Rolling principal component analysis for dynamic process monitoring and end point detection
AssigneeViavi Solutions Inc.
InventorsLan Sun, Chang Meng Hsiung, Edward Gooding
Life of this patent
Estimating the term…
Abstract
In some implementations, a device may receive spectroscopic data associated with a dynamic process. The device may generate a principal component analysis (PCA) model based on a first block of spectra from the spectroscopic data. The device may project a second block of spectra from the spectroscopic data to the PCA model generated based on the first block of spectra. The device may determine a value of a metric associated with the second block based on projecting the second block of spectra to the PCA model. The device may determine whether the dynamic process has reached an end point based on the value of the metric associated with the second block.
Claims20
A method, comprising:
- receiving, by one or more devices, spectroscopic data during a performance of a blending process;
- projecting, by the one or more devices and based on a first block of spectra from the spectroscopic data, a second block of spectra from the spectroscopic data,
- where the first block of spectra includes spectra collected during one or more revolutions of a blender performing the blending process;
- determining, by the one or more devices and based on projecting the second block of spectra, whether the blending process has reached a steady state,
- where the steady state indicates that blending has been achieved;
- and
- one or more of:
- causing, by the one or more devices and based on whether the blending process has reached the steady state, the blender to stop the blending process or restart the blending process on new materials,
- causing, by the one or more devices and based on whether the blending process has reached the steady state, a next step of a manufacturing process to be initiated,
- providing, by the one or more devices, an indication of whether the blending process has reached the steady state, or
- enabling visualization of an evolution of the blending process.
A system, comprising:
- one or more memories;
- and
- one or more processors, coupled to the one or more memories, configured to cause the system to:
- receive spectroscopic data during a performance of a blending process;
- project, based on a first block of spectra from the spectroscopic data, a second block of spectra from the spectroscopic data,
- where the first block of spectra includes spectra collected during one or more revolutions of a blender performing the blending process;
- determine, based on projecting the second block of spectra, whether the blending process has reached a steady state;
- and
- one or more of:
- control the blending process based on whether the blending process has reached the steady state,
- provide an indication of whether the blending process has reached the steady state, or
- enable visualization of an evolution of the blending process.
A system, comprising:
- means for receiving spectroscopic data during a performance of a blending process;
- means for projecting, based on a first block of spectra from the spectroscopic data, a second block of spectra from the spectroscopic data,
- where the first block of spectra includes spectra collected during one or more revolutions of a blender performing the blending process;
- means for determining, based on projecting the second block of spectra, whether the blending process has reached a steady state;
- and
- means for controlling the blending process based on whether the blending process has reached the steady state.
Dependencies are inferred from the claim text.