Method and device for predicting process quality in a process control system

Abstract

A method for predicting process quality in a process control system, comprising: receiving process control information relating to a process at a first time, including a first value related to a first measured variable and a second value related to a second measured variable; calculating a variation based on the received process control information by comparing the received process control information with values ​​associated with previous batches and / or with specified target values ​​for current batch data; determining whether the variation exceeds a limit based on the received process control information associated with the process; if the variation exceeds the limit, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation; determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and calculating a predicted process quality based on the at least one corrective action at a time after the first time.

Description

AREA OF REVELATION

[0001] The present disclosure relates generally to process control systems and, more particularly, to methods and apparatus for predicting process quality in a process control system. GENERAL STATE OF THE ART

[0002] Process control systems, such as those used in chemical, petroleum, or other processes, typically include one or more process control devices and input / output (I / O) devices that communicate with at least one host or operator workstation and with one or more field devices via analog, digital, or combined analog / digital buses. The field devices, which may include valves, valve positioners, switches, and transmitters (e.g., temperature, pressure, and flow sensors), perform process control functions within the process, such as opening or closing valves and measuring process control parameters.The process control devices receive signals representative of the process measurements taken by the field devices, process this information to implement a control program, and generate control signals that are sent to the field devices via the buses or other communication lines to control the operation of the process. In this way, the process control devices can execute and coordinate control strategies using the field devices via the buses and / or other communication links.

[0003] Process information from the field devices as well as control devices can be made available to one or more applications (e.g., software programs, programs, etc.) executed by the operator workstation (e.g., a processor-based system) to enable an operator to perform the desired functions related to the process, such as viewing the current status of the process (e.g., via a graphical user interface), evaluating the process, modifying the process operation (e.g., via a visual object diagram), etc. Many process control systems also include one or more user stations (e.g., workstations).Typically, these user stations are implemented using a personal computer, laptop, or similar device that communicates via a local area network (LAN) with the control devices, operator workstations, or other systems within the process control system. Each user station may have a graphical user interface that displays process control information, including the values ​​of process variables, the values ​​of process-related quality parameters, process fault diagnostic information, and / or process status information.

[0004] Typically, the display of process information on the graphical user interface is limited to displaying a value for each process variable associated with the process. Furthermore, some control systems may characterize the relationships between some process variables to determine the quality metric associated with the process. However, in cases where a product resulting from the process does not conform to the predetermined quality control metric, the process and / or process variables may not be analyzed until after the completion of a batch, process, and / or kit of the resulting product. As a result of reviewing the process and / or quality variables after the completion of a process and / or batch, improvements may be implemented to enhance the manufacturing and / or processing of subsequent products.However, these improvements may not be able to correct current products that do not meet specifications.

[0005] Document D1 (JP 2005-242818 A) concerns a method and system for analyzing and controlling product quality in manufacturing. Multivariate analysis is used to investigate how different manufacturing conditions influence product quality. In D1, real, often interdependent manufacturing data is first converted into a smaller number of independent components. The influence of each of these components on product quality is then analyzed. The results are then linked back to the original manufacturing parameters to specifically identify which conditions are critical. Based on this analysis, quality forecasts can be created and quality-improvement measures can be implemented in production.

[0006] Document D2 (DE 11 2004 000 432 T5) discloses a method for processing information relating to an operational state of machines in a process plant, the method comprising: receiving a first value relating to a monitored machine, the first value being selected from a range of values ​​of a first index, and at least a part of the range of values ​​of the first index being indicative of varying degrees of deviation from an acceptable operational state of the monitored machine; generating a second value based on the first value, the second value being selected from a range of values ​​of a second index different from the first index, the second index being indicative of varying degrees of relative functional reliability of a process entity adapted for use in process plants; and generating a message relating to the monitored machine containing the second value. SUMMARY

[0007] It is the primary object of the present invention to provide an improved method and apparatus for predicting process quality in a process control system. This object is achieved by a method according to independent claim 1 and an apparatus according to independent claim 6. Further advantageous embodiments of the invention are specified in the dependent claims. In one example, a method comprises receiving process control information relating to a process at a first time, including a first value associated with a first measured variable and a second value associated with a second measured variable.The example method further includes determining whether a variation associated with the process based on the received process control information exceeds a threshold, and if the variation exceeds the threshold, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation. The example method also includes determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value, and calculating a predicted process quality based on the at least one corrective action at a time after the first time.

[0008] An exemplary apparatus includes a batch data receiver for receiving process control information relating to a process at a first time, including a first value associated with a first measured variable and a second value associated with a second measured variable.The exemplary apparatus further comprises a processor for determining whether a variation associated with the process based on the received process control information exceeds a threshold, and if the variation exceeds the threshold, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of a second measured variable to the variation, determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value, and calculating a predicted process quality based on the at least one corrective action at a time after the first time.

[0009] Preferably, a method according to the invention further comprises receiving a selection of the first measured variable and / or the second measured variable from a portion of the second graph; graphically displaying the value associated with the selected measured variable and corresponding previous values ​​over a period of the process in a fourth graph via the user interface; displaying an average of previous values ​​associated with the selected measured variable in the fourth graph over the period of the process; and displaying a standard deviation of previous values ​​associated with the selected measured variable in the fourth graph over the period of the process.

[0010] According to another preferred embodiment of the method, the method further comprises displaying the predicted process quality in relation to a confidence interval of the predicted process quality and / or a limit value of the predicted process quality; displaying the predicted process quality over a period of time of the process; applying the at least one corrective action to the process.

[0011] According to yet another preferred embodiment of the method, the method further comprises calculating a third value associated with a calculated quality variable based on the first measured variable and / or the second measured variable; calculating a third contribution value based on a contribution of the calculated quality variable to the variation; and determining the at least one corrective action based on the third contribution value.

[0012] According to another preferred embodiment of the method, the method further comprises determining whether the first value exceeds a first limit; determining whether the second value exceeds a second limit; determining whether the third value exceeds a third limit; if the first value, the second value and / or the third value exceeds the respective one of the limit values, indicating in a graph via the user interface a process error, a deviation of the first measured variable, a deviation of the second measured variable, a deviation of the quality variable, a process deviation, a duration of the error, a cause of the process error and / or the variation of the process; and displaying the first graph upon receiving a selection of a portion of the graph.

[0013] According to another preferred embodiment of the method, the method further comprises displaying the first value and the second value in a fourth graphic via the user interface, wherein the fourth graphic can be selected from a part of the first graphic, a part of the second graphic, or a part of the diagram; and displaying in the fourth graphic a time axis corresponding to a period of the process, a bar chart indicating the first value and the second value normalized in relation to a mean and a standard deviation, and / or sparklines (word graphics) corresponding to the first variable and the second variable and showing the previous process values ​​of the first variable and the second variable.

[0014] According to another preferred embodiment of the method, the time axis contains an indication of an error associated with the variation of the process, a current course of the process in relation to the period of time of the process, an activatable arrow that changes at least one of the values ​​displayed in the bar chart to values ​​that correspond to a time position of the arrow on the time axis, and / or a course of the sparklines corresponding to a time position of the arrow on the time axis.

[0015] According to another preferred embodiment of the method, the sparklines contain an indication of a deviation if previous values ​​associated with the first measured variable and / or the second measured variable exceed a second limit value.

[0016] According to another preferred embodiment of the method, the method further comprises receiving process control information relating to the process at a second time, including a third value associated with the first measured variable and a fourth value associated with the second measured variable; determining whether a second variation associated with the process exceeds the threshold value based on the received process control information at the second time, and graphically displaying the second variation in the first graphic; if the second variation exceeds the threshold value, calculating a third contribution value based on a contribution of the first measured variable to the second variation at the second time, and a fourth contribution value based on a contribution of the second measured variable to the second variation at the second time;graphically displaying the third contribution value and the fourth contribution value in the second graph, which can be selected from a portion of the first graph; determining at least one corrective action based on the third contribution value or the fourth contribution value, and displaying the at least one corrective action in the second graph; and calculating the predicted process quality based on the at least one corrective action at a time after the second time, and graphically displaying the predicted process quality in the third graph, which can be selected from a portion of the second graph.

[0017] According to another preferred embodiment of the method, the first measured variable and the second measured variable comprise a measured process variable and / or a measured quality variable.

[0018] According to another preferred embodiment of the method, the variation is based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproduction analysis and / or a time series analysis of the first measured variable and / or the second measured variable.

[0019] According to another preferred embodiment of the method, calculating the first contribution value and the second contribution value comprises determining a relationship between the first and second contribution values ​​and the variation based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproductive analysis, and / or a time series analysis.

[0020] According to another preferred embodiment of the method, calculating the predicted process quality comprises: determining a ratio of an overall quality variable related to the predicted process quality with the first measured variable and / or the second measured variable based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproductive analysis, and / or a time series analysis; applying the at least one corrective action to the ratio; and calculating the predicted process quality from the ratio.

[0021] A preferred apparatus according to the invention comprises a display manager, the display manager being operable to: receive a selection of the first measured variable and / or the second measured variable from a portion of the second graph; graphically displaying, via the user interface, the value associated with the selected measured variable and corresponding previous values ​​over a period of the process in a fourth graph; displaying an average of previous values ​​associated with the selected measured variable in the fourth graph over a period of the process; and displaying a standard deviation of previous values ​​associated with the selected measured variable in the fourth graph over the period of the process.

[0022] According to another preferred embodiment of the device, the display manager is for the following: graphically displaying the predicted process quality in relation to a confidence interval of the predicted process quality and / or a limit value of the predicted process quality; and graphically displaying the predicted process quality over a period of time of the process.

[0023] According to yet another preferred embodiment of the device, the processor serves to apply the at least one corrective action to the process.

[0024] According to another preferred apparatus, the processor is operable to: calculate a third value associated with a calculated quality variable based on the first measured variable and / or the second measured variable; calculate a third contribution value based on a contribution of the calculated quality variable to the variation; determine the at least one corrective action based on the third contribution value; and calculate the predicted process quality based on the at least one corrective action.

[0025] According to another preferred apparatus, the processor is operable to: determine whether the first value exceeds a first limit; determine whether the second value exceeds a second limit; determine whether the third value exceeds a third limit; and if the first value, the second value, and / or the third value exceeds each of the limits, indicate in a graph via the user interface a process error, a deviation of the first measured variable, a deviation of the second measured variable, a deviation of the quality variable, a process deviation, a duration of the error, a cause of the process error, and / or the variation of the process.

[0026] According to another preferred device, the display manager is operable to: display the first graph upon receiving a selection of a portion of the graph; display the first value and the second value in a fourth graph via the user interface, which can be selected from a portion of the first graph, a portion of the second graph, and / or a portion of the graph; and display in the fourth graph a time axis corresponding to a period of the process, a bar chart indicating the first value and the second value normalized in relation to a mean and a standard deviation, and / or sparklines (word graphics) corresponding to the first variable and the second variable and showing the previous process values ​​of the first variable and the second variable.

[0027] According to another preferred apparatus, the batch data receiver is to receive process control information relating to the process at a second time, including a third value associated with the first measured variable and a fourth value associated with the second measured variable; the processor is to perform the following: determining whether the second variation associated with the process exceeds the threshold value based on the received process control information at the second time; if the second variation exceeds the threshold value, calculating a third contribution value based on a contribution of the first measured variable to the second variation at the second time and a fourth contribution value based on a contribution of the second measured variable to the second variation at the second time;determining at least one corrective action based on the third contribution value or the fourth contribution value; and calculating the predicted process quality by applying the at least one corrective action at a time after the second time; and the display manager is to perform the following: graphically displaying the second variation in the first graphic via the user interface; graphically displaying the third contribution value and the fourth contribution value in the second graphic via the user interface, wherein the second graphic is selectable from a portion of the first graphic; graphically displaying the at least one corrective action in the second graphic; and graphically displaying the predicted process quality in the third graphic via the user interface, wherein the third graphic is selectable from a portion of the second graphic.

[0028] According to another preferred apparatus, the processor is to calculate the predicted process quality by determining a ratio of an overall quality variable related to the predicted process quality to the first measured variable and / or the second measured variable based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproductive analysis, and / or a time series analysis; and applying the at least one corrective action to the ratio.

[0029] According to another preferred apparatus, the processor determines the variation based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproductive analysis, and / or a time series analysis of the first measured variable and / or the second measured variable.

[0030] According to another preferred device, the device further comprises a process model generator for determining a relationship between the first and second contribution values ​​and the variation based on a multivariate analysis, an algebraic analysis, an optimization analysis, a previous batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproduction analysis and / or a time series analysis.

[0031] According to another preferred device, the processor calculates the first contribution value and the second contribution value based on the determined ratio. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram illustrating an example process control system, including an example operations management system. Fig. Figure 2 shows a data structure for an example batch, including process variables and quality variables. Fig. Figure 3 shows a data structure for example batches, including process variables and corresponding quality variables. Fig. 4 is a functional diagram of the exemplary operations management system of Fig. 1. Fig. 5 is a diagram illustrating charts and / or graphs generated by the operations management system of Fig. 1 can be created which can be displayed to indicate a fault within the process control system. Fig. 6 illustrates a user interface that displays an example process overview diagram of the example process control system of Fig. 1. Fig. 7 presents the user interface of Fig. 6, which shows an example process variation graph, including unexplained and explained variation graphs for the example batch process of Fig. 6 displays. Fig. 8 presents the user interface of Fig. 7, which shows the example process variation graph, including a contribution summary panel. Fig. 9 presents the user interface of Fig. 6, which shows an example contribution graph, including unexplained and explained process variations for the processes implemented with the process control system of Fig. 1 related variables. Fig. 10 presents the user interface of Fig. 6, which shows a trend graph of the media flow variable of Fig. 8 and Fig. 9. Fig. 11 presents the user interface of Fig. 6, which shows a quality prediction graph for the exemplary batch of Fig. 6 displays. Fig. 12 presents the user interface of Fig. 6, which shows an example micrograph at a first time point, including the sparklines, as well as a bar chart for some of the variables of Fig. 9. Fig. 13 presents the user interface of Fig. 6, which shows the exemplary micro diagram of Fig. 12 at a selected second time. Fig. 14A and Fig. 14B represent the user interface of Fig. 6, which shows exemplary sparklines for the mixer temperature process variable of Fig. 12 and Fig. Show 13. Fig. 15, 16A-16F and 17A-17B are flow diagrams of example methods that may be employed to configure the example operations management system, example user interface, example analytics processor, process model generator and / or display manager of Fig. 1 and / or 4 to be implemented. Fig. 18 is a block diagram of an example processor system that may be employed to implement the example methods and apparatus described herein. DETAILED DESCRIPTION

[0032] Although example methods and apparatus, including additional components, software, and / or firmware embodied in hardware, are described below, it should be noted that these examples are for illustrative purposes only and should not be considered limiting. It is also contemplated that any or all of the hardware, software, and firmware components may be embodied exclusively in hardware, exclusively in software, or in any combination of hardware and software. Accordingly, those skilled in the art will appreciate that the examples described below do not represent the only manner in which the recited methods and apparatus may be implemented.

