Abstract: A method of predicting the properties (e.g., mechanical and/or processing properties) of an injection-molded article is disclosed. The method makes use of a hybrid model which includes at least one neural network. In order to forecast (or predict) properties with respect to the manufacture of a plastic molded article, a hybrid model is used in the present invention, which includes: one or more neural networks NN1, NN2, NN3, NN4, . . . , NNk; and optionally one or more rigorous models R1, R2, R3, R4, . . . , which are connected to one another. The rigorous models are used to map model elements which can be described in mathematical formulae. The neural networks are used to map processes whose relationship is present only in the form of data, as it is in effect impossible to model such processes rigorously. As a result, a forecast relating to properties including the mechanical, thermal and Theological processing properties and relating to the process time of a plastic molded article is obtained.

Abstract: A control system is provided for controlling a process for making paper or paper pulp. The process has a process product output at an end of the process. The controls include a process variable sensor input configured to receive a process variable related to the process. The controller is configured to provide a control signal to control the process. A process model has a model product output which is a model representation of the process product output. The model process output is a function of the sensed process variable and a product output setpoint representative of a desired process product output. The control signal is a function of the product output setpoint and the modeled product output.

Abstract: In an output controlling method for controlling output of a driving power source or motor installed in a vehicle, the relationship between a primary output-controller, which is manipulated by the user, and a secondary output-controller, which directly operates the motor, is regulated by control parameter subjected to evolution by using evolutionary computing based on at least one of the following: the user's characteristics, driving conditions, environmental changes, and deterioration of the drive power source with time. The evolution is conducted on-line or on a real-time basis. The primary output-controller includes an acceleration pedal or grip, and the secondary output-controller includes a throttle valve for an internal combustion engine or a running current controller for an electric motor.

Abstract: A cooperative camera network involving segmenting an image of an object and extracting a color signature from it. The color signature may be matched with the other signatures in an attempt to identify the object. There also can be combining and fusing the tracking of people and objects with image processing and the identification of the people and objects being tracked in an area or facility.

Abstract: A method of determining efficiently parameters in chemical-mechanical polishing, especially a method includes a Neural-Taguchi method to seek the best parameter set to increase the quantity of output. This invention involves a time parameter to achieve end-point detection on line during the CMP procedure and also completes a maximum material removal rate (MRR) and a minimum within wafer non-uniformity (WIWNU) at the same time.

Abstract: The provides improved control devices, systems and methods for operation thereof. These rely on control devices that provide virtual machine environments in which Java objects, or other such software constructs, are executed to implement control (e.g., to monitor and/or control a device, process or system). These objects define blocks which are the basic functional unit of the control. They also define the input, output and body parts from which blocks are formed, and the signals that are communicated between blocks. The objects also define nested and composite groupings of blocks used to control loops and higher-level control functions.

Abstract: A method of determining properties relating to the manufacture of an injection-molded article is described. The method makes use of a hybrid model which includes at least one neural network and at least one rigorous model. In order to forecast (or predict) properties relating to the manufacture of a plastic molded part, a hybrid model is used which includes: one or more neural networks NN1, NN2, NN3, NN4 , . . . , NNk; and one or more rigorous models R1, R2, R3, R4, . . . , which are connected to one another. The rigorous models are used to map model elements which can be described in mathematical formulae. The neural model elements are used to map processes whose relationship is present only in the form of data, as it is typically impossible to model such processes rigorously. As a result, a forecast (or prediction) relating to properties including, for example, the mechanical, thermal and rheological processing properties and relating to the cycle time of a plastic molded part can be made.

Abstract: A method and apparatus for regulating Internet or Intranet access to selected functions of a machine controller based upon a user network address.

Abstract: A predictive failure system for a power delivery system. The power delivery system includes a number of modules which are interconnected to a system monitor. The system monitor collects data on operating parameters of each module and environmental parameters. The system monitor analyzes the data in order to define conditions based upon the parameters. The parameters are then applied against a set of rules to determine whether a warning or a fault indicator should be generated. The system monitor may be implemented locally as part of the power delivery system or may be located remotely from the power delivery system to enable off site data collection and analysis.

