Parameter Estimation Or Identification Patents (Class 700/52)
-
Patent number: 6748279Abstract: A factorial experiment is conducted on a manufacturing process to generate a response matrix. The responses are used to calculate individual contrasts in a document as well as replicates effects. The contrast sums are also calculated and displayed in the document. The largest of the contrast sums are identified, and effects associated with those contrast sums are tested for significance using an end count method. The information from the process transformed into “significant effects” information is used to adjust process variables to improve the manufacturing process by avoiding the effect or imparting it to a measurable response of the process.Type: GrantFiled: January 31, 2001Date of Patent: June 8, 2004Assignee: Red X Technologies, Inc.Inventor: Carl Bennett
-
Publication number: 20040107012Abstract: Real-time control of a dynamical system is provided by determining control variables that get as close as possible to producing a desired response. Additional consideration of physical limitations leads to a convex Quadratic Program with inequality constraints that needs to be solved in real-time. A new active set algorithm is described to solve the convex Quadratic Program efficiently that meets real-time requirements. Based on the key observation that the physical limitations of the system translate to optimal active sets that remain relatively unchanged over time (even though the actual optimal controls may be varying), starting guesses for the active set obtained from the final iterate in the previous time period greatly reduces the number of iterations and hence allows the Quadratic Programs to be solved to convergence in real-time.Type: ApplicationFiled: December 2, 2002Publication date: June 3, 2004Inventors: Indraneel Das, James W. Fuller
-
Publication number: 20040107013Abstract: A method of controlling a multivariable system includes the step of receiving a plurality of sensor signals indicating current conditions of the system and receiving a plurality of commands. The desired dynamic response of the system is then determined based upon the commands and the sensor signals. The problem of controlling the system to achieve the desired dynamic response without violating numerous actuator and physical constraints is then formulated as a quadratic programming problem. By solving the quadratic programming problem, the effector commands are determined and the physical constraints are enforced.Type: ApplicationFiled: December 2, 2002Publication date: June 3, 2004Inventors: James W. Fuller, Indraneel Das
-
Patent number: 6745089Abstract: An autonomous controller, comprised of a state knowledge manager, a control executor, hardware proxies and a statistical estimator collaborates with a goal elaborator, with which it shares common models of the behavior of the system and the controller. The elaborator uses the common models to generate from temporally indeterminate sets of goals, executable goals to be executed by the controller. The controller may be updated to operate in a different system or environment than that for which it was originally designed by the replacement of shared statistical models and by the instantiation of a new set of state variable objects derived from a state variable class. The adaptation of the controller does not require substantial modification of the goal elaborator for its application to the new system or environment.Type: GrantFiled: February 1, 2001Date of Patent: June 1, 2004Assignee: California Institute of TechnologyInventors: Robert D. Rasmussen, Daniel L. Dvorak, Kim P. Gostelow, Thomas W. Starbird, Erann Gat, Steve Ankuo Chien, Robert M. Keller
-
Patent number: 6738682Abstract: A method for scheduling activities in a manufacturing system includes defining a plurality of observed states associated with the manufacturing system. State estimates are generated for the observed states. Uncertainty values for the state estimates are generated. A plurality of candidate schedules for performing activities in the manufacturing system is identified. Changes to the uncertainty values are predicted based on the candidate schedules. One of the candidate schedules is selected based on the predicted changes to the uncertainty values.Type: GrantFiled: July 31, 2002Date of Patent: May 18, 2004Assignee: Advances Micro Devices, Inc.Inventor: Alexander J. Pasadyn
-
Patent number: 6731990Abstract: Communications data such as traffic levels in a communications network is analysed using techniques adapted from the study of chaos. Future values of a series of communications data are predicted and an attractor structure is determined from the communications data. This enables the communications processes to be monitored, controlled and analysed. Action can be taken to modify the communications process using the results from the prediction and attractor structure to reduce costs and improve performance and efficiency. These methods may also be used for product data from manufacturing processes. An algorithm bank is compiled containing prediction algorithms suitable for different types of data series, including those exhibiting deterministic behaviour and those exhibiting stochastic behaviour. Recent past values of a data series are taken and assessed or audited in order to determine which of the algorithms in the bank would provide the optimal prediction.Type: GrantFiled: January 27, 2000Date of Patent: May 4, 2004Assignee: Nortel Networks LimitedInventors: Malcolm Edward Carter, Otakar Fojt, Michael Maurice Dodson, Jason Levesley, Christopher Hobbs
-