[0033] Currently, process control systems provide analytical and / or statistical analysis of process control information. However, these systems implement offline tools to determine the cause and potential corrective actions for process control failures that may impact final product quality. These offline tools may include process studies, laboratory studies, business studies, troubleshooting, process improvement analyses, and / or Six Sigma analyses. Although these tools can correct the process for subsequent products, they are unable to improve or correct process quality while the failure is occurring. Therefore, these offline tools react to process control conditions and may result in the production of products of reduced quality until the process can be corrected.

[0034] The exemplary methods and apparatus described herein can be employed within a process control system to provide in-process troubleshooting, analysis, and / or corrective information to enable an operator to correct a process error while the process is occurring or running. In other words, process corrections can be made in response to predicted errors at the time a error occurs or substantially immediately after a error occurs. The methods and apparatus described herein can be employed to predict and / or correct process errors to improve the process quality of a batch and / or an ongoing process.Furthermore or alternatively, the exemplary methods and apparatus may be used to improve product quality by predicting product quality and correcting the corresponding process defects and / or by correcting detected process defects.

[0035] An exemplary operations management system (OMS) described herein manages the analysis, organization, and display of process control information generated by the process control system. The exemplary OMS provides process control information to workstations authorized to retrieve the information. A process control system may include any type of batch processing system, continuous processing system, automated system, and / or manufacturing system.

[0036] The process control information is generated by field devices within the process control system and can be used for, for example, process environment measurements (such as temperature, concentration, or pressure sensing), field device measurements (such as pump speed, valve position, or machine speed), process status measurements, and / or process throughput measurements. The process control information can be received from a control device as variable output data (e.g., the value associated with variables coming from the field devices). The control device can then forward this variable data to the OMS for analytical and statistical processing based on the process models.

[0037] The exemplary OMS described herein uses descriptive modeling, predictive modeling, and / or optimization to generate feedback regarding the status and / or quality of the process control system. Furthermore, the OMS predicts, detects, identifies, and / or diagnoses process operation faults and predicts the impact of any faults on quality variables related to the quality of a final product. A fault may include a process quality fault that is a multivariate and / or statistical combination of outputs from one or more field devices that exceeds a threshold. Additionally, a fault may include output values ​​from one or more field devices that exceed respective thresholds. Furthermore, the exemplary OMS manages access to and control of process control data using a graphical user interface.This user interface can be used to alert a process control operator to the occurrence of any process control fault. Additionally, the user interface can guide the operator through an analysis process to determine the source of a process fault and predict the impact of correcting the process fault on the quality of the final product(s). The user interface can be accessed by any process control operator with access to the information, including process operators who can access the user interface remotely via the internet. The OMS can provide a process control operator with process information while the process is in progress, allowing the operator to make adjustments to the process to correct faults.By making error corrections, the operator can maintain the quality of the final product and subsequent products.

[0038] The example OMS can display the capture, analysis, corrective action, and quality prediction information in one or more diagrams and / or graphs within the graphical user interface. For example, a process overview diagram can display the status of one or more processes being monitored. From this overview diagram, an operator can select a process variation graph that displays any explained (e.g., modeled) and / or unexplained (e.g., unmodeled) variations within the process. The operator can select a point on the process variation graph that displays the contributions of unexplained and explained variations of each process and / or the quality associated with the overall process variation.The exemplary OMS may determine the contribution relationships between the process and / or quality variables based on modeling and / or analyzing the process control system.

[0039] From the contribution chart, an operator can identify one or more variables contributing to the process defect. The operator can then select one or more variables to display a variable trend graph, which shows values ​​associated with each variable during the process relative to previous averages and / or deviations of the variables during previous processes. The operator can then access a process quality prediction graph to determine whether the defect will significantly affect the resulting quality of a product resulting from the process. If quality has not been significantly affected, the operator can apply the variable trend information to make appropriate adjustments to the process to correct defects.

[0040] In an alternative implementation, an operator can display a microchart showing the defect types associated with each of the process and / or quality variables. The microchart can also include sparklines that display summary information about previous values ​​of each process and / or quality variable during the process. Furthermore, the microchart can include a timeline displaying previous and / or current defects and / or deviations. The operator can scroll along the timeline at a time to view variable contributions, the values ​​of the process and / or quality variables at the selected time. The microchart can also display predicted, calculated, and / or overall quality variables for a selected time. The operator can select any variable to view a variable trend graph for that variable or, alternatively, navigate to a process quality prediction graph.Therefore, by providing multivariable analytical data and / or predictive data within easy-to-read and linked charts and / or graphs (e.g., graphs), the exemplary methods and apparatus described herein may enable the process control operator to relatively easily address process control errors before they adversely affect the quality of a final product.

[0041] While the exemplary methods and apparatus mentioned herein refer to exemplary microdiagrams, sparklines, combination graphs, overview graphs, process variation graphs, process quality prediction graphs, and / or variable trend graphs, the microdiagrams, sparklines, combination graphs, overview graphs, process variation graphs, process quality prediction graphs, and / or variable trend graphs may be graphically implemented in any manner such that they display the same type of process control information and / or relationships between the process control information.Furthermore, while the example OMS generates the example microdiagrams, sparklines, combination graphs, overview graphs, process variation graphs, process quality prediction graphs, and / or variable trend graphs, any additional process controller, server, workstation, and / or component may also generate these microdiagrams, sparklines, combination graphs, overview graphs, process variation graphs, process quality prediction graphs, and / or variable trend graphs.

[0042] Fig. 1 is a block diagram illustrating an example process control environment 100 including an example operations management system (OMS) 102. In further examples, the OMS 102 may be known as a process monitoring and quality prediction system (PMS). The example OMS 102 is located in a plant 104 that includes a process control system 106. The example plant 104 may be any type of manufacturing facility, processing facility, automation facility, and / or any type of process control structure or system. In some examples, the plant 104 may include multiple facilities located at different locations. Although the example plant 104 includes the process control system 106, the plant 104 may also include additional process control systems.

[0043] The exemplary process control system 106 is communicatively coupled to a controller 108 via a data bus 110. The process control system 106 may include any number of field devices (e.g., input and / or output devices). The field devices may include any type of process control component capable of receiving input, generating output, and / or controlling a process. The field devices may include, for example, input devices such as valves, pumps, fans, heaters, chillers, and / or mixers to control a process. The field devices may also include output devices such as thermometers, pressure gauges, concentration gauges, liquid level gauges, flow meters, and / or vapor sensors to measure portions of a process. The input devices may receive instructions from the controller 108 to execute specific commands to effect a change in the process.Furthermore, the output devices measure process data, environmental data, and / or input device data and transmit the measured data to the controller 108 as process control information. This process control information may include the values ​​of the variables (e.g., the measured process variables and / or the measured quality variables) that correspond to a measured output from each field device.

[0044] In the example shown by Fig. 1, the exemplary controller 108 can communicate with the field devices within the process control system 106 via the data bus 110. This data bus 110 can be coupled to intermediate communication components within the process control system 106. These communication components can include field terminal boxes for coupling the field devices in a command domain in a communication connection to the data bus 110. Furthermore, the communication components can include marshalling cabinets for organizing the communication paths to the field devices and / or to the field terminal boxes. In addition, the communication components can include I / O cards for receiving data from the field devices and converting the data into a communication medium capable of being received by the exemplary controller 108.These I / O cards can also convert data from the controller 108 into a data format that can be processed by the corresponding field devices. In one example, the data bus 110 can be implemented using the Fieldbus protocol or other types of wired and / or wireless communication protocols (e.g., Profibus protocol, HART protocol, etc.).

[0045] The exemplary control unit 108 of Fig. 1 manages one or more control routines for managing the field devices within the process control system 106. The control routines may include process monitoring applications, alarm management applications, process trend and / or history applications, batch processing and / or action management applications, statistical applications, video streaming applications, advanced control applications, etc. Furthermore, the controller 108 forwards process control information to the example OMS 102. The control routines may ensure that the process control system 106 produces specific quantities of a desired product within a certain quality limit. The process control system 106 may, for example, be configured as a batch system that produces a product at the end of a batch process.In further examples, the process control system 106 may include a continuous process manufacturing system that continuously manufactures products.

[0046] The process control information from the controller 108 may include values ​​corresponding to measured process and / or quality variables originating from field devices within the process control system 106. In further examples, the OMS 102 may parse values ​​within the process control information into the corresponding variables. The measured process variables may be associated with the process control information originating from field devices that measure portions of the process and / or characteristics of the field devices. The measured quality variables may be associated with the process control information associated with the measured characteristics of the process associated with at least a portion of a finished product.

[0047] For example, the process control system 106 may include a chemical reaction in a tank that produces a concentration of a chemical in a liquid. In this example, the concentration of the chemical in the liquid may be a quality variable. A temperature of the liquid and the rate of liquid flow into the tank may be process variables. The example OMS 102 may determine, via process control modeling and / or monitoring, that the concentration of the liquid in the tank is based on the temperature of the liquid in the tank and the flow rate of the liquid into the tank. Therefore, not only is the concentration a quality variable, but the flow rate of the liquid and the temperature of the liquid also contribute to or affect the quality of the concentration.In other words, the measured process variables contribute to or influence the quality of the measured quality variables. The OMS 102 can use statistical processing to determine the extent of influence and / or contribution of each process variable to or for a quality variable.

[0048] Additionally, the example OMS 102 may model and / or determine the relationships between the measured process variables and / or the quality variables associated with the process control system 106. These relationships between the measured process and / or quality variables may generate one or more calculated quality variables. A calculated quality variable may be a multivariate and / or linear algebraic combination of one or more measured process variables, measured quality variables, and / or other calculated quality variables. Further, the OMS 102 may determine an overall quality variable from a combination of the measured process variables, the measured quality variables, and / or the calculated quality variables. The overall quality variable may correspond to a determination of the quality of the entire process and / or may correspond to a predicted quality of an end product of the process.

[0049] The exemplary OMS 102 from Fig. 1 includes an analytics processor 114. The exemplary analytics processor 114 employs descriptive modeling, predictive modeling, and / or optimization to generate feedback about the status and / or quality of the process control system 106. The analytics processor 114 may detect, identify, and / or diagnose process operation errors and predict the impact of any errors on the quality variables and / or an overall quality variable related to a quality of an end product of the process control system 106. Further, the analytics processor 114 may monitor the quality of the process by statistically and / or logically combining the quality and / or process variables into an overall quality variable related to the overall quality of the process.The analytics processor 114 can then compare the values ​​calculated for the overall quality variable and / or the values ​​associated with the other quality variables with the corresponding threshold values. These threshold values ​​can be based on the predetermined quality limits of the overall quality variable at various times within the process. For example, if an overall quality variable associated with a process exceeds a threshold value for a period of time, the predicted final quality of the end product may not meet the quality metrics associated with the end product.

[0050] If the overall quality variable and / or other quality variables deviate from the respective limits, the analytics processor 114 may generate a fault indication within a process overview chart and / or a process variation graph that shows an explained and / or unexplained variation (or deviation) related to the overall quality variable and / or that may show a variable that generated the process fault. The exemplary analytics processor 114 manages the analysis to determine a cause of one or more process faults by providing functionality that allows an operator to generate process quality graphs (e.g., combination graphs, micrographs, process variation graphs, variable trend graphs, graphs, etc.) that may display current and / or previous values ​​of the measured process variable, the measured quality variable, and / or the calculated quality variable.Furthermore, the analytics processor 114 generates these graphs while the process is operating and continuously updates and / or recalculates multivariate statistics associated with each of the graphs as additional process control information is received from the OMS 102.

[0051] The analytics processor 114 can generate a contribution graph by calculating the contributions of the process variables and / or the quality variables to the overall quality variable or the quality variable causing the defect. The contributions of the process and / or quality variables can be displayed as an explained and / or unexplained variation of each variable as a contribution to the variation associated with the overall quality and / or the quality variable associated with the defect.

[0052] Furthermore, the example analytics processor 114 may generate variable trend graphs for any selected process and / or quality variables that may exhibit variations greater than a defined threshold. The variable trend graph may display values ​​associated with the variable over a period of the process relative to the values ​​of the variable during similar periods in previous processes. By generating the contribution graph and / or the variable trend graphs, the analytics processor 114 may also identify possible corrections to the process to eliminate the detected error. The variable trend graph may assist an operator in determining a cause of a process error by providing an overlay of graphical representations of the history with associated variations (e.g., standard deviations) from the current value.

[0053] The analytics processor 114 can generate a quality prediction graph to determine the impact of the correction(s), if implemented, on the overall quality of the process. If the correction(s) maintain or improve overall quality within prescribed limits, the analytics processor 114 can indicate to the OMS 102 to implement the correction(s). Alternatively, the analytics processor 114 can send instructions to the controller 108 to implement the process correction(s).

[0054] Furthermore, the example analytics processor 114 may generate a microchart upon detecting a failure associated with an overall quality variable and / or any other quality variable. The microchart may include values ​​of the process and / or quality variables at a prescribed time (e.g., a time associated with the process failure) relative to an average and / or standard deviation for each of the variables. Additionally, the microchart may include sparklines indicating previous values ​​associated with each of the process and / or quality variables.From the micrograph, the example analytics processor 114 may enable an operator to determine and / or select one or more corrective actions in the process and / or determine whether the corrections will improve the process such that the overall quality variable is expected to be within the prescribed limits.

[0055] The exemplary OMS 102 manages access and control to the process control data, including the process variation graphs, contribution graphs, variable trend graphs, quality prediction graphs, and / or the micrographs, via an online data processor 116. In addition, the online data processor 116 provides the process control operator with access to view the process control data, change and / or modify the process control data, and / or generate instructions for field devices within the process control system 106.

