Back to results
US 12,613,231 B2Granted
System and method for performing industrial processes across facilities
AssigneeHartford Steam Boiler Inspection and Insurance Company
InventorRichard B. Jones
Life of this patent
Estimating the term…
Abstract
A system and method is described herein for performing at least one industrial process at each facility of a plurality of facilities based on an industrial process standard generated by reducing functional, and trend line outlier bias in data of one or more process parameters as measured by one or more sensors. Outliers are removed from the data set through an objective method. Bias is determined based on absolute, relative error, or both. Error values are computed from the data, model coefficients, or trend line estimates. Outlier data records are removed when the error values are greater than or equal to one or more criteria.
Claims22
A system, comprising:
- at least one industrial plant facility performing at least one industrial process;
- at least one sensor configured to measure one or more target variables of at least process parameter of the at least one industrial process performed at the at least one industrial plant facility, wherein the at least one sensor is configured to generate a corresponding current target data set for each target variable of the one or more target variables of the at least one process parameter, wherein the at least one process parameter is associated with at least one chemical compound indicative of a greenhouse gas emitted from the at least one industrial plant facility, and wherein the at least one sensor is further configured to detect and quantify the at least one chemical compound;
- a computing unit configured to execute a computer program that is stored by a non-transient storage subsystem, wherein the computer program, when executed by at least one processor, causes the at least one processor of the computing unit to at least:
- receive the corresponding current target data set for each target variable of the one or more target variables of the at least one process parameter from the at least one sensor;
- wherein the one or more target variables are measured by the at least one sensor that feeds a model for a plurality of industrial plant facilities,
- wherein the model comprises a data-driven model;
- generate a random process data set from the corresponding current target data set;
- perform a dynamic outlier bias reduction on the corresponding current target data set for one or more bias criteria values of a set of bias criteria values to generate one or more outlier bias reduced target data sets;
- perform the dynamic outlier bias reduction on the random process data set for the one or more bias criteria values of the set of bias criteria values to generate one or more outlier bias reduced random data set;
- generate a first bias criteria curve for the corresponding current target data set and a second bias criteria curve for the random process data set from the one or more bias criteria values, a set of target error values, a set of random error values, a set of target correlation coefficients, and a set of random correlation coefficients;
- dynamically determine, based on the first bias criteria curve and the second bias criteria curve, a non-biased viability of the corresponding current target data set;
- wherein the non-biased viability is an indicator of whether the corresponding current target data set is representative of the at least one process parameter of the at least one industrial process being modeled by the data-driven model;
- dynamically generate, when the non-biased viability identifies that the corresponding current target data set is representative of the at least one process parameter of the at least one industrial process, a plurality of industrial process standards configured to be utilized to monitor a performance consistency of the at least one industrial process of the at least one industrial plant facility based on the detected and quantified at least one chemical compound,
- wherein the plurality of industrial process standards comprise a plurality of thresholds associated with the at least one chemical compound that assists in the monitoring of the performance consistency of the at least one industrial process of the at least one industrial plant facility;
- dynamically adjust the at least one industrial process performed at the at least one industrial plant facility based at least in part on the plurality of industrial process standards according to the non-biased viability so as to enable the performance consistency of the at least one industrial process of the at least one industrial plant facility based on the at least one chemical compound;
- utilize the data-driven model to update the plurality of industrial process standards based on at least one adjustment to the at least one industrial process,
- wherein the update to the plurality of industrial process standards are applied to each industrial plant facility of the plurality of industrial plant facilities for subsequent performance consistency;
- and
- terminate the at least one industrial process performed at the at least one industrial plant facility based on a particular error value associated with the at least one process parameter exceeding a predetermined error threshold to form a termination process, the termination process comprising performance of a termination operation due to a determination of the particular error value exceeding a predetermined percent change of a previous target variable of the at least one industrial process.
16 more elements
A method, comprising:
- placing at least one sensor in an industrial plant facility that performs at least one industrial process, wherein the at least one sensor measures one or more target variables of at least one process parameter of the at least one industrial process performed at the industrial plant facility, wherein the at least one sensor further generates a current target data set for each target variable of the one or more target variables of the at least one process parameter, wherein the at least one process parameter is associated with at least one chemical compound indicative of a greenhouse gas emitted from the industrial plant facility, and wherein the at least one sensor is further configured to detect and quantify the at least one chemical compound;
- receiving by a computer unit, the current target data set for each target variable of the one or more target variables of the at least one process parameter from the at least one sensor;
- wherein the one or more target variables measured by the at least one sensor feed a model for a plurality of industrial plant facilities including the industrial plant facility,
- generating, by the computer unit, a random process data set from the current target data set;
- performing, by the computer unit, a dynamic outlier bias reduction on the current target data set for one or more bias criteria values of a set of bias criteria values to generate one or more outlier bias reduced target data set;
- performing, by the computer unit, the dynamic outlier bias reduction on the random process data set for the one or more bias criteria values of the set of bias criteria values to generate one or more outlier bias reduced random data set;
- generating, by the computer unit, a first bias criteria curve for the current target data set and a second bias criteria curve for the random process data set from the one or more bias criteria values, a set of target error values, a set of random error values, a set of target correlation coefficients, and a set of random correlation coefficients;
- dynamically determining, by the computer unit, based on the first bias criteria curve and the second bias criteria curve, a non-biased viability of the current target data set;
- wherein the non-biased viability is an indicator of whether the current target data set is representative of the at least one process parameter of the at least one industrial process being modeled by the model;
- wherein the model comprises a data-driven model,
- dynamically generating, by the computer unit, when the non-biased viability identifies that the current target data set is representative of the at least one process parameter of the at least one industrial process, at least one industrial process standard configured to be utilized to monitor a performance consistency of the at least one industrial process of the industrial plant facility based on the detected and quantified at least one chemical compound;
- wherein a plurality of industrial process standards comprise a plurality of thresholds associated with the at least one chemical compound that assists in the monitoring of the performance consistency of the at least one industrial process of the industrial plant facility;
- dynamically adjusting, by the computer unit, the at least one industrial process performed at the industrial plant facility based at least in part on the plurality of industrial process standards according to the non-biased viability so as to enable the performance consistency of the at least one industrial process of the industrial plant facility based on the at least one chemical compound;
- utilizing the data-driven model to update the plurality of industrial process standards based on at least one adjustment to the at least one industrial process,
- wherein the update to the plurality of industrial process standards are applied to the industrial plant facility and each facility of the plurality of industrial plant facilities for subsequent performance consistency;
- and
- terminating the at least one industrial process performed at the industrial plant facility based on a particular error value associated with the at least one process parameter exceeding a predetermined error threshold to form a termination process, the termination process comprising performance of a termination operation due to a determination of the particular error value exceeding a predetermined percent change of a previous target variable of the at least one industrial process.
14 more elements
Dependencies are inferred from the claim text.