Abstract: A system for selling, manufacturing and distributing a custom digital data product from retail stores, over the Internet, over the telephone, or by electronic means (e.g., fax, e-mail, and the like) wherein a customer is provided (e.g., by electronic mail verification) order tracking information. After a customer selects a “set” of sound recordings or data from a library or catalog of such recordings or data and payment or credit is received or verified, an image of the “set” is assembled from a storage or “disk” farm. The image is preferably assembled at a manufacturing facility, e.g., a CD-ROM burner farm, where the product is then made. Every data object on the product may have a code associated therewith for later reference. The disk and burner farms communicate via a high speed communications subsystem to facilitate continuous processing. Upon assembly and manufacture, the product is packaged and shipped.

Abstract: A classification apparatus, notably for uses in recognition or characterisation of odors, comprises a plurality of sensors for generating raw data representing a plurality of instances of a plurality of different classes; and a processing unit for processing said raw data so as to determine an identification model. The identification model comprises definitions of the classes and a particular allocation rule selected dependent upon the application. The class definitions are established by analysing data obtained during a learning phase, the analysis being performed according to a particular information extraction method. During a later identification phase, the identification model enables an instance of unknown class to be allocated to an appropriate class amongst those defined during the learning phase. The information extraction method can be selected dependent upon the application.

Abstract: The invention concerns a classifying apparatus, used in particular for recognising or characterising odours, applying a combination of statistical methods and neuronal networks to classify into the appropriate class instances of a plurality of different classes presented to the apparatus. The apparatus comprises a processing unit for determining for each class (j) a subspace (SEj) wherein the instances of said class are best separated from instances of other classes, said subspace being defined by synthetic variables (VDj), and for determining a discriminating subspace (SED) defined by the whole set of synthetic variables identified for the whole set of classes. Each neuron in the neuronal input layer corresponds to one of the variables defining the discriminating space (SED) and each neuron of the output layer corresponds to one of the classes.

Abstract: In a computer implemented learning and/or process control system, a computer model is constituted by the most currently fit entity in a population of computer program entities. The computer model defines fitness as a function of inputs and outputs. A computing unit accesses the model with a set of inputs, and determines a set of outputs for which the fitness is highest. This associates a sensory-motor (input-output) state with a fitness in a manner that might be termed “feeling”. The learning and/or control system preferably utilizes a Compiling Genetic Programming System (CGPS) in which one or more machine code entities such as functions are created which represent solutions to a problem and are directly executable by a computer. The programs are created and altered by a program in a higher level language such as “C” which is not directly executable, but requires translation into executable machine code through compilation, interpretation, translation, etc.

Abstract: A fuzzy-neuro method uses a neuro network to discriminate optical disk type. The fuzzy-neuro method establishes a plurality of characteristic signal sets corresponding to the optical disks of various formats; and uses the neuro network to match the optical disk under question to the characteristic signal sets for finding the type thereof.

Abstract: A fuzzy data record pointer is utilized for identification of both a target file and a target data record within the target file. A target data record is accessed from a target file, selected from a set of N related files, utilizing a fuzzy data record pointer (“fuzzy”, as used herein, means that the data record pointer need not be coincident with the actual data record address). A modulus for the data record pointer divided by N is computed. This modulus is used to select the target file. A data record address is computed for the target data record utilizing the data record pointer and modulus. In this manner a fuzzy data record pointer is utilized to determine both the target file from a set of N related files and the target data record to be accessed within the target file.

Abstract: A verification method and a verification apparatus of a neural network for guaranteeing the operation of the neural network to any input signals which might be inputted.

Abstract: An orthogonal functional basis method for function approximation is disclosed. Starting with the orthogonal least squares method, a new subset selection method for selecting a set of appropriate basis functions is explained where, instead of picking a subset from a given functional basis, the subset is selected from a combination of functional basis evolved from a set of heterogeneous basis functions. The method results in a more efficient neural network.