[0056] The exemplary system 104 of Fig. 1 includes a router 120 and a local workstation 122, which are communicatively connected to the online data processor 116 via a local area network (LAN). Furthermore, the exemplary router 120 may couple any other workstations (not shown) within the facility 104 to the LAN 124 and / or the online data processor 116 in a communicative connection. The router 120 may be communicatively connected to the other workstations wirelessly and / or via a wired connection. The router 120 may comprise any type of wireless and / or wired router as an access network node to the LAN 124 and / or the online data processor 116.

[0057] The LAN 124 can be implemented using any desired communication medium and protocol. For example, the LAN 124 can be based on a wired or wireless Ethernet communication scheme. However, any other suitable communication medium and protocol can also be used. Furthermore, although a single LAN is illustrated, more than one LAN and corresponding communication hardware can be used within the workstation 122 to provide redundant communication paths between the workstation 122 and a corresponding similar workstation (not illustrated).

[0058] The LAN 124 is also communicatively connected to a firewall 128. The firewall 128 determines, based on one or more rules, whether to permit communication from remote workstations 130 and / or 132 within the facility 104. The example remote workstations 130 and 132 may provide access to resources within the facility 104 to operators who are not located within the facility 104. The remote workstations 130 and 132 communicate with the firewall 128 via a wide area network (WAN) 134.

[0059] The example workstations 122, 130, and / or 132 may be configured to view, modify, and / or correct one or more processes within the process control system 106. The workstations 122, 130, and / or 132 may, for example, include a user interface 136 that prepares and / or displays process control information generated by the OMS 102. The user interface 136 may, for example, receive graphics and / or diagrams, or alternatively, data for generating a process control graphic and / or diagram from the OMS 102. Upon receipt of the graphic and / or diagram data at the respective workstation 122, 130, and / or 132, the user interface 136 may generate a display of a graphic and / or diagram 138 that can be relatively easily understood by the operator. The example in Fig. 1 shows workstation 132 with user interface 136. However, workstations 122 and / or 130 may include user interfaces 136.

[0060] The exemplary user interface 136 can alert a process control operator to the occurrence of any process control errors within the process control system 106 and / or any other process control system within the plant 104. Furthermore, the user interface 136 can guide a process control operator through an analysis process to determine a source of a process error and predict an impact of the process error on the quality of the final product. The user interface 136 can provide the operator with statistical information about the process control while the process is running, allowing the operator to make adjustments to the process to correct errors. By correcting errors during the process, the operator can maintain the quality of the final product.

[0061] The example user interface 136 may display the capture, analysis, corrective action, and quality prediction information via the example OMS 102. The user interface 136 may display, for example, a process overview diagram, a process variation graph, a micrograph, a contribution graph, a variable trend graph, and / or a quality prediction graph (e.g., graph 138). After viewing these graphs 138, the operator may select additional graphs 138 to view multivariate and / or statistical process information to determine a root cause of a process failure. Furthermore, the user interface 136 may display possible corrective actions for a process failure. The user interface 136 may then allow an operator to select one or more corrective actions.After selecting a correction, the user interface 136 can transmit the correction to the OMS 102, which then sends an instruction to the controller 108 to make the corresponding correction in the process control system 106.

[0062] The exemplary workstations 122, 130 and / or 132 of Fig. 1 may include any computing device, including a workstation, a laptop, a server, a controller, a personal digital assistant (PDA), a microcomputer, etc. The workstations 122, 130 and / or 132 may be implemented using a suitable computer system or processing system (e.g., the processor system P10 of Fig. 18). Workstations 122, 130, and / or 132 could be implemented, for example, using a single processor workstation, single or multiprocessor workstations, etc.

[0063] The exemplary process control environment 100 is provided to illustrate one type of system in which the exemplary methods and apparatus described in more detail below may be advantageously employed. However, if desired, the exemplary methods and apparatus described herein may also be advantageously employed in other systems that are more or less complex than those described in Fig. 1 and / or the process control system 106 and / or in systems used in connection with the process control processes, operations management processes, communication processes, etc.

[0064] Fig. 2 illustrates a data structure 200 for an exemplary batch (e.g., Batch No. 1), including the measured variables 202 and the calculated quality variables 204. The exemplary data structure 200 may also include an overall quality variable (not shown). Batch processing is a type of product manufacturing wherein a relatively large number of products and / or portions of products are produced in parallel at one or more locations controlled by a routine. The routine may include one or more process operations, each operation comprising one or more operations and each operation comprising one or more phases. In further examples, a process control system for producing products may employ continuous processing.Continuous processing is similar to assembly line manufacturing, where processes and / or operations are performed serially as a single function on a single product (or only a few products) at a time. Although the methods and equipment described here refer to batch processes, any type of process can be implemented.

[0065] The exemplary measured variables 202 include measured process and / or quality variables. For example, the variable P1 may correspond to a liquid flow rate (e.g., a process variable) and the variable P2 may correspond to a liquid concentration (e.g., a quality variable). The measured variables 202 are shown in connection with the batch process BATCH No. 1. The batch process takes place during a period of time shown along the z-axis (e.g., TIME). Furthermore, the batch process of Fig. 2 eight measured variables. However, in further examples, the batch process may contain fewer or more measured variables.

[0066] Fig. Figure 2 shows that each of the measured variables 202 is only relevant for specific points in time during the batch process. For example, the variable P1 is relevant from the start of the batch to a midpoint during the batch. Therefore, if the variable P1 is associated with a liquid flow rate, liquid can only flow from the start of the batch to a midpoint during the batch process. After this point, the batch may not utilize a liquid flow, and therefore the variable P1 is no longer relevant to the batch process beyond this point. In contrast, the variable P4 is relevant to the entire batch process.

[0067] The exemplary calculated quality variables 204 are related to the entire batch process. The calculated quality variables 204 may be the result of a multivariate, statistical, and / or algebraic relationship between the measured variables 202 and / or the quality variables 204. The quality variable Q1 204 may, for example, correspond to a composition quality of a product resulting from the batch process. The composition quality Q1 may be a quality variable because it may not be directly measurable in the process control system 106. Instead, the composition quality Q1 may be modeled and / or determined from a multivariate combination of the measured variables 202 P1, P3, P4, and P7.Therefore, if the composition quality Q1 exceeds a defined limit, any and / or a combination of the measured variables P1, P3, P4 and / or P7 may be a contributing factor to the deviation.

[0068] Fig. 3 illustrates a data structure 300 for exemplary batches, including process variables 302 and corresponding quality variables 304. The batches (e.g., BATCH 1-7) indicate that a batch process comprises operations (e.g., OPERATIONS 1-4) performed in serial order. OPERATION 1 may, for example, correspond to a combination and mixture of chemicals in a batch, while OPERATION 2 corresponds to a curing of these mixed chemicals in a batch. These operations can be further divided into operations, phases, and / or stages. Furthermore, the calculated quality variables 306 correspond to the measured variables 302 at each batch.

[0069] The example in Fig. Figure 3 shows that each batch may differ in terms of time period, with the start and completion of each operation differing from batch to batch. For example, BATCH 2 is completed in a shorter time period than BATCH 2, but BATCHES 3 and 4 are completed after a longer time period than BATCH 1. Furthermore, BATCH 1 requires a longer time period to complete OPERATION 1 than BATCH 2. However, the relevant time period of each variable (not shown) may be proportional to the time period for the corresponding operation(s). Therefore, the varying time to complete batches and / or operations can be decided by the measured variables 302 in each batch. As a result of the proportional time period of the measured variables, comparisons can be made between the measured variable values ​​between batches.For example, the value of the measured variable P1 at 50% across OPERATION 1 of BATCH 1 should be essentially the same as the variable P1 at 50% across OPERATION 1 of BATCHES 2 - 7.

[0070] Fig. 4 is a functional diagram of the exemplary operations management system (OMS) 102 of Fig. 1. The exemplary OMS 102 processes process control information from the control unit 108 of Fig. 1, establishes models related to variables associated with the process control system 106, calculates quality variables from the process control information, determines whether variables exceed corresponding limits, generates display information for user interfaces, and / or manages access to the process control information. While the exemplary OMS 102 of Fig. 4 further includes functional blocks configured to execute processes, the OMS 102 may also connect functional blocks or include additional functional blocks. In some examples, the OMS 102 may be associated with a single process control system (e.g., the process control system 106), while in other examples, the OMS 102 may process data from a plurality of process control systems. While the OMS 102 is further described as processing and managing batch data, the OMS 102 may be capable of processing and / or managing data associated with continuous, automated, and / or manufacturing-type processes.

[0071] To receive and process data from the controller 108, the example OMS 102 includes a batch data receiver 402. The example batch data receiver 402 receives process control information from the controller 108 via a communication path 404. The example communication path 404 may include any type of wired and / or wireless communication path. The process control information may include output data from the field devices within the process control system 106. This output data may be received by the batch data receiver 402 as values ​​corresponding to the measured process and / or quality variables originating from the field devices. In further examples, the process control data may be received as a file containing values ​​corresponding to the outputs of the field devices.In these examples, batch data receiver 402 may determine the appropriate measured variables corresponding to the values ​​by noting the field devices from which they originate. Furthermore, batch data receiver 402 may receive the process control information from controller 108 at regular times as controller 108 receives the process control information and / or by querying the process control information from controller 108.

[0072] After receiving the process control information, the exemplary batch data receiver 402 organizes Fig. 4 the values ​​from the field devices according to the corresponding variable and / or according to the time at which the data was generated by the field devices. The batch data receiver 402 can also organize values ​​according to an operation, an operation, a process, a batch number, and / or an event within the process control system 106. For example, the batch data receiver 402 can organize the values ​​according to a batch identification (e.g., BATCH No. 7), a batch operation (e.g., OPERATION 2), a batch operation within the operation (e.g., heat), etc. After organizing the values ​​according to the measured variables, the batch data receiver 402 stores the organized information in a batch data database 406. Furthermore, the batch data receiver 402 can retrieve stored batch data and / or transmit batch data to the analytics processor 114.The batch data database 406 may be implemented by electronically erasable read-only memory (EEPROM), random access memory (RAM), read-only memory (ROM), and / or any other type of memory.

[0073] The exemplary analytics processor 114 manages the calculation, modeling, and / or fault detection associated with the process control system 106. The analytics processor 114 includes an analytical process modeler 408 for calculating values ​​associated with calculated variables from values ​​associated with measured variables. The analytical process modeler 408 utilizes models of the process control system 106 to determine the relationships among the measured process and / or quality variables and / or to determine the relationships between the measured variables and the calculated and / or total variables.The analytical process modeler 408 calculates the values ​​for the calculated and / or overall quality variables by including the values ​​associated with the measured variables in the process models, processing the models with the measured values, and / or receiving the values ​​associated with the calculated quality and / or overall quality variables as an output of the model. In further examples, the measured data may be organized within a graphical space that employs multivariate, optimization, geometric, and / or algebraic projections of the measured values ​​to extract the values ​​and / or generalizations associated with the calculated and / or overall quality variables.

[0074] For example graphics that employ variations, the example analytical process modeler 408 may calculate the variation of the current batch data by comparing the values ​​associated with the current batch data (e.g., measured, calculated, and / or total variables) with values ​​associated with previous batches and / or with specified target values ​​for the current batch data. Calculating the variation may involve applying a T 2 and / or a Q statistical test on the values ​​related to the measured and / or calculated variables. The T 2 For example, the Q statistic can be used to determine explained variation in the process control variables, while the Q statistic can be used to determine the acceptability of outlying values ​​associated with the variables that represent unexplained variation.

[0075] Additionally, the exemplary analytical process modeler 408 may determine the contribution amounts of measured variables with variations related to measured, calculated, and / or overall quality variables. The analytical process modeler 408 may determine the contribution amounts of the measured variables based on process models of the process control system 106. Based on these models, the analytical process modeler 408 may apply the measured variable values ​​to the models, process the models, and / or identify a contribution factor for each variable.

[0076] Furthermore, the analytical process modeler 408 can use values ​​related to measured, calculated, and / or overall variables to predict the quality of a process after implementation of a correction. The analytical process modeler 408 can predict the quality by examining the current process variables and using predictive models that account for the correction and / or quality of previous batches with similar data characteristics. Furthermore, the analytical process modeler 408 can calculate statistical confidence intervals for the predicted quality, which can, for example, indicate with 95% certainty what the quality of a final product will be.

[0077] The exemplary analytical process modeler 408 of Fig. 4 may calculate total and / or calculated variables, determine amounts of variable contributions, predict process quality, and / or calculate variation statistics as process control information received from controller 108. In further examples, analytical process modeler 408 may calculate total and / or calculated variables, determine amounts of variable contributions, predict process quality, and / or calculate variation statistics at predetermined time periods upon query by a process control operator. Additionally, analytical process modeler 408 may calculate total and / or calculated variables, determine amounts of variable contributions, predict process quality, and / or calculate variation statistics as a corresponding graphic that is used and / or queried by a process control operator viewing the process control information on a user interface. For example, analytical process modeler 408 maypredict process quality only when a process control operator selects to view a quality prediction graph. Alternatively, the analytical process modeler 408 may provide the total and / or calculated variables, the variable contribution quantities, the predicted process quality, and / or the variation statistics on further functional blocks within the OMS 102 for display to a process control operator. Further, the analytical process modeler 408 may transmit the total and / or calculated quality variables, the variable contribution quantities, the predicted process quality, and / or the variation statistics to a user interface (e.g., user interface 136) associated with access to the process control information by a process control operator.

[0078] The exemplary analytics processor 114 of Fig. 4 includes an evaluation process modeler 410 to determine whether the calculated quality values, process variations, variable contribution amounts, and / or predicted quality values ​​exceed a corresponding predetermined threshold. The exemplary evaluation process modeler 410 may receive the thresholds from a model associated with the process control system 106. Alternatively, the evaluation process modeler 410 may receive the thresholds from a process control operator. These thresholds may be based on maximum and / or minimum metrics and / or values ​​to ensure that a product of the process control system 106 meets the quality standards.

