Process stabilizing process controller
First Claim
1. An automated control processing method comprising the steps of:
- conducting a process which is controlled as said process is performed;
establishing, within a learning means, at least one learned relationship between data for at least a first process control parameter for controlling said process, and data for at least a first process influencing factor, said learning means receiving process training data output in said step of conducting;
generating at least a first data cluster, said first data cluster generated from an output of said learning means, said first data cluster including first control data related to said first process control parameter and including first applicability data related to said first process influencing factor for determining when said control data is to be used in controlling said process;
entering said first data cluster into a plurality of stored data clusters, each said data cluster having control data related to said first process control parameter and applicability data related to said first process influencing factor for determining when said control data is to be used in controlling said process;
controlling said process using said first data cluster as an accessed data cluster obtained from said plurality of stored data clusters;
post-filtering said plurality of stored data clusters to retain said first data cluster when used as said accessed data cluster and said process has acceptable performance.
12 Assignments
0 Petitions
Accused Products
Abstract
A process controller is disclosed for controlling a process. The process controller includes a process supervisor for accessing process control data clusters from a library. Each data cluster provides control data, e.g., setpoints, deadbands, execution frequencies and alarm ranges for process actuator controllers. Each actuator controller maintains a related process attribute within a deadband by controlling at least one process actuator. Data clusters for the library are provided in substantially real-time by a learning unit. Data continuously output from process related sensors are transformed and supplied to the learning unit as training data for learning at least one relationship between process control data and data for at least one substantially uncontrollable process influencing factor (e.g., processing site environmental factors such as temperature and barometric pressure). Prior to inserting a newly provided data cluster into the library, the data cluster is evaluated for predicted successful performance when accessed by the supervisor. Data clusters are also evaluated after each use by the supervisor, with non-performing clusters becoming inaccessible for further use. The supervisor can also fine tune or adjust control data to more closely fit the present behavior of the process. The process controller is embodied in controlling a coal drying process using a combustion gas heat transfer technique.
54 Citations
71 Claims
-
1. An automated control processing method comprising the steps of:
-
conducting a process which is controlled as said process is performed; establishing, within a learning means, at least one learned relationship between data for at least a first process control parameter for controlling said process, and data for at least a first process influencing factor, said learning means receiving process training data output in said step of conducting; generating at least a first data cluster, said first data cluster generated from an output of said learning means, said first data cluster including first control data related to said first process control parameter and including first applicability data related to said first process influencing factor for determining when said control data is to be used in controlling said process; entering said first data cluster into a plurality of stored data clusters, each said data cluster having control data related to said first process control parameter and applicability data related to said first process influencing factor for determining when said control data is to be used in controlling said process; controlling said process using said first data cluster as an accessed data cluster obtained from said plurality of stored data clusters; post-filtering said plurality of stored data clusters to retain said first data cluster when used as said accessed data cluster and said process has acceptable performance. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30)
-
-
31. A coal drying control processing method comprising the steps of:
-
controlling a coal drying process in real time using control data from a first data cluster derived from an output of a learning unit; obtaining a second data cluster from a plurality of stored data clusters for controlling the coal drying process, said second data cluster also having control data to be used in said controlling step, each data cluster of said plurality of data clusters being derived from an output of said learning unit; said step of obtaining including comparing a first value obtained from control data in said first data cluster with a second value obtained from control data in said second data cluster; supplying said second data cluster to said controlling step. - View Dependent Claims (32, 33, 34, 35, 36, 37, 38, 39, 40)
-
-
41. A coal drying control processing method comprising the steps of:
-
controlling a coal drying process in real time; supplying a data cluster to said controlling step, said data cluster derived from an output of a learning unit; wherein said step of supplying includes a first filtering of said data cluster prior to becoming accessible for use in said step of controlling and a second filtering of said data cluster after termination use in said controlling step, wherein said first filtering includes applying heuristic rules to said data cluster.
-
-
42. A coal drying control processing method comprising the steps of;
-
controlling a coal drying process in real time; supplying a data cluster to said controlling step, said data cluster derived from an output of a learning unit; wherein said step of supplying includes a first filtering of said data cluster prior to becoming accessible for use in said step of controlling and a second filtering of said data cluster after termination of use in said controlling step, wherein said second filtering includes preparing an air stability measurement, related to said data cluster, with a predetermined threshold, the comparison used in determining whether said data cluster is to become inaccessible for further use in said step of controlling. - View Dependent Claims (43)
-
-
44. A coal drying control processing method comprising the steps of:
-
controlling a coal drying process in real time; supplying a data cluster to said controlling step, said data cluster derived from an output of a learning unit; wherein said step of supplying includes a first filtering of said data cluster prior to becoming accessible for use in said step of controlling and a second filtering of said data cluster after termination of use in said controlling step.
-
-
45. A coal drying control processing method comprising the steps of:
-
controlling a coal drying process in real time; supplying a data cluster to said controlling step, said data cluster derived from an output of a learning unit; wherein said step of supplying includes a first filtering of said data cluster prior to becoming accessible for use in said stem of controlling and a second filtering of said data cluster after termination of use in said controlling step, wherein said first filtering includes determining confidence prediction data for said data cluster, said confidence prediction data a composite of data obtained from said cold drying process.
