Industrial Algorithms

Toronto, Canada

Industrial Algorithms

Toronto, Canada
Time filter
Source Type

Kelly J.D.,Industrial Algorithms | Hedengren J.D.,Brigham Young University
Journal of Process Control | Year: 2013

Detecting windows or intervals of when a continuous process is operating in a state of steadiness is useful especially when steady-state models are being used to optimize the process or plant on-line or in real-time. The term steady-state implies that the process is operating around some stable point or within some stationary region where it must be assumed that the accumulation or rate-of-change of material, energy and momentum is statistically insignificant or negligible. This new approach is to assume the null-hypothesis that the process is stationary about its mean subject to independent and identically distributed random error or shocks (white-noise) with the alternative-hypothesis that it is non-stationary with a detectable and deterministic slope, trend, bias or drift. The drift profile would be typical of a time-varying inventory or holdup of material with imbalanced flows or even an unexpected leak indicating that the process signal is not steady. A probability of being steady or at least stationary over the window is computed by performing a residual Student t test using the estimated mean of the process signal without any drift and the estimated standard-deviation of the underlying white-noise driving force. There are essentially two settings or options for the method which are the window-length and the Student t critical value and can be easily tuned for each process signal that are included in the multivariate detection strategy. © 2012 Elsevier Ltd. All rights reserved.

Menezes B.C.,Petrobras | Kelly J.D.,Industrial Algorithms | Grossmann I.E.,Carnegie Mellon University
Industrial and Engineering Chemistry Research | Year: 2013

Nonlinear planning and scheduling models for crude-oil atmospheric and vacuum distillation units are essential to manage increased complexities and narrow margins present in the petroleum industry. Traditionally, conventional swing-cut modeling is based on fixed yields with fixed properties for the hypothetical cuts that swing between adjacent light and heavy distillates, which can subsequently lead to inaccuracies in the predictions of both its quantity and quality. A new extension is proposed to better predict quantities and qualities for the distilled products by taking into consideration that we require corresponding light and heavy swing-cuts with appropriately varying qualities. By computing interpolated qualities relative to its light and heavy swing-cut quantities, we can show an improvement in the accuracy of the blended or pooled quality predictions. Additional nonlinear variables and constraints are necessary in the model, but it is shown that these are relatively easy to deal with in the nonlinear optimization. © 2013 American Chemical Society.

Kelly J.D.,Industrial Algorithms | Menezes B.C.,Petrobras | Grossmann I.E.,Carnegie Mellon University
Industrial and Engineering Chemistry Research | Year: 2014

A novel technique using monotonic interpolation to blend and cut distillation temperatures and evaporations for petroleum fuels in an optimization environment is proposed. Blending distillation temperatures are well-known in simulations whereby cumulative evaporations at specific temperatures are mixed together; these data points are used in piece-wise cubic spline interpolations to revert back to the distillation temperatures. Our method replaces the splines with monotonic splines to eliminate the well-known oscillation effect called Runges phenomenon, and to allow the distillation curve itself to be adjusted by optimizing its initial and final boiling points known as cutpoints. By optimizing both the recipes of the blended material and its blending component distillation curves, very significant benefits can be achieved, especially given the global push toward ultralow sulfur fuels (ULSF), because of the increase in natural gas plays, reducing the demand for other oil distillates. Four examples are provided to highlight and demonstrate the technique, where we have good agreement between the predicted and actual evaporation curves of the blends. © 2014 American Chemical Society.

Menezes B.C.,Petrobras | Kelly J.D.,Industrial Algorithms | Grossmann I.E.,Carnegie Mellon University
Computer Aided Chemical Engineering | Year: 2015

We propose a mixed-integer nonlinear optimization for process design synthesis of oil-refinery units that includes crude-oil mixing, unit processing and product blending. The quantity-logic-quality phenomena involving a non-convex mixed-integer nonlinear problem is decomposed into a two-stage stochastic programming model with complete recourse considering, in a first stage, quantity and logic variables in a mixed-integer linear model and, in a second stage, quantity and quality variables in a nonlinear programming formulation. Iteratively, nonlinear models of each demand scenario are restricted by the multi-scenario process design results. An industrial-sized example that is not solved in a full space model demonstrates our tailor-made decomposition scheme, which yields within 5% gap between the first and the average second stage results. © 2015 Elsevier B.V.

Castillo P.A.C.,McMaster University | Mahalec V.,McMaster University | Kelly J.D.,Industrial Algorithms
AIChE Journal | Year: 2013

Current gasoline blend scheduling practice is to optimize blend plans via fixed duration (e.g., days) multiperiod NLP or MINLP models and schedule blends via interactive simulation. Solutions of multiperiod models typically have different blend recipes for each time period. We introduce inventory pinch points and use them to construct an algorithm based on single-period nonlinear model to minimize the number of different blend recipes. The algorithm optimizes multigrade blend recipes for each period delimited by the inventory pinch points and then uses a fine-grid multiperiod fixed-recipe MILP to compute blend volumes profile. If MILP is infeasible, a corresponding period between the pinch points is subdivided and recipes are reoptimized. In our case studies, solutions are computed in significant less time and are most often within 0.01% of the solutions by multiperiod MINLP. Reduced number of blend recipes makes it easier for the blend scheduler to create a schedule by interactive simulation. © 2013 American Institute of Chemical Engineers.

Kelly J.D.,Industrial Algorithms | Zyngier D.,Hatch Ltd.
Optimization and Engineering | Year: 2016

The focus of this paper is to detail the quantity and quality modeling aspects of production flowsheets found in all process industries. Production flowsheets are typically at a higher-level than process flowsheets given that in many cases more direct business or economic related decisions are being made such as maximizing profit and performance for the overall plant and/or for several integrated plants together with shared resources. These decisions are usually planning and scheduling related, often referred to as production control, which require a larger spatial and temporal scope compared to more myopic process flowsheets which detail the steady or unsteady-state material, energy and momentum balances of a particular process unit-operation over a relatively short time horizon. This implies that simpler but still representative mathematical models of the individual processes are necessary in order to solve the multi time-period nonlinear system using nonlinear optimizers such as successive linear programming and sequential quadratic programming. In this paper we describe six types of unit-operation models which can be used as fundamental building blocks or objects to formulate large production flowsheets. In addition, we articulate the differences between continuous and batch processes while also discussing several other important implementation issues regarding the use of these unit-operation models within a decision-making system. It is useful to also note that the quantity and quality modeling system described in this paper complements the quantity and logic modeling used to describe production and inventory systems outlined in Zyngier and Kelly (Optimization and logistics challenges in the enterprise, Springer, New York 61–95, 2009). © 2016 Springer Science+Business Media New York

Loading Industrial Algorithms collaborators
Loading Industrial Algorithms collaborators