[0079] The exemplary evaluation process modeler 410 may also generate fault indications if a measured, calculated, and / or total variable value exceeds a threshold. In addition, the evaluation process modeler 410 may generate a fault indication if a variation associated with a measured, calculated, and / or total variable exceeds a threshold. After generating a fault indication, the evaluation process modeler 410 may transmit the fault indication to another functional block within the OMS 102 and / or to a user interface within a workstation that controls the process control data related to the fault. Furthermore, the evaluation process modeler 410 may predict process faults based on trends in the measured and / or calculated variable values ​​and / or the variations associated with these variables. The evaluation process modeler 410 may, for example,determine that a process failure is occurring based on a long-term increase in the values ​​of a measured variable. If the evaluation process modeler 410 determines that a process failure may occur, the evaluation process modeler 410 may generate a predictive failure indication.

[0080] Additionally, the evaluation process modeler 410 may determine whether a quality prediction calculation from the analytical process modeler 408 exceeds or falls short of quality limits. Furthermore, the evaluation process modeler 410 may determine whether confidence intervals of the predicted quality exceed or otherwise fall short of a corresponding limit. If any of the confidence intervals exceed a quality limit, the evaluation process modeler 410 may send a predictive failure indication. The predictive failure indication may alert a process control operator that corrective action may not bring the process within the quality control limits or boundaries.Based on a predictive failure indication, a process control operator may decide to abort a process due to time constraints and / or restart the process because the final product may not meet quality standards, metrics, and / or limits.

[0081] The exemplary evaluation process modeler 410 from Fig. 4 may determine whether variable values, variations, and / or quality predictions exceed or fail to meet a threshold after each occurrence of process control data being received by the OMS 102 from the controller 108. Alternatively, the evaluation process modeler 410 may determine whether variable values, variations, and / or quality predictions exceed a threshold after predetermined time periods and / or upon query by a process control operator.

[0082] In order to determine process models of the process control system 106, the exemplary OMS of Fig. 4, a process model generator 412. The exemplary process model generator 412 determines models used by the analytical process modeler 408, the analytical processor 114, and / or the evaluation process modeler 410. The models may define relationships between measured variables, quality variables, and / or overall process quality, threshold values ​​for the measured variables, quality variables, and / or overall process quality, and / or graphical display styles for the measured variables, quality variables, and / or overall process quality.

[0083] The example process model generator 412 may construct and / or determine a process model based on receiving a list from the field devices within the process control system 106, the inputs and / or outputs associated with each field device, the arrangement of the field devices, and / or the interconnection among the field devices. Additionally, the process model generator 412 may utilize prior process relationships between some field devices to predict relationships between the field devices and / or the outputs of the field devices. These relationships may also be determined by applying multivariate statistical methods to analyze the data from the field devices to determine relationships and / or contributions based on interconnections between the field devices.Multivariate analyses may include various analyses, principal component analysis, projection to latent structures analysis, and / or multivariate process control analyses. In addition to multivariate analyses, the process model generator 412 may employ exploratory data analyses, control and capacity analyses, regression analyses, correlation analyses, analysis of variance (e.g., ANOVA), repeatability analyses, reproductive analyses, and / or time series analyses.

[0084] The prior relationships, multivariate analysis, and / or statistical analysis may be used by the process model generator 412 to determine threshold values ​​for each of the measured variables, calculated variables, predicted quality values, variations, etc. The threshold values ​​may also be determined through experimentation, design prototyping, draft disturbance effects analysis, pre-production process control system prototyping, process control operator calculations, and / or any other method that can accurately predict and / or calculate process quality.

[0085] In addition, the process model generator 412 can determine relationships between variables by analyzing functional diagrams, algorithms, routines, and / or any other type of command structure that defines the control and / or operation of the process control system 106. Process control operators can also define relationships between variables based on experience and / or calculations performed outside of the process model generator 412. Furthermore, the process model generator 412 can also determine conditions for applying corrective actions to correct a process error by analyzing previous batch data with errors and effects on the corrective action(s). In addition, corrective actions can be determined by linking inputs to a field device with measured variables that may represent the cause of the error. For example, ifIf a flow rate is determined to be the cause of a fault, the process model generator 412 may determine that the flow rate may be adjusted by a valve and / or pump. Therefore, changing an input to the valve and / or pump may be determined as a corrective action for an inconsistent flow rate. Alternatively, the process model generator 412 may receive corrective action from a process control operator.

[0086] Furthermore, the process model generator 412 may employ any multivariate and / or statistical methods to determine the relationship between measured and / or calculated variables and overall process quality. The process model generator 412 may also employ any multivariate and / or statistical methods to determine a model that predicts process quality based on corrections to the process control system 106. The quality prediction models may be based on applying a corrective action to previous data using a projection analysis based on previous batches with similar data characteristics. The process model generator 412 may also employ any multivariate and / or statistical methods to determine a model that predicts whether variables will exceed a threshold based on previous batch data trends.

[0087] Furthermore, the process model generator 412 may estimate variations and / or contributions of measured variables to quality variables from the modeled data by determining measured process disturbances obtained by applying the Hotelling T 2 -statistic and / or of unmeasured disturbances quantified by an application of the Q statistic (e.g., the mean square error (SPE)). Additionally, the models determined by the process model generator 412 may group values ​​associated with measured variables by individual, normally adjusted batches that cluster within a prescribed range of values ​​and / or by unadjusted batches that cluster outside the clustered range.

[0088] In addition to determining models for the relationships between measured and / or calculated variables, the exemplary process model generator 412 determines graph types based on a modeled relationship type. The graph types can generally be constructed by a process control operator who determines an appearance, data display, data types, and / or interface option for each graph. The exemplary process model generator 412 can then populate each type of graph with the appropriate data types. The process model generator 412 includes, for example, measured variables to be displayed in a micrograph, process variation data and / or corresponding variables to be displayed in a process variation graph, and measured and / or calculated variables and contribution ratios to be displayed in a contribution graph.

[0089] The exemplary process model generator 412 from Fig. 4 may store the models, ratios, and / or limits in a process model database 416. The process model database 416 may be accessed by a process control operator via a communication path 418 to view, modify, and / or add information to any models, ratios, and / or limits. The communication path 418 may be any type of wired and / or wireless path. The process model database 416 may be implemented using EEPROM, RAM, ROM, and / or any other suitable type of memory.

[0090] The exemplary process model generator 412 may provide the models, ratios, and / or limits to the analytical process modeler 408 and / or the evaluation process modeler 410 upon request from an operator. Alternatively, the analytical process modeler 408 and / or the evaluation process modeler 410 may access the models, ratios, and / or limits stored in the process model database 416 via the process model generator 412 as needed. Furthermore, the process model generator 412 may update each model, ratio, and / or limit as process control information is received from the OMS 102.

[0091] To manage the display of variable values ​​within a user interface, the exemplary OMS 102 from Fig. 4 a display manager 420. The exemplary display manager 420 receives measured variable values, calculated variable values, calculated variations, calculated contributions, limits, graphical information, and / or error indications from the analytics processor 114. Upon receiving the information, the display manager 420 organizes the information by graphic type for display within a user interface. If a workstation (e.g., workstation 130 of Fig. 1) accesses the OMS 102 using a web application, the exemplary display manager 420 displays one or more graphics within the user interface.

[0092] The display manager 420 can display a graphic by connecting generated graphic information from the analytical process modeler 408 and / or the analytics processor 114 to a user interface application. For example, the display manager 420 can insert a generated graphic into a rich Internet application (RIA) built using Silverlight™, Adobe Flash™, Hypertext Mark-Up Language (HTML) 5, or any other similar rich Internet application technology based on a plug-in to provide a web server framework for process control operators, engineers, and / or administrative personnel. The display manager 420 enables interactive, browser-based applications that do not require the use of specialized process software beyond any workstation containing a Silverlight and / or Flash plug-in.The display manager 420 enables direct manipulation of the data within the graphics from a workstation without the workstation having to continuously access the OMS 102. Direct manipulation of the graphics may include visual feedback of points on a graphic, panning and / or zooming a graphic, and / or displaying related graphics. Alternatively, in examples where a specialized application is required to view the graphics, the display manager 420 may provide that application with data related to the generated graphic.

[0093] To manage which process control operator accesses which process control data, the example OMS 102 includes a session controller 422. In some examples, the session controller 422 may be combined with the display manager 420 as a web server. The example session controller 422 initiates a new session for each process control operator that accesses the data and / or graphics generated by the analytics processor 114. By creating a new session for each operator, the session controller 422 ensures that only process control data related to the process controlled by the operator is transmitted to the operator. For example, four different operators may access the OMS 102 to view data related to four different processes.Furthermore, the session controller 422 may manage sessions when more than one operator is viewing process data and / or graphics related to the same process.

[0094] Additionally, the session controller 422 can track the data displayed within the user interface at a workstation. In this way, if a workstation loses connection to the OMS 102, the session controller 422 saves the workstation's last location for later use when connection to the workstation can be re-established. Furthermore, the session controller 422 can manage the transmission of information to a user interface. For example, after the operator initiates a session, the session controller 422 can transmit an overview chart and / or any process variation graphs to the workstation. The session controller 422 can then transmit any contribution graphs, variable trend graphs, and / or micrographs.

[0095] To manage the communication of process control information to workstations, the exemplary OMS 102 from Fig. 4 illustrates the online data processor 116. The exemplary online data processor 116 may resolve intra-plant communications and / or web-based communications into a single protocol for communicating with the session controller 422, the display manager 420, and / or the analytics processor 114. Additionally, the online data processor 116 may functionally include security and / or user authentication to ensure that only registered workstations and / or process control operators have access to the process control data and associated graphics.

[0096] To communicate with the workstations within the facility 104, the exemplary OMS 102 includes an on-site access server 424. The on-site access server 424 may include components and / or connectivity to the LAN 124 of Fig. 1. Additionally, the plant-internal access server 424 may include encryption and / or any other data transmission security to ensure that the transmitted data cannot be viewed by unauthorized persons. The plant-internal access server 424 may be accessed from any workstation, laptop, personal digital assistant (PDA), smartphone, and / or any other device capable of displaying the process control data and associated graphics.

[0097] For communication with workstations outside the facility 104, the exemplary OMS 102 includes a web access server 428. The web access server 428 may include components and / or connectivity to the WAN 134 of Fig. 1. In addition, the web access server 428 may include encryption and / or any other data transmission security to ensure that the transmitted data cannot be viewed by unauthorized persons.

[0098] Although an exemplary way of implementing the OMS 102 in Fig. 4, one or more of the Fig. 4 illustrated interfaces, data structures, elements, processes and / or devices may be combined, subdivided, rearranged, omitted, eliminated and / or implemented in any way. Fig. 4, the exemplary batch data receiver 402, the exemplary analytics processor 114, the exemplary analytical process modeler 408, the exemplary evaluation process modeler 410, the exemplary process model generator 412, the exemplary display manager 420, the exemplary session controller 422, the exemplary online data processor 116, the exemplary in-plant access server 424, and / or the exemplary web access server 428 may be implemented separately and / or in any combination using, for example, machine-accessible or human-readable instructions executed by one or more computing devices and / or computing platforms (e.g., the exemplary processing platform P10 of Fig. 18) are carried out.

[0099] Furthermore, the example batch data receiver 402, the example analytics processor 114, the example analytical process modeler 408, the example evaluation process modeler 410, the example process model generator 412, the example display manager 420, the example session controller 422, the example online data processor 116, the example on-site access server 424 and / or the example web access server 428 and / or more generally the OMS 102 may be implemented by hardware, software, firmware and / or any combination of hardware, software and / or firmware.Therefore, the example batch data receiver 402, the example analytics processor 114, the example analytical process modeler 408, the example evaluation process modeler 410, the example process model generator 412, the example display manager 420, the example session controller 422, the example online data processor 116, the example plant-internal access server 424 and / or the example web access server 428 and / or more generally the OMS 102 may each be implemented by one or more of integrated circuits, programmable processors, application specific integrated circuits (ASIC(s)), programmable logic device(s) (PLD(S)) and / or programmable logic field device(s) (FPLD(S)), etc.

[0100] Fig. 5 is a diagram 500 taken from the OMS 102 of Fig. 1 and / or 4, which may be displayed to indicate a fault within the process control system 106. The diagram 500 illustrates how relatively easily a process control operator can logically navigate from an initial detection of a process fault to determining a cause of the fault and then to determining whether a corrective action for the fault would improve overall process quality. The diagram 500 illustrates two possible flows for detecting a fault and / or predicting process quality with the correction of the fault. However, other lists, graphs, charts, and / or data may also be displayed. For each of the graphs and / or charts 502-512, in conjunction with Fig. 6 - 14B an explanation is provided.

[0101] A process overview diagram 502 shows the status of one or more process control systems (e.g., process control system 106). The overview diagram 502 may be displayed when a process control operator first opens a user interface at a workstation and selects a process control environment and / or equipment. If the OMS 102 detects a fault and / or a prediction of a process fault, the OMS 102 may display the fault within the overview diagram 502.

[0102] After viewing the indication of a fault in the summary chart 502, a process control operator may select the indication to open a process variation graph 504. Alternatively, the operator may select the indication to open a microchart 508. The process variation graph 504 may show the variation associated with the indicated fault. For example, if the process fault is associated with overall process quality, the process variation graph 504 may display the unexplained and / or explained variation associated with overall process quality. Alternatively, if the fault is associated with a measured and / or calculated variable, the process variation graph 504 may display the explained and / or unexplained variation associated with the measured and / or calculated variable.Furthermore, the process variation graph 504 may display the current and previous values ​​of the variable and / or the process quality associated with the error with respect to an average value, a standard deviation of the value, and / or a limit of the value. The process variation graph 504 may also display a bar chart with measured and / or calculated variable contributions to the variation.

[0103] After viewing the process variation graph 504, the process control operator can select a point on the graph corresponding to a time. After selecting a point, a contribution graph 506 can be displayed. The contribution graph 506 shows variables that contribute to the process and / or a variable variation. The operator can use the contribution graph 506 to determine which variables are contributing to the identified fault. Therefore, the contribution graph 506 shows the causes potentially responsible for the detected fault. Further, the process control operator can change a time in the process to view the values ​​of the variables at the selected time. Additionally, the contribution graph 506 can display suggested corrective actions for the fault.The operator may select any of the variables listed in the contribution chart 504 to view a history of the variables for the current batch process in a variable trend graph 510.

