Showing papers by "Debi Prasad Das published in 2020"

Monitoring the fill level of a ball mill using vibration sensing and artificial neural network

Abstract: Ball mills are extensively used in the size reduction process of different ores and minerals. The fill level inside a ball mill is a crucial parameter which needs to be monitored regularly for optimal operation of the ball mill. In this paper, a vibration monitoring-based method is proposed and tested for estimating the fill level inside a laboratory-scale ball mill. A vibration signal is captured from the base of a laboratory-scale ball mill by using a ± 5 g accelerometer. Features are extracted from the vibration signal by using different transforms such as fast Fourier transform, discrete wavelet transform, wavelet packet decomposition, and empirical mode decomposition. These features are given as input to an artificial neural network which is used to predict the percentage fill level inside the ball mill. In this paper, the predicted fill level obtained by using different features are compared. It is found that the predicted fill level due to features obtained after fast Fourier transform outperforms other transforms.

Estimation of hydrogen flow rate in atmospheric Ar:H 2 plasma by using artificial neural network

Abstract: Atmospheric Ar:H2 plasma is an eco-friendly option for the reduction of metal oxides. For better reduction performance and safety concern, the hydrogen gas injected into the reactor should be monitored. A hydrogen flow rate estimation system is presented in this paper by using an artificial neural network (ANN) model fed with features of optical emission spectra of the plasma. ANN models are studied with two different sets of input, i.e. for the first case the inputs to the model are the three features of Hα line such as the peak intensity count, full-width half maximum and area under Hα line, while for the second case, the peak intensity count of a group of emission lines like Hα, Ar I, O I, K I, Na D lines are considered as the inputs. ANN model is developed for estimating four different sets of hydrogen flow rates 5, 8, 10 and 12 litres per minute (lpm) when the argon flow rate is constant at 10 lpm. For both the input features, the model performances are compared, and it is shown that improved estimation accuracy is observed from the second case, i.e. from peak intensity count of a group of emission lines instead of only hydrogen emission line.

An Efficient Narrowband Active Noise Control System for Accommodating Frequency Mismatch

Abstract: The conventional narrowband active noise control (ANC) is a popular method for reducing narrowband noise generated from rotating machines like engines, fans, blowers, and power transformers. The narrowband active noise control works efficiently only when the internal reference frequency of the controller and the frequency of the primary noise remains the same. Any change in the frequency of the primary noise from that of the reference is termed as frequency mismatch (FM), which degrades the narrowband ANC performance. In this paper, a filtered-x weighted-frequency Fourier linear combiner least mean square (FX-WFLC-LMS) algorithm is developed for narrowband ANC system. This algorithm is capable of adapting to both frequency and amplitude variations in the primary noise. To reduce the computational burden of the proposed FX-WFLC-LMS algorithm, a computationally efficient filtered-error weighted-frequency Fourier linear combiner least mean square (FE-WFLC-LMS) algorithm is also proposed. The comparative performance of these proposed algorithms is evaluated through extensive simulation and real-time experiments. It was found that both these proposed algorithms are capable of correcting any amount of frequency mismatch and are suitable for narrowband ANC systems.

Prevention of Hydrocyclone Choking Through Detection of Sub-Hz Frequency Shift of Vibration Signal

Abstract: Hydrocyclones are extensively used in mineral processing and other industries to separate heavy and coarse particles from light and fine particles, respectively. Due to variation in the density of feed which contains water and particles, the hydrocyclone gets choked during continuous operation. Clearing the choking is a time-consuming and expensive affair which requires shutdown of the hydrocyclone. To prevent this, the hydrocyclone operation needs to be monitored continuously to alarm well before choking happens. A vibration sensing and analysis-based method is proposed in this paper to predict the choking condition well in advance. Vibration signals were collected by varying the solid percentage in the feed input to create choking condition. Vibration signal was processed using Fourier transform to observe the variations under different feed concentrations. It was observed that higher slurry density in the feed input leads to choking of hydrocyclone. Vibration signal which is captured at the apex of the hydrocyclone is recommended here to be used to monitor the operational conditions of the hydrocyclone. After analysis of the captured vibration signals in frequency domain, it was found that there is shift of specific frequencies due to an increase in slurry density. By monitoring the shift in peak frequencies in those frequency ranges, the choking can be detected and cleared well in advance to avoid a shutdown of hydrocyclone.

Detection of Economic Minerals in Beach Placer Samples by Machine Learning-Based Microscopic Image Processing

Abstract: Beach sand minerals (BSM) contains different economic minerals, out of which seven minerals viz. Ilmenitev, Rutile, Leucoxene, Zircon, Sillimanite, Garnet and Monazite are important ones. These minerals have different physical properties based on which these are separated conventionally. In this paper, a separation process using visual information is proposed. Accordingly, microscopic images of sand particles are taken and using machine learning algorithm, the particular economic mineral is accurately detected in the microscopic images. Aggregated Channel Features (ACF) detector-based learning algorithm is used in this paper.