ABSTRACT

In India, coal is one of the prime sources of energy used in the power generation and metallurgy sector. The processing of coal below 3 mm is not successfully carried out in India. The quality of coal below 3 mm can be improved by decreasing the coal’s particle size, which reduces the ash percentage of coal. Screening is one of the significant beneficiation techniques used to reduce the size fraction of coal. The difficult to process coal of size −3 + 1 mm was selected in the present work. In this work, an attempt has been made to screen the coal of size −2 + 1 mm from −3 + 1 mm using a 2 mm screen mesh in the vibrating screen generated at different moisture content, angle, and frequency of the deck. The performance of the vibrating screen was evaluated using screening efficiency. Furthermore, prediction using a feed backward artificial neural network (ANN) model was developed on the experimental results for ten different neuron conditions. From the results, it was clear that the prediction results obtained from the ANN model were in good correlation with the experimental results.

摘要

在印度，煤炭是发电和冶金行业使用的主要能源之一。印度未能成功加工 3 毫米以下的煤。3毫米以下的煤质量可以通过降低煤的粒度来改善，从而降低煤的灰分。筛选是用于降低煤粒度分数的重要选矿技术之一。在目前的工作中选择了尺寸为-3 + 1 mm 的难以加工的煤。在这项工作中，尝试使用在不同含水量、角度和甲板频率下产生的振动筛中的 2 mm 筛网从 -3 + 1 mm 筛分尺寸为 -2 + 1 mm 的煤。使用筛分效率评估振动筛的性能。此外，基于十种不同神经元条件的实验结果，开发了使用反馈人工神经网络 (ANN) 模型的预测。从结果可以看出，从人工神经网络模型得到的预测结果与实验结果具有很好的相关性。