<p>In the field of smart grid, power suppliers control the allocation and pricing of electricity by real-time aggregation of meter readings from different regions. However, these meter readings often contain sensitive data of users. If the meter readings are directly used for data aggregation, the user’s lifestyle and identity information can be exposed. Therefore, we propose a blockchain-based privacy-preserving data aggregation scheme with robustness. In our scheme, we first use random numbers to encrypt the meter data to ensure the privacy of meter readings during data aggregation. Then, we implement the robustness mechanism based on Shamir threshold secret sharing scheme. When some smart meters fail to transmit data normally due to disconnection, damage, or other faults, or some cloud servers are attacked and unable to operate normally, the scheme can still correctly complete the data aggregation work. We also propose a distributed storage architecture using blockchain, which effectively avoids the risk of a single point of attack on the server. Compared with other data aggregation scheme with robustness, the security and performance analysis shows that our scheme is more efficient.</p>