Achieving Maximum Value from Steam at a Pulp & Paper Mill by Implementing Advanced Control, 2015 PEERS Conference

ARE YOU A TAPPI MEMBER? TAPPI members have exclusive, FREE access to technical conference papers and presentations six months after the conference in TAPPI's e-library. The e-Library offers:

• Unlimited access to more than 18,000+ documents
• Fast, robust search engine for fast search results
• Members get FREE access to conference proceedings, TAPPI JOURNAL articles, Paper360º articles, archived Solutions! articles, and much more
• View thousands of technical paper abstracts
• Ability to search by keyword, title, author, events or industry segment

*Technical papers and presentations are available for sale immediately following the conferences before the 6 month embargo period.

Pulp and paper mills are some of the largest energy consumers of all industries. In recent decades, significant progress has been made to reduce fuel costs by switching to renewable biomass sources. Mills strive to become more energy efficient and constantly search for new technologies to maximize utilization of fuel and steam. In particular, the complexity of mill steam networks yields a challenging optimization problem in effective control. Therefore, steam network control is one of the most beneficial applications for the latest evolution of advanced process control. The goal of this study was to optimize production and distribution in the steam network at a pulp and paper mill by applying a multivariable model-based controller with both predictive and adaptive natures. This paper describes the installation, commissioning, and operational performance of a multivariable predictive controller at a pulp and paper mill in South America. Results included over 80 % reduction in steam losses (released to atmosphere) due to improved distribution of the steam throughout the network, reduction of start-up time, reduction of fuel consumption in auxiliary boilers, reduction of the need for manual operator adjustment with automatic control, and ROI measured in a few months.