Abstract In order to test the performance of pump turbine, it is necessary to collect and analyze the performance indexes such as energy characteristics, stability performance, cavitation performance and force characteristics of the unit through the pump turbine model test test system. however, the current test systems all use manual operation based on the standard specification requirements to handle the data collection and analysis of the test, and the accuracy of the data relies on human experience and the efficiency is not high. The method described in this paper uses emerging computer test technology means such as big data technology, noise and image recognition technology to automatically and efficiently complete the test data test, and quickly complete the test test, while the test results are more accurate.