Publication number: 20040078097Abstract: In a motor vehicle air conditioning installation comprising a central control unit (12), a computer network (14) connected to the central control unit (12) and a plurality of peripherals (101, 102, 103, . . . , 10n) connected to nodes (N1, N2, N3, . . . , Nn) of the computer network, the peripherals are connected in series via a common conductor (18). The installation comprises means (120) for inputting at the level of each peripheral an electric signature on a common conductor, means (130) for detecting an electric signal depending upon the electric signatures of the peripherals situated upstream on the common conductor (18), means for generating a data item representing the relative position of the peripheral on the basis of the electric signal detected, and means for identifying the node of the network to which it is connected on the basis of the relative position data item generated.Type: ApplicationFiled: August 26, 2003Publication date: April 22, 2004Inventors: Christophe Bruzy, Vincent Aubry, Bertrand Puzenat, Laurent Rede
-
Publication number: 20040078098Abstract: A computer system for assessing workplace hazards presents potential hazards to a user as a series of questions, which a user can select as being relevant or not to a particular task being carried out. All the data is stored in a database (10) with the different categories of hazards being associated with different work areas. Potential hazards are presented to the user and selected, if relevant. Where potential hazards are identified appropriate exposure avoidance controls are identified. A reporter generates a report setting out identified hazards and their avoidance controls. Permits and equipment data are also stored in the database and included in the report. Risk ratings are also provided for each identified risk. By providing a predefined database of hazards, avoidance controls and associated risk rating for each hazard, user subjectivity and human error are significantly reduced.Type: ApplicationFiled: October 14, 2003Publication date: April 22, 2004Inventor: Owen Jeffries
-
Publication number: 20040078099Abstract: The present invention discloses system and method for analyzing operational parameter data from a malfunctioning locomotive or other large land-based, self-powered transport equipment. The method allows for receiving new operational parameter data comprising a plurality of anomaly definitions from the malfunctioning equipment. The method further allows for selecting a plurality of distinct anomaly definitions from the new operational parameter data. Respective generating steps allow for generating at least one distinct anomaly definition cluster from the plurality of distinct anomaly definitions and for generating a plurality of weighted repair and distinct anomaly definition cluster combinations. An identifying step allows for identifying at least one repair for the at least one distinct anomaly definition cluster using the plurality of weighted repair and distinct anomaly definition cluster combinations.Type: ApplicationFiled: October 16, 2003Publication date: April 22, 2004Inventors: Anil Varma, Nicholas Edward Roddy, David Richard Gibson
-
Patent number: 6721610Abstract: The invention relates to a method for pre-calculating the parameters of industrial processes. According to said method, an admissible process input variable vector is determined with definition ranges allocated to each variable and a process output variable vector is determined with the pre-calculable process parameters. Known information on the process is stored in a data bank and ranges of validity for the process input variables are allocated to said information. For each process input vector inputted from an admissible definition range provided with valid information, exactly one process output vector is determined according to the information valid therefor.Type: GrantFiled: December 26, 2002Date of Patent: April 13, 2004Assignee: Siemens AktiengesellschaftInventors: Dirk Gade, Thomas Peuker
-
Publication number: 20040049296Abstract: A control system for a plant is disclosed. The control system includes a controller for controlling the plant based on a controlled object model which is obtained by modeling the plant. The controlled object model is modeled using an input and an output of the plant which are sampled at intervals of a sampling period which is longer than a control period of the controller. The sampled input of the plant is a filtered control output which is obtained by filtering an output of the controller. The controller carries out a control process of the plant at intervals of the control period.Type: ApplicationFiled: September 5, 2003Publication date: March 11, 2004Applicant: Honda Giken Kogyo Kabushiki KaishaInventors: Eiji Hashimoto, Tetsuya Ishiguro, Masahiro Sato, Shusuke Akazaki, Yoshihisa Iwaki, Yuji Yasui
-
Publication number: 20040049295Abstract: An optimization technique for use in driving a process plant controller, such as a model predictive controller, uses an organized, systematic but computationally simple method of relaxing or redefining manipulated, control and/or auxiliary variable constraints when there is no feasible optimal solution within pre-established constraints, to thereby develop an achievable solution for use by the controller. The optimization routine uses penalized slack variables and/or redefines the constraint model in conjunction with the use of penalty variables to develop a new objective function, and then uses the new objective function to determine a control solution that bests meets the original constraint limits.Type: ApplicationFiled: June 19, 2003Publication date: March 11, 2004Inventors: Wilhelm Wojsznis, Terrence Blevins, Mark Nixon, Peter Wojsznis
-
