Abstract: System and method for asynchronous distributed optimization of an enterprise. The system includes multiple computer systems coupled over a network, which store and implement multiple models, including one or more dynamic models representing respective sub-systems or processes of the enterprise. At least two of the models are interdependent. The system also includes an optimizing system that includes multiple optimizers, at least two of which are interdependent, and constraints and/or objectives, and is operable to receive information related to the enterprise from multiple information sources, and use one or more of the plurality of models to generate a solution subject to the one or more constraints and/or objectives, which is usable in managing the enterprise. Some or all of the system operates in an asynchronous manner. Various portions of the system, e.g., the models, data sources, optimizers, constraints and/or objectives, etc., may be updated, e.g., asynchronously, as desired.

Abstract: A method for building a model of a system includes first extracting data from a historical database (310). Once the data is extracted, a dataset is then created, which dataset involves the steps of preprocessing the data. This dataset is then utilized to build a model. The model is defined as a plurality of transforms which can be utilized to run an on-line model. This on-line model is interfaced with the historical database such that the variable names associated therewith can be downloaded to the historical database. This historical database can then be interfaced with a control system to either directly operate the plant or to provide an operator an interface to various predicted data about the plant. The building operation will create the transform list and then a configuration step is performed in order to configure the model to interface with the historical database. When the dataset was extracted, it is unknown whether the variables names are still valid.

Abstract: A predictive network is disclosed for operating in a runtime mode and in a training mode. The network includes a preprocessor (34') for preprocessing input data in accordance with parameters stored in a storage device (14') for output as preprocessed data to a delay device (36'). The delay device (36') provides a predetermined amount of delay as defined by predetermined delay settings in a storage device (18). The delayed data is input to a system model (26') which is operable in a training mode or a runtime mode. In the training mode, training data is stored in a data file (10) and retrieved therefrom for preprocessing and delay and then input to the system model (26'). Model parameters are learned and then stored in the storage device (22). During the training mode, the preprocess parameters are defined and stored in a storage device (14) in a particular sequence and delay settings are determined in the storage device (18).

Abstract: A predictive network is disclosed for operating in a runtime mode and in a training mode. The network includes a preprocessor (34') for preprocessing input data in accordance with parameters stored in a storage device (14') for output as preprocessed data to a delay device (36'). The delay device (36') provides a predetermined amount of delay as defined by predetermined delay settings in a storage device (18). The delayed data is input to a system model (26') which is operable in a training mode or a runtime mode. In the training mode, training data is stored in a data file (10) and retrieved therefrom for preprocessing and delay and then input to the system model (26'). Model parameters are learned and then stored in the storage device (22). During the training mode, the preprocess parameters are defined and stored in a storage device (14) in a particular sequence and delay settings are determined in the storage device (18).

Abstract: A preprocessing system for preprocessing input data to a neural network includes a training system for training a model (20) on data from a data file (10). The data is first preprocessed in a preprocessor (12) to fill in bad or missing data and merge all the time values on a common time scale. The preprocess operation utilizes preprocessing algorithms and time merging algorithms which are stored in a storage area (14). The output of the preprocessor (12) is then delayed in a delay block (16) in accordance with delay settings in storage area (18). These delayed outputs are then utilized to train the model (20), the model parameter is then stored in a storage area (22) during run time, a distributed control system (24) outputs the data to a preprocess block (34) and then preprocesses data in accordance with the algorithms in storage area (14). These outputs are then delayed in accordance with a delay block (36) with the delay settings (18).

Abstract: A neural network system is provided that models the system in a system model (12) with the output thereof providing a predicted output. This predicted output is modified or controlled by an output control (14). Input data is processed in a data preprocess step (10) to reconcile the data for input to the system model (12). Additionally, the error resulted from the reconciliation is input to an uncertainty model to predict the uncertainty in the predicted output. This is input to a decision processor (20) which is utilized to control the output control (14). The output control (14) is controlled to either vary the predicted output or to inhibit the predicted output whenever the output of the uncertainty model (18) exceeds a predetermined decision threshold, input by a decision threshold block (22).

