APL Edge ML Inference Under Industrial Power Constraints

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Advanced Physical Layer (APL) edge devices in industrial systems face power constraints, limiting their ability to perform machine learning tasks due to energy restrictions, which hampers their capability to analyze and make inferences from sensor data within hazardous environments.

Innovation Solution

An event-driven framework is applied to APL-based edge devices to invoke and deploy machine learning models, allowing them to analyze input data, make inferences, and take actions within the power constraints, enabling the devices to detect normal and abnormal conditions in industrial systems and control actuators or communicate with cloud-based processors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If machine learning tasks are performed at the edge device, then communication bandwidth consumption is reduced, but power consumption increases beyond APL constraints

Engineering Contradiction:

Improvecommunication bandwidth consumptionVSAvoidpower consumption

Core Design Contradiction:

Loss of informationVSUse of energy by moving object

Solution Approach 1:

The machine learning workload is segmented between cloud processors and edge devices. Complex model training and heavy inference tasks are performed in the cloud, while simplified inference tasks are executed at the edge device. This segmentation allows the edge device to perform local analysis with reduced power consumption while still achieving bandwidth savings compared to full cloud processing.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The edge device performs partial machine learning inference tasks locally rather than complete processing. By executing only the inference phase with pre-trained models and sending only essential results to the cloud, the system achieves partial local processing that reduces communication bandwidth while keeping power consumption within APL limits.

Inventive Principle:

Principle #16Partial or excessive action

2Use of energy by moving object

If APL edge devices have minimal processing resources, then power consumption is reduced, but machine learning capability is insufficient

Engineering Contradiction:

Improvepower consumptionVSAvoidmachine learning capability

Core Design Contradiction:

Use of energy by moving objectVSAdaptability or versatility

Solution Approach 1:

Cloud-based processors serve as intermediaries that perform complex machine learning model training and provide pre-trained models to edge devices. The edge devices then use these pre-trained models for local inference tasks, enabling machine learning capability at the edge without requiring the edge device to perform computationally intensive training operations.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

Machine learning models are pre-trained in the cloud before being deployed to edge devices. This preliminary action of training allows the edge device to perform only the inference phase with minimal computational resources, maintaining power consumption within APL constraints while still providing machine learning functionality for industrial monitoring and control.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If all sensor data is sent to the cloud for analysis, then processing accuracy is improved, but communication bandwidth consumption increases

Engineering Contradiction:

Improveprocessing accuracyVSAvoidcommunication bandwidth consumption

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

Different processing approaches are applied locally at the edge device versus remotely in the cloud. The edge device performs local preprocessing, filtering, and simple inference tasks on sensor data, while the cloud receives only essential data or results for complex analysis. This local quality differentiation reduces communication bandwidth consumption while maintaining processing accuracy for critical functions.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20240118681A1Systems and methods for implementing machine learning in a local APL edge device with power constraints

Publication Date: 2024.04.11 SCHNEIDER ELECTRIC SYSTEMS USA INC

AI summary

A method performed by an Advanced Physical Layer (APL)-based edge device with power constraints is provided. The method includes applying an event-driven framework that is compliant with power constraints of the APL-based edge device to receive input data; applying the event-driven framework to the input data to invoke a machine learning (ML) model that is trained to analyze the input data and make inferences about one or more aspects of an industrial system based on the input data, and applying the invoked machine learning model to analyze the input data and make an inference about the one or more aspects of the industrial system based on the input data. The input data is received by the APL-based edge device from one or more source field devices of the industrial system and/or the inference is used make a decision and cause an action to be applied to the industrial system.