Abstract: A neural network for predicting values in non-linear functional mappings having a single hidden layer function generator (12) and an output layer (40). The single hidden layer function generator (12) is operable to receive one or more mapping inputs (x1) and generate a plurality of terms (14) from each mapping input. The plurality of terms generated by the single hidden layer function generator (12) includes at least one trigonometric term selected from the group comprising sin(x1), sin(2x1), sin(3x1), cos(x1), cos(2xl), cos(3xl), cosec(xl), cotan(xl), and being free of Gaussian and Sigmoidal terms.

Abstract: The present invention provides for a method and an apparatus for using equipment state data for controlling a manufacturing process. Initial equipment state data is acquired. At least one semiconductor device is processed using the initial equipment state data is performed. Equipment and wafer state data processing is performed using data from the processing of the semiconductor device and the initial equipment state data. A determination is made whether at least one control input parameter used for processing of the semiconductor device is to be modified in response to performing the equipment and wafer state data processing. The control input parameter is modified in response to determining that at least one the control input parameter is to be modified.

Abstract: An auto-tuner for use in tuning a control element in a process control network having distributed control functions includes a first tuning element located in the field device in which the control element is operating and a second tuning element located in a different device that communicates with the first device via a communication network. The first tuning element controls the operation of the control element during the dynamic data capture phase of an auto-tuning procedure, collects data during this phase of the auto-tuning procedure and determines one or more process characteristics from the collected data. The first tuning element then communicates the determined process characteristics to the second tuning element via the communication network.

Abstract: A method and system for commissioning industrial plants, in particular in the basic materials industry, having a plant control system which carries out both non-control functions and control functions and whose control system operates with process models, in particular control engineering models, for example in the form of mathematical models, neural network models, expert systems etc., in a control system computing unit. The commissioning is carried out in subdivided fashion into commissioning the non-control functions with extensive initialization of the control functions, by means of personnel located on site, and extensive commissioning of the control functions by means of remotely-transmitted data via data lines from at least one site remote from the plant, preferably from an engineering center.

Abstract: A method of presetting the roll nip profile of a roll stand for rolling a rolled strip is provided. The roll nip profile is influenced by output values for the roll nip profile and the tensile stress distribution being set over the roll nip profile. The output values for the roll nip profile is determined by using a roll nip profile model which calculates the roll nip profile. The calculated roll nip profile or an equivalent quantity is linked to a correction value, in particular by addition or multiplication, to form a corrected calculated roll nip profile, so that the roll nip profile model is adapted to the actual roll nip profile of the roll stand by using the correction value.

Abstract: A network-embedded device controller for industrial process control comprises a processor for running a real-time operating system, input devices measuring parameters related to an industrial process and operatively connected with the processor, an embedded service stack with a local database containing data related to the industrial process, and a dynamic service stack including a program executable by the processor to control the industrial process in a predetermined manner. The device has particular applicability for use with an in-motion weighing system.

Abstract: The width of bands to be laminated on a mill train is adjusted by vertical upsetting rollers, resulting, however, in a narrowing at the band ends due to the asymetric material flow there. In order to solve the problem, the upsetting rollers are so designed as to move at the passage of the band ends in keeping with a curve defined according to specified parameters. The parameters are based on neuro-computer made predictions related to the milling process.

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: An integrated control system for a machine such as an engine. The machine is composed of plurality of components such as a throttle valve and a fuel injector. The control system has a plurality of control modules for controlling the respective components and finally controlling the machine through the control of the components. Each control module is associated with at least one parameter for controlling each component. The parameter is evolved under genetic algorithm so as to be adapted to at least one of a predetermined characteristic that is a target of the machine, a characteristic of a user who uses the machine, a using condition and an environmental condition of the machine.

Abstract: A method for transformation of fuzzy logic (FS) into a neural network (NN), in which, in order to form a defuzzified output value (y2) from normalized single-element functions (F1 . . . Fm), the single-element functions (F1 . . . Fm) are each assigned a singleton position (A1 . . . Am) and at least one singleton weighting factor (R1 . . . Rn), those singleton weighting factors (R1 . . . Rn) which are assigned to the same single-element function (F1 . . . Fm) are additively linked, and the singleton weighting factors (R1 . . . Rn) and the additively linked singleton weighting factors (R1 . . . Rn) are weighted via the corresponding singleton positions (A1 . . . Am) and are additively linked in order to form the defuzzified output value (y2). One advantage of the method according to the invention is that the singleton positions (A1 . . .

