Blockchain-based electricity charge settlement method and system for energy storage station

A blockchain-based electricity charge settlement method for an energy storage station, comprising:

• collecting, by a trusted terminal, two-way electricity quantity data of an energy storage station, and distributing the two-way electricity quantity data to a blockchain, wherein the collecting, by a trusted terminal, two-way electricity quantity data of an energy storage station, and distributing the two-way electricity quantity data to a blockchain comprises:
• collecting, by the trusted terminal, the two-way electricity quantity data of the energy storage station, and determining whether the two-way electricity quantity data meets a precision requirement;
• and if the two-way electricity quantity data meets the precision requirement, distributing the two-way electricity quantity data to the blockchain;
• or if the two-way electricity quantity data does not meet the precision requirement, compensating the two-way electricity quantity data, replacing the two-way electricity quantity data with compensated two-way electricity quantity data, and returning to perform the step of determining whether the two-way electricity quantity data meets the precision requirement, wherein the determining whether the two-way electricity quantity data meets a precision requirement comprises:
• obtaining a sampling error value of the trusted terminal, a sampling error value of a current transformer in the energy storage station, and a sampling error value of a voltage transformer in the energy storage station, inputting the sampling error value of the trusted terminal, the sampling error value of the current transformer, and the sampling error value of the voltage transformer into a pre-trained back propagation (BP) neural network model, to obtain a predicted electricity quantity output by the pre-trained BP neural network model, and determining whether an error between the predicted electricity quantity and the two-way electricity quantity data collected by the trusted terminal exceeds a specified threshold;
• a process of training a BP neural network model comprises:
• initializing parameters of each layer in the BP neural network model;
• selecting a training sample from a sample set;
• inputting the training sample into the BP neural network model to obtain a predicted electricity quantity output by the BP neural network model, wherein the training sample comprises the sampling error value of the trusted terminal, the sampling error value of the current transformer in the energy storage station, and the sampling error value of the voltage transformer in the energy storage station;
• inputting the predicted electricity quantity and the two-way electricity quantity data of the energy storage station into a neural network error function
• E
• =
• 1
• 2
• ⁢
• ∑
• q
• =
• 1
• p
• ∑
• K
• =
• 1
• L
• (
• T
• k
• -
• O
• k
• )
• 2
• to obtain an error, wherein Ok represents the predicted electricity quantity, Tk represents the two-way electricity quantity data of the energy storage station, p represents a quantity of training samples, L represents a quantity of nodes at an output layer in the BP neural network model, K represents a node at the output layer in the BP neural network model, and E represents the error;
• determining whether an error between the predicted electricity quantity and a reference electricity quantity is within a specified range;
• and if the error between the predicted electricity quantity and the reference electricity quantity is within the specified range, ending the training;
• or if the error between the predicted electricity quantity and the reference electricity quantity is not within the specified range, transmitting the error to each layer in the BP neural network model based on an order from the output layer to an input layer in the BP neural network model, to update the parameters of each layer in the BP neural network model, and returning to perform the step of selecting the training sample from the sample set;
• and the compensating the two-way electricity quantity data comprises:
• compensating the two-way electricity quantity data by using a compensation formula
• C
• k
• =
• O
• k
• +
• (
• -
• η
• ⁢
• ∂
• E
• ∂
• W
• ki
• ⁢
• (
• T
• k
• -
• O
• k
• )
• )
• ,
• wherein Ok represents the predicted electricity quantity, Tk represents the two-way electricity quantity data of the energy storage station, Wki represents a weight of an ith node from a hidden layer to the output layer, η represents a network learning rate, and Ck represents the compensated two-way electricity quantity data;
• performing, by a consensus node of a power-consuming enterprise and a consensus node of an energy storage station investment operator in the blockchain, consensus calculation on the two-way electricity quantity data, and after reaching a consensus, selecting one of the consensus node of the power-consuming enterprise and the consensus node of the energy storage station investment operator as a first accounting node;
• processing, by the first accounting node, the two-way electricity quantity data to obtain a first target block, and writing the first target block into the blockchain;
• executing a smart settlement contract, obtaining basic settlement data from the blockchain, performing electricity charge settlement based on the basic settlement data to obtain an electricity charge settlement result, and distributing the electricity charge settlement result to the blockchain, wherein the basic settlement data at least comprises:
• two-way electricity quantity data in the first target block;
• selecting, by the consensus node of the power-consuming enterprise and the consensus node of the energy storage station investment operator in the blockchain after reaching a consensus on the electricity charge settlement result, one of the consensus node of the power-consuming enterprise and the consensus node of the energy storage station investment operator as a second accounting node;
• and
• processing, by the second accounting node, the electricity charge settlement result to obtain a second target block, and writing the second target block into the blockchain.