Abstract: A plant (72) is operable to receive control inputs c(t) and provide an output y(t). The plant (72) has associated therewith state variables s(t) that are not variable. A control network (74) is provided that accurately models the plant (72). The output of the control network (74) provides a predicted output which is combined with a desired output to generate an error. This error is back propagated through an inverse control network (76), which is the inverse of the control network (74) to generate a control error signal that is input to a distributed control system (73) to vary the control inputs to the plant (72) in order to change the output y(t) to meet the desired output. The control network (74) is comprised of a first network NET 1 that is operable to store a representation of the dependency of the control variables on the state variables. The predicted result is subtracted from the actual state variable input and stored as a residual in a residual layer (102).

Abstract: A continuous emission monitoring system for a manufacturing plant (10) includes a control system (16) which has associated therewith a virtual sensor network (18). The network (18) is a predictive network that receives as inputs both control values to the plant (10) and also sensor values. The network (18) is then operable to map the inputs through a stored representation of the plant (10) to output a predicted pollutant sensor level. This predicted pollutant sensor level is essentially the prediction of an actual pollutant sensor level that can be measured by a pollutant sensor (14). The network (18) therefore is a substitute for the pollutant sensor (14), thus providing a virtual sensor. The sensor values from the plant (10) are first processed through a sensor validation system (22).

Abstract: An internal combustion engine [(360)] is provided with a plurality of sensors to monitor the operation thereof with respect to various temperature measurements, pressure measurements, etc. A predictive model processor [(322)] is provided that utilizes model parameters stored in the memory [(324)] to predict from the sensor inputs a predicted emissions output. The model is trained with inputs provided by the sensor and an actual emissions sensor output. During operation, this predicted output on line [(326)] can be utilized to provide an alarm or to be stored in a history database in a memory [(328)]. Additionally, the internal combustion engine [(260)] can have the predicted emissions output thereof periodically checked to determine the accuracy of the model. This is effected by connecting the output of the engine to an external emissions sensor [(310)] and taking the difference between the actual output and the predicted output to provide an error.

Abstract: A predictive network is disclosed for operating in a runtime mode and in a training mode. The network includes a preprocessor (34') for preprocessing input data in accordance with parameters stored in a storage device (14') for output as preprocessed data to a delay device (36'). The delay device (36') provides a predetermined amount of delay as defined by predetermined delay settings in a storage device (18). The delayed data is input to a system model (26') which is operable in a training mode or a runtime mode. In the training mode, training data is stored in a data file (10) and retrieved therefrom for preprocessing and delay and then input to the system model (26'). Model parameters are learned and then stored in the storage device (22). During the training mode, the preprocess parameters are defined and stored in a storage device (14) in a particular sequence and delay settings are determined in the storage device (18).

Abstract: A continuous emission monitoring system for a manufacturing plant (10) includes a control system (16) which has associated therewith a virtual sensor network (18). The network (18) is a predictive network that receives as inputs both control values to the plant (10) and also sensor values. The network (18) is then operable to map the inputs through a stored representation of the plant (10) to output a predicted pollutant sensor level. This predicted pollutant sensor level is essentially the prediction of an actual pollutant sensor level that can be measured by a pollutant sensor (14). The network (18) therefore is a substitute for the pollutant sensor (14), thus providing a virtual sensor. The sensor values from the plant (10) are first processed through a sensor validation system (22).

Abstract: A plant (72) is operable to receive control inputs c(t) and provide an output y(t). The plant (72) has associated therewith state variables s(t) that are not variable. A control network (74) is provided that accurately models the plant (72). The output of the control network (74) provides a predicted output which is combined with a desired output to generate an error. This error is back propagated through an inverse control network (76), which is the inverse of the control network (74) to generate a control error signal that is input to a distributed control system (73) to vary the control inputs to the plant (72) in order to change the output y(t) to meet the desired output. The control network (74) is comprised of a first network NET 1 that is operable to store a representation of the dependency of the control variables on the state variables. The predicted result is subtracted from the actual state variable input and stored as a residual in a residual layer (102).

Abstract: A neural network includes an input layer comprising a plurality of input units (24) interconnected to a hidden layer with a plurality of hidden units (26) disposed therein through an interconnection matrix (28). Each of the hidden units (26) is a single output that is connected to output units (32) in an output layer through an interconnection matrix (30). Each of the interconnections between one of the hidden units (26) to one of the output units (32) has a weight associated therewith. Each of the hidden units (26) has an activation in the i'th dimension and extending across all the other dimensions in a non-localized manner in accordance with the following equation: ##EQU1## that the network learns by the Back Propagation method to vary the output weights and the parameters of the activation function .mu..sub.hi and .sigma..sub.hi.