The conventional methods for verifying the integrity of financial audit data mainly rely on manual review and simple statistical methods, resulting in a small amount of verification data. To address this issue, a financial audit data integrity verification method based on differential evolution algorithm was designed. This method first involves comprehensive collection of financial audit data to ensure its comprehensiveness and accuracy. Based on differential evolution algorithm, feature extraction is performed on the collected financial audit data. Through algorithm optimization and iteration, key features and patterns in the data are accurately identified. Use the extracted data characteristics to build a financial audit data integrity verification model, which will be used as the core tool of data verification to detect whether there is any abnormality or tampering in the data. Once again, achieve integrity verification of financial audit data to ensure its completeness and authenticity, providing strong data support and guarantee for financial audit work. Finally, the recovery process of financial audit data is studied, including data block acquisition, starting position localization, unit data segment partitioning, and selecting the optimal recovery result through SVM classifier to ensure the accuracy and completeness of audit data. The experimental results show that the designed financial audit data integrity verification method based on differential evolution algorithm has a higher number of verifications on multiple datasets than Method 1 and Method 2, with a maximum of 980 verifications. This result indicates that the design method has high verification efficiency and is more suitable for verifying the integrity of financial audit data.