Elsevier

Minerals Engineering

Volumes 43–44, April 2013, Pages 112-120
Minerals Engineering

Modelling and simulation of dynamic crushing plant behavior with MATLAB/Simulink

https://doi.org/10.1016/j.mineng.2012.09.006Get rights and content

Abstract

Every process is subjected to changes in performance and efficiency over time. These dynamics can originate upstream and be inherent through the process or occur anywhere in the downstream process. Traditional plant simulations are performed with steady-state simulation, which are limited to give the performance in an ideal situation. However, plant performance usually tends to deviate away from the predicted plant performance. These dynamics are usually consequences of an altered state of the plant due to factors such as natural variation, unmatched, inappropriate or degrading equipment performance and/or stochastic events.

This paper presents a novel approach for simulating dynamic plant behavior and evaluating effects from process modification through dynamic simulations with MATLAB/Simulink. An example of an existing crushing circuit is used to illustrate the functionality and the advantage of using a dynamic simulator. The results and knowledge gained from the simulation can provide a base for optimizing a robust production output in the form optimal utilization, energy efficiency or higher product quality.

Highlights

► Every process is subjected to changes in performance and efficiency over time. ► Identifying problems requires detailed information about the process. ► Plant performance saturation depends on how it is configured and controlled. ► Process simulations revealed possible improvements, 4.7% in one scenario. ► Empirical study confirms simulation result with approx. 4.9% increase in throughput.

Introduction

Crushing plant’s design relay on accurate plant simulations. Crushing plants are designed to be able to produce certain throughput on predefined specification (i.e. a certain particle size distribution) and a certain particle size distribution while operating at a reasonable cost and at efficient energy consumption.

Equipment manufactures as well as plant designers use software packages for predicting the plant performance. There are a number of software packages available that are able to predict plant performance. The most widely used type of simulations is steady state simulations, meaning that the system is considered to be at equilibrium with all time derivatives exactly zero. Examples of steady state simulation packages include: Plantdesigner (Sandvik), Bruno (Metso), JKSimMet (JKSimMet), Aggflow (BedRock Solution) and UsimPac (Caspeo).

An interest in more dynamic simulations has been growing in minerals processing (Napier-Munn and Lynch, 1992, Liu and Spencer, 2004, Smith, 2005, Reynolds, 2010). Examples of available software that can perform dynamic simulations include Simulink (Mathworks) SysCAD (Kenwalt), Aspen Dynamics (Aspentech) and Dymola (Dassault Systémes). Even though plants experience a steady-state condition under certain circumstances, it is inaccurate to assume that the system is steady under all circumstances. It is the authors’ opinion that crushing plants seldom operate under steady conditions during longer time periods. Crushing is a continuous process; as a continuous system, equipment is subjected to variations and changes over time. These variations can be caused by: natural variation, unmatched, inappropriate or degrading equipment performance, stochastic events and more which are common in daily operations.

A development of a simulator which is capable of representing the dynamic behavior in crushing plant is ongoing at Chalmers University of Technology. The purpose of the simulator is to get more detailed simulation tool which can be used for: evaluating plant performance, control development and operator training. This paper aims to describe the developed simulator and the methodology for evaluating dynamic plant performance by introducing mechanical process modifications. All models and layouts have been modelled using the MATLAB/Simulink software.

Section snippets

Method

Crushing plants like any other production process are affected by changes over time. To be able to predict the dynamic behavior of any system an understanding about the entities and interaction there in between is essential. System complexity is depending on the level of detail. Simple models are single input single output (SISO) but that is seldom the case in reality, actual systems are often complex with multiple input, where an output (variable x) is linked to multiple input variables (u1,  , u

Modelling

In order to simulate an entire system, a plant, the models are connected together according to the user preference and configured with a set of defined parameters. The models share the same type of connection, so any unit can be connected together with ease and material properties are inherent for subsequent units. The modelling has been done by using MATLAB/Simulink.

Simulink is a commercial simulation software developed for simulating and analyzing dynamic and discrete systems, which is widely

Simulation

To illustrate the possibilities with dynamic simulation a reference plant was modelled. This particular section shown in Fig. 4 displays the dry section of an actual platinum plant. This plant was design an constructed to be able to handle 1400 TPH but due to number of factors the plant is only able the handle in average 700–1200 TPH. A steady-state simulation did not give any indication of problems with the process, probably since the three crushers are all feed from the same source of

Validation

The simulations results were validated with actual process modifications with the suggested changes. The process modifications were made and the process allowed to normalize. The process performance was later analyzed from when the process is operating. Increasing the throw of crusher 3 from 38 mm to 44 mm, enabled higher capacity of that particular crusher and better equipped it to handle the amount of recirculating load. By increasing the throw the plant was able to process approximately 1351

Conclusions

As can be seen in the plant example it was demonstrated that for this particular plant the plant reaches performance saturation under specific load. By evaluating and simulating process modification the theoretical plant performance was increased by up to 13.3%. The empirical test revealed increased plant performance of the magnitude of 1.6% resp 4.9% for the two different scenarios, increasing the actual maximum plant performance.

One of the main sources of dynamics in the simulation was caused

Acknowledgements

This work has been performed within the Sustainable Production Initiative and the Production Area of Advance at Chalmers; this support is gratefully acknowledged.

The authors wish to thank the Hesselman Foundation for Scientific Research and the Swedish national research program MinBaS (Minerals, Ballast and dimensional Stone) for its financial support.

Anglo Platinum and their personnel at Mogalakwena are gratefully acknowledged for all of their support and efforts to make this work possible.

References (19)

There are more references available in the full text version of this article.

Cited by (0)

View full text