[0104] Alternatively, the microchart 508 may be displayed after the operator selects a point on the process variation graph 504. The example microchart 508 shows values ​​associated with measured and / or calculated variables based on a selected time. The operator may use the microchart 508 to examine the values ​​of the measured and / or calculated variables beyond the history of the batch process to determine if and / or by how much one or more variables exceed limits and / or standard deviations. The microchart 508 may also include sparklines that display a history of the measured and / or calculated values. The operator may select any sparklines to view a more detailed description of the selected sparklines. Further, the microchart 508 may include a timeline that displays the time at which a process error occurred.The operator can select an arrow within the timeline and scroll through and / or select a point in time within the timeline to display the corresponding variable values ​​measured and / or calculated at that point in time. The operator can select any of the displayed variables to open the variable trend graph 510. Alternatively or additionally, the operator can open a quality prediction graph 512 to display the predicted process quality based on an implemented corrective action of one or more corrective actions.

[0105] The example variable trend graph 510 displays the variable values ​​during the duration of a batch process. In some examples, the variable trend graph 510 may display variable values ​​beyond one operation and / or one operation of the batch process, depending on the resolution selected by the operator. The example variable trend graph 510 may show the history of the selected variable during the batch process in relation to an average and / or standard deviation of previous batch processes. The variable trend graph 510 may also display limit values ​​and / or calculated standard deviations for the selected variable. The operator may use the variable trend graph 510 to view a history of a variable during the batch process. After viewing the variable trend graph 510, the operator may select the quality prediction graph 512.

[0106] The example quality prediction graph 512 displays the predicted quality of the process based on the implementation of one or more corrective actions. The operator can use the quality prediction graph 512 to determine whether a corrective action will improve the process quality sufficiently to continue execution of the process. If the predicted process quality has not improved and / or if the process quality has already deviated significantly from acceptable limits, the operator can decide to abort the process. Additionally, if a selected corrective action is not predicted to improve the process, the operator can select additional corrective actions to determine whether one or more corrective actions may improve the quality of the process.If a prediction of a corrective action improves process quality within acceptable variations, the operator may select within the quality prediction graph 512 to transmit the corrective action to the OMS 102 and / or the controller 108.

[0107] The quality prediction graph 512 may also show the predicted variation of the overall quality variable and / or the measured and / or calculated variables associated with a defect. Furthermore, the quality prediction graph 512 may show a predicted quality of the process as well as the current and / or previous variations in the process quality. After viewing the quality prediction graph 512, the operator may return to the summary graph 502. Additionally, after viewing any of the graphs 504-510, the operator may also return to the summary graph 502 and / or open any of the additional graphs 504-512. Furthermore, the graphs and / or graphs 502-512 may be updated with the current variable values ​​and / or quality predictions as the OMS 102 receives and / or processes process control information.

[0108] Fig. 6 shows the user interface 136 that displays the example process overview diagram 502 of Fig. 5, which is connected to the exemplary process control system 106 of Fig. 1. The overview chart 502 may be used to provide an overview of the multiple processes of a plant (e.g., plant 104) where more than one batch may be active at the same time. The overview chart 502 is organized by process area and includes a status of a first process area (e.g., process area 1) and a second process area (e.g., process area 2). The first process area may correspond to the process control system 106. Each process area includes information related to the current and / or previous batches that may be used to alert a process control operator that a defect has been detected and / or when one end of a batch quality predictor variable is outside a prescribed limit.

[0109] The example overview graphic 502 contains batch information for batches 12359-12369. However, only the batch information for batch 12369 is shown. The example overview graphic 502 contains fields for a batch identification number (e.g., Batch ID), an operation of the batch (e.g., Operation), a status of the operation (e.g., Status), a predicted quality value (e.g., Prediction), and / or a defect status (e.g., Defect). The predicted quality value may correspond to a predicted variation of an overall quality variable that is related to a quality of a final product. The defect field may be displayed if a defect has been detected and / or if a defect is predicted. The defect field may also be displayed at a time at which the defect occurred and / or any other process information related to the defect (e.g.,measured and / or calculated variables that may be a cause of the error). Additionally, the error field may include a symbol 604 to graphically alert an operator to an error.

[0110] An operator may view additional information associated with the fault by selecting icon 604, text in the fault field (e.g., True), the batch number, and / or any other text in the fields associated with batch 12369. Alternatively, an operator may view additional information associated with the fault by selecting a forward arrow (not shown) and / or a graphic tab (not shown) that may be located within the summary diagram 502. Furthermore, an operator may alert other process control operators and / or managers to the detected fault by selecting text and / or icon 604 associated with the fault.

[0111] Fig. 7 shows the user interface 136 of Fig. 6, which shows the exemplary process variation graph 504 of Fig. 5, including an explained variation graph 710 and an unexplained variation graph 720 for an overall quality of the batch process 12369. The process variation graph 504 can be viewed by clicking on the information of the batch 12369 within the overview diagram 502 of Fig. 6 are displayed.

[0112] The example process variation graph 504 shows calculated statistics for batch 12369 for a point in time associated with a process time period (e.g., T1 to T4). A process control operator can select a time period to encompass a batch, an operation, an operation, and / or a period surrounding a detected fault. The example in Fig. Figure 7 shows a process quality variation. However, other process variation graphs show variations related to measured, calculated, and / or overall quality variables. Additionally, instead of showing variations in the variables, the process variation graph can also show values ​​of the variables over the process period.

[0113] The explained variation graph 710 shows a variation graph 712 of explained variation values ​​for the process period. Similarly, the unexplained variation graph 720 shows a variation graph 722 of the unexplained variation values ​​for the process period. The point selections 714 and 724 point to a point on the corresponding variation graphs 710 and 720 that are selected by an operator. The point selection 724 in Fig. 7 shows, for example, that the operator has selected the highest point of unexplained variation. Alternatively, upon selection of a fault in the summary chart 502, the process variation graph 504 may open with point selections 714 and / or 724 displayed at a highest point of variation. A limit line 726 includes a limit of 1.0. A fault may be indicated if the unexplained variation graph 722 exceeds the limit line 726. The fault indicated in the summary chart 502 may, for example, be associated with the unexplained variation graph 722 of the overall quality of the batch, which exceeds the 1.0 limit at 12:38:26 p.m. just before T3. A time indication line 728 references a time during the time period associated with the time corresponding to the point selections 714 and / or 724.

[0114] The process variation graphic 504 further includes a process information panel 730 that includes navigation arrows, a time corresponding to the time of the point selections 714 and / or 724 (e.g., 12:38:26 p.m.), a batch identification number, material being processed by the batch, equipment currently being used by the batch, an operation of the batch, and a status of the batch. Additionally, the panel 730 includes numerical values ​​of the explained and unexplained variations at the corresponding point selections 714 and / or 724. An operator can move the point selections 714 and / or 724 along the corresponding graphics 712 and 722 and view the numerical variation values ​​in the panel 730.

[0115] The operator can select one or more arrows to switch between the graphics and / or diagrams 502 - 512 of Fig. 5. By clicking, for example, a forward arrow (e.g., the arrow above PM), the contribution graph 506 can be displayed, or by clicking the up and / or back arrows, the overview graph 502 can be displayed. In addition, the operator can navigate between graphs 504 - 512 by selecting the appropriate tab located at the top of the process variation graph 504. The operator can, for example, view a micrograph (e.g., micrograph 508) by selecting the micrograph tab. Furthermore, the operator can navigate to the contribution graph 506 and / or micrograph 508 by selecting a point on the graphs 712 and / or 722 (e.g., point selections 714 and / or 724). Furthermore, the selected point in time associated with the point in time on the graphs 722 and / or 724 can be forwarded to the contribution graph 506 and / or micrograph 508.

[0116] Fig. 8 shows the user interface 136 of Fig. 7, which displays the exemplary process variation graph 504 including a contribution summary panel 802. The process variation graph 504 includes the explained variation graph 710, the unexplained variation graph 720, the variation graphs 712 and 722, and the process information panel 730 similar to that shown in Fig. 7. Panel 802 of the contribution summary contains a contribution graph with variables contributing to explained and / or unexplained variation. The contribution bar chart shows the value of the variable relative to how much the variable contributes to the explained and / or unexplained portion of the total quality variation. Additionally, the contribution graph shows a value associated with a contribution amount and / or a variation amount of that variable.

[0117] The media flow variable can, for example, be a measured process variable that contributes most to the unexplained variation. The unexplained contribution can be indicated by a solid black block portion, and the explained contribution can be indicated by the striped block portion. Additionally, the value of the media flow variable can indicate that the media flow is 2.33 gallons (gal)-second (sec) below a media flow threshold. Alternatively, the value 2.33 can indicate a statistical confidence value and / or a statistical contribution value of the media flow variable to the overall process quality variation. The contribution of each variable can be determined by the OMS 102 of Fig. 1 and / or 4 are modeled and / or calculated.

[0118] An operator can select a point in the graphs 712 and / or 722. If the operator selects a point in the graph, the contribution information within the summary contribution panel 802 can change to reflect the contributions to the variations at the selected time. For example, if a point is selected at time T2, the media flow may show lower explained and / or unexplained contribution values ​​because the variation in the overall process quality is lower at time T2. Additionally, similar to Fig. 7, an operator can select the forward arrow in the process information panel 730, the tabs, and / or by selecting the summary contribution panel 802, open the contribution graphic 506 and / or the microchart 508.

[0119] Fig. 9 shows the user interface 136 of Fig. 6, which shows the exemplary contribution graphic 506 of Fig. 5, including unexplained and explained process variations for variables associated with the process control system 106 of Fig. 1. The example contribution graphic 506 includes a bar chart 910 that displays explained variations (e.g., solid black bars or bar segments) as well as unexplained variations (e.g., striped bars or bar segments) for each variable that contributes to the variation in overall process quality. In further examples, the bar charts may be shown in different colors and / or shapes. Additionally, the variables (e.g., media flow, mixer temp., water temp., etc.) are ordered such that the variable with the largest contribution to the overall process variation is displayed first. In further examples, the variables may be ordered according to the preferences of a process control operator.

[0120] The variables in Bar Chart 910 contain a numerical value associated with a variation of the variable. For example, the value -2.33 associated with the media flow variable may indicate that a fluid is flowing 2.33 gal / sec slower than an average or limit. Alternatively, the value -2.33 may indicate a statistical contribution that the media flow makes to the variation of the overall process quality variation of Fig. 7 and Fig. 8. The contribution quantities of the variables in the bar chart 910 can be determined by the OMS 102 from Fig. 1 can be determined.

[0121] The explained and unexplained variations of each variable are overlaid or shown together to provide an easy-to-understand graphical display for a process control operator. The explained variations of each variable can be calculated by the OMS 102 by applying a model variation test. 2-statistics can be calculated. The unexplained variation of each variable can be calculated by the OMS 102 using a statistical Q test. In further examples, the bar chart 910 can display values ​​of the variables at selected time points, mean values ​​of the variables, and / or standard deviations for each of the variables.

[0122] The example contribution graphic 506 includes a process information panel 920 that displays process information, process times of the displayed variations for each variable, and numerical values ​​of explained and unexplained variations for a selected variable. Fig. 9, the process information panel 920 shows, for example, that at 12:38:26 the media flow variable has a variation of 0.26 and an unexplained contribution to the total process variation of 2.89, as in Fig. 7 and Fig. 8. The contribution graph 506 may display the variations of the variables for the time 12:38:26 PM based on a time received from the process variation graph 504. Alternatively, an operator may change the process time by selecting one of the arrows to advance or regress the time. By changing the time, the bar chart 910 may display variable variation values ​​corresponding to the selected time.

[0123] Furthermore, the contribution graphic 506 contains a panel 930 with recommended measures. The panel 930 with recommended measures can display recommendations for process correction in order to correct a detected process error (e.g. the detected error in Fig. 6). The recommended action orifice 930 recommends, for example, correcting the error by increasing the media flow by 1.85 gal / s in the field device FIC3. The FIC3 may correspond to a valve or pump within the process control system 106 that is capable of changing the media flow rate. In addition, the recommended action orifice 930 contains a recommendation to check the flow meter FIT3. By checking the flow meter FIT3, an operator can determine whether the flow meter is outputting a correct media flow value. The recommended actions in orifice 930 can be determined through prior analysis of similar process control systems with similar errors. In addition, the recommended actions can be determined by the OMS 102 based on the modeling of the process control system 106.

[0124] An operator can view the history of values ​​for each variable by selecting the desired variable in the bar chart 910. Upon selecting a variable, the variable trend graph 510 is displayed, showing a history of the selected variable. Alternatively, the operator can select a variable and then select the variable trend tab and / or the forward arrow in the process information panel 920. Additionally, by selecting one or more of the recommended actions in the recommended actions panel 930, the operator can display the quality prediction graph 512 to view the predicted quality when one or more corrective actions are performed.

[0125] Fig. 10 shows the user interface 136 of Fig. 6, which shows the variable trend graph 510 of Fig. 5 for the media flow variable of Fig. 8 and Fig. 9. The variable trend graph 510 can be used by an operator to compare a process variable trend during the current batch process with the trends of the variable during previous batch processes that ended with product within quality limits. Abnormal deviations of a variable from previous batches can be a cause of detected defects and / or deviations of the overall quality variable and / or overall process quality variable. The variable trend graph 510 shows how values ​​associated with a current batch of a variable vary from previous batches. An operator can open multiple variable trend graphs 510 to view the historical trends of multiple variables. The example variable trend graph 510 can display a history of values ​​for any measured variable, calculated quality variable, and / or overall process quality variable.

[0126] The example variable trend graph 510 includes a current variable graph 1002 with a selection point 1004. The current variable graph 1002 shows the values ​​of the variables over a period of the process (e.g., T1 to T4). The period can be adjusted by the operator to display only one operation, one operation, and / or a time around a fault detection. The current variable graph 1002 shows that the media flow in the batch process began before time T2 and had a flow rate of between 1.7 and 2.4 gal / sec until approximately the midpoint of time T3. At time T4, the media flow in the batch process 12369 was stopped.

[0127] The variable trend graph 510 includes a variable average 1006, an upper standard deviation 1008, and a lower standard deviation 1010 corresponding to the media flow values ​​from previous processes. In addition, the variable trend graph 510 may include a calculated average, calculated standard deviations, and / or a limit for the media flow. By displaying the variable average 1006 and the standard deviations 1008 and 1010, the variable trend graph 510 allows an operator to compare the trend of a variable with previous trends of the same variable during different batches, each of which produced an acceptable product. The variable average 1006 and / or the standard deviations 1008 and 1010 may be provided by the OMS 102 from Fig. 1 can be determined.

