Modular Neural Networks for Adaptive Industrial Sensor Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing industrial environments face challenges in efficiently collecting and processing data from multiple sensors due to varying computing resources, network capabilities, and challenging environmental conditions, leading to conservative sensing configurations and limited real-time data management.
Innovation Solution
An expert system utilizing a modular neural network is employed to process inputs from sensors in industrial environments, enabling pattern recognition, self-organization of data collection activities, and autonomous control of sensors and data distribution across multiple storage devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is collected from multiple sensors in industrial environments, then monitoring and diagnostic capabilities are improved, but device complexity and data processing burden increase
Solution Approach 1:
The patent segments the data processing task by deploying distributed edge computing nodes at different locations in the industrial environment. Each edge node processes data from local sensors independently, dividing the overall processing burden into manageable segments that can be handled locally rather than requiring centralized processing of all sensor data
Solution Approach 2:
The patent introduces edge computing nodes as intermediary components between sensors and central systems. These edge nodes act as mediators that pre-process, filter, and aggregate sensor data before transmission, reducing the complexity burden on central processing systems while maintaining comprehensive monitoring capabilities
2Speed
If real-time data processing is implemented, then response speed and control capability are improved, but computing resource requirements and energy consumption increase
Solution Approach 1:
The patent applies local quality by implementing edge computing nodes with varying processing capabilities at different locations. Each edge node is configured with computing resources matched to its specific needs and local data volume, allowing real-time processing where critical while conserving energy in less demanding areas
Solution Approach 2:
The patent implements partial real-time processing by prioritizing processing of critical sensor data that requires immediate response while allowing non-critical data to be processed with lower priority or in batches, achieving real-time response for essential functions without the full energy cost of universal real-time processing
3Productivity
If sensing configurations are made adaptive to environmental conditions, then data collection efficiency is improved, but system complexity and control difficulty increase
Solution Approach 1:
The patent implements self-service by enabling edge computing nodes to automatically adapt their sensing configurations based on local environmental conditions and data priorities. The system autonomously adjusts sampling rates, sensor activation, and data filtering parameters without requiring manual reconfiguration, improving data collection efficiency while managing complexity through automation
Solution Approach 2:
The patent applies dynamics by making sensing configurations flexible and adjustable rather than fixed. Edge computing nodes dynamically modify their operational parameters in response to changing environmental conditions, data quality requirements, and priority levels, allowing the system to optimize performance across varying operational contexts
Data Source
AI summary
Methods and an expert system for processing a plurality of inputs collected from sensors in an industrial environment are disclosed. A modular neural network, where the expert system uses one type of neural network for recognizing a pattern relating to at least one of: the sensors, components of the industrial environment and a different neural network for self-organizing a data collection activity in the industrial environment is disclosed. A data communication network configured to communicate at least a portion of the plurality of inputs collected from the sensors to storage device is also disclosed.