Publication number: 20040030422Abstract: The invention relates to a field device for automation systems, comprising a unit for internal data organisation, means for recording process parameters and/or means for outputting regulation parameters, means for processing signals according to the recorded regulation parameters or the regulation parameters to be output, and at least one communication interface. The means and the communication interface thus communicate with a unit for internal data organisation in order to exchange data between each other. A transaction manager unit, which is provided to control blockings, transactions and/or users, is used to control the access via the means and via the at least one communication interface to the data of the unit for internal data organisation between the means, the at least one communication interface and the unit for internal data communication.Type: ApplicationFiled: October 9, 2003Publication date: February 12, 2004Inventor: Rene Simon
-
Publication number: 20040027235Abstract: The invention relates to a method for cyclical communication between communications stations (1, 2, 3) which are provided for controlling or surveying a technical process (50), through a bus (4). According to the invention, communications relations (12, 13, 21, 31) which are provided for the communication stations (1, 2, 3) are executed during each bus cycle of predetermined duration (&Dgr;t) and if a communications relation is disturbed, its repetition (21W) is scheduled for a succeeding bus cycle and the disturbed communications relation is acknowledged with a special acknowledgement code.Type: ApplicationFiled: February 7, 2003Publication date: February 12, 2004Inventor: Gunter Steindl
-
Patent number: 6687649Abstract: A system, method and computer program for facilitating the operation of a control system utilizing one or more input devices is disclosed. The control system provides the ability to monitor and modify operating parameters of the input devices. A function integrated within a software code is to be executed by the control system. The function is responsive to the input device. The input device includes at least one condition comprising a state wherein all of the conditions must be met prior to the execution of the function. A display of an unmet condition is exhibited to operator personnel of the control system during the execution of any function having an unmet condition prohibiting the execution of the function. The operator personnel is capable of overriding the unmet condition by selecting the state of the condition that allows the condition to be met, thus allowing the function to be executed. A maintenance log contains a record of the conditions that were bypassed or overridden.Type: GrantFiled: April 30, 2001Date of Patent: February 3, 2004Assignee: Schneider Automation Inc.Inventors: John Ditter, Scott Walczak
-
Patent number: 6684112Abstract: A Robust Model-Free Adaptive controller is disclosed for effectively controlling simple to complex systems including industrial processes, equipment, facilities, devices, engines, robots, vehicles, and appliances. Without the need of re-designing a controller or re-tuning the controller parameters, the inventive controller is able to provide a wide robust range and keep the system under automatic control during normal and extreme operating conditions when there are significant disturbances or changes in system dynamics. Because of its simplicity and capability, the control system is useful for building flexible and adaptive production systems to fulfill the on demand manufacturing needs of the new e-commerce environment.Type: GrantFiled: April 11, 2001Date of Patent: January 27, 2004Inventor: George Shu-Xing Cheng
-
Publication number: 20040015247Abstract: There are provided a method and system for processing information indicative of a frequency of reproduction of information recorded on mediums, such as music data recorded on CDs. A reproduction-frequency information processing system has a plurality of terminals each configured to record pieces of information and to reproduce the recorded information and a server system communicably connected with terminals via communication means and configured to perform statistic processing on a reproduced condition of the recorded information at each terminal. Each terminal obtains reproduction-frequency information, every piece of the recorded information, depending on the reproduced condition, and transmits the obtained reproduction-frequency information to the server system in response to a request issued from the server system.Type: ApplicationFiled: July 8, 2003Publication date: January 22, 2004Applicant: PIONEER CORPORATIONInventor: Kazuhiro Kamiya
-
Publication number: 20040002776Abstract: A system and method which partitions a parameter estimation model, a fault detection model, and a fault classification model for a process surveillance scheme into two or more coordinated submodels together providing improved diagnostic decision making for at least one determined operating mode of an asset.Type: ApplicationFiled: June 20, 2003Publication date: January 1, 2004Inventor: Randall L. Bickford
-
Publication number: 20030225465Abstract: A system and method 100 of using residual feedback in a control loop in a manner that substantially eliminates the steady state error in the predicted states that results from the mismatch in gain between the plant and the model. The control effort used by the estimator to predict the next position is modified to compensate for this difference in gain. By integrating the residual, and modifying the apparent control effort accordingly, the residual is driven to have a mean value of zero. When the residual goes to zero, by definition, the steady state error in the position state goes to zero; and to the extent that the model matches the plant, the velocity state also approaches zero such that the steady state error in the predicted states are substantially eliminated, allowing for improved control.Type: ApplicationFiled: May 31, 2002Publication date: December 4, 2003Inventor: Eric G. Oettinger
