Coal Preparation » Process Control
The objective of this project is to develop an effective and inexpensive Electrical Impedance Spectroscopy (EIS) based technique for on-line monitoring of the operation of DMCs in order to improve timely detection of poor operating conditions and reduce coal loss.
Due to the lack of appropriate instrumentation which requires minimum operator/maintenance attention, dense medium cyclones are usually operated without monitoring density differential, ratio of medium to coal and concentration of other fine solids in the medium. In order to identify problem areas and improve separation efficiencies, density tracer techniques have been utilized as a tool to detect faults and unusual operating conditions. However, density tracer tests cannot be performed routinely due to cost and time restraints. Therefore, if there are faults and/or unusual conditions found by the tracer test, a significant amount of product coal would have been lost. It is apparent that there is a need for an online monitoring instrumentation for dense medium cyclone circuits, which is low cost and requires little attention.
In this project, an innovative measurement technique based on the intelligent data analysis of electrical impedance spectra has been developed for on-line monitoring systems of the phase composition of slurry mixtures in coal preparation processes. The electrical impedance spectrum, on which the measurement technique is based, has been found to be a measurement variable sensitive to the phase composition of slurry mixtures. It has been identified that a Back Propagation Artificial Neural Network (BPANN) combined with a principal component analysis (PCA) network for feature extraction is an effective and reliable algorithm for estimating the phase composition from the measured electrical impedance spectrum. It has been demonstrated in experiments carried out in both a desk-top stirred tank and a pilot-plant scale DMC circuit that the EIS based technique is capable of measuring the medium density of slurry mixtures containing water, magnetite, coal and/or sand and the volumetric fraction of coal or sand in the dense medium with a high accuracy.
Due to the relatively simple hardware involved, the EIS based technique would have advantages of high reliability, low cost, low maintenance and amenability to multiplexing. The benefit to coal producers is an improvement of the operating efficiency of the dense medium cyclone (DMC) circuit through
- improved monitoring and control of medium density and stability
- improved monitoring and control of the ratio of medium to coal
- timely detection of poor operating conditions
In addition to the above-mentioned benefits, information from the EIS based technique would also permit diagnostic work to be conducted on plant issues, such as bias in distribution of feed to DMCs, high level of contaminating fine solids in circulation medium, excessive wear in DMC, or pump and drain screen, and problems in the magnetite recovery circuit.
The second stage of this project involved substantial modification to the electrodes and approaches to calibration.
In order to minimize the calibration error, the EIS data for electrode calibration need to be measured over a wide range of well controlled set points of operating conditions, including temperature and medium density. The initial design of electrode assembly and the associated calibration rig make it difficult to efficiently change the set points and control the variations of operating conditions within allowable limits (e.g. ±0.5 0 C/min around a set point of temperature) during the measurement of an EIS spectrum. These problems were then solved by redesigning the electrode assembly and the associated calibration test rig.
The new design of the electrode assembly consists of a flow guide and a 4-electrode configuration. The EIS technique for measuring medium density and other solids content has been significantly improved by using this 4-electrode configuration as it simplifies the approach for EIS data analysis and has the following advantages:
- easily fitted into existing plants without modifications to existing equipment
- low cost due to its simple configuration and no need for pipe sections
- fast and reliable calibration
- standardized design applicable to any similar locations
It should be noted that the temperature variation during the measurement of an EIS spectrum in coal preparation plants (each measurement of an electrical impedance spectrum takes 20 to 60 seconds) is not an issue as the temperature variation rate in the plants is hardly higher than 1 0 C/min due to the large heat capacity of plant slurry system. Moreover, the new approach for analyzing EIS data from 4-electrode configuration can remove the influence of temperature on the prediction of the volumetric fraction of magnetite.
A new approach for analyzing EIS data from 4-electrode configuration has been developed. In this approach, the volumetric fraction of magnetite can be directly correlated with the slope of normalized impedance versus log(f) at the frequency of 100kHz without influence by temperature and water conductivity. The nearly linear relationship between the volumetric fraction of magnetite and the slope even makes it possible to calibrate the electrode assembly on plant sites.