System and method for controlling a machine using dimensionally aware rule extraction

A method of determining quality of a production event performed at a machine comprising:

• obtaining, at a computer, production data comprising time series data from the machine;
• filtering anomalous data from the production data to form a filtered set of data;
• adding synthetic data to the filtered set of data;
• extracting, at the computer, a set of salient features from the filtered set of data of the time series data corresponding to a dimensionally aware model;
• determining, at the computer, a boundary equation for the production event;
• obtaining, at the computer a classification value based on calculating the set of salient features into the boundary equation;
• when the classification value is within the boundary equation classifying the set of salient features as unacceptable;
• when the classification value is outside the boundary equation classifying the set of salient features as acceptable;
• adjusting the dimensionally aware model only when classifying the set of features as acceptable;
• generating updated machine parameters when the set of features is classified as unacceptable;
• electronically communicating the updated machine parameters to a machine controller of the machine;
• automatically controlling the machine based on the updated machine parameters;
• and
• generating, by the computer, an indicator on a display corresponding to the classification value.

A system for classifying weld quality comprising:

• a computer having at least one processor;
• a display;
• and a memory coupled to the at least one processor, wherein the memory stores:
• a dimensionally aware model generated based on a training set and guided by feature dimensions and instructions for execution by the computer and the at least one processor and
• wherein the instructions include,
• in response to receiving a set of time series data comprising power data corresponding to power consumed by an ultrasonic transducer, movement data corresponding to a sonotrode tip movement along a direction of a clamping force and acoustic data corresponding to a fixed ultrasonic microphone from a machine, causing the computer to perform:
• identifying a set of features from the set of time series data;
• applying the dimensionally aware model to the set of features by implementing a boundary representation;
• classifying the set of features as acceptable in response to the implementation of the boundary representation indicating the set of features are outside the boundary representation and adjusting the dimensionally aware model only when the set of features is classified as acceptable;
• classifying the set of features as unacceptable in response to the implementation of the boundary representation indicating the set of features are inside the boundary representation;
• generating, for display on the display of a user device, an alert based on the classification;
• generating updated machine parameters when the set of features is classified as unacceptable;
• electronically communicating the updated machine parameters to a machine controller of the machine;
• and
• automatically controlling the machine based on the updated machine parameters.