PLC Design Environment Views for Machine Learning Model Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial automation environments struggle to effectively integrate machine learning models into control programs, making it difficult to target specific variables and optimize industrial processes efficiently.
Innovation Solution
A system and method for surfacing machine learning models in industrial automation environments, allowing programmers to integrate machine learning models into control programs by selecting tags that represent industrial assets, and connecting these models with other elements within the programming environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine learning models are integrated into control programs, then advanced control and process optimization are enabled, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary layer between machine learning models and control programs that facilitates integration without direct complex coupling. This intermediary enables advanced control capabilities while managing system complexity by providing a standardized interface and abstraction layer.
Solution Approach 2:
The system is divided into distinct modular components including machine learning model modules, control program modules, and integration layers. This segmentation allows each component to be developed, tested, and maintained independently, reducing overall system complexity while enabling advanced control through composability.
2Productivity
If machine learning models are integrated into control programs, then industrial processes are optimized, but integration difficulty increases
Solution Approach 1:
The patent creates a universal integration framework that can accommodate multiple types of machine learning models and control programs through standardized interfaces. This multi-functional approach enables process optimization across different industrial applications while simplifying integration through consistent methods rather than custom solutions for each case.
Solution Approach 2:
The system incorporates automated features that reduce manual integration effort, such as automatic model selection, configuration propagation, and validation. These self-service capabilities enable process optimization while reducing integration difficulty by minimizing manual programming and configuration tasks.
3Reliability
If machine learning models are integrated into control programs, then process control is enhanced, but programming complexity increases
Solution Approach 1:
The patent employs template-based approaches where standardized patterns for integrating machine learning models into control programs are pre-defined and can be copied or instantiated multiple times. This reduces programming complexity by eliminating the need to reinvent integration logic for each case while maintaining reliable process control through proven patterns.
Solution Approach 2:
The system performs preliminary actions such as automatic model compilation, validation, and configuration generation before deployment. This preliminary processing reduces programming complexity by preparing components in advance with proper configurations, ensuring reliable process control without requiring complex runtime programming decisions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments of the present technology generally relate to industrial automation environments. More specifically, embodiments include systems and methods to surface machine learning systems in a design application of an industrial automation environment. In some examples, a design component generates a control program configured for implementation by a Programmable Logic Controller (PLC). The design component receives a user input that selects a program tag that represents a target variable in an industrial automation process. In response to the user selection, the design component identifies one or more machine learning models associated with the target variable and displays the one or more machine learning models. The design component receives a user input that selects one of the one or more machine learning models and responsively integrates another program tag that represents the selected machine learning model into the control program.