Mental health analysis plays a significant role in the process of comprehensive judgment, which lies in students’ mental health and future learning. Therefore, it is necessary to analyze it in depth, especially when dealing with complex problems. In this paper, this paper proposes an applied research scheme in mental health analysis based on cluster analysis algorithm to better solve this problem. The scheme accurately locates the influencing factors through the similarity measurement theory, so as to determine the division of indicators, and uses the cluster analysis algorithm to construct the scheme. Experimental results show that under certain evaluation criteria, the accuracy and speed of the scheme is significantly improved for different problems, and it has better performance. Therefore, the simulation scheme based on cluster analysis algorithm can better solve the limitations of the traditional improved whale algorithm and improve the simulation accuracy and efficiency in the application research of college students’ mental health analysis.