基于线性SVM 算法的云数据中心蓄电池状态预测.

Aimed at the problem that valve regulated lead-acid (VRLA) battery aging threatening power supply reliability in cloud data centers, a method of an online battery state prediction method based on machine learning algorithms was proposed. Predicting battery status through classic machine learning algorithm models, comparative experiments were conducted on the prediction accuracy between models without extracted feature values and models with extracted feature values. Compare the prediction accuracy of the model using unrefined feature values with the model extracting feature values. The prediction accuracy of the model without extracted feature values ranges from 71. 72% to 81. 82%, and after extracting feature values, the prediction accuracy has improved by 10. 10% to 20. 20%, extracting feature values can improve prediction accuracy. The prediction model based on linear SVM method for extracting feature values is superior to other algorithms, with an accuracy of 96. 46%. Experiment results show that machine learning algorithm based prediction models can be used for online VRLA battery state prediction in cloud data centers. [ABSTRACT FROM AUTHOR]