Abstract: An on-line optimizer is comprised of a nonlinear dynamic model (702) which is operable to provide an estimation of the output of a plant. This receives manipulated variables (MV), disturbance variables (DV), and computed disturbance variables (CDB). The estimated output of the model is then compared to the actual output measured by virtual on-line analyzer (VOA) (616). This is compared is a difference block 618 to generate a bias which is then filtered by a filter(620). The output thereof is then provided to an output block (672) in a steady state optimizer (700) to offset the desired setpoints. These set points are input to a steady state nonlinear model which is operable to optimize the inputs to the plants for use for writing new set points in accordance with a predetermined cost function. This cost function is utilized to optimize the new inputs with the use of the steady state model in accordance with various constraints and target values.

Abstract: A method for computing model error bounds for system identification of stochastic systems is disclosed. The model error bounds take the form of additive frequency-weighted singular value bounds such that they are directly used in H∞ and &mgr;-synthesis robust control design methods. The largest singular value of the additive uncertainty bound is determined by performing a high number of simulations for the model uncertainty. Simulated values of the uncertainty are computed for a large data population, such that each candidate entry of simulated value lies on the 3-sigma ellipsoids defined by the covariance functions. For each simulated value of uncertainty, the maximum singular values are then determined. In order to determine the scalar uncertainty function needed for robust control design, the maximum over the population of the maximum singular values of uncertainty simulated values is then computed.

Abstract: A scheme for management of uncertainty in control systems that require adaptation is described. The scheme removes as much uncertainty as possible at design time and manufacturing time by taking advantage of available compute resources in pre-computing a large set of robustness-performance tradeoff based model controllers at different operating conditions for different values of uncertainty parameters that occur in-use and in manufacturing. A group of control models, and corresponding model controllers with an uncertainty bound larger than the best-tuned controller, are generated. A subset of the model controllers is implemented in the system at manufacturing time based on characterization of the system. The subset of model controllers is switched at run time based on transient information received during operation of the system.

Abstract: A deviation between a target value of a quantity of state and an actual value of the quantity of state that is caused to follow the target value or a time-integral of the deviation is filtered. Based on the filtered value, a switching surface &sgr; is calculated. Based on a value of the switching surface &sgr;, a control input value u is outputted. The filter is set through comparison in Bode diagrams between a design model of a control system based on an ordinary sliding mode control method and a characteristic variation model of the control system, and by performing compensation in such a direction as to cancel out the variation. The filtering process makes it possible to properly control the control system having a dead time by the sliding mode control method.

Abstract: A legged moving robot has an action plan function to operate according to a predetermined action plan. The legged moving robot has a fuel cell for supplying operating electric energy for the legged moving robot, an operation control unit for controlling operation of the legged moving robot according to the action plan, and an electric generation managing unit for monitoring a state of the fuel cell and contents of the action plan and for regulating an amount of electric energy generated by the fuel cell depending on the action plan.

Abstract: A device for the simulation of spinning machines with a view to their optimum economic use and operation. Starting from material properties of a preproduct (1) determined by measurement, as well as from configuration and setting parameters (3) of such a spinning machine, material properties (2) of the intermediate or output products which are being produced, which properties can be determined by measurement, are predicted using a process model describing the behavior of the spinning machine, by a simulation device. The process model on which a method being based is presented by a neural network. The coefficients determining the actual behavior of this neural network area calculated from a set of sample data in a training phase. This sample data and/or from otherwise conclusively predicted properties of the behavior of such a spinning machine.

Abstract: The present invention provides a processing system that is capable of achieving substantially optimal control performance of a process facility on a “worst case” process system by accounting for changing system dynamics. The processing system includes a storage device and a processor. The storage device is operable to represent (i) at least one of a plurality of associated processes mathematically to define the various relationships among different inputs and outputs of the one or more represented associated processes, and (ii) uncertainty factors that are associated with these defined relationships. The uncertainty factors define a range of dynamics across which the one or more represented associated processes operate, an error in the mathematical representation, or, alternatively, some combination of the same.

Abstract: In current manufacturing practices, if a process results in a partial product which is outside its specification, it is either sent back to be reworked, or is scrapped. This results in unacceptable waste. The present invention comprises a method for minimizing this wasted work and materials, by corrective actions by subsequent processes. This approach is general, and is capable of correcting the effects of out-of-specs manufacturing process conditions, including the salvaging of partial product, thereby obviating the need for rework or scrap.

Abstract: A distributed control system (14) receives on the input thereof the control inputs and then outputs control signals to a plant (10) for the operation thereof. The measured variables of the plant and the control inputs are input to a predictive model (34) that operates in conjunction with an inverse model (36) to generate predicted control inputs. The predicted control inputs are processed through a filter (46) to apply hard constraints and sensitivity modifiers, the values of which are received from a control parameter block (22). During operation, the sensitivity of output variables on various input variables is determined. This information can be displayed and then the user allowed to select which of the input variables constitute the most sensitive input variables. These can then be utilized with a control network (470) to modify the predicted values of the input variables. Additionally, a neural network (406) can be trained on only the selected input variables that are determined to be the most sensitive.

Abstract: A control system and a method of operating the same are introduced that globally optimize associated processes within a process facility. The control system includes a global controller and many local controllers. The global controller monitors characteristics of the associated processes and generates, in response thereto, control data for optimizing the process facility. The local controllers monitor the associated processes and operate in accordance with the control data to varying degrees to control the monitored associated processes, thereby cooperating with the global controller to optimize the process facility.

Abstract: The present invention includes a method and system for nonlinear feedforward control for ramp following and overshoot minimization in a plant under control. The present invention augments a feedforward system with a nonlinear compensation mechanism for minimizing the overshoot that occurs in a conventional feedforward system.

Abstract: A feedback control system which includes a disturbance estimator and controller which takes account of and corrects for non-linear disturbances. A simple state model representative of the controlled plant is provided. A plant controller responds to a command signal and a process feedback signal for producing a controller output. The controller output drives both the plant and the state model. A sensed signal representing a state of the plant is compared with a corresponding output state of the model to produce an error signal which is representative of disturbances applied to the system. The disturbance estimator and controller responds to the error signal to produce a disturbance control output which is adapted to compensate for the non-linear disturbance. In a preferred embodiment the plant is an actuator and the non-linear disturbances are friction related.

Abstract: A hybrid predictor for predicting the output of a process, and a hybrid prediction method using the hybrid predictor. Also, a system for and a method of controlling a process using the hybrid predictor and hybrid prediction method are provided. In order to obtain a shifted prediction vector and step response coefficients, the hybrid predictor uses a parameter model which can be updated in accordance with a variation in the process. The control method includes the steps of predicting process output signals generated in a prediction horizon, based on input signals applied to the process, correcting the predicted values, based on output signal values measured at the next scan, comparing the corrected prediction value with a reference value to derive an error vector, calculating a control signal to minimize the error vector, and applying the control signal to the process. The hybrid predictor is characterized by an addition of a model parameter estimator and a model response converter.

Abstract: A method for modeling a steady-state network in the absence of steady-state historical data. A steady-state neural network can be tied by impressing the dynamics of the system onto the input data during the training operation by first determining the dynamics in a local region of the input space, this providing a set of dynamic training data. This dynamic training data is then utilized to train a dynamic model, gain thereof then set equal to unity such that the dynamic model is now valid over the entire input space. This is a linear model, and the historical data over the entire input space is then processed through this model prior to input to the neural network during training thereof to remove the dynamic component from the data, leaving the steady-state component for the purpose of training. This provides a valid model in the presence of historical data that has a large content of dynamic behavior.