Univariate method for monitoring and analysis of multivariate data
First Claim
1. A method of monitoring multivariate process data in a process plant, wherein the multivariate process data comprises a plurality of process variables each having a plurality of observations, wherein at least one process variable comprises an unknown process variable control limit, the method comprising:
- collecting a set of data for the at least one process variable, wherein the set of collected process variable data is representative of a normal operation of the process when the process is operating normally and comprises a plurality of observations of the at least one multivariate process variable;
determining the unknown process variable control limit under the normal operation of the process;
defining each process variable as a process variable vector comprising a set of observation components, wherein the set of observation components comprises time dependent process data corresponding to the observations of the process variable;
normalizing each of a plurality of process variable vector transformations to the same process variable control limit using a scaling parameter unique to each process variable to generate a common process variable control limit for the process variables, wherein each process variable vector transformation is a function of a univariate variable unifying the process variables;
calculating a multivariable transformation as a function of the plurality of process variable vector transformations each corresponding to one of the process variables; and
representing an operation of a process based on the multivariable transformation, wherein the representation of the operation of the process designates a multivariate projection of the process data by the univariate variable for each of the process variables, wherein representing the operation of the process based on the multivariable transformation comprises monitoring each process variable in relation to the determined process variable limit.
1 Assignment
0 Petitions
Accused Products
Abstract
A method and system of monitoring multivariate process data in a process plant, where the multivariate process data comprises a plurality of process variables each having a plurality of observations, includes defining each process variable as a process variable vector comprising a set of observation components, where the set of observation components comprises time dependent process data corresponding to the observations of the process variable, calculation, a multivariable transformation as a function of a plurality of process variable vector transformation each corresponding to one of the process variables, where each process variable vector transformation is a function of a univariate variable unifying the process variables, and representing the operation of the process based on the multivariable transformation, where the representation of the operation of the process designates a multivariate projection of the process data by the univariate variable for each of the process variables.
-
Citations
43 Claims
-
1. A method of monitoring multivariate process data in a process plant, wherein the multivariate process data comprises a plurality of process variables each having a plurality of observations, wherein at least one process variable comprises an unknown process variable control limit, the method comprising:
-
collecting a set of data for the at least one process variable, wherein the set of collected process variable data is representative of a normal operation of the process when the process is operating normally and comprises a plurality of observations of the at least one multivariate process variable; determining the unknown process variable control limit under the normal operation of the process; defining each process variable as a process variable vector comprising a set of observation components, wherein the set of observation components comprises time dependent process data corresponding to the observations of the process variable; normalizing each of a plurality of process variable vector transformations to the same process variable control limit using a scaling parameter unique to each process variable to generate a common process variable control limit for the process variables, wherein each process variable vector transformation is a function of a univariate variable unifying the process variables; calculating a multivariable transformation as a function of the plurality of process variable vector transformations each corresponding to one of the process variables; and representing an operation of a process based on the multivariable transformation, wherein the representation of the operation of the process designates a multivariate projection of the process data by the univariate variable for each of the process variables, wherein representing the operation of the process based on the multivariable transformation comprises monitoring each process variable in relation to the determined process variable limit. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A computer readable medium having computer executable instructions stored in memory for monitoring a process in a process plant, wherein the process generates multivariate process data comprising a plurality of process variables each having a plurality of observations, wherein each process variable is defined as a process variable vector comprising a set of observation components having time dependent process data corresponding to the observations of the process variable, and wherein at least one process variable comprises an unknown process variable control limit, the computer executable instructions comprising:
-
a data collection tool adapted to collect a set of data for the at least one process variable, wherein the set of collected process variable data is representative of a normal operation of the process when the process is operating normally and comprises a plurality of observations of the at least one multivariate process variable; an analysis tool adapted to determine the unknown process variable control limit under the normal operation of the process; a normalization tool adapted to normalize each of a plurality of process variable vector transformations to the same process variable control limit using a scaling parameter unique to each process variable to generate a common process variable control limit for the process variables, wherein each process variable vector transformation is a function of a univariate variable unifying the process variables; a process variable transformation tool adapted to calculate a multivariable transformation as a function of the plurality of process variable vector transformations each corresponding to one of the process variables; and a monitoring tool adapted to represent an operation of the process based on the multivariable transformation, wherein the representation of the operation of the process designates a multivariate projection of the process data by the univariate variable for each of the process variables wherein the monitoring tool is adapted to monitor each process variable in relation to the common process variable limit. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32)
-
-
33. A computer readable medium having computer executable instructions stored in memory for monitoring a process in a process plant, the computer executable instructions comprising:
-
a first analysis tool adapted to collect a set of data for at least one multivariate process variable, wherein at least one of the multivariate process variables comprises an unknown process variable control limit, and wherein the set of collected multivariate process variable data is representative of a normal operation of the process when the process is operating normally and comprises a plurality of observations of the at least one multivariate process variable; a second analysis tool adapted to determine the unknown process variable control limit under the normal operation of the process;
,a third analysis tool adapted to generate a univariate statistical representation for each of the plurality of multivariate process variables each having a plurality of observations based on a univariate variable unifying the multivariate process variables; a fourth analysis tool adapted to normalize each univariate statistical representation of a multivariate process variable to a common process variable control limit; a fifth analysis tool adapted to generate a univariate statistical representation of an operation of the process based on the plurality of univariate statistical representations of the plurality of multivariate process variables; and a sixth analysis tool adapted to monitor the operation of the process based on the univariate statistical representation of the operation of the process as a function of the univariate variable, and adapted to monitor each multivariate process variable as represented by the univariate statistical representation of the operation of the process in relation to the common process variable control limit. - View Dependent Claims (34, 35, 36, 37, 38, 39, 40, 41, 42, 43)
-
Specification