Abstract: A blockchain-based electricity trading method and system are disclosed. Based on the initial trading plan of the user in the virtual power plant, a non-cooperative game model among multiple users and the virtual power plant is constructed, a purchase price, a sale price and a demand response compensation price of a trading between the user and the virtual power plant during t period are determined; according to the above results, a final load demand of user i, a charging capacity or discharging capacity of the energy storage device of user i, as well as a purchase electricity quantity or sale electricity quantity of a trading with the virtual power plant during t period are determined, and a final trading scheme is formed; and based on the final trading scheme, both trading parties are matched, a trading contract is generated and a verification is performed.

Abstract: The invention provided an electric vehicle charging scheduling method, apparatus and system based on cloud-edge collaboration, a storage medium and an electronic device. In the present invention, a charging request of an electric vehicle user is accepted and processed by an edge computing unit, and a target charging station for a to-be-charged electric vehicle is determined with a minimum traveling cost as a target, so that a data transmission distance is reduced, and the electric vehicle user is timely assisted in selecting the target charging station and completing a charging appointment. After the charging appointment is made, charging data is uploaded to a charging optimization scheduling model pre-trained by a cloud platform for obtaining an electric vehicle charging scheduling strategy, so that powerful cloud platform computing abilities and rapid response advantages of the edge computing unit are fully utilized, the problem of network congestion is avoided, and timeliness is improved.

Abstract: A blockchain-based electricity trading method and system are disclosed. Based on the initial trading plan of the user in the virtual power plant, a non-cooperative game model among multiple users and the virtual power plant is constructed, a purchase price, a sale price and a demand response compensation price of a trading between the user and the virtual power plant during t period are determined; according to the above results, a final load demand of user i, a charging capacity or discharging capacity of the energy storage device of user i, as well as a purchase electricity quantity or sale electricity quantity of a trading with the virtual power plant during t period are determined, and a final trading scheme is formed; and based on the final trading scheme, both trading parties are matched, a trading contract is generated and a verification is performed.

Abstract: Provided are a method and a system for optimizing charging and discharging behaviors of a BESS based on a SOH, relating to charging and discharging optimization. The number of cycles of the battery pack and corresponding DODs are obtained based on the curve of the SOC of the battery pack. Then, the SOH of the battery pack is obtained. A charging index sequence and a discharging index sequence of battery packs are obtained based on the SOH, the SOC and a charging and discharging state of the battery pack. The optimal number of the charging and discharging battery packs and optimal DODs are determined. Charging and discharging tasks are carried out according to the charging and discharging index sequences of the battery packs based on the optimal number of the charging and discharging battery packs and the optimal DODs.

Abstract: The disclosure provides a method, a system and a storage medium for predicting power load probability density based on deep learning. The method comprises: S101, collecting power load data of a user, meteorological data and air quality data in a preset historical time period, and dividing the collected data into a training set and a test set; S102, determining a deep learning model for predicting power load; S103, inputting the test set into the deep learning model for predicting power load, and obtaining power load prediction data of the user at different quantile points in a third time interval; S104, performing kernel density estimation and obtaining a probability density curve of the power load of the user in the third time interval.

Abstract: The invention provided an electric vehicle charging scheduling method, apparatus and system based on cloud-edge collaboration, a storage medium and an electronic device. In the present invention, a charging request of an electric vehicle user is accepted and processed by an edge computing unit, and a target charging station for a to-be-charged electric vehicle is determined with a minimum traveling cost as a target, so that a data transmission distance is reduced, and the electric vehicle user is timely assisted in selecting the target charging station and completing a charging appointment. After the charging appointment is made, charging data is uploaded to a charging optimization scheduling model pre-trained by a cloud platform for obtaining an electric vehicle charging scheduling strategy, so that powerful cloud platform computing abilities and rapid response advantages of the edge computing unit are fully utilized, the problem of network congestion is avoided, and timeliness is improved.

