This paper presents the development trend and application prospects of big data-based condition monitoring of power apparatus. Based on the analysis of big data attributes of condition monitoring data, three big data application directions of condition monitoring system are discussed, including big data technologies, theories and algorithms. A universal architecture of a big data-based condition monitoring system is presented. Data acquisition, data denoising, feature extraction, pattern recognition and data mining are discussed in detail to present the relationships of big data and all procedures of a condition monitoring system. An application example of a comprehensive condition monitoring system is presented, including big data-based data fusion, condition evaluation and failure-prediction. The application of big data based condition monitoring has great potential to overcome the challenges of condition evaluation and failure prediction and will contribute to the development of intelligent condition monitoring systems of power apparatus. This work is supported by National Natural Science Foundation of China (No. 51807072).