Abstract: A temperature sensor used to control a steel rolling mill includes a housing arranged adjacent to the mill in which a temperature detector is arranged. The detector generates a signal corresponding to a detected temperature. A microprocessor receives the signal and processes it in accordance with a programmed characteristic. The processed signal is delivered to an output switch which produces a control signal for controlling the delivery of roll material to a roller of the rolling mill.

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 flexible multifunction model-free adaptive controller capable of controlling a very broad range of processes uses storage and selective use of multiple controller parameter sets, measurement filtering, transient prediction and use of extra controllers to dynamically set constraints for the output of the process controller in order to deal with transients resulting from sudden input changes, yet allow the process to run close to its physical limitations under dynamically varying operating conditions and periodic large processing parameter changes.

Abstract: Setting parameters of a PID controller are obtained by feeding a step signal or another input signal to an assigned controlled system. The response signal emitted by the controlled system is sampled and the characteristics of the Bode diagram are generated from the input signal and the response signal by using a smoothing method and elementary correspondences. The characteristics are normalized and input values are derived therefrom for a neural network which is trained on the properties of the controlled systems. The neural network directly generates the setting parameters for the controller.

Abstract: A method for performance envelope boundary cueing for a vehicle control system comprises the steps of formulating a prediction system for a neural network and training the neural network to predict values of limited parameters as a function of current control positions and current vehicle operating conditions. The method further comprises the steps of applying the neural network to the control system of the vehicle, where the vehicle has capability for measuring current control positions and current vehicle operating conditions. The neural network generates a map of current control positions and vehicle operating conditions versus the limited parameters in a pre-determined vehicle operating condition. The method estimates critical control deflections from the current control positions required to drive the vehicle to a performance envelope boundary.

Abstract: Pattern recognition of common modes by neural networks and other techniques are used to monitor and determine or predict the state of networks, computers, software systems, logical networks or other components of an information system, to report actual or predicted states, and to report other state characteristics.

Abstract: A robot with multi-joint arms, wherein articulated first and second arms 9, 10 are movable in a horizontal plane as well as in a vertical direction, the second arm having a free end to which a supplementary unit A is removably connected, the supplementary unit A including a joint member 17 adjustably connected to the free end of the second arm 10 and a working member 18 removable connected to the joint member 17. The working member 18 has a predetermined length Lx which determines the effective length L2 of the second arm 10, and wherein the data representing the effective length L2 of the second arm is entered to renew the existing effective length of the second arm to produce new data representing the value of the newly entered effective length L2 of the second arm 10, the new data being employed to calculate out the value to control the operations of the first and second arms 9, 10.

Abstract: The invention relates to a control system to which a state vector representing the states of a controlled system is applied. The control system provides a correcting variables vector of optimized correcting variables. The relation between state vector and correcting variables vector is defined by a matrix of weights. These weights depend on the solution of the state dependent Riccati equation. An equation solving system for solving the state dependent Riccati equation in real time is provided. The state vector is applied to this equation solving system. The solution of the state dependent Riccati equation is transferred to the control system to determine the weights.

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 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: Process and system for programmed control of a home and building automation system, for saving energy and improving comfort, with the process being performed by means of software and presence and activity monitoring sensors. The software includes first adaptive algorithms controlled by signals from the presence and activity monitoring sensors, for continuously detecting and storing systematic and stochastic behavior of at least one person in a room and over several rooms, and including second modifiable deterministic algorithms for adapting and triggering actuators of installations or groups of installations. The algorithms are combined with external parameters into a third algorithm to optimize the automation system as a whole and to adjust the first and second algorithms by means of feedback.

Abstract: An Advanced Process Control (APC) Framework performs automatic process control operations through the design and development of a software framework that integrates factory, process, and equipment control systems. The APC Framework benefits semiconductor-manufacturing factories, or “fabs,” throughout the development of the APC Framework by using an iterative development approach. The APC Framework is designed to integrate seamlessly with commercially-available APC tools. The APC Framework specifies components and a component structure that enable multiple vendors to build and sell framework-compatible products using an open architecture that accommodates plug-and-play components. The APC Framework advantageously increases product yield distributions and equipment utilization, and lowers defect densities.

