ML Power Prediction for Digital Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack the ability to accurately assess and reduce the electrical consumption of digital display devices used for displaying digital content, as the energy consumption varies based on the content, device, and environment, making it difficult to evaluate and minimize power usage effectively.
Innovation Solution
A supervised machine learning engine is trained using a dataset of images, device characteristics, and environmental factors to predict the electrical consumption of digital display devices, allowing for the selection of optimal content and device configurations to minimize energy usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If digital signage devices are deployed in urban environments to display content, then information dissemination capability is improved, but electrical energy consumption increases significantly
Solution Approach 1:
The system performs preliminary analysis of image characteristics (luminance, contrast, complexity) before displaying them on digital signage devices. By predicting power consumption in advance based on these characteristics, the system can pre-select optimal display configurations or alternative content to minimize energy consumption while maintaining information dissemination effectiveness.
Solution Approach 2:
The system dynamically adjusts display parameters (brightness, contrast, refresh rate) based on the analyzed characteristics of each image and environmental conditions. This allows the digital signage to adapt its power consumption to the actual content being displayed, reducing energy waste while preserving the quality of information delivery.
2Loss of energy
If power consumption of digital signage is reduced by optimizing content display, then energy efficiency is improved, but the ability to accurately assess power consumption in advance deteriorates due to variability from content, device, and environment
Solution Approach 1:
The system segments the power consumption assessment into multiple independent components: image characteristics analysis (luminance, contrast, complexity), device characteristics (display technology, size, efficiency), and environmental factors (ambient light, temperature). By analyzing each segment separately and combining their effects, the system achieves accurate overall power consumption prediction despite the variability of individual factors.
Solution Approach 2:
The system incorporates feedback mechanisms where actual power consumption measurements from digital signage devices are compared with predicted values. This feedback loop allows the system to refine its prediction models continuously, improving measurement precision while maintaining energy optimization capabilities.
3Measurement precision
If multiple features including image characteristics, device characteristics, and environmental factors are analyzed to predict power consumption, then prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The system introduces an intermediary image analysis module that processes visual content and extracts relevant characteristics (luminance, contrast, complexity) before feeding them to the power consumption prediction engine. This intermediary layer simplifies the overall system architecture by preprocessing complex image data into standardized features, making the prediction process more manageable while maintaining high accuracy.
Data Source
Figure 1
Figure 2~3
Figure 4~5a
AI summary
Method for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device, said method comprising: obtaining a training set from a plurality of training images, each sample of said training set comprising: at least one feature belonging to a group of features comprising: features of a digital display device; features of the environment of the digital display device; features of a training image; a measured value of power consumption induced by the display of said training image, said induced power consumption value being measured on the digital display device during the display of the training image;training said supervised machine learning engine on said training basis to minimize a loss function between the induced power consumption predictions made by the supervised machine learning engine and the measured values of induced consumption;