Process Observers and Data Reconciliation Using Mass and Energy Balance Equations - Advanced Control and Supervision of Mineral Processing Plants - page 15

Chapter 2
Process Observers and Data Reconciliation
Using Mass and Energy Balance Equations

Daniel Hodouin

Abstract This chapter is devoted to data reconciliation for process audit, diagno-
sis, monitoring, modeling, advanced automatic control, and real-time optimization
purposes. The emphasis is put on the constraints of mass and energy conservation,
which are used as a foundation for measurement strategy design, measured value
upgrading by measurement error filtering techniques, and unmeasured process vari-
ables estimation. Since the key variables in a mineral processing unit are usually
flowrates and concentrations, their reconciliation with the laws of mass conserva-
tion is central to the discussed techniques. Tools are proposed for three different
kinds of operating regimes: steady-state, stationary and dynamic. These reconcilia-
tion methods are based on the usual least squares and Kalman filtering techniques.
Short examples involving grinding, flotation, leaching and thermal processes are
presented to illustrate the problems of data reconciliation, sensor placement, fault
detection and diagnosis. Strategies for coupling data reconciliation with real-time
optimization and automatic control techniques are also proposed. A nomenclature
section is included at the end of the chapter

2.1 Introduction

The production goal of a mineral or metallurgical plant (MMP) is ideally to maintain
complex unit operating conditions at values where some plant performance index is
optimized. The performance index could be expressed either by technical factors,
such as the tonnage of valuable material produced, or by the quality of the material
produced ( e.g. , concentrate grade or metal purity). More globally, since a trade-off
between the productivity, the material quality, and the production costs is required,