In bio-inspired design, identifying keywords is an important step of biological information searching. Based on existing information retrieval approaches, the amount of relevant biological information in a search result largely depends on the identified keywords. Due to the limitation of biological knowledge, design engineers are difficult to identify the appropriate keywords that can find the biological information related to engineering requirements. To address this issue, we present an algorithm that can calculate the Composite Correlation Intension of functionally combined words, which integrates semantic similarity computation, data normalization, and corpus technology. Based on the algorithm, a method that automatically pushes keywords for biological information searching in bio-inspired design is also proposed. The method decomposes engineering requirements and structures functionally combined words, calculates the Composite Correlation Intension values of all functionally combined words, ranks the functionally combined words by the Composite Correlation Intension values, takes the functionally combined words with larger Composite Correlation Intension values as keywords, and pushes them. Through these keywords, design engineers retrieve the relevant biological information and discover the required design knowledge. In order to facilitate the use of proposed method, an auxiliary tool was developed in Python environment. Finally, the possibility of proposed method was demonstrated by a preliminary validation and an application case. The results show that the proposed method would be a promising alternative to identify keywords for biological information searching in bio-inspired design.