Machine Operation Classifier Using On-Board Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting machine operations using on-board engineering channel data are burdensome and computationally inefficient, requiring manual input and rule establishment for accurate predictions.
Innovation Solution
Developing machine operation classifiers using machine learning algorithms that receive training data from on-board engineering channels, determine training features and labels, and build predictive models to accurately classify machine operations with improved efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual methods are used to establish rules for predicting machine operations, then prediction accuracy can be achieved, but the process becomes burdensome and computationally inefficient
Solution Approach 1:
The patent replaces manual rule-establishment methods with machine learning algorithms. The system automatically learns prediction rules from training data consisting of machine sensor data and corresponding operation labels, eliminating the need for manual rule creation while maintaining or improving prediction accuracy and significantly enhancing computational efficiency.
Solution Approach 2:
The machine learning model performs self-training by automatically learning from labeled training data. The system autonomously identifies patterns and relationships between machine sensor readings and operations without requiring continuous manual intervention, enabling the model to improve its own predictive capabilities through exposure to more training examples.
2Ease of manufacture
If manual rule establishment is used for machine operation prediction, then interpretable rules can be created, but the process is burdensome to the user
Solution Approach 1:
The patent substitutes manual rule-creation processes with automated machine learning algorithms. The system automatically generates prediction models from training data, eliminating the burden of manual rule establishment while reducing overall system complexity by consolidating multiple manual steps into an automated training process.
Solution Approach 2:
The system performs preliminary action by pre-training machine learning models using labeled training data before deployment. This upfront training phase automatically establishes the prediction rules needed for operational use, eliminating the need for continuous manual rule creation and simplifying the overall process.
Data Source
AI summary
A method for developing machine operation classifiers for a machine is disclosed. The method includes receiving training data associated with the machine from one or more on-board engineering channels associated with the machine and determining one or more training features based on the training data values. The method also includes determining one or more training labels associated with the one or more training features and building a predictive model for determining machine operation classifiers using a computer. Building the predictive model may include feeding the one or more training features and the one or more training labels associated with the one or more training features to a machine learning algorithm and determining a predictive model from the machine learning algorithm. The predictive model may be used for receiving new data associated with the machine and determining a predicted label based on the new data.