Biomedical event representation can be applied to various language tasks. A biomedical event often involves multiple biomedical entities and trigger words, and the event structure is complex. However, existing research on event representation mainly focuses on the general domain. If models from the general domain are directly transferred to biomedical event representation, the results may not be satisfactory. We argue that biomedical events can be divided into three hierarchies, each containing unique feature information. Therefore, we propose the Triple-views Event Hierarchy Model (TEHM) to enhance the quality of biomedical event representation. TEHM extracts feature information from three different views and integrates them. Specifically, due to the complexity of biomedical events, We propose the Trigger-aware Aggregator module to handle complex units within biomedical events. Additionally, we annotate two similarity task datasets in the biomedical domain using annotation standards from the general domain. Extensive experiments demonstrate that TEHM achieves state-of-the-art performance on biomedical similarity tasks and biomedical event casual relation extraction.