Huabao New Energy has applied for a patent on a method for training electricity prediction models for home energy storage systems, aimed at improving prediction accuracy. According to the National Intellectual Property Administration, the company submitted this patent application titled “Method for Training Electricity Prediction Models for Home Energy Storage, Electronic Devices, and Media” on December 2024, with the publication number CN119852997A.
The patent summary indicates that this application reveals a method designed for training electricity prediction models specifically for household energy storage, falling under the energy management technology domain. The method involves gathering and aggregating the encrypted model parameters of various electricity prediction models to construct a global optimization model. This model is then deployed across multiple objects, allowing any target object to conduct local training based on a preset reinforcement learning algorithm, subsequently updating the model parameters of the global optimization model.
Furthermore, the method retrieves electricity prediction models from other objects that match the current object, utilizing these to build a corresponding electricity prediction model for the current object. This process enhances prediction accuracy for home energy storage while ensuring the protection of household privacy.
According to Tianyancha, Huabao New Energy Co., Ltd. was established in 2011 and is located in Shenzhen. The company primarily engages in the manufacturing of electrical machinery and equipment, with a registered capital of 124.8 million RMB. Data analysis has revealed that the company has invested in two other enterprises, participated in nine bidding projects, holds 151 trademark records, and possesses 1,201 patents, in addition to having 34 administrative licenses.
Original article by NenPower, If reposted, please credit the source: https://nenpower.com/blog/huabao-new-energy-files-patent-for-improved-home-energy-consumption-prediction-model/
