Chevron U.s.a. Inc.

University of Southern California

Lisa Ann Brenskelle

Alisha Deshpande

Si-Zhao J Qin

Yingying Zheng

Yining Dong

Jerry M Mendel

Mohammad Reza Rajati

Andrei Popa

Martin Antonio Morles Bermudez

Methods and systems for building and maintaining model(s) of a physical process are disclosed. One method includes receiving training data associated with a plurality of different data sources, and performing a clustering process to form one or more clusters. For each of the one or more clusters, the method includes building a data model based on the training data associated with the data sources in the cluster, automatically performing a data cleansing process on operational data based on the data model, and automatically updating the data model based on updated training data that is received as operational data. For data sources excluded from the clusters, automatic building, data cleansing, and updating of models can also be applied.

Automated model building and updating environment

Methods and systems for detecting a fault in a data set from an industrial process are disclosed. One method includes forming a first data matrix at a data processing framework from time-series training data, and performing a principal component pursuit on the first data matrix to form an uncorrupted, unscaled matrix and a sparse matrix in the memory, and scaling the uncorrupted, unscaled matrix to form an uncorrupted scaled matrix. The method also includes performing a dynamic principal component analysis (DPCA) on the uncorrupted scaled matrix to form a DPCA model, and determining a squared prediction error from the DPCA model. Based on the squared prediction error, faults are detected in a different data set from operation of the industrial process. At least one of (1) correcting the one or more faults in the different data set or (2) performing a repair operation on a sensor is performed.

Fault detection system utilizing dynamic principal components analysis

Systems and methods for processing online data are disclosed. One such method includes receiving a plurality of data points in a time-series at a short term storage. The method also includes calculating at least one approximation coefficient based on the plurality of data points using a wavelet transform, including calculating a highest level approximation coefficient, and calculating estimated value based on the highest level approximation coefficient. The method further includes calculating differences between the estimated value and the plurality of data points of the short term storage, and determining whether a maximum difference among the calculated differences is less than a predetermined threshold. The method further includes, based on the maximum difference being greater than or equal to the predetermined threshold, storing the oldest data point of the short term storage in a long term storage.

Automated wavelet-based data compression systems and methods

A method of attaining a linguistic goal output of a process using a linguistic goal oriented decision making system, comprising setting the output value to a linguistic goal, and establishing a set of descriptive rules using historical data, wherein each descriptive rule has at least one variable, a decision variable, and at least one linguistic output value. The method further comprises grouping the set of descriptive rules into subsets of descriptive rules according to the linguistic output value of each descriptive rule and establishing decision rules based on comparing the grouped subsets of descriptive rules, the one or more decision rules describing a modification of a decision variable associated with at least one of the descriptive rules to affect the process.

Linguistic goal oriented decision making

A method and system are provided for reducing the volume and/or frequency of slugging in a fluid processing system that includes a pipeline for conveying produced fluids and a vessel for receiving the produced fluids from the pipeline. A control valve is provided in the pipeline upstream of the vessel. A pressure sensor and/or a level sensor is coupled to the vessel. Pressure information from the pressure sensor and/or level information from the level sensor is sent to at least one master control loop in a cascade control scheme. The master control loop output determines a set point of a slave control loop coupled to the control valve to achieve a pressure setpoint or a level setpoint. The slave control loop, also referred to as a pseudo-flow controller, determines whether the control valve opening needs be modulated to achieve the setpoint of the slave control loop. A method is also provided for retrofitting an existing fluid processing system.

Method and system for controlling slugging in a fluid processing system

A method and a control system are provided for reducing the size and/or the frequency of hydrodynamic slugging in a fluid processing system. The fluid processing system includes a pipeline for conveying produced fluids and a vessel for receiving the produced fluids from the pipeline. A control valve is provided in the pipeline upstream of the vessel. A pressure sensor is provided upstream of the control valve. Pressure information from the pressure sensor is sent to a master control loop in a cascade control scheme in which the master control loop controls a slave control loop which in turn controls the control valve. The master control loop determines a set point of the slave control loop coupled to the control valve to achieve a pressure setpoint. The slave control loop, also referred to as a pseudo-flow controller, determines whether the control valve opening needs be modulated to achieve the setpoint of the slave control loop. A method is also provided for retrofitting an existing fluid processing system.

Growing community of inventors

Houston, TX, United States of America

Lisa Ann Brenskelle