[0128] The variable average value 1006 can be a specific average value of the media flow variable for previous batches, and the standard deviations 1008 and 1010 can be determined from standard deviations of the media flow variable of previous batches. The variable trend graph 510 in Fig. Figure 10 shows that the current variable graph 1002 is approximately 2.0 gal / sec below the variable mean 1006 and 1.6 gal / sec below the lower standard deviation.

[0129] Furthermore, the variable trend graph 510 can include a process information panel and / or a recommended action panel. An operator can access the quality prediction graph 512 by selecting a point on the variable graph 1002 (e.g., selection point 1004), by selecting the Predictions tab, and / or by selecting a corrective action in a recommended action panel. Alternatively, the operator can navigate back to the contribution graph 506 by selecting the Contribution tab and / or by closing the variable trend graph 510.

[0130] Fig. 11 shows the user interface 136 of Fig. 6, which shows the quality prediction graph 512 of Fig. 5 for the exemplary batch 12369. The exemplary quality prediction graph 512 shows an effect on possible corrections of an error in a process variation (e.g., the error shown in the process variation graph 504 of Fig. 7). The quality prediction graph 512 may show a predicted quality of a product resulting from the process and / or a predicted quality associated with the end of the batch process 12369. Additionally, the quality prediction graph 512 may show a predicted quality of the batch process after implementation of a corrective action to eliminate a detected defect. An operator may use the quality prediction graph 512 to determine whether a corrective action is sufficient to bring the process quality within defined limits. If the quality prediction graph 512 shows that a corrective action is not sufficient to produce a quality product, the operator may select other corrective actions and / or abort the process.

[0131] Furthermore, the quality prediction graph 512 may correspond to an overall quality variable. In further examples, the quality prediction graph 512 may show predicted values ​​for a measured and / or calculated variable after implementation of a corrective action. The exemplary quality prediction graph 512 of Fig. 11 contains a quality prediction graph 1102 over a complete period of the batch process 12369 (e.g., from T1 to T7). The quality prediction graph 1102 may be represented as a statistical process quality calculation, which may be normalized. The statistical process quality calculation and / or a model of the statistical process quality calculation may be implemented by the exemplary OMS 102 of Fig. 1 can be generated.

[0132] The example quality prediction graph 512 also includes a time marker line 1104 indicating the current time and / or progress during the batch process. Additionally, the time marker line 1104 may represent a point in time at which the quality prediction graph 512 calculates the impact of applying a corrective action on the batch process. Furthermore, the quality prediction graph 512 includes confidence limits 1108 representing a calculated confidence interval for the predicted quality. For example, the confidence limits 1108 may indicate with 95% certainty that the actual quality of the batch process will be within the confidence limits 1108. In other words, there is a 95% chance that the quality of the batch process will be at some point between the confidence limits 1108.The exemplary confidence limits 1108 are expanded at the beginning of the batch process 12369 because the quality of the process may be more uncertain at the beginning of the process. Thereafter, as the batch process progresses, the quality of the process may be more certain, which is indicated by tighter confidence limits 1108 at time T7.

[0133] Furthermore, the quality prediction chart contains 512 of Fig. 11, a threshold limit 1110 (e.g., specification) and a target limit 1111. The threshold limit 1110 can indicate a maximum value that the quality prediction graph 1102 can have, so that the batch process is considered to be of acceptable quality. The target limit 1111 can indicate an ideal value of the quality prediction graph 1102 to obtain an acceptable quality product.

[0134] The example quality prediction graph 512 also includes a process information panel 1112 that includes navigation arrows, batch process information, a predicted numerical value of a selected point on the batch process quality graph 1102, and numerical values ​​of the confidence limits 1108 associated with the selected point on the graph 1102. The selected point on the predicted quality graph 1102 may be represented by a black circle at time T7. Additionally, the quality prediction graph 512 includes a recommended action panel 1114 that includes a selectable corrective action. In the example of Fig. 11, the corrective action of increasing the FIC3 flow by 1.85 gal / s is selected, indicating that the quality prediction graphic 512 has applied the corrective action to the batch process quality prediction. In further examples, the recommended action panel 1114 may include additional corrective actions and / or functionality that allows the operator to enter corrective actions.

[0135] If a process control operator determines that an applied corrective action resolves a detected error, the process control operator may transfer the corrective action to the process control system 106 of Fig. 1 by selecting the action within the recommended action panel 1114. Alternatively, the operator may select to apply the corrective action through other functionality contained within the user interface 136 and / or the quality prediction graphic 512.

[0136] Fig. 12 shows the user interface 136 of Fig. 6, which shows the exemplary micro diagram 508 of Fig. 5 at a first point in time. The exemplary microchart includes a time axis 1202 showing a batch process time 1204 and a current progress 1206 up to the batch process time 1204. In addition, the time axis 1202 includes detected errors 1208 - 1214 and a graphically displayed time position arrow 1216. The process time 1204 corresponds to a point in time for the completion of a batch process (e.g., the batch process 12369). The current progress 1206 shows the status of the batch process with respect to the process time 1204. In Fig. 12, the current history is, for example, approximately two-thirds of the process time 1204. In further examples, the process time 1204 may correspond to an operation time, an operating time and / or a time specified by the operator.

[0137] The detected errors 1208 - 1214 correspond to those within the process control system 106 of Fig. 1 detected errors. The time period of the detected errors 1208 - 1214 may be indicated by the width of the detected error bars 1208 - 1214. Additionally, the shading of the detected errors 1208 - 1214 may correspond to a type of error. For example, errors 1208 and 1210 may correspond to an operating condition (e.g., a quality variable) of the batch process that exceeds a limit. Error 1212 may correspond to an error associated with a measurement, a control loop, and / or a field device (e.g., measured variables). Error 1214 may correspond to a combination of an error corresponding to an operating condition and an error corresponding to a measurement, a control loop, and / or a field device. In further examples, the detected errors 1208 - 1214 may be represented by colored bars, by shapes, and / or by symbols.

[0138] The time axis 1202 includes the graphically displayed time position arrow 1216 to allow an operator to select a time during the batch process to view the variable value information. The variable value information may be displayed in the bar chart 1218 and / or sparklines 1220-1226. The microchart 508 of Fig. 12 shows a graphically represented time position arrow positioned at a first time point, and the variable values ​​shown in bar chart 1218 correspond to the first time period. Additionally, the history of sparklines 1220 - 1226 up to the first time period is shown. Furthermore, the variables contain a name of the variable and a numerical value of the variable at the first time point. The first variable, media flow, for example, contains sparkline 1220, which shows a history of the values ​​for the media flow variable during the batch process, a value at a first time point of 2.8 gal / s, and a bar chart 1218 that is normalized and shows how the 2.8 gal / s value compares to an average value for the media flow variable.

[0139] The example microchart 508 provides a process control operator with current variable values ​​(e.g., via bar chart 1218) as well as a history of variable values ​​(e.g., via sparklines 1220-1226) for each of the variables that may contribute to batch process variations. Therefore, the example microchart 508 combines the functionality of the contribution graph 506 and the variable trend graph 510.

[0140] The example microchart 508 may normalize and / or rescale the mean and / or standard deviation of each of the variables so that the variables can be displayed within an overall mean and / or standard deviation of the bar chart 1218. The bars in the bar chart 1218 may be colored differently to indicate that one or more variables exceed the standard deviation but do not cause a process error. Additionally, the bars in the bar chart 1218 may change color to indicate that one or more variables exceed the standard deviation and thus cause a process error. Further, the bars in the bar chart 1218 may be shaded to indicate a statistical history of each of the variables. Similar to an average and standard deviation from previous batches in the variable trend chart 510 of Fig. 10, for example, the bars in bar chart 1218 may be shaded to indicate an average and / or standard deviation from previous batches for each variable. Furthermore, an operator may select any of the variables in bar chart 1218 to view the selected variable in variable trend graph 510.

[0141] The example sparklines 1220–1226 show a history of previous failures for the current batch for each of the variables. Sparklines 1220–1226 can be plotted along an average value for each variable and / or shown as an absolute value of the variable. Furthermore, the sparklines can contain notes indicating where the values ​​of a variable exceed a limit during the process. A process control operator can view detailed information about sparklines 1220–1226 by selecting a desired sparkline 1220–1226.

[0142] Fig. 13 shows the user interface 136 of Fig. 6, which shows the exemplary micro diagram 508 of Fig. 12 at a selected second point in time. In the example of Fig. 13, the graphically represented time position arrow 1216 is moved to the second time point on the current history 1206 of the batch process time 1204. The graphically represented time position arrow 1216 is positioned when the error 1214 occurs. Therefore, the bar chart 1218 shows variable values ​​associated with the second time point, and the variables show numerical values ​​associated with the second time point (e.g., media flow at 3.3 gal / s). Additionally, the sparklines 1220 - 1226 show previous values ​​for each variable up to the second time point. In further examples, the sparklines 1220 - 1226 may contain the previous values ​​of each variable up to the current time point of the batch process, but may show an indication of a second time point. The example of Fig. Figure 13 shows that error 1214 may be related to media flow exceeding the standard deviation shown in bar chart 1218. Additionally, the bar in media flow-related bar chart 1218 may change color and / or shading to indicate that the variable is a contributing factor to error 1214.

[0143] Fig. 14A and Fig. 14B show the user interface 136 of Fig. 6, which shows the exemplary sparklines 1220 for the mixer temperature process variable of Fig. 12 and Fig. 13. Sparkline graphics 1400 and / or 1420 can be displayed within the user interface 136 by selecting the sparkline 1220 in Fig. 12 and / or 13. The example sparkline graphics 1400 and 1420 of Fig. 14A and Fig. 14B shows two different configurations for displaying previous variable values ​​in a batch process. However, other sparkline graphs can also be configured in various ways to display previous variable values ​​in a batch process.

[0144] The example Sparklines graphic 1400 from Fig. Figure 14A contains the mixer temperature sparkline 1220 for batch process times T1 - T4. The sparkline 1220 is shown along an absolute temperature scale. For example, at T1, the mixer temperature is 40 °C, and at T4, the mixer temperature is 180 °C. Therefore, the sparkline 1220 increases with temperature, as shown.

[0145] Additionally, the sparkline graph 1400 contains deviation hints 1404 and 1406. A length of deviation hints 1404 and 1406 corresponds to a point in time at which variable values ​​exceed a limit. A height of deviation hints 1404 and 1406 from the sparkline can indicate by how much the variable values ​​exceed the limit. Deviation hint 1404 positioned above sparkline 1220 can indicate that the variable values ​​exceed an upper standard deviation limit, while deviation hint 1406 positioned below sparkline 1220 can indicate that the variable values ​​exceed a lower standard deviation limit. Furthermore, deviation hints 1404 and 1406 can be color-coded to indicate the severity of a deviation, the amount of deviation, and / or a type of deviation.

[0146] The example Sparklines graphic 1420 from Fig. 14B contains the sparkline mixer temperature line 1220 for the batch process times T1-T4. The sparkline 1220 is shown along an average value and a standard deviation scale. The sparkline graph 1420, for example, shows the variable values ​​of the sparkline 1220 in relation to a calculated average value for the mixer temperature during the batch process from times T1 - T4. In addition, the sparkline graph 1420 contains a deviation note 1424. The deviation note 1424 does not correspond to the values ​​shown in the sparkline graph 1400 of Fig. 4A. The example deviation indicator 1424 provides a visual indication to an operator that a portion of the sparkline 1220 exceeds a standard deviation (e.g., a threshold). The deviation indicator 1424 may be visually indicated by a dashed line, a bold line, line colors, shapes, and / or icons.

[0147] Fig. 15, 16A-16F and 17A-17B are flowcharts of example methods that may be performed to configure the example OMS 102, the example controller 108, the example user interface 136, the example batch data receiver 402, the example analytics processor 114, the example analytical process modeler 408, the example evaluation process modeler 410, the example process model generator 412, the example display manager 420, the example session controller 422, the example online data processor 116, the example web access server 428 and / or the example on-site access server 424 of Fig. 1 and / or 4. The example methods of FIGS. 15, 16A-16F and / or 17A-17B may be executed by a processor, a controller, and / or other suitable processing device. The example methods of Fig. 15, 16A - 16F and / or 17A - 17B may, for example, be embodied as coded instructions stored on any tangible computer-readable medium, such as a flash memory, a CD, a DVD, a floppy disk, a ROM, a RAM, a programmable ROM (PROM), an electronically programmable ROM (EPROM), an electronically erasable ROM (EEPROM), an optical disk storage, an optical disk storage device, a magnetic disk storage, a magnetic disk storage device and / or any medium that can be used to carry and store program code and / or instructions in the form of methods or data structures and that can be executed by a processor, general purpose computer or a special purpose computer or other machine having a processor (e.g., those described below in connection with Fig. 18, the exemplary processor platform P10). Combinations of these are also included within the scope of computer-readable media.

[0148] The methods include, for example, instructions and / or data that cause a processor, general-purpose computer, special-purpose computer, or special-purpose processing machine to implement one or more particular methods. Alternatively, some or all of the exemplary methods may be Fig. 15, 16A - 16F and / or 17A - 17B using any combination(s) of ASIC(s), PLD(s), FPLD(s), discrete logic, hardware, firmware, etc.

[0149] Some or all of the exemplary procedures of Fig. 15, 16A-16F and / or 17A-17B may also be implemented using manual operations or any combination of the aforementioned techniques, such as any combination of firmware, software, discrete logic and / or hardware. Furthermore, many other methods for implementing the exemplary operations of Fig. 15, 16A - 16F and / or 17A - 17B. For example, the order of execution of the blocks may be changed and / or one or more of the described blocks may be changed, eliminated, subdivided, or combined. In addition, one or all of the exemplary methods of Fig. 15, 16A - 16F and / or 17A - 17B are executed sequentially and / or in parallel, e.g. by separate processing threads, processors, devices, discrete logic, control circuits, etc.

[0150] The exemplary procedure 1500 of Fig. 15 models relationships between measured, calculated, and / or overall quality variables based on the characteristics of a process control system. Multiple example methods 1500 may be executed in parallel or sequentially to implement portions of a process control system and / or to model other process control systems.

