US Patent 11792136 Systems and methods for blockchain network optimization

In some aspects, a computing system may use time series data and machine learning to determine an efficient time to send data to a blockchain. A machine learning model may use a variety of blockchain related data to predict a network usage costs for different times in the future. The predicted network usage costs may be used to determine when data should be sent for storing on the blockchain to reduce network resource costs for a computing system. For example, based on inputting the blockchain data into the machine learning model, a computing system may generate output indicating network usage costs for a future time period, with each network usage cost corresponding to a timestamp within the future time period. The computing system may determine a minimum network usage cost of the network usage costs and send data to a node in the blockchain network at the corresponding timestamp.