Abstract
Four machine learning algorithms (artificial neural networks, a naive Bayes' classifier, a support vector machines and decision trees) wwere applied for a single pump station mine to monitor and predict the dam levels and energy consumption.