Cloud-Guided Industrial Data Transmission Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for data transmission from industrial control devices to cloud systems are inflexible and inefficient, often transmitting unnecessary data, leading to resource consumption and high costs, particularly in terms of usage-related costs and data connection limitations.
Innovation Solution
Employing machine learning to analyze user behavior within the cloud system to optimize data transmission from industrial control devices by adjusting the data sent based on predicted usage patterns, using commands to modify the transmission parameters such as sampling rates and data selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data transmission is defined programmatically by a PLC developer at development time, then the data transmission can be configured, but the transmission is inflexible and laborious to change, and unnecessary data is transmitted
Solution Approach 1:
The patent applies dynamics by transitioning from static, development-time configured data transmission to dynamic, runtime-adjustable transmission. The system enables users to modify data selection and transmission parameters during operation through a user interface, allowing the transmission configuration to adapt to changing needs without requiring program reconfiguration.
Solution Approach 2:
The patent implements feedback mechanisms where the system monitors actual data usage patterns and provides information back to users. This enables users to adjust transmission settings based on real usage data, optimizing the balance between transmitting sufficient data for cloud processing and avoiding unnecessary data transmission.
2Reliability
If all data is transmitted to the cloud system, then complete data availability is ensured, but resource consumption and costs increase due to unnecessary data transmission
Solution Approach 1:
The patent extracts only the necessary data for cloud processing by implementing selective data transmission. Users can define which data points should be transmitted to the cloud and which should be processed locally, extracting only the essential information needed for cloud-based analysis and avoiding transmission of unnecessary data.
Solution Approach 2:
The patent applies local quality by enabling different data handling strategies at different locations. Local processing occurs for data that doesn't need cloud transmission, while selective transmission occurs for data that requires cloud processing. This allows optimization of data quality and quantity based on specific local needs and cloud requirements.
3Quantity of substance
If data transmission rate is increased to transmit more data, then data completeness improves, but the physical data transmission rate limit is exceeded and connection efficiency decreases
Solution Approach 1:
The patent segments data transmission by dividing data into different categories based on priority and requirements. Critical data can be transmitted at higher rates while less important data is transmitted at lower rates or batched, allowing the system to manage transmission volume within physical rate limits while maintaining data completeness.
Solution Approach 2:
The patent implements periodic action by enabling batched data transmission instead of continuous transmission. Data can be accumulated and transmitted in periodic batches, allowing the system to manage transmission rate by controlling the frequency and size of transmission cycles, thereby avoiding saturation of the physical connection.
Data Source
AI summary
A method for adjusting the data transmission from an industrial control device to a cloud system, the method comprising steps carried out by a cloud application, the steps comprising: determining usage behavior by means of machine learning, by the cloud system or an associated computer system connected thereto, of data, more particularly process data, transmitted from the industrial control device; and sending at least one command to the industrial control device in order to adjust the data transmission.