Machine Learning Variable Discovery for Industrial Control Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial automation environments face challenges in extracting enterprise-level insights from vast operational data due to the complexity of control code programming, limited access to runtime data, and the difficulty in integrating thousands of relevant variables, which hinders effective control logic programming and data science applications.
Innovation Solution
Implementing a machine learning-based analysis engine within industrial programming and data science environments to identify underutilized variables, provide recommendations for their integration into control logic, and offer data science tools for enhanced analytics, thereby improving industrial automation programming and data science capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If control programmers manually analyze and integrate thousands of operational variables into control logic, then control logic completeness may improve, but the time and complexity required increases dramatically
Solution Approach 1:
The system enables self-service by automatically analyzing operational data and generating control logic recommendations without requiring manual intervention. The machine learning engine autonomously identifies patterns, relationships, and optimal control logic from the operational data, freeing programmers from time-consuming manual analysis while maintaining high control logic completeness.
Solution Approach 2:
The patent replaces the mechanical manual analysis process with an automated machine learning system. Instead of programmers manually examining thousands of operational variables, the system uses computational algorithms to automatically process operational data, identify meaningful patterns, and generate control logic recommendations, dramatically reducing programming time while maintaining reliability.
2Loss of information
If data scientists access enormous amounts of operational data for analysis, then data mining capabilities improve, but the complexity of data processing and understanding control logic increases
Solution Approach 1:
The system introduces an intermediary machine learning engine that bridges data scientists and operational data. This intermediary automatically processes the enormous amounts of operational data, performs data mining, and presents results in a simplified format that data scientists can easily interpret, reducing the complexity of data processing while maintaining full data mining capability.
Solution Approach 2:
The machine learning engine performs self-service by autonomously analyzing operational data, identifying patterns, and generating insights without requiring data scientists to manually process the data. The system automatically handles data cleaning, feature extraction, and pattern recognition, allowing data scientists to focus on interpreting results rather than processing raw data.
3Reliability
If control programmers are provided with runtime data and statistics, then control logic quality may improve, but the accessibility and usability of this information for programmers remains limited
Solution Approach 1:
The system introduces an intermediary layer that translates complex runtime data and statistics into programmer-friendly formats. The machine learning engine processes raw operational data and presents simplified, actionable insights that control programmers can easily understand and apply, improving data accessibility while maintaining control logic quality.
Solution Approach 2:
The system transforms the parameters of data presentation by converting raw operational data into processed, meaningful metrics and recommendations. The machine learning engine changes the form and structure of data to match programmer needs, presenting information in formats that are easily accessible and actionable, thereby improving ease of operation without sacrificing control logic quality.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments of the present technology generally relate to solutions for improving industrial automation programming and data science capabilities with machine learning. More specifically, embodiments of the present technology include systems and methods for implementing machine learning engines within industrial programming and data science environments to improve performance, increase productivity, and add functionality. In an embodiment, a system comprises a machine learning-based analysis engine configured to identify a variable that is available to be utilized in control logic for controlling an industrial automation environment. The machine learning-based analysis engine is further configured to determine that the variable is not utilized in the control logic. A recommendation component of the system is configured to, in an industrial programming environment, surface a recommendation to add the variable to the control logic.