[0151] The exemplary procedure 1500 of Fig. 15 begins with the identification of field devices within a process control system (block 1502). The example method 1500 may identify field devices by identifying serial numbers, process control identification numbers, and / or other identification methods. The example method 1500 may then identify the inputs to the field devices (block 1504). Next, the example method 1500 identifies the outputs of the field devices and identifies a measured process and / or quality variable associated with each output (block 1506).

[0152] The exemplary method 1500 proceeds by determining the calculated quality variables from the measured process and / or quality variables (block 1508). The exemplary method 1500 then determines the overall quality variables for the process based on the measured and / or calculated variables (block 1510). Next, the exemplary method 1500 calculates the ratios (e.g., the contribution ratios) between the measured, calculated, and / or overall quality variables (block 1512). The ratios may be determined by the process model generator 412 of Fig. 4 be determined and / or modeled using any multivariable, statistical, algebraic, process control history analysis and / or optimization techniques.

[0153] The exemplary procedure 1500 of Fig. 15 then determines the limit values ​​for each of the variables (block 1514). The exemplary method 1500 may utilize any multivariate, statistical, algebraic, prior process analysis and / or optimization techniques and / or in conjunction with the process model generator 412 of Fig. 4 to determine the calculated and / or overall quality variables from the measured variables. In addition, the exemplary method 1500 may utilize any multivariate, statistical, algebraic, prior process analysis and / or optimization techniques and / or in conjunction with the process model generator 412 of Fig. 4 to determine the relationships between the variables and the limit values ​​for each of the variables. After determining the limit values, the exemplary method 1500 stores the variables, the original source of the values ​​for each of the variables (e.g., the field devices), the relationships between the variables, and / or the limit values ​​for each of the variables in the process model database 414 of Fig. 4 (Block 1516). Additionally, the example method 1500 may determine predictive models from the process control system and store the predictive models in the database 414. After storing the process control variables, limits, and / or ratios, the example method 1500 terminates.

[0154] The exemplary procedure 1600 of Fig. 16A - 16F creates and / or manages navigation between process control graphics and / or the diagrams (e.g., the graphics and / or diagrams 502 - 512 of Fig. 5-14B). Multiple example methods 1600 may be executed in parallel or sequentially to create and / or navigate from the diagrams and / or graphs. Additionally, multiple example methods 1600 may be implemented for each session initiated by a process control operator. The example method 1600 begins when a process control operator opens a session and selects a process control environment, a plant, and / or a process control system. Additionally, the example method 1600 may start by displaying an overview diagram (e.g., the overview diagram 502 of Fig. 6) is displayed.

[0155] The exemplary procedure 1600 of Fig. 16A begins by receiving process control information from a controller (e.g., controller 108 of Fig. 1) (block 1602). In addition, the example method 1600 may combine the obtained process control information from previously obtained process control information associated with the same batch and / or process. Next, the example method 1600 determines the values ​​associated with measured values ​​from the obtained process control information (block 1604). The example method 1600 then calculates the values ​​associated with calculated and / or overall quality variables from the values ​​associated with the measured variables (block 1606). The example method 1600 stores the values ​​associated with the variables in the batch data database 416 (block 1608). The values ​​may be stored based on a time at which the values ​​were determined by the field devices (e.g.,using a timestamp), by batch number and / or by an operation within the batch.

[0156] The example method 1600 continues by determining whether any values ​​associated with the variables exceed corresponding limits (block 1610). If one or more values ​​exceed a corresponding limit, the example method 1600 indicates a process control fault corresponding to one or more values ​​exceeding the corresponding limits (block 1612). The example method 1600 then determines whether a selection of a graphic has been received (block 1614). The selection of a graphic may be made by a process control operator. Additionally, if one or more values ​​do not exceed the corresponding limit (block 1610), the example method 1610 determines whether a selection of the graphic has been received (block 1614).

[0157] If a selection of a graphic has not been received, the example method 1600 continues to receive process control information (block 1602). However, if a selection of a graphic has been received, the example method 1600 determines the type of selected graphic (e.g., microchart, sparkline, contribution graph, variable trend graph, process variation graph, and / or quality prediction graph) (block 1614). Additionally, the example method 1600 may continue to receive process control information and / or may incorporate the received variable values ​​into the batch data while an operator selects and views graphics.

[0158] The exemplary method 1600 runs in Fig. 16B if a selection of a process variation graph has been received. The example method 1600 generates a representation of the process variation graph, including an unexplained and / or an explained variation associated with a selected variable (block 1616). The selected variable may correspond to a fault detected by the example method 1600, including an overall quality variable, a measured variable, and / or a calculated variable. Alternatively, the process variation graph may include a history of the values ​​and a corresponding limit value associated with the selected variables. The example method 1600 then displays values ​​for the unexplained and / or explained variation for a selected time and / or the most recently received process control information (block 1618). For example, if an operatorselects a fault in the summary chart, the example method 1600 may display the unexplained and / or explained variation values ​​associated with a time at which the fault was detected.

[0159] The example method 1600 then determines whether a selection of a time point has been received from the process variation graph (block 1620). The selection of a time point may coincide with the instant at which the operator selects a point in a graphically represented variation corresponding to a process time. If the example method 1600 has received a selection of a time point, the example method 1600 displays unexplained and / or explained variation values ​​associated with the selected time point (block 1622). Next, the example method 1600 determines whether a selection has been received to view variable contributions to the variation (block 1624). Even if a selection of a time point in the process variation graph has not been received (block 1620), the example method 1600 determines whether a selection has been received to view variable contributions to the variation (block 1624).

[0160] If a selection has not been received to view variable contributions to variation, the example method 1600 may determine whether the operator wishes to proceed with an analysis of the batch data (block 1626). For example, after viewing the process variation graph, the operator may determine that the process is acceptable and may wish to return to a display of the summary chart. If the operator intends to proceed with the batch analysis, the example method 1600 may display the process summary chart and continue receiving process control information (block 1602). If the operator does not wish to proceed with the batch analysis (block 1626), the operator may abort the process, and the example method 1600 terminates.

[0161] However, if a selection has been received to view variable contributions to variation (block 1624) and if the contributions have been selected for display in a micrograph, the example method 1600 proceeds to Fig. 16C by generating a display of the microchart, including measured process and / or quality variables (block 1628). The microchart may also display calculated quality variables. Additionally, the microchart may be displayed by the example method 1600 upon receiving a selection of a microchart (block 1614). The example method 1600 displays the values ​​of the variables within the microchart based on a selection of a time point by the operator (block 1630). The time selection may, for example, be a time point associated with a process fault if the operator selects the process fault in the summary chart and / or may be a time point selected to view the unexplained and / or explained variation in the process variation graph.After displaying the microchart, the example method 1600 determines whether another time point selection has been received (block 1632). A time point selection may be made in a microchart by moving an arrow along a time axis to a desired point in the process. Alternatively, a time point selection may be made by selecting a point on a sparkline associated with a variable within the microchart.

[0162] If the example method receives a selection of a time point, the example method 1600 retrieves values ​​from the batch data database 416 corresponding to the selected time point for each of the variables displayed within the microchart (block 1634). Alternatively, the values ​​may already be contained within the generated microchart. In cases where the values ​​are already contained within the microchart (e.g., in a Flash plug-in application), the example method 1600 updates the display of the microchart to reflect the values ​​associated with the selected time point.

[0163] The example method 1600 then displays the values ​​within the bar chart and / or sparklines associated with each of the displayed variables (block 1636). Next, the example method 1600 determines whether a selection associated with a sparkline of at least one of the variables has been received (block 1638). If the example method 1600 has not received a selection of a time point (block 1632), the example method 1600 additionally determines whether a selection associated with a sparkline of at least one of the variables has been received (block 1638).

[0164] If the example method 1600 receives a selection of a variable within a sparkline, the example method 1600 generates Fig. 16D, a detailed display of the selected sparkline for the selected variable (block 1646). The example method 1600 may also generate a sparkline by receiving a selection of a graph from the operator (block 1614). The sparkline may start at a time from a start of the batch and end at a time associated with the most recent batch data. Alternatively, the sparkline may represent a time period prescribed by the operator (e.g., an operation, an operation, a prescribed time around the occurrence of a fault, etc.). The example method 1600 may then indicate any deviations and / or faults of any values ​​within the sparkline that may exceed an associated threshold (block 1648).

[0165] Next, the example method 1600 determines whether a selection of a point in time on the displayed sparkline (block 1650) has been received. If no point has been selected, the example method 1600 may determine whether the operator intends to proceed with an analysis of the batch data (block 1652). If the operator intends to proceed with the analysis, the example method 1600 may display the process overview diagram and proceed to receive process control information (block 1602). If the operator does not intend to proceed with the batch analysis (block 1652), the operator may abort the process, and the example method 1600 terminates. If the example method 1600 Fig. 16D receives a selection of a point on the sparkline (block 1650), the example method 1600 may return to displaying the micrograph with the variable values ​​associated with the time on the sparkline selected by the operator (blocks 1654, 1628, and 1630).

[0166] If the exemplary method 1600 of Fig. 16C does not receive a selection of a sparkline (block 1638), the example method 1600 determines whether a selection of a variable indicated in the bar chart of the microchart (block 1640) has been received. The example method 1600 of Fig. 16D continues if a selection of a variable displayed in the bar chart of the microchart has been received by determining an average value and / or standard deviation from the previous batch data associated with the selected variable (block 1656). The example method 1600 then generates a display of a variable trend graph for the selected variable, including the determined average value and / or standard deviation from the previous batch data (block 1658). The example method 1600 may also generate a variable trend graph upon receiving a selection of the variable trend graph from the operator (block 1614). Next, the example method 1600 determines whether a selection has been received to enable display of a quality prediction graph (block 1660).

[0167] If no quality prediction graph has been selected, the example method 1600 may determine whether the operator intends to proceed with an analysis of the batch data (block 1652). If the operator intends to proceed with the batch analysis, the example method 1600 may display the process overview diagram and continue receiving process control information (block 1602). If the operator does not intend to proceed with the batch analysis (block 1652), the operator may abort the process, and the example method 1600 terminates.

[0168] However, if a quality prediction graph has been selected (block 1660), the exemplary method 1600 proceeds to Fig. 16F. A quality prediction graph can also be displayed by the operator in Fig. 16A (block 1614). Furthermore, if the exemplary method 1600 does not receive a selection of a variable in the bar chart (block 1640), but receives a selection to view a quality prediction graph (block 1642) in FIG. 16C, the exemplary method 1600 continues in Fig. 16F continued.

[0169] If the example method 1600 does not include a selection to view a quality prediction graph (block 1642) in Fig. 16B, the example method 1600 may alternatively determine whether the operator intends to proceed with an analysis of the batch data (block 1644). If the operator intends to proceed with the batch analysis, the example method 1600 may display the process overview diagram and proceed to receive the process control information (block 1602). If the operator does not intend to proceed with the batch analysis (block 1644), the operator may abort the process, and the example method 1600 terminates.

[0170] However, if a selection has been received to view the variable contributions to the variation and if the contributions have been selected to be viewed in a contribution graph (Bock 1624), the exemplary method 1600 proceeds to Fig. 16E by calculating unexplained and / or explained variables related to the error for the selected time period (Block 1662). The explained variation can be calculated using a T 2 -statistic may be calculated, and the unexplained variation may be calculated using a Q-statistic. The selected time period may correspond to a point in time on a process variation graph and / or an operator-selected summary chart. The example method 1600 then generates a display of the contribution graph, including the contribution of the explained and / or unexplained variations for each of the associated variables (block 1664). Alternatively, the contribution graph may display values ​​associated with variables contributing to the process error, an average value for each of the values, and / or a standard deviation for each of the values.

[0171] The example method 1600 continues by detecting and displaying a recommended action message (block 1666). The recommended action may include corrections that can be applied to the process control system to correct a detected error. The example method 1600 then determines whether a selection of a time point within the contribution graph has been received. The selection of a time point corresponds to a time point during the current process. If a selection of a time point has been received, the unexplained and / or explained variations are calculated for the variables associated with the error for the newly selected time period (block 1662). If the example method 1600 does not receive a selection of the time point, the example method 1600 determines whether a selection of a variable in the contribution graph has been received (block 1670).If the selection of a variable has been received, the example method generates a variable trend graph for the selected variable (blocks 1656 and 1658 of . Fig. 16D).

[0172] However, if no selection of a variable has been received (block 1670), the example method 1600 determines whether a selection to view a quality prediction graph has been received (block 1672). If a selection to view a quality prediction graph has been received, the example method generates a quality prediction graph in Fig. 16F. If a quality prediction graph has not been selected, the example method 1600 may determine whether the operator intends to proceed with an analysis of the batch data (block 1674). If the operator intends to proceed with the batch analysis, the example method 1600 may display the process overview chart and continue receiving process control information (block 1602). If the operator does not intend to proceed with the batch analysis (block 1674), the operator may abort the process, and the example method 1600 terminates.

[0173] If the example method 1600 receives a quality prediction graph (blocks 1614, 1642, 1660, 1672), the example method 1600 proceeds to Fig. 16F by receiving a selection of a correction to the process error (block 1676). The process control operator may select a correction to the process error by selecting a correction from the list of possible corrections or by entering the correction into the example method 1600. Alternatively, the example method 1600 may select a correction to the process error from the list of possible corrections provided by the example OMS 102 of Fig. 1 and Fig. 4 generated correction models.

[0174] The example method 1600 continues with the prediction of process quality with the selected correction applied to the process (block 1678). The example method 1600 may be performed using any model and / or relationship provided by the process model generator 412 of Fig. 4 using any multivariable, statistical, algebraic, process control history analysis, and / or optimization method. The predicted process quality may correspond to an overall quality variable, calculated quality variations, and / or measured variables. Additionally, the example method 1600 may predict the process quality based on predicted process models generated by an example OMS 102. Next, the example method 1600 calculates confidence limits (e.g., ranges) for the predicted quality (block 1680). The example method 1600 then displays the quality prediction graph with the predicted process quality and the corresponding confidence limits (block 1682).