-
Patent number: 6658305Abstract: A method is disclosed for automatic synthesis of multi-variable controllers. A design sequence for designing a controller for a system is selected, wherein the controller includes one or more controller components. A first set of parameters associated to the one or more controller components is then selected. A second set of parameters associated to the first set of parameters is subsequently selected. One or more parameters of the second set are then iteratively modified to obtain a set of modified parameters, until a plurality of performance characteristics calculated using the second set of parameters meets a performance objective for the system.Type: GrantFiled: March 10, 2000Date of Patent: December 2, 2003Assignee: Tokyo Electron LimitedInventors: Thorkell Gudmundsson, Mark Erickson
-
Patent number: 6658304Abstract: A computer based method of controlling an industrial process, including the steps of measuring the values of at least one process variable (y1, y2), and predicting future deviations of one process variable (y1) with regard to the measured value of said at least one process variable (y1, y2), and providing a control signal (uFB) based on said predictions, and by means of a first control law. The method comprises the further steps of measuring a measurable disturbance (d) in the process, predicting future deviations of said process variable (y1) with regard to said disturbance but without regard to the measured value of said at least one process variable (y1, y2), and providing a control signal (ud) based on said predictions by means of a second control law.Type: GrantFiled: October 5, 2000Date of Patent: December 2, 2003Assignee: ABB ABInventors: Mats Molander, Michael Lundh
-
Patent number: 6640152Abstract: Headbox transient responses for sheet weight and moisture are modeled as a combination of two sets of time constants and dead time delays. One set represents a shorter delay with faster response dynamics, the fast mode weight and moisture responses, and the other models the longer delay with slower dynamics, the slow mode weight and moisture responses. A weight and/or moisture transient model is then formed for headbox changes by combining the fast mode weight and moisture responses and the slow mode weight and moisture responses. Stock weight and moisture dynamic and delay time models are determined for operation of stock flow of the paper making machine and the stock flow is controlled in accordance with the stock weight and/or moisture models and the headbox weight transient and/or moisture transient model to compensate for weight and moisture changes in a web of paper being manufacture which weight and moisture changes result from headbox changes.Type: GrantFiled: March 24, 2000Date of Patent: October 28, 2003Assignee: ABB Automation, Inc.Inventors: Shih-Chin Chen, Timothy F. Murphy
-
Patent number: 6633782Abstract: A diagnostic system for use in a process control system collects and stores in a database information pertaining to the operation of the process control system, and that uses an expert engine to apply rules for analysis to the information in the database to determine solutions to problems. The database stores various types of information such as event and alarm data, notices of scheduled maintenance and changes to operating parameters, and historical data related to previous changes to the process control system that are relevant to determining both the source of the problems detected in the process control system and the steps necessary to either further analyze or correct the detected problems. The diagnostic system identifies the source of the problem and identifies and runs the appropriate analytical tools or takes remedial measures based on the rules for analysis for the expert engine.Type: GrantFiled: February 7, 2000Date of Patent: October 14, 2003Assignee: Fisher-Rosemount Systems, Inc.Inventors: Trevor D. Schleiss, Mark Nixon, Terrence Lynn Blevins, Todd B. Brase, Suresh Ganesamoorthi
-
Patent number: 6633826Abstract: The circuit arrangement serves to identify an input signal (s1) assigned to a predetermined signal class. It comprises signal-matching electronics (1) responsive to a current component and/or a voltage component of the input signal (s1) and providing a corresponding output signal (s2). By means of signal recognition electronics (2) of the circuit arrangement, the output signal (s2) is transformed into a recognition signal (c1) representative of the signal class.Type: GrantFiled: October 4, 2000Date of Patent: October 14, 2003Assignee: Endress + Hauser Wetzer GmbH + Co. KGInventor: Dieter Schmidt
-
Patent number: 6628995Abstract: Aircraft flight data collection is enhanced by changing the data sampling rate as a function of an operational condition of the aircraft, such as its flight phase. An algorithm is used to determine the flight phase. Then, the data is collected at a first sampling rate when the flight phase is one of a first set of flight phases and collected at a second sampling rate when the flight phase is one of a second set of flight phases. The second sampling rate is greater than the first sampling rate to maximize the data storage capacity of the system. Generally, the higher sampling rate is used with transient flight phases such as takeoff and thrust reverse.Type: GrantFiled: August 11, 2000Date of Patent: September 30, 2003Assignee: General Electric CompanyInventors: Shannon Lee Korson, Bruce Gunter Schings, Packey Pasquale Velleca, Mary Beth Golding
