, integrated strategy for computing and data analysis in the undergraduate curriculum. A key focus is retooling the sophomore-level “Numerical and Statistical Analysis” course (required) to provi... [more]
Performance Evaluation of Real Industrial RTO Systems
Maurício M. Câmara, André D. Quelhas, José Carlos Pinto
July 30, 2018 (v1)
Subject: Other
Keywords: industrial RTO systems, Numerical Methods, on-line optimization, optimizing control, repeated identification and optimization, static real-time optimization (RTO)
The proper design of RTO systems’ structure and critical diagnosis tools is neglected in commercial RTO software and poorly discussed in the literature. In a previous article, Quelhas et al. (Can J Chem Eng., 2013, 91, 652⁻668) have reviewed the concepts behind the two-step RTO approach and discussed the vulnerabilities of intuitive, experience-based RTO design choices. This work evaluates and analyzes the performance of industrial RTO implementations in the face of real settings regarding the choice of steady-state detection methods and parameters, the choice of adjustable model parameters and selected variables in the model adaptation problem, the convergence determination of optimization techniques, among other aspects, in the presence of real noisy data. Results clearly show the importance of a robust and careful consideration of all aspects of a two-step RTO structure, as well as of the performance evaluation, in order to have a real and undoubted improvement of process operation.
Deterministic Global Optimization with Artificial Neural Networks Embedded
Global deterministische Optimierung von Optimierungsproblemen mit künstlichen neuronalen Netzwerken
Artur M Schweidtmann, Alexander Mitsos
October 15, 2018 (v2)
Subject: Optimization
Artificial neural networks (ANNs) are used in various applications for data-driven black-box modeling and subsequent optimization. Herein, we present an efficient method for deterministic global optimization of ANN embedded optimization problems. The proposed method is based on relaxations of algorithms using McCormick relaxations in a reduced-space [\textit{SIOPT}, 20 (2009), pp. 573-601] including the convex and concave envelopes of the nonlinear activation function of ANNs. The optimization problem is solved using our in-house global deterministic solver MAiNGO. The performance of the proposed method is shown in four optimization examples: an illustrative function, a fermentation process, a compressor plant and a chemical process optimization. The results show that computational solution time is favorable compared to the global general-purpose optimization solver BARON.
[Show All Keywords]