Fluid quality control using broad spectrum impedance spectroscopy

A quality control system controlling properties of a fluid during manufacture, comprising:

• a) an impedance spectroscopy (IS) probe operatively disposed in contact with the fluid;
• b) an impedance spectrometer, operatively connected to the IS probe, wherein the impedance spectrometer performs impedance measurements on the fluid, and wherein the impedance spectrometer provides IS data for at least three distinct frequencies;
• c) a data processing system, operatively coupled to the impedance spectrometer receiving the IS data, wherein the data processing system provides first IS signatures indicative of the fluid properties, and wherein the data processing system provides quality control information responsive to a comparison of the first IS signatures to a set of second IS signatures, wherein the set of second IS signatures are indicative of properties of a set of calibration fluids;
• and
• d) a Quality Control Operator, operatively disposed to receive the quality control information from the data processing system, and wherein the Quality Control Operator modifies the fluid properties in response to the quality control information;
• wherein the data processing system employs at least one statistical technique, and wherein the at least one statistical technique includes at least one of the following statistical techniques:
• Principal Component Analysis, Multivariate Least Squares Regression, Principal Component Regression, Pattern Recognition analysis, Cluster analysis, Neural Net analysis, and Group Methods of Data Handling.

A quality control system controlling properties of a manufactured fluid, comprising:

• a) means for obtaining a set of calibration fluids representative of a manufacturing process;
• b) means for performing impedance measurements on each of the calibration fluids to obtain first impedance spectroscopy (IS) data, wherein the first IS data include data for at least three frequencies;
• c) means for analyzing the first IS data providing first IS signatures indicative of properties for each of the calibration fluids;
• d) means for performing impedance measurements on a test fluid, wherein the test fluid is derived from the manufactured fluid, providing second IS data, wherein the second IS data include data for at least three frequencies;
• e) means for analyzing the second IS data providing second IS signatures indicative of properties of the test fluid;
• f) means for comparing the first IS signatures and the second IS signatures providing quality control information relating to the properties of the test fluid;
• and
• g) means for modifying the properties of the manufactured fluid responsive to the quality control information;
• wherein the means for analyzing the first IS data and the means for analyzing the second IS data include means for using at least one statistical technique, and wherein the at least one statistical technique includes at least one of the following statistical techniques:
• Principal Component Analysis, Multivariate Least Squares Regression, Principal Component Regression, Pattern Recognition analysis, Cluster analysis, Neural Net analysis, and Group Methods of Data Handling.

A fluid property quality control method, comprising the steps of:

• a) obtaining a set of calibration fluids representative of a first product;
• b) performing impedance measurements on each of the calibration fluids to obtain first impedance spectroscopy (IS) data, wherein the first IS data include data for at least three frequencies;
• c) using an at least one statistical technique to analyze the first IS data providing first IS signatures indicative of properties for each of the calibration fluids;
• d) performing impedance measurements on a test fluid providing second IS data, wherein the test fluid is representative of a second product, and wherein the second IS data include data for at least three frequencies;
• e) using the at least one statistical technique to analyze the second IS data providing second IS signatures indicative of properties of the test fluid;
• f) comparing the first IS signatures and the second IS signatures providing quality control information relating to the properties of the test fluid;
• and
• g) selectively accepting the second product, responsive to the quality control information;
• wherein the at least one statistical technique includes at least one of the following statistical techniques:
• Principal Component Analysis, Multivariate Least Squares Regression, Principal Component Regression, Pattern Recognition analysis, Cluster analysis, Neural Net analysis, and Group Methods of Data Handling.