2results about How to "Solve communication bottlenecks" patented technology

The present invention relates to the technical field of asynchronous federated learning, and provides an asynchronous federated learning method and system based on a multi-agent model, including: randomly selecting a number of pre-trained clients in each group of clients, and obtaining a decision result of whether each pre-trained client participates in the training and uploading of the model; receiving a local model obtained by training the clients participating in the training and uploading of the model in each group of clients, and obtaining a group model; performing weighted aggregation on the group model to obtain a global model. It can not only solve the long waiting delay problem in synchronous federated learning, but also solve the communication bottleneck problem in completely semi-asynchronous federated learning.

The utility model relates to energy storage battery technology field provides a kind of BMS topological device, the device includes multi-channel AI analog input unit, multi-channel DI digital input unit, main control MCU chip, DO digital output unit and RTC real-time clock submodule, wherein: the multi-channel AI analog input unit, multi-channel DI digital input unit and RTC real-time clock submodule are connected with main control MCU chip, and the main control MCU chip is connected with DO digital output unit. The utility model is highly integrated and optimized connection by multi-channel AI analog input unit, multi-channel DI digital input unit, main control MCU chip, DO digital output unit and RTC real-time clock submodule, realizes the integrated processing of battery management system key function, simplifies the redundant problem of the quantity of controller in traditional three-level architecture BMS system, reduces the complexity of hardware configuration and communication network, effectively reduces system total cost and fault point quantity.