In the era of information networking, search engine has become an indispensable tool for people to get information. Additionally, building the evaluation system of search engine can not only provide reference for users to choose a search engine which is most suitable for themselves, but also provide direction guidance for search engine providers to optimize search engine. In this paper, the evaluation of search engine information retrieval ability is studied from the point of view of search engine service quality. A set of more objective and effective search engine evaluation tools is established by integrating grey relational analysis (GRA) and fuzzy AHP. Among the rule hierarchy indicators of search engines, information quality has the greatest weight, followed by system quality, and value-added service quality has the lowest proportion. In terms of specific search engine selection, in general, college students choose Google the most whose scores in information quality and value-added service quality are higher than those of Baidu and Sogou.