-
Publication number: 20030171829Abstract: Apparatus for control of a complex process, said process being described by a plurality of input variables, a plurality of intermediate variables and a plurality of output variables having relationships therebetween such that ones of said inputs and said intermediate variables effect respectively different output variables, each of said output variables having a target, said apparatus comprising an optimizer for finding an optimum value for respective ones of said input and intermediate variables to maximize a summed convergence of said output variables to said targetsType: ApplicationFiled: March 7, 2002Publication date: September 11, 2003Applicant: Insyst Ltd.Inventors: Yossi Fisher, Jehuda Hartman, Yuri Kokotov, Jacques Seror, Efim Entin
-
Patent number: 6618681Abstract: A method of predicting by non-invasive testing the available energy of a battery at any state of charge by acquiring data of the parameters of internal resistance (IR), open circuit voltage (OCV) and temperature (T), the points of voltage and current of the slope on a positive current ramp of Vup and Iup at the transition from charge to overcharge and on a negative current ramp of Vdn and Idn at the transition from overcharge to charge for a plurality of batteries. Next an algorithm in the form of a linear equation is developed using this data. The available energy of a battery under test is predicted by acquiring from it the numerical data values of these parameters and applying them to the algorithm.Type: GrantFiled: May 2, 2001Date of Patent: September 9, 2003Assignee: Honeywell International Inc.Inventors: Steven Hoenig, Harmohan Singh, Thirumalai G. Palanisamy, Margaret Eagan
-
Patent number: 6611726Abstract: A method for forecasting a value of a dependent variable, such as product demand, in a future time period later than the next, upcoming future time period. The method includes selecting a dependent variable for which a value is to be forecast, gathering historical data on values of the dependent variable and explanatory variables in prior time periods, and determining a forecasting equation based on the gathered historical data. The method includes selecting a future time period that is a number of time periods beyond the next, upcoming time period. The forecasting method continues with calculating a forecasted value of the dependent variable for the selected future time period, then determining an error value by comparing the forecasted value with the historical data and based on the error value, modifying the forecasting equation to reduce the error value.Type: GrantFiled: September 15, 2000Date of Patent: August 26, 2003Inventor: Carl E. Crosswhite
-
Patent number: 6606525Abstract: A system and method for merging static data and dynamic data downloaded to a browser from a network site of a network, preferably being a website on the Internet. The system includes at least one server in communication with the network, and the server hosts one or more network sites that include static data and dynamic data. The system further includes at least one user computer in communication with the network with a browser on the user computer that selectively interacts with network sites and selectively downloads the static data and dynamic data from the interacted network site. The user computer stores the static data and dynamic data in a local data store and recognizes the static data and dynamic data, preferably through marking with pointers, and upon refreshing of the downloaded data being requested, the user computer downloads dynamic data from the network site.Type: GrantFiled: December 27, 1999Date of Patent: August 12, 2003Assignee: Motorola, Inc.Inventors: Sivakumar Muthuswamy, Anupama Ramamurthy
-
Patent number: 6601116Abstract: A device for writing descriptors, the device including a local memory comprising a multiplicity of mini-queues, wherein each of the mini-queues temporarily stores a plurality of descriptors, wherein each of the descriptors is associated with one of the data packets. Additionally including an output memory comprising a multiplicity of output queues, wherein each of the output queues in output memory is associated with one of the queues in said local memory, and a burst writer which writes N descriptors simultaneously from the mini-queue in the local memory to its associated output queue in output memory.Type: GrantFiled: July 26, 1999Date of Patent: July 29, 2003Assignee: Marvell Semiconductor Israel Ltd.Inventors: David Shemla, Rami Rozensvaig
-
Publication number: 20030139828Abstract: A system and method for preprocessing input data to a support vector machine (SVM). The SVM is a system model having parameters that define the representation of the system being modeled, and operates in two modes: run-time and training. A data preprocessor preprocesses received data in accordance with predetermined preprocessing parameters, and outputs preprocessed data. The data preprocessor includes an input buffer for receiving and storing the input data. The input data may be on different time scales. A time merge device determines a desired time scale and reconciles the input data so that all of the input data are placed on the desired time scale. An output device outputs the reconciled data from the time merge device as preprocessed data. The reconciled data may be input to the SVM in training mode to train the SVM, and/or in run-time mode to generate control parameters and/or predictive output information.Type: ApplicationFiled: January 18, 2002Publication date: July 24, 2003Inventors: Bruce Ferguson, Eric Hartman