Abstract: A Virtual Design Module (VDM) used in a networked design environment generates manufactured product designs that are near optimal in terms of cost and production cycle time by using design data files containing alternative parts and manufacturers information. Numerous product design alternatives are considered and evaluated in terms of design-manufacturing-parts-supplier feasibility and real-time information on cost and production cycle time for realization. The VDM generates a population of new designs with appropriate board design information to allow for design-manufacturer-supplier decision making and determines the feasibility of each member of the current generation of designs and rejects designs that are not feasible. The VDM triggers Mobile Software Agents (MSA) that obtain data for parts availability, cost, lead time and manufacturer data for manufacturing availability, cost and lead time for each feasible member of the current generation of designs and return the data.

Abstract: A method and apparatus of correcting for saturation in a current transformer, which outputs a current measurement, is provided. A switching algorithm receives a value of the current measurement from the current transformer and determines within which of three ranges the value falls. If the value falls in a first range, the current measurement is provided to a protective device such as a relay. If the value falls in a second range, the current measurement is provided to an artificial neural network that produces an output that accounts for saturation of the current transformer. If the value falls in a third range, the current measurement is provided to another artificial neural network that produces an output that accounts for saturation of the current transformer.

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 process model of an industrial process or system is generated. The model correlates a first number M of process parameters forming input values with a second number L of quality characteristics forming output values, which are processed to form feedback control signals for the process or system. A third number N of training data sets of the industrial process are first gathered and processed during a learning phase of the model with the help of a central processing unit, whereby a preliminary approximately model is used including a neural network with local approximation characteristics. The neural network is connected in parallel with a linear network. Both networks are connected to the same inputs. The neural network initially has a number N of neural cells corresponding to the number of training data sets. A weighted linear combination of the M process parameters is performed. The linear network and the neural network are connected with their outputs through weighting circuits to a common summing point.

Abstract: A system and method for monitoring an industrial process and/or industrial data source. The system includes generating time varying data from industrial data sources, processing the data to obtain time correlation of the data, determining the range of data, determining learned states of normal operation and using these states to generate expected values, comparing the expected values to current actual values to identify a current state of the process closest to a learned, normal state; generating a set of modeled data, and processing the modeled data to identify a data pattern and generating an alarm upon detecting a deviation from normalcy.

Abstract: An apparatus and method for controlling a process using a neural network which operates as part of a closed loop control system. The state of the control system is defined by one or more process condition signals and monitored for a predetermined set of controller parameters. The output of the control system is one or more device control signals, used by a control device to alter a process being controlled. The neural network uses normalized values of process condition signals in combination with a predetermined set of controller parameters to produce correction control signals. The correction control signals are then used to the create device control signals. Proper normalization of at least one of the process condition signals using the throttling range set by the controller parameters is necessary. The remaining input signals must be normalized as well, but the method of normalization is not as critical, except to create a common range for all process condition signals input to the neural network.

Abstract: A highly efficient, accurate and high performance modeling support system 100 for modeling a control model necessary for simulating a control object is provided, in which a plurality of data sets stored in a first data base 121 for use in modeling a plurality of control models is divided into a plurality of sub data groups each having least statistical disparity from each other by data divider 114, and stored in a plurality of sub data bases, respectively in second data base 122. Modeling means 115 constructs a control model using data in one of the plurality of sub data groups, which is then evaluated by model evaluator 116 using data in another one of said plurality of sub data groups.

Abstract: A deadband control assistant, operating as part of a control system, eliminates oscillation of a controlled process output around a desired setpoint. The deadband control defines a deadband range, and steady-state-stabilization-time for the process being controlled. When the process output is within the deadband range and oscillation around the setpoint occurs for the steady-state-stabilization-time, the error is set to zero. The control signal may simultaneously be set, or may settle to a fixed value. The deadband control assistant thus allows the process output to stabilize as close as possible to the desired setpoint. Normal process control may resume once the process output strays outside the deadband range.