Abstract: Provided are a method and a system for optimizing charging and discharging behaviors of a BESS based on a SOH, relating to charging and discharging optimization. The number of cycles of the battery pack and corresponding DODs are obtained based on the curve of the SOC of the battery pack. Then, the SOH of the battery pack is obtained. A charging index sequence and a discharging index sequence of battery packs are obtained based on the SOH, the SOC and a charging and discharging state of the battery pack. The optimal number of the charging and discharging battery packs and optimal DODs are determined. Charging and discharging tasks are carried out according to the charging and discharging index sequences of the battery packs based on the optimal number of the charging and discharging battery packs and the optimal DODs.

Abstract: The disclosure provides a method, a system and a storage medium for predicting power load probability density based on deep learning. The method comprises: S101, collecting power load data of a user, meteorological data and air quality data in a preset historical time period, and dividing the collected data into a training set and a test set; S102, determining a deep learning model for predicting power load; S103, inputting the test set into the deep learning model for predicting power load, and obtaining power load prediction data of the user at different quantile points in a third time interval; S104, performing kernel density estimation and obtaining a probability density curve of the power load of the user in the third time interval.

Abstract: The present invention relates to a method and a system for compressing data from a smart meter. The method comprises: LZ-encoding electricity load data collected by the smart meter whenever the smart meter collects the electricity load data; storing the LZ-encoded electricity load data in a temporary database through a smart grid communication channel; reading the electricity load data from the temporary database every preset second duration, wherein the read electricity load data is electricity load data stored in the temporary database within the second duration before a corresponding reading time point; and LZ-decoding the read electricity load data, SAX-compressing the LZ-decoded electricity load data, and storing the SAX-compressed electricity load data in a data center. The present invention has high compression rate, reduces the transmission burden for communication lines and storage burden for the data center, and improves the efficiency of smart electricity data analysis and mining.

Abstract: The present invention relates to a method and a system for compressing data from a smart meter. The method comprises: LZ-encoding electricity load data collected by the smart meter whenever the smart meter collects the electricity load data; storing the LZ-encoded electricity load data in a temporary database through a smart grid communication channel; reading the electricity load data from the temporary database every preset second duration, wherein the read electricity load data is electricity load data stored in the temporary database within the second duration before a corresponding reading time point; and LZ-decoding the read electricity load data, SAX-compressing the LZ-decoded electricity load data, and storing the SAX-compressed electricity load data in a data center. The present invention has high compression rate, reduces the transmission burden for communication lines and storage burden for the data center, and improves the efficiency of smart electricity data analysis and mining.

Abstract: A charging and discharging scheduling method for electric vehicles in microgrid under time-of-use price includes: determining the system structure of the microgrid and the characters of each unit; establishing the optimal scheduling objective function of the microgrid considering the depreciation cost of the electric vehicle (EV) battery under time-of-use price; determining the constraints of each distributed generator and EV battery, and forming an optimal scheduling model of the microgrid together with the optimal scheduling objective function of the microgrid; determining the amount, starting and ending time, starting and ending charge state, and other basic calculating data of the EV accessing the microgrid under time-of-use price; determining the charge and discharge power of the EV when accessing the grid, by solving the optimal scheduling model of the microgrid with a particle swarm optimization algorithm.

Abstract: A charging and discharging scheduling method for electric vehicles in microgrid under time-of-use price includes: determining the system structure of the microgrid and the characters of each unit; establishing the optimal scheduling objective function of the microgrid considering the depreciation cost of the electric vehicle (EV) battery under time-of-use price; determining the constraints of each distributed generator and EV battery, and forming an optimal scheduling model of the microgrid together with the optimal scheduling objective function of the microgrid; determining the amount, starting and ending time, starting and ending charge state, and other basic calculating data of the EV accessing the microgrid under time-of-use price; determining the charge and discharge power of the EV when accessing the grid, by solving the optimal scheduling model of the microgrid with a particle swarm optimization algorithm.