[0175] The example method 1600 then determines whether the correction should be implemented within the process control system (block 1684). If the correction should be implemented, the example method 1600 transmits the correction to the process via the OMS 102 and / or the controller 108 (block 1686). By transmitting the correction, the process control system can receive instructions associated with the correction to change the operating characteristics of the field devices that are, in turn, associated with the correction. The example method 1600 then determines whether batch analysis should proceed (block 1688). Additionally, if the correction should not be implemented (block 1684), the example method 1600 determines whether batch analysis should proceed (block 1688).If the operator intends to proceed with the batch analysis, the example method 1600 may display the process overview diagram and proceed to obtain process control information (block 1602). If the operator does not intend to proceed with the batch analysis (block 1608), the operator may abort the process, and the example method 1600 terminates.

[0176] The exemplary procedure 1700 of Fig. 17A-17B determines a corrective action for a detected process error. The example method 1700 may be employed to display the example diagram 500 of Fig. 5. Multiple example methods 1700 may be executed in parallel or sequentially to correct multiple process errors from conventional processes. Additionally, multiple example methods 1700 may be executed for an error detected by another process. The example method 1700 begins when a process control operator logs on to a workstation configured to display a user interface (e.g., the user interface 136 of Fig. 1) to display.

[0177] The exemplary procedure 1700 of Fig. 17A begins by receiving an initiation of a session from a process control operator (block 1702). The example method 1700 then displays a user interface and prompts the operator to select a process environment (block 1704). The example method 1700 receives a selection of a process environment from the operator (block 1706). Alternatively, the operator may select a plant, a process control system, and / or multiple process control systems. Next, the example method 1700 generates and / or displays a summary chart associated with the selected process environment (block 1708). The example method 1700 then receives and processes measured variable values ​​from the selected process control system, environment, and / or plant (block 1710).

[0178] The example method 1700 continues by determining whether a detection of a process error has occurred (block 1712). The process error may be a result of a variable exceeding a threshold, a prediction of a variable exceeding a threshold, and / or a quality of the process exceeding a threshold. If the example method 1700 determines that no error has been detected, the example method continues to receive measured variable values ​​from the process control system (block 1710). However, if the example method 1700 detects a process error, the example method 1700 indicates the detected error within the summary chart (block 1714). Additionally, the example method 1700 may continue to receive and / or process variable values ​​while the example method 1700 analyzes variable values ​​to determine a cause of the error.

[0179] Next, the example method 1700 receives a selection of the fault in the overview chart (block 1716). The example method 1700 then generates and / or displays a process variation graph associated with the detected fault (block 1718). After enabling the operator to view the information on the process variation graph, the example method 1700 may receive a selection of a time point in the process variation graph (block 1720). The selection of a time point may correspond to a time point in the process variation graph at which the variation during the detected fault exceeds a threshold. Upon receiving a selection of a time point, the example method 1700 determines whether the operator selects to view contribution variables of the variation in a micrograph and / or in a magnitude graph (block 1722).

[0180] If the operator selects a microchart, the example method 1700 generates and displays a microchart with variables associated with the selected variation in the process variation graph (block 1724). The values ​​associated with the variables may be obtained from a bar chart and / or sparklines in the microchart. While viewing the microchart, the operator may view detailed views of selected sparklines and / or change a time in the process corresponding to the displayed variable values.

[0181] However, if the operator selects a contribution graph, the example method 1700 generates and displays the contribution graph with variables related to the selected variation in the process variation graph (block 1726). The operator can select a graph by selecting a forward arrow in a process analysis section and / or selecting a tab in the corresponding graph on the user interface.

[0182] After enabling the operator to view the variable values ​​in the microdiagram and / or the contribution graph, the example method 1700 determines whether the operator has selected one or more variables in the microdiagram and / or the contribution graph (block 1728). If a selection of at least one variable has been received, the example method 1700 generates and / or displays a variable trend graph for each selected variable (block 1730). The example method 1700 then determines whether a selection of a quality prediction graph has been received (block 1732). If the operator has not selected a variable in the microdiagram and / or the contribution graph, the example method 1700 additionally determines whether a selection of a quality prediction graph has been received (block 1732).An operator can choose to view a quality prediction graph by selecting an appropriate tab on the user interface and / or by selecting the quality prediction graph from a variable trend graph.

[0183] If the example method 1700 has not received a selection of a quality prediction graph, the example method 1700 may return to the overview chart and continue receiving and / or processing measured variable values ​​(block 1710). However, if the example method 1700 has received a selection of a quality prediction graph (block 1732), the example method generates and / or displays the quality prediction graph (block 1734). After displaying the quality prediction graph 1700, the example method determines whether a correction has been performed (block 1736). If the correction is to be performed, the example method 1700 transmits the correction to the process via the OMS 102 and / or the controller 108 (block 1738).By transmitting the correction, the process control system may receive instructions associated with the correction to change the operating characteristics of the field devices associated with the correction. The example method 1700 then determines whether to proceed with the batch analysis (block 1740). If the correction is not to be made (block 1736), the example method 1700 further determines whether to proceed with the batch analysis (block 1740). If the operator intends to proceed with the batch analysis, the example method 1700 may display the process overview diagram and continue receiving process control information (block 1710). If the operator does not intend to proceed with the batch analysis (block 1740), the example method aborts the process (block 1742), and the example method 1700 exits.

[0184] Fig. 18 is a block diagram of an exemplary processor system P10 that can be used to implement the exemplary methods and apparatus described herein. For example, processor systems can be employed to implement the exemplary OMS 102, the exemplary batch data receiver 402, the exemplary analytics processor 114, the exemplary analytics process modeler 408, the exemplary evaluation process modeler 410, the exemplary process model generator 412, the exemplary display manager 420, the exemplary session controller 422, the exemplary online data processor 116, the exemplary on-site access server 424, and / or the exemplary web access server 428 of Fig. 1 and / or 4 that are similar or identical to the example processor system P10. Although the example processor system P10 is described below as including a variety of peripherals, interfaces, chips, memories, etc., one or more of these elements may be omitted from other example processor systems used to implement one or more of the example OMS 102, the example batch data receivers 402, the example analytics processors 114, the example analytics process modelers 408, the example evaluation process modelers 410, the example process model generators 412, the example display managers 420, the example session controllers 422, the example online data processors 116, the example on-site access servers 424, and / or the example web access servers 428.

[0185] As in Fig. 18, the processor system P10 includes a processor P12 coupled to an interconnect bus P14. The processor P12 includes a register set or register space P16, which is Fig. 18 is shown as being positioned entirely on a chip, which may alternatively be positioned entirely or partially separately from the chip and directly coupled to the processor P12 via a separate electrical connection and / or via the interconnect bus P14. The processor P12 may be any suitable processor, processing unit, or microprocessor. Although not shown in Fig. 18, the system P10 may be a multiprocessor system and may therefore include one or more additional processors identical or similar to the processor P12 and in communication with the interconnection bus P14.

[0186] The P12 processor from Fig. 18 is coupled to a chipset P18, which contains a memory controller P20 and a peripheral input / output (I / O) controller P22. As is well known, a chipset typically provides I / O and memory management functions, as well as a variety of general-purpose and / or special-purpose registers, clocks, etc., accessible or used by one or more processors coupled to the chipset P18. The memory controller P20 performs functions that enable the processor P12 (or processors, in the case of multiple processors) to access a system memory P24 and a mass storage device P25.

[0187] The system memory P24 may comprise any type of volatile and / or non-volatile memory, such as random access memory (SRAM), random access memory (DRAM), flash memory, read-only memory (ROM), etc. The mass storage P25 may comprise any type of mass storage device. For example, if the exemplary processor system P10 is used to implement the OMS 102 ( Fig. 2), the mass storage P25 may include a hard disk drive, an optical drive, a tape storage device, etc. Alternatively, if the example processor system P10 is used to implement the process model database 416 and / or the batch data database 406, the mass storage P25 may include semiconductor memory (e.g., flash memory, RAM memory, etc.), magnetic memory (e.g., a hard disk drive), or any other memory suitable for mass storage in the process model database 416 and / or the batch data database 406.

[0188] The peripheral input / output (I / O) controller P22 performs functions that enable the processor P12 to communicate with the input / output (I / O) devices P26 and P28 and a network interface P30 via an I / O peripheral bus P32. The I / O devices P26 and P28 can be any type of I / O device, such as a keyboard, a display (e.g., a liquid crystal display (LCD), a cathode ray tube display (CRT), etc.), a navigation device (e.g., a mouse, a trackball, an integrated touch pad, a joystick, etc.), etc. The network interface P30 can be, for example, an Ethernet device, an asynchronous transfer mode (ATM) device, an 802.11 device, a DSL modem, a cable modem, a radio modem, etc., which enables the processor system P10 to communicate with other processor systems.

[0189] While the memory control unit P20 and the I / O control unit P22 are in Fig.18 are shown as separate functional blocks within the chipset P18, the functions performed by these blocks can be integrated within a single semiconductor control circuit or implemented using two or more separate integrated control circuits.

[0190] At least some of the example methods and / or devices described above are implemented by one or more software and / or firmware programs running on a computer processor. However, stand-alone hardware implementations, including, but not limited to, application-specific integrated circuits, programmable logic arrays, and other hardware devices, may similarly be constructed to implement some or all of the example methods and / or devices described herein, either in whole or in part. Furthermore, alternative software implementations, including, but not limited to, distributed processing or component / object distributed processing, parallel processing, or virtual machine processing, may also be constructed to implement the example methods and / or systems described herein.

[0191] It should be noted that the example software and / or firmware implementations described herein are stored on a tangible storage medium, such as: a magnetic medium (e.g., a magnetic disk or tape); a magneto-optical or optical medium, such as an optical disk; or a solid-state medium, such as a memory card or other package incorporating one or more read-only (non-volatile) memories, random access memory, or other rewritable (volatile) memories. Accordingly, the example software and / or firmware described herein may be stored on a tangible storage medium, such as the media described above or successor storage media. In this sense, while the above specification describes components and functions with respect to particular standards and protocols, it is to be understood that the scope of this patent is not limited to such standards and protocols.For example, each of the standards for Internet and other packet-switched network transmission (e.g., TCP / IP, Internet Protocol (IP), User Datagram Protocol (UDP), and IP, HyperText Markup Language (HTML), and HyperText Transfer Protocol (HTTP)) represent examples of the current state of the art. Such standards are regularly replaced by faster and more powerful equivalents with the same general functionality. Accordingly, replacement standards and protocols with the same functions are contemplated by this patent as equivalents and are therefore deemed to be within the scope of the appended claims.

[0192] Further, while this patent discloses example methods and devices, including software or firmware embodied in hardware, it is understood that such systems are for illustrative purposes only and are not limiting. For example, it is contemplated that any or all of these hardware and software components could be embodied exclusively in hardware, exclusively in software, exclusively in firmware, or any combination of hardware, firmware, and / or software. Accordingly, while the above specification describes example methods, systems, and a machine-accessible medium, the examples are merely one way to implement such systems, methods, and the machine-accessible medium.Although certain exemplary methods, systems, and machine-accessible medium described herein have been described, the scope of this patent is in no way limited thereto.

Claims

[1] A method for predicting process quality in a process control system, comprising: receiving process control information relating to a process at a first time, including a first value related to a first measured variable and a second value related to a second measured variable; calculating a variation based on the received process control information by comparing the received process control information with values ​​associated with previous batches and / or with specified target values ​​for current batch data; determining whether the variation exceeds a limit based on the received process control information associated with the process; if the variation exceeds the limit, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation; determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and calculating a predicted process quality based on the at least one corrective action at a time after the first time. [2] The method of claim 1, further comprising: graphically displaying the variation in a first graphic via a user interface; graphically displaying the first contribution value and the second contribution value in a second graphic via the user interface, wherein the second graphic can be selected from a part of the first graphic; displaying the at least one corrective action via the user interface; and graphically displaying the predicted process quality in a third graphic via the user interface, wherein the third graphic can be selected from a part of the second graphic. [3] The method of claim 2, further comprising: graphically displaying the variation as an explained variation and as an unexplained variation in the first graphic; and graphically displaying the first contribution value and the second contribution value in the first graphic. [4] The method of claim 2, further comprising: predicting the predicted process quality and / or variation based on the first value and / or the second value; specifying a predicted error if the predicted process quality and / or variation exceeds a second limit; and displaying the predicted error in the second graphic and displaying the at least one corrective action in the third graphic. [5] The method of claim 2, further comprising: calculating a first unexplained variation and a first explained variation related to the first contribution value; calculating a second unexplained variation and a second explained variation related to the second contribution value; displaying the first unexplained variation and the first explained variation in a bar chart in the second graph by superimposing the first unexplained variation over the first explained variation and / or showing the first unexplained variation together with the first explained variation; and displaying the second unexplained variation and the second explained variation in the bar chart by superimposing the second unexplained variation over the second explained variation and / or showing the second unexplained variation together with the second explained variation. [6] Apparatus for predicting process quality in a process control system, comprising: a batch data receiver for receiving process control information relating to a process at a first time, including a first value associated with a first measured variable and a second value associated with a second measured variable; and a processor to: Calculating a variation based on the received process control information by comparing the received process control information with values ​​associated with previous batches and / or with specified target values ​​for current batch data; Determine whether the variation exceeds a limit based on the received process control information associated with the process; if the variation exceeds the limit, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation; Determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and Calculating a predicted process quality based on the at least one corrective action at a time after the first time. [7] The device of claim 6, further comprising a display manager for: graphically displaying the variation in a first graphic via a user interface; graphically displaying the first contribution value and the second contribution value in a second graphic via the user interface, wherein the second graphic can be selected from a part of the first graphic; graphically displaying the at least one corrective action via the user interface; and graphically displaying the predicted process quality in a third graph via the user interface, where the third graph can be selected from a part of the second graph. [8] The device of claim 7, wherein the display manager is for: graphically displaying the variation as an explained variation and an unexplained variation in the first graphic; and graphically displaying the first contribution value and the second contribution value in the first graphic. [9] The apparatus of claim 7, wherein the processor is for: calculating a first unexplained variation and a first explained variation that are related to the first contribution value; and calculating a second unexplained variation and a second explained variation that are related to the second contribution value. [10] The device of claim 9, wherein the display manager is for: displaying the first unexplained variation and the first explained variation in a bar chart in the second graph by superimposing the first unexplained variation over the first explained variation and / or showing the first unexplained variation together with the first explained variation; and displaying the second unexplained variation and the second explained variation in the bar chart by superimposing the second unexplained variation over the second explained variation and / or showing the second unexplained variation together with the second explained variation.

Images