-
Patent number: 6597957Abstract: A system for consolidating and sorting event data include a computing platform communicatively coupled to a computer readable medium and a network. The computer readable medium may store an application that includes at least one node mapped into a tree. The at least one node may have a data element reference including a pointer to a data element that includes event data received via the network. In addition, the node may have a row indicator node count, a least child reference, a greatest child reference, a lesser sibling reference, a greatest sibling reference, a parent reference, and a status manager reference.Type: GrantFiled: December 20, 1999Date of Patent: July 22, 2003Assignee: Cisco Technology, Inc.Inventor: Jonathan G. Beakley
-
Patent number: 6597958Abstract: A method for determining the measure of control provided to a process by a control system. The determines process model parameters for a simple and complex model of the process and uses those parameters along with the value of the process variable and the final control element position to predict the off control data. The method also uses the process model parameters to determine the optimal tuning and then forecasts the optimal process performance from the predicted off control data and the determined optimal tuning.Type: GrantFiled: March 22, 2001Date of Patent: July 22, 2003Assignee: ABB Automation Inc.Inventor: Kevin D. Starr
-
Patent number: 6584365Abstract: An electronic trip device including an analog measuring for measuring electrical signals, analog-to-digital conversion having an input connected to the analog measuring an output having a first output signal, a digital processing means connected to the conversion for performing protection functions, and an offset correction. The offset corrections means are connected to the output of the conversion, and include high-pass digital filtering for computing an output signal value for each sample signal supplied by the conversion, according to an equation.Type: GrantFiled: July 26, 2000Date of Patent: June 24, 2003Assignee: Square D CompanyInventor: Pierre Blanchard
-
Publication number: 20030109940Abstract: Methods are described with which, from a large number of objects and in an efficient way, a search can be made for the objects which best resemble a sample object. For this purpose, the number of objects to be considered is restricted via efficiently calculated limiting values. In addition, the methods have search strategies which use the values of the characteristics of the objects considered for an efficient search strategy.Type: ApplicationFiled: October 30, 2002Publication date: June 12, 2003Inventors: Ulrich Guntzer, Wolf-Tilo Balke, Werner Kiessling
-
Patent number: 6577908Abstract: A system and method for adaptively designing self-tuning controllers, specifically PID controllers for process control systems. The method is based on a model-parameter interpolation, according to which a candidate process model is defined by a predetermined, limited set of models. Each of the models is characterized by a plurality of parameters, and, for each model, each of the parameters has a respective value that is selected from a set of predetermined initialization values corresponding to the parameter. Evaluation of each of the models includes computation of a model squared error and computation of a Norm that is derived from the model square errors calculated for the models. The Norm value is assigned to every parameter value represented in the model that is represented in an evaluation scan. As repeated evaluations of models are conducted, an accumulated Norm is calculated for each parameter value.Type: GrantFiled: June 20, 2000Date of Patent: June 10, 2003Assignee: Fisher Rosemount Systems, IncInventors: Wilhelm K. Wojsznis, Terrence L. Blevins
-
Patent number: 6574516Abstract: One or more signals are optimized in many industrial applications, one or more input quantities are combined according to a predetermined process to produce an output as a function of the inputs. For many types of functions, an optimum output value is possible, corresponding to a predetermined optimum input value or, where there is more than one input, to a set of such input values, which may be considered to define an input vector. A problem with many search techniques is that a local optimum rather than the global optimum is found. However, in the present optimization process, test values are selected, and the most optimal function values generated from the selected test values are stored. When a local optima is detected stored optimal function values are used to move away from the local optima.Type: GrantFiled: August 17, 2000Date of Patent: June 3, 2003Assignee: British Telecommunications public limited companyInventor: Shara Jalal Amin
-
Publication number: 20030097194Abstract: The invention relates to a method for pre-calculating the parameters of industrial processes. According to said method, an admissible process input variable vector is determined with definition ranges allocated to each variable and a process output variable vector is determined with the pre-calculable process parameters. Known information on the process is stored in a data bank and ranges of validity for the process input variables are allocated to said information. For each process input vector inputted from an admissible definition range provided with valid information, exactly one process output vector is determined according to the information valid therefor.Type: ApplicationFiled: December 26, 2002Publication date: May 22, 2003Inventors: Dirk Gade, Thomas Peuker