-
-
46. A coal drying control processing method comprising the steps of:
-
controlling a coal drying process in real time; supplying a data cluster to said controlling step, the data cluster derived from an output of a learning unit, the data cluster obtained from a plurality of data clusters by selecting the data cluster such that performance target data related to the data cluster is a function of a previous performance target data being used in said step of controlling; wherein said performance target data includes at least one of;
moisture content of coal to be dried, a dried coal tonnage rate, the percentage of free oxygen in a heat transport gas, a drying chamber temperature and an exhaust fan temperature.
-
-
47. An automated control processing method for controlling a process in real time comprising:
-
supplying control data to a plurality of attribute controllers, each attribute controller used in controlling a related process attribute, each related process attribute having a current operating value; stabilizing the process being controlled so that a measurement of process stability is within a predetermined acceptable range; adjusting said control data supplied to a first attribute controller of said plurality of attribute controllers for controlling a first process attribute, said control data including a deadband and a setpoint; wherein said step of adjusting includes; (A1) obtaining a direction for shifting said setpoint toward a current operating value for said first process attribute; (A2) performing at least one of;
shifting said setpoint in said direction and shifting said deadband in said direction. - View Dependent Claims (48, 49, 50, 51)
-
-
52. A control processing method comprising the steps of:
-
providing access to a stored plurality of data clusters, each data cluster for providing process control data to a process manager for controlling a process, each data cluster related to; (A1) performance target data for specifying target values for one or more processing characteristics of the process; (A2) applicability data for specifying when the data cluster is applicable for use by the process manager, said applicability data including data related to a measurement of one or more process influencing factors having an uncontrolled variation that influences the process sufficiently so that the process is controlled to accommodate the uncontrolled variation; entering a first data cluster, related to first performance target data, into said stored plurality of data clusters, said step of entering including associating said first data cluster with a process commencing data cluster contained in said stored plurality of data clusters, said process commencing data cluster used for commencing the process, and said process commencing data cluster being related to a second performance target data, wherein said first and second performance target data have only differences that are inconsequential to a controlling of the process; accessing said first data cluster from said stored plurality of data clusters for controlling the process, said step of accessing including determining whether a data cluster for obtaining process control data immediately previously to said first data cluster is said process commencing data cluster.
-
-
53. An apparatus for controlling a coal drying process comprising:
-
supervisor means for determining process control data used in controlling the coal drying process; learning means for; (a) learning at least a first process relationship using; i) first data, output by said supervisor means, for one or more process control parameters, and ii) second data related to at least a first process influencing factor, said first process influencing factor has an uncontrolled variation that influences the coal drying process sufficiently so that the coal drying process is controlled to accommodate the uncontrolled variation; and (b) outputting learned data relating to said first process relationship; library means for providing access to a plurality of data clusters having process control data, a first data cluster in said plurality of data clusters is; (a) generated using said learned data, and (b) accessed by said supervisor means for obtaining process control data; filter means for accessing said first data cluster by said supervisor means, said filter means using at least a first measurement for said first process influencing factor. - View Dependent Claims (54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65)
-
-
66. An apparatus for controlling a process comprising:
-
manager means for using process control data in controlling the process; learning means for; (A1) learning at least a first process relationship between; (i) first data, derived from output by the process related to one or more process control parameters, and (ii) second data related to at least a first process influencing factor, said first process influencing factor has an uncontrolled variation that influences the process sufficiently so that the process is controlled to accommodate the uncontrolled variation; and (A2) outputting learned data related to said first process relationship; access means for supplying a first data cluster having process control data to said manager means, said first data cluster generated from the output learned data, said access means including filter means for determining access by said manager means to said first data cluster, said filter means using at least a measurement of said first process influencing factor in determining access to said first data cluster. - View Dependent Claims (67, 68)
-
-
69. An automated control processing method for controlling a process in real time comprising:
-
activating each attribute controller of a plurality of attribute controllers, each said attribute controller used in controlling the process, each said attribute controller having an activation frequency or execution frequency for determining a time interval between successive activations of the attribute controller; stabilizing the process being controlled so that a measurement of process stability for the process is within a predetermined acceptable range; adjusting a first execution frequency for a first attribute controller of said plurality of controllers wherein said step of adjusting includes; (A1) determining an execution frequency increment for adjusting said first execution frequency; (A2) adding said execution frequency increment to said first execution frequency. - View Dependent Claims (70)
-
-
71. An automated control processing method comprising the steps of:
-
conducting a process which is controlled as the process is performed; providing a stored plurality of data clusters wherein each data cluster of said stored plurality of data clusters is used in controlling the process; accessing a first data cluster of said stored plurality of data clusters by a process manager for controlling the process; replacing access of said first data cluster by the process manager with access to a second data cluster of said stored plurality of data clusters for controlling the process; post-filtering said stored plurality of data clusters for retaining said first data cluster in said stored plurality of data clusters for a subsequent step of accessing by the process manager, said post-filtering including determining whether the process manager provides acceptable performance in controlling the process during said step of accessing.
-
Specification