Electric-quantity prediction model train method, electric-quantity prediction method, intelligent device and system

Abstract

The invention belongs to the field of intelligent equipment and discloses an electric quantity prediction model training method, an electric quantity prediction method, intelligent equipment and a system. The method comprises the following steps: acquiring a first number of historical power consumption data as training data of the date type for a certain date type among a plurality of date types;training a power quantity prediction model corresponding to a date type by using training data of different date types and outputting predicted values of all sample dates in training data of each datetype; and calculating an error between a predicted value and a real value of all sample dates of each date type output by each training, and obtaining a power quantity prediction model correspondingto the smallest error. The invention generates a corresponding power consumption prediction model according to different date types, realizes the effective prediction of the power consumption of the intelligent device, and can effectively improve the accuracy and usability of the power consumption prediction.

Description

technical field [0001] The invention belongs to the field of smart devices, and in particular relates to a power prediction model training method, a power prediction method, a smart device and a system. Background technique [0002] With the rapid development of technology, smart devices including children's phones, watches, and smart bracelets are becoming more and more popular. In the actual application of smart devices with mobile networking functions, due to the size limitation of smart The battery capacity of the device is limited and the power consumption is relatively fast. As the applications of smart devices become more and more abundant, their power consumption is also changing. How to predict the power consumption of smart devices has always been a problem in various fields. The key issues that major smart device companies are concerned about are also the modules that major companies want to vigorously tune. [0003] At present, there are few algorithms for predi...

AI Technical Summary

Problems solved by technology

[0003] At present, there are few algorithms for predicting the power consumption of smart devices. Therefore, an effective power prediction method is needed to provide convenience for users to use smart devices

Images