-
Publication number: 20030093162Abstract: Generally, an Eigen network and system using same are disclosed that use Principal Component Analysis (PCA) in a middle (or “hidden”) layer of a neural network. The PCA essentially takes the place of a Radial Basis Function hidden layer. A classifier comprises inputs that are routed to a PCA device. The PCA device performs PCA on the inputs and produces outputs (entitled “PCA outputs” for clarity). The PCA outputs are connected to output nodes. Generally, each output is connected to each output node. Each connection is multiplied by a weight, and each output node uses weighted PCA outputs to produce an output (entitled a “node output” for clarity). These node outputs are then generally compared in order to assign a class to the input. A system uses the PCA classifier to classify input patterns. In a third aspect of the invention, a PCA classifier is trained in order to determine weights for each of the connections that are connected to the output nodes.Type: ApplicationFiled: November 13, 2001Publication date: May 15, 2003Applicant: Koninklijke Philips Electronics N.V.Inventors: Srinivas Gutta, Vasanth Philomin, Miroslav Trajkovic
-
Publication number: 20030088322Abstract: A kiln thermal and combustion control. A predictive model is provided of the dynamics of selected aspects of the operation of the plant for modeling the dynamics thereof The model has at least two discrete models associated therewith that model at least two of the selected aspects, the at least two discrete models having different dynamic responses. An optimizer receives desired values for the selected aspects of the operation of the plant modeled by the model and optimizes the inputs to the model to minimize error between the predicted and desired values.Type: ApplicationFiled: August 14, 2002Publication date: May 8, 2003Inventors: Gregory D. Martin, Eugene Boe, Stephen Piche, James David Keeler, Douglas Timmer, Mark Gerules, John P. Havener
-
Patent number: 6560493Abstract: An apparatus for optimal control of distributed actuator and sensor arrays includes a control system having a state estimator including an operator which acts on estimation error through convolution with respect to a spatial variable to generate a state estimate, and a control output generator which applies a control operator to the state estimate through convolution with respect to the spatial variable. A method of designing an optimized control system includes the steps of obtaining a system model, computing a transform of the model with respect to the spatial domain, solving linear matrix inequalities to generate auxiliary variables, performing an inverse transform on the auxiliary variables, and computing an estimator operator and a control operator for a model-based estimator control system.Type: GrantFiled: February 26, 1999Date of Patent: May 6, 2003Assignee: Massachusetts Institute of TechnologyInventors: Munther A. Dahleh, Bassam Bamieh, Fernando Paganini
-
Patent number: 6556876Abstract: Variation in results of a semiconductor fabrication process is minimized by adjusting process parameters utilizing a fuzzy-controlled learning feedback system. The fuzzy-controlled learning system receives as inputs the results of the fabrication process and then converts these results into a fuzzy set defined by a membership function. An inference engine applies a fuzzy rule base to map an output fuzzy set from the input fuzzy set. The fuzzy output set is then converted to crisp outputs which automatically adjusts parameters of the fabrication process in order to minimize variation.Type: GrantFiled: October 12, 2000Date of Patent: April 29, 2003Assignee: National Semiconductor CorporationInventors: Hank Prosack, Patrick H. Ferland, Kenneth Swan
-
Publication number: 20030078683Abstract: A system and method for on-line training of a support vector machine (SVM). The SVM is trained with training sets from a stream of process data. The system detects availability of new training data, and constructs a training set from the corresponding input data. Over time, many training sets are presented to the SVM. When multiple presentations are needed to effectively train the SVM, a buffer of training sets is filled and updated as new training data becomes available. Once the buffer is full, a new training set bumps the oldest training set from the buffer. The training sets are presented one or more times each time a new training set is constructed. An historical database of time-stamped data may be used to construct training sets for the SVM. The SVM may be trained retrospectively by searching the historical database and constructing training sets based on the time-stamped data.Type: ApplicationFiled: September 5, 2001Publication date: April 24, 2003Inventors: Eric Hartman, Bruce Ferguson, Doug Johnson, Eric Hurley
-
Patent number: 6549817Abstract: A process for regulating a centrifuge used for solid/liquid separation of sewage sludge commences by measuring the input variables of the operating point ot the centrifuge, the variables including 1) suspended matter content of the centrate; 2) the flow rate of sludge; 3) the flow rate of reagent; and 4) the value of torque of the motor of a conventional centrifuge. Operating zones are established, each of which is two dimensional relative to the input variables. Fuzzy logic rules are established, qualifying the operation of the centrifuge, the rules respectively corresponding to the zones. The rules are implemented in response to the input variable measurements, where actions on the sludge flow rate and reagent flow rate make it possible to bring the operating point into a particular zone of optimal and stable centrifuge operation.Type: GrantFiled: July 19, 2000Date of Patent: April 15, 2003Assignee: DegremontInventor: Jacques Audibert
