In experimentally intensive disciplines such as bioinformatics and medicine, scientific research often relies on experimental data from other teams or researchers for further investigation. However, the current experimental data sharing scheme is not yet mature and has difficulty effectively protecting the privacy and security of data. This may not only lead to the misuse of research data and privacy breaches for researchers but also limit the efficient sharing and reuse of data. Traditional data sharing methods face risks of privacy breaches, data abuse, and tampering and have difficulty meeting the personalized needs of different data requesters. This article proposes a fine-grained experimental data sharing scheme based on blockchain, aiming to achieve flexible and secure data sharing. This solution combines blockchain technology, function encryption, proxy re-encryption, and distributed storage IPFS to build a data sharing framework with high privacy protection and dynamic access control capabilities. Through smart contracts, data requesters can autonomously choose function encryption or proxy re encryption according to their needs, thereby achieving secure access to data calculation results or raw data. The experimental results show that the model improves the flexibility and efficiency of data sharing while ensuring data integrity and privacy. This approach can meet the needs of experimental data sharing in multiple scenarios and has broad application potential. This study provides new ideas and technical support for the secure sharing of future data and has important theoretical and practical value.