Mass flow estimation in mineral processing applications
Väyrynen, Teemu (2013)
Väyrynen, Teemu
2013
Automaatiotekniikan koulutusohjelma
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2013-05-08
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201305221146
https://urn.fi/URN:NBN:fi:tty-201305221146
Tiivistelmä
Development and implementation of automated monitoring, control and optimization systems for mineral processing applications require accurate online mass flow measurements from the processes. The mass flow sensors used in mineral processing plants are designed to monitor the production volumes of the end products.
Belt scale is the most common online bulk material mass flow sensor used in the mineral processing. The belt scale provides an accurate online mass flow measurement from the process. However, the high unit price of the sensor prevents the installation of multiple belt scales in a single mineral processing plant. In addition to the online mass flow measurements of the belt scales, offline mass flow measurements might be also carried out at the plants. Offline mass flow sensors include wheel loader scales and truck scales. The high unit price, low measurement frequency and variable measurement delays prevent the offline mass flow sensors to be used in automated process control.
The main goals of this work are determined by the three research questions formulated for this work. The first goal is to analyse the correlations of four measurement signals of the online mass flow sensors against the reference mass flow measurement of the belt scale. The fitting of the measurement signals is performed by linear regression method. The analysis of the signals is performed by measures of fit methods, the root mean square RMSE and correlation analysis method R-squared. The second goal is to analyse the accuracies of three mass flow estimation models. The third goal is to perform comprehensive analysis of the features, benefits and restrictions of each online mass flow sensor. This analysis can be utilized, if the presented online mass flow sensors are implemented in a mineral processing plant. The online mass flow sensors used in this work are a power transducer, laser profilometer, ultrasonic sensor and strain gauge.
In order to answer the research questions, an experimental measurement setup was designed and installed at the test plant. An aggregate production plant is used as an example of mineral processing application in this work. The presented online mass flow sensors can also be applied to other mineral processing applications handling solid materials on belt conveyors.
The results of this work indicate that all of the measurement signals of the online mass flow sensors correlate well with the reference mass flow measurement of the belt scale. The mass flow estimation models of the power transducer and laser profilometer were proven accurate. The results indicate that the online mass flow measurements can be utilized more effectively in the process control and optimization of the mineral processing plants. However, a reference mass flow measurement is required for calibration of the presented online mass flow sensors. Future research proposals are presented in the field of online mass flow estimation in mineral processing.
Belt scale is the most common online bulk material mass flow sensor used in the mineral processing. The belt scale provides an accurate online mass flow measurement from the process. However, the high unit price of the sensor prevents the installation of multiple belt scales in a single mineral processing plant. In addition to the online mass flow measurements of the belt scales, offline mass flow measurements might be also carried out at the plants. Offline mass flow sensors include wheel loader scales and truck scales. The high unit price, low measurement frequency and variable measurement delays prevent the offline mass flow sensors to be used in automated process control.
The main goals of this work are determined by the three research questions formulated for this work. The first goal is to analyse the correlations of four measurement signals of the online mass flow sensors against the reference mass flow measurement of the belt scale. The fitting of the measurement signals is performed by linear regression method. The analysis of the signals is performed by measures of fit methods, the root mean square RMSE and correlation analysis method R-squared. The second goal is to analyse the accuracies of three mass flow estimation models. The third goal is to perform comprehensive analysis of the features, benefits and restrictions of each online mass flow sensor. This analysis can be utilized, if the presented online mass flow sensors are implemented in a mineral processing plant. The online mass flow sensors used in this work are a power transducer, laser profilometer, ultrasonic sensor and strain gauge.
In order to answer the research questions, an experimental measurement setup was designed and installed at the test plant. An aggregate production plant is used as an example of mineral processing application in this work. The presented online mass flow sensors can also be applied to other mineral processing applications handling solid materials on belt conveyors.
The results of this work indicate that all of the measurement signals of the online mass flow sensors correlate well with the reference mass flow measurement of the belt scale. The mass flow estimation models of the power transducer and laser profilometer were proven accurate. The results indicate that the online mass flow measurements can be utilized more effectively in the process control and optimization of the mineral processing plants. However, a reference mass flow measurement is required for calibration of the presented online mass flow sensors. Future research proposals are presented in the field of online mass flow estimation in mineral processing.