Abstract: With the continuous development of mine automation and intelligence,smart mines have become an important direction for the transformation and upgrading of the mining industry. Energy consumption optimization and anomaly detection are key links in the construction of smart mines,which are crucial for the safety production and economic benefits of mines. A method for mine energy consumption prediction and anomaly detection based on ensemble learning strategy is proposed. Firstly, an energy consumption prediction model based on ensemble learning is constructed. This model uses historical energy consumption data and real-time monitoring data,and predicts and optimizes mine energy consumption through integrating algorithms such as support vector machine,random forest,and neural network. Secondly,an anomaly detection method based on ensemble learning is proposed. This method integrates algorithms such as isolation forest,local anomaly factor and autoencoder to detect anomalies in mine energy consumption data. The proposed method is applied to a mine,and the results show that compared with a single method,this method can effectively reduce mine energy consumption and improve the accuracy of anomaly detection, providing technical support for the intelligent construction of mines. 

Key words: smart mine,integrated learning,anomaly detection,support vector machine,random forest,neural network