-
Patent number: 6549816Abstract: A plumbing control system includes a plurality of plumbing fixtures operated by an electronic control board having a microprocessor. The control board can form a node on a network that monitors and controls the functions of multiple boards throughout a facility from a central computer. Network software running on the central computer communicates with the control boards. The software includes a binding feature that requires reporting of a network operating variable at a first control board to the central computer or a second control board. The network software also has site and device setup forms.Type: GrantFiled: December 21, 2000Date of Patent: April 15, 2003Assignee: Sloan Valve CompanyInventors: Jerome M. Gauthier, Nhon T. Vuong, Mark J. Sippel
-
Patent number: 6540473Abstract: A robot vehicle for hot-line job for transmission and distribution lines which comprises an insulating boom formed of an insulating material at a tip end stage of multiple stage booms, a base end of which is supported on an elevated work vehicle to be capable of swinging, pivoting up and down, and extending and contracting, and a frame provided on the tip end of the insulating boom, the frame mounting thereon double arm manipulators of multishaft construction for performing a distribution work, a slide device for sliding independently these right and left double arm manipulators front to rear, and lifting arms of multishaft construction having a function of lifting a heavy article, so as to allow an operator to perform remote control. An electric actuator is used for driving the double arm manipulators and the slide device, and hydraulic manipulators for driving the lifting arms.Type: GrantFiled: June 13, 2001Date of Patent: April 1, 2003Assignees: Kabushiki Kaisha Yaskawa Denki, Kyushu Electric Power Co., Inc.Inventors: Hirofumi Inokuchi, Toshihide Tomiyama, Yoshihiro Yurita, Shinji Murai, Yusuke Hirano, Shinji Harada, Kengo Tsuruta, Youichi Motomura, Kazutoshi Imai, Yoshiaki Haga, Yoshikatsu Nakamura, Yukio Hashiguchi, Mitsuhiro Matsuzaki, Hiromichi Yamada, Akihiko Mishima, Yoshinaga Maruyama, Kyoji Yano, Moriyuki Nakashima, Osamu Yamashita
-
Patent number: 6539311Abstract: An apparatus and method for measuring chemical, biological, nuclear agents in an environment includes several detectors capable of measuring concentrations of the agents in the environment and a processor capable of operating an algorithm which, based on two sequential measures of concentration of the agent, estimates decay or elevation rate of the concentrations of the agent and feeds this estimated change rate to a Kalman filter which predicts the next measurement. The predicted next measurement of concentration is compared with the measured concentration and the error is calculated based on the discrepancy between the measured and predicted concentrations of the agent in the environment. Using this error, the attributes of the Kalman filter are adjusted to lower noise signals and improve further predictions of the concentration of the agent during the mission time.Type: GrantFiled: October 30, 2000Date of Patent: March 25, 2003Assignee: The Mitre CorporationInventor: Eugene L. Berger
-
Publication number: 20030050715Abstract: The invention relates to a communication station (S32-S34), which as a passive station in a communication system which is not authorized to transmit but listens in on all communication processes and can receive only data (D) actually intended for it, which fact can be conclusively verified by means of a destination point code (T) assigned to the data (D). According to the invention, said communication station is additionally upgraded to receive data (D), the destination (T) and originating point code (S) of which matches an entry (FT, FS) in a filter table (F) provided for in a communication station (S32-S34) which has been upgraded in accordance with the invention.Type: ApplicationFiled: October 27, 1999Publication date: March 13, 2003Inventor: GUNTER STEINDL
-
Patent number: 6529783Abstract: A method for estimating an interference phase shift when receiving a multicarrier signal formed of a time series of symbols modulating a plurality of carrier frequencies, at least some of the carrier frequencies of at least some of the symbols bearing reference elements, whereof the value at transmission level is known by the receiver carrying out the reception. The phase variation &dgr;n between at least two symbols bearing reference elements is estimated by analysing the reference elements. Each of the reference elements in the estimation is weighted by information representing the noise affecting the carrier frequency bearing the reference element.Type: GrantFiled: March 7, 2000Date of Patent: March 4, 2003Assignees: France Telecom, Telediffusion de FranceInventors: Pierre Combelles, Bertrand Sueur, Dominique Lacroix
-
Publication number: 20030040816Abstract: A control system having an input and an output module connected to a communication bus or network. The input module is responsive to a signal that represents a condition. In response to the condition, the input module transmits the representative signal onto the communication bus. The output module includes a reflex function. The reflex function converts the representative signal into a state signal. The output module is capable of transmitting the state signal onto the communication bus.Type: ApplicationFiled: July 12, 2001Publication date: February 27, 2003Inventors: Paul Wolejko, Steven Webster, Allan Pfeiffer, Jean-Francois Rolland