SYSTEM AND METHOD FOR ONLINE AUTOMATION

摘要

A changepoint detector for modeling data received from at least one sensor in a process in the hydrocarbon industry. The data is segmented into a plurality of segments and for each segment a model is assigned and the data corresponding to the segment fit to that model. A plurality of segmentations are thus provided and these segmentations are evaluated and assigned weights indicative of the fit of the models of the segmentation to the underlying data. The segmentation models are further used to calculate a result that may be input to a process control program.

主权项

1. A method for controlling an automated hydrocarbon industrial process having a desired performance goal and at least one controllable parameter, wherein the automated process is subject to at least one dynamic constraint, comprising:
receiving a stream of measurement data from a sensor system comprising at least one sensor configured to measure a property from which a functional relationship between the desired performance goal and the at least one controllable parameter is determined; postulating that the stream of measurement data is segmented according to a plurality of possible segmentations each comprising a plurality of segments divided by changepoints, wherein each changepoint is indicative of a change in operating condition; fitting portions of the stream of measurement data corresponding to each segment in the plurality of segments to a model corresponding to the each segment in the plurality of segmentations; evaluating each of the plurality of segmentations by determining how well the models corresponding to the each segment in the plurality of segmentations fit the portions of the stream of measurement data corresponding to each segment of each segmentation; and using at least one of a most likely segmentation and models corresponding to the most likely segmentation to determine the at least one dynamic constraint and the functional relationship between the desired performance goal and the at least one controllable parameter, wherein the most likely segmentation comprises a one of the plurality of possible segmentations having a best fit between the models corresponding to the each segment in the one of the plurality of possible segmentations and corresponding portions of the stream for the one of the plurality of possible segmentations; using the functional relationship between the desired performance goal and the at least one dynamic constraint to determine a suggested parameter setting for the at least one controllable parameter, wherein the suggested parameter is determined such that according to the determined functional relationship between the at least one controllable parameter and the desired performance goal an improved performance is achieved using the suggested parameter setting while operating within the at least one dynamic constraint; and controlling the automated process by adjusting the controllable parameter to the suggested parameter.