US 12,277,549 B2
Blockchain-based transaction system for green certificate
Dong Wang, Beijing (CN); Wei Jiang, Beijing (CN); Da Li, Beijing (CN); Jiaxing Xuan, Beijing (CN); Guomin Li, Beijing (CN); Hejian Wang, Beijing (CN); Xin Shi, Beijing (CN); Jiangtao Li, Beijing (CN); Zhan Su, Beijing (CN); Lei Zhou, Beijing (CN); Lihua Zhao, Beijing (CN); and Fan Jia, Beijing (CN)
Assigned to STATE GRID BLOCKCHAIN TECHNOLOGY (BEIJING) CO., LTD., Beijing (CN); State Grid Digital Technology Holding CO., LTD., Beijing (CN); and State Grid Corporation of China, Beijing (CN)
Appl. No. 17/784,627
Filed by STATE GRID BLOCKCHAIN TECHNOLOGY (BEIJING) CO., LTD., Beijing (CN); State Grid Digital Technology Holding CO., LTD., Beijing (CN); and State Grid Corporation of China, Beijing (CN)
PCT Filed Jan. 21, 2022, PCT No. PCT/CN2022/073319
§ 371(c)(1), (2) Date Jun. 10, 2022,
PCT Pub. No. WO2022/206143, PCT Pub. Date Oct. 6, 2022.
Claims priority of application No. 202110331886.8 (CN), filed on Mar. 29, 2021.
Prior Publication US 2024/0185233 A1, Jun. 6, 2024
Int. Cl. G06Q 20/38 (2012.01); G06F 16/23 (2019.01); G06F 16/27 (2019.01); G06F 21/60 (2013.01); G06F 21/64 (2013.01); G06Q 20/40 (2012.01); G06Q 30/0201 (2023.01); G06Q 30/08 (2012.01); G06Q 40/04 (2012.01); G06Q 50/06 (2024.01)
CPC G06Q 20/38215 (2013.01) [G06F 16/2393 (2019.01); G06F 21/602 (2013.01); G06F 21/64 (2013.01); G06Q 20/389 (2013.01); G06Q 20/401 (2013.01); G06Q 30/0206 (2013.01); G06Q 30/08 (2013.01); G06Q 50/06 (2013.01); G06F 16/27 (2019.01); G06Q 40/04 (2013.01)] 9 Claims
OG exemplary drawing
 
1. A blockchain-based transaction system for a green certificate, comprising an on-chain node and an off-chain node, wherein the on-chain node comprises an audit node and a proxy node, and the proxy node is used by a power producer and a purchaser for login;
the audit node is configured to verify qualification information of the power producer on the proxy node, and send a green certificate to the power producer on the proxy node when the qualification information satisfies a preset condition;
the off-chain node is configured to: perform calculation by using a preset model based on a historical transaction price, to obtain a first eigenvector for predicting a transaction price of a next transaction; map target features of each power producer on a current blockchain platform into implicit vectors, wherein the target features comprise a power generation type, current transaction information, and current information of the green certificate; aggregate the implicit vectors by using an objective function, to obtain a second eigenvector, wherein the objective function is a function for ensuring permutation invariability; and calculate the first eigenvector and the second eigenvector by using a fully connected neural network, to obtain the transaction price of the next transaction, wherein the transaction price is used to provide a reference for the power producer and the purchaser to determine a price;
the proxy node is configured to: when receiving sale information of the power producer and purchase information of the purchaser, send the sale information, a digital signature of the power producer, the purchase information, and a digital signature of the purchaser to the off-chain node;
the off-chain node is configured to match the sale information and the purchase information, send successfully matched transaction information to the proxy node, and return the sale information and the purchase information to a front end in a visual manner for display;
the proxy node is further configured to send the successfully matched transaction information to a transaction smart contract, and the transaction smart contract performs transaction processing based on the transaction information; and
when calculating the first eigenvector and the second eigenvector by using the fully connected neural network, the off-chain node is specifically configured to:
stitch the first eigenvector and the second eigenvector to obtain a stitched eigenvector, and input the stitched eigenvector into the fully connected neural network to obtain a predicted transaction price.