Modeling, control, and optimization of an industrial austenitization furnace

Date

2018-12

Authors

Ganesh, Hari Sai

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Steel production and processing is both energy-intensive (2% of overall energy consumption) and one of the biggest contributors to CO₂ emissions. Its use is projected to increase by 1.5 times that of present levels (around 1.6 billion metric tonnes per year) by 2050 to meet the needs of a growing population. The main goal of this research is to minimize the energy consumption of a steel quench hardening (or heat treating) process, currently in operation at an industrial partner, by mathematical modeling, optimization, advanced control, and heat integration. The quench hardening processes consists of heating pre-finished metal parts to a certain temperature in a continuously operating furnace (austenitization), followed by rapid cooling (quenching) in water, brine or oil to induce desired metallurgical properties like hardness, toughness, shear strength, tensile strength, etc. The novelty of this work lies in the two scale modeling approach considered to solve the furnace energy consumption minimization problem. We improve a previously developed two-dimensional (2D) physicsbased model of the heat treating furnace that computes the energy usage of the furnace and the part temperature distribution as a function of time and position within the furnace under temperature feedback control. We predict the effect of process variables on microstructural evolution of the parts using an empirical relation reported in the literature and their consequent effects on the metallurgical properties of the quenched product. The physics-based model combined with the empirical model is used to simulate the furnace operation for a batch of parts processed sequentially under heuristic temperature set points with a simple linear control strategy suggested by the operators of the plant. We then minimize the energy consumption of the furnace without compromising the product quality by real-time optimization (RTO), model predictive control (MPC), and heat integration using radiant recuperators. Energy savings of 3.7%, 15.93%, and 20.88% were obtained under model predictive control, heat integration, and optimized set points respectively compared to reference heuristic operation case without heat integration and MPC.

Description

LCSH Subject Headings

Citation