John P. Eason
Carnegie Mellon University
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by John P. Eason.
Computers & Chemical Engineering | 2017
Haoshui Yu; John P. Eason; Lorenz T. Biegler; Xiao Feng
Abstract Low and medium temperature energy utilization is one way to alleviate the energy crisis and environmental pollution problems. In the past decades, Organic Rankine Cycles (ORCs) have become a very promising technology for low and medium temperature energy utilization. When an ORC is used to recover waste heat in chemical plants, heat integration between the ORC and the process streams should be performed to save more utilities and generate more power. This study aims to integrate an ORC into a background process to generate maximum electricity without increasing the hot utility usage. We propose a two-step method to integrate an ORC to a background process, optimally considering the modifications of the ORC to increase the thermal efficiency and heat recovered by the working fluid simultaneously. The first step is to determine the configuration (turbine bleeding, regeneration, superheating) and operating conditions (working fluid flowrate, evaporation and condensation temperatures, turbine bleed ratio, degree of superheat, bleeding pressure). The second step is to synthesize the heat exchanger network by minimizing the number of heat exchangers that keep the hot utility unchanged. A well-studied example from the literature is solved to demonstrate the effectiveness of the proposed model for industrial waste heat recovery. The net power output in this paper is improved by 13% compared with the best known previous literature design for this system. The proposed method is also useful for quickly screening working fluids while considering integration potential. Screening of several working fluids revealed that using R601 (n-pentane) in place of the original working fluid (n-hexane) can increase the power output of the example system by an additional 14%.
Computers & Chemical Engineering | 2017
Wei Wan; John P. Eason; Bethany L. Nicholson; Lorenz T. Biegler
Abstract Direct transcription of dynamic optimization problems, with differential-algebraic equations discretized and written as algebraic constraints, can create very large nonlinear optimization problems. When this discretized optimization problem is solved with an NLP solver, such as IPOPT, the dominant computational cost often lies in solving the linear system that generates Newton steps for the KKT system. Computational cost and memory constraints for this linear system solution raise many challenges as the system size increases. On the other hand, the linear KKT system for our dynamic optimization problem is sparse and structured, and can be permuted to form a block tridiagonal matrix. This study explores a parallel decomposition strategy for block tridiagonal systems that is based on cyclic reduction (CR) factorization of the KKT matrix. The classical CR method has good observed performance, but its numerical stability properties need further study for our KKT system. Finally, we discuss modifications to the CR decomposition that improve performance, and we apply the approach to four industrially relevant case studies. On the largest problem, a parallel speedup of a factor of four is observed when using eight processors.
Archive | 2016
Alexander W. Dowling; John P. Eason; Jinliang Ma; David C. Miller; Lorenz T. Biegler
The application of “systems-based tools’’ including exergy/pinch analysis and process simulation has facilitated increases in the thermal efficiency of ambient pressure oxycombustion coal-fired power systems with carbon capture from 36 to 39–40 %LHV, while also considering capital costs. This corresponds to a decrease in the energy penalty 10 %-points to 6–7 %-points (absolute), relative to reference air-fired coal power plants without CO2 capture (46 %LHV). These efficiency improvements are primarily due to tailored next-generation air separation systems and plant-wide heat integration. Furthermore, oxycombustion power systems are an ideal candidate for numerical optimization, given the complex interactions between its five subsystems. This chapter extensively surveys the oxycombustion literature and summarizes four key design questions. A new, fully equation-based, flowsheet optimization framework is then introduced and applied to three oxycombustion-related case studies: design of a minimum energy air separation unit to produce an O2 enriched stream for the boiler, optimization of the CO2 polishing unit and compression train to minimize specific energy, and maximization of thermal efficiency in the oxy-fired steam cycle using a hybrid 1D/3D boiler model.
Computer-aided chemical engineering | 2015
John P. Eason; Lorenz T. Biegler
Abstract As advanced simulation technologies mature, it is desirable to utilize their predictive power in process design and optimization. However, these models present unique challenges for traditional optimization frameworks. Simulations can be very computationally expensive and derivative information may be unavailable. A common approach is to construct reduced order models (RMs) that approximate the behavior of the simulation with a simple algebraic form. By substituting a RM in place of the simulation, one may simultaneously optimize the entire flowsheet with traditional, derivative-based techniques. Using trust region concepts from nonlinear programming, we systematically construct a series of local RMs to help ensure convergence of the overall problem with the complex simulation. We compare two methods of handling the constraints: a penalty method capable of handling inequality constraints, and a novel approach using a filter method that is capable of handling both inequality and equality constraints. The advantage of the filter method is demonstrated and this method is applied to two flowsheet optimization examples: the Williams-Otto process and the ammonia synthesis process.
Archive | 2018
John P. Eason; Jiayuan Kang; Xi Chen; Lorenz T. Biegler
Abstract In equation-oriented process optimization, thermodynamic properties are often challenging to model. Thermodynamic property models tend to be highly nonlinear and must handle complex issues such as root selection and phase identification. In this work, we propose a method for using simple surrogate equations of state (SEOS), fit locally, to derive the thermodynamic state functions of the system. This can be fit to data derived from more complex thermodynamic models or directly from experiments. The SEOS are built to complement the equation-oriented optimization paradigm, while enabling flexibility to model diverse thermodynamic behavior. While still a data-driven model, surrogate equations of state offer better extrapolation potential than general regression of thermodynamic properties. A parameter estimation problem is formulated to determine the parameters of the SEOS, and the resulting model is applied to optimize a flash model encountered in a polymerization process. Finally, we discuss the potential to incorporate SEOS in a trust region framework to adaptively manage the approximation error.
Energy | 2017
Haoshui Yu; John P. Eason; Lorenz T. Biegler; Xiao Feng
Aiche Journal | 2016
John P. Eason; Lorenz T. Biegler
Applied Energy | 2016
Haoshui Yu; Xiao Feng; Yufei Wang; Lorenz T. Biegler; John P. Eason
International Journal of Greenhouse Gas Control | 2016
Jinliang Ma; John P. Eason; Alexander W. Dowling; Lorenz T. Biegler; David C. Miller
Energy Procedia | 2014
Alexander W. Dowling; John P. Eason; Jinliang Ma; David C. Miller; Lorenz T. Biegler