Abstract: A method of controlling a valve of a cam phaser of an internal combustion engine includes determining a delta duty cycle from a position error and a rate of change of error (error dot). A fuzzy gain value is calculated from the delta duty cycle and a correction term. A duty cycle is calculated from an integral gain value and the fuzzy gain value. The duty cycle is then used to control the valve of the cam phaser.

Abstract: Data from a monitoring system is analyzed for monitoring characteristics of a dynamic system. Characteristic data in respect of a dynamic system is monitored and provided with at least one known normal state. Characteristic data is received from a monitoring system and confirmation information from an operator when the dynamic system is in a known normal state. A normality model is derived including comprising data indicative of known normal states in response to received characteristic data and confirmation information. Future characteristic data is predicted from the normality model a difference function is provided indicating an acceptable difference between predicted and received characteristic data. Predicted characteristic data is compared with received characteristic data in conjunction with the difference function and an abnormality signal is produced if the difference exceeds the difference function.

Abstract: System and method for guiding a driver of a vehicle (20) from a starting point to a destination point using an assignment of levels of urgency of navigation cues. The system comprises a navigation unit (38), a controller (22) and a user interface (42). The navigation unit (38) is capable of selecting a route between the starting point and the destination point. The navigation unit (38) is further capable of generating a plurality of navigation cues to guide the driver along the selected route. The controller (22) is connected to the navigation unit (38) and receives navigation cues. The controller (22) has an urgency calculator (40) that is capable of assigning a level of urgency to the navigation cues. The controller (22) may also have a workload manager (60) that is capable of classifying an activity state of the driver as well as updating the level of urgency of the navigation cue. The user interface (42) presents the navigation cues based on the level of urgency of the navigation cues.

Abstract: Method for controlling the production of one or more liquid products produced by one or more plants, that can be an air separation plant, that has a liquid storage capacity and that consumes electrical power. Linguistic values related to current liquid inventory, a rate of change of liquid inventory, projected demand requirements and projected unit power costs are inputted into a expert system controller having one or more rule sets that apply such input linguistic values to produce output linguistic values that are converted back to an output numerical value of a forecasted differential production rate to be applied during a forecast period. The output numerical value of differential production is added to the current production rate of the plants to obtain a new production rate which is applied during the forecast period.

Abstract: The invention concerns heuristic algorithms for the classification of Objects. A first learning algorithm comprises a genetic algorithm that is used to abstract a data stream associated with each Object and a pattern recognition algorithm that is used to classify the Objects and measure the fitness of the chromosomes of the genetic algorithm. The learning algorithm is applied to a training data set. The learning algorithm generates a classifying algorithm, which is used to classify or categorize unknown Objects. The invention is useful in the areas of classifying texts and medical samples, predicting the behavior of one financial market based on price changes in others and in monitoring the state of complex process facilities to detect impending failures.

Abstract: A system for determining the state-of-health of batteries has an adaptive component. The system tests a battery by measuring a number of electrochemical parameters and using fuzzy logic to compute a state-of-health for the battery. The fuzzy logic computations are based upon membership functions which are preferably triangular. The membership functions may be stored in a matrix. The adaptive component adapts the system for use with new models of battery by obtaining a reliable measurement of SoH for a number of calibration batteries. Electrochemical parameters of the calibration batteries are measured. The adaptive component modifies a prototype matrix selected to match the chemistry type of the calibration batteries by scaling, translating and flexing its membership functions to achieve a matrix which specifies membership functions suitable for assessing state-of-health of the new model of battery.

Abstract: A method for controlling a vehicle semi-active suspension system comprising at least one suspension, providing for: detecting vehicle dynamic quantities during the vehicle ride; using the detected dynamic quantities, determining an index of ride comfort and an index of roadholding; applying a weight factor to the index of ride comfort and to the index of roadholding and, based on a Sky Hook control model, determining a target damping force characteristics for the at least one suspension of the suspension system; controlling the at least one suspension to put the respective damping force characteristics in accordance with the calculated target damping force characteristics. The weight factor is calculated dynamically during the vehicle ride, by means of a fuzzy calculation on the detected vehicle dynamic quantities.

Abstract: A system includes a sensor section which receives fuzzy information inputs X containing a plurality of components and converts the plurality of components into a plurality of measurable input physical quantities, a converter which receives a plurality of input physical quantities and converts the input physical quantities into a plurality of pulses having pulse widths corresponding to the magnitudes thereof, and a feature extraction section (NF) which receives a plurality of pulses, selects the plurality of pulses by using a set pulse width as a reference, and extracts feature information items y0, y1, y2, . . . which express the features of the fuzzy information inputs X from the fuzzy information inputs X according to the number of selected pulses.

Abstract: A method of controlling production of a gaseous product produced by one or more production plants such as an air separation plant, having a control system to vary production of the gaseous product. The gaseous product is consumed by a customer in accordance with a varying demand pattern. A storage capacity for the gaseous product which may be the pipeline itself or a gaseous receiver tank is provided. Control is exercised by a fuzzy logic program that utilizes the average rate of change of customer demand and storage capacity pressure to produce control data to vary the production rate of the plant.

Abstract: A method, system and apparatus for performing peritoneal dialysis are provided. To this end, in part, a method of controlling a medical fluid heater is provided. The method includes the steps of determining a first heater control output based on a number of measured inputs for the heater and at least one mathematical relationship between at least two of the measured inputs, determining a second heater control output based on at least one fuzzy logic membership function and at least one fuzzy logic rule, and determining a third heater control output based on the first and second outputs and using the third heater control output to control the heater.

Abstract: A system for separating a multi phase mixture into a first liquid phase component, a second liquid phase component and a solid phase component includes a three phase centrifuge and a control system for the centrifuge. The control system includes a fuzzy soft sensor programmed with fuzzy logic rules and a feed forward controller in signal communication with the fuzzy soft sensor. The feed forward controller is configured to adjust a feed rate and a feed temperature of the mixture based on the rules, the cold feed temperature, the percent change of water in the mixture, and the percent change of solids in the mixture. The system also includes a feedback controller configured to adjust the feed rate and the feed temperature of the mixture based on the rules, and the basic water and solid (BS&W) content of the first liquid phase component.

Abstract: Control apparatus for a horizontal boring tool is described which drives a boring lance via boring rods. The tool includes an input interface for receiving actual values of controlled variables of the horizontal boring tool, and an output unit for issuing control signals for controlling the horizontal boring tool. Between the input interface and the output unit is provided a fuzzy control unit which determines the control signals for activation of the horizontal boring tool from the actual values of the controlled variables and the desired values for the controlled variables by fuzzy logic while taking into account heuristic process values. The control apparatus permits automatic operation of the horizontal boring tool with good course steering and high precision.

Abstract: A filtering process is adapted for eliminating the need of prediscretizing a continuous-time differential model into a discrete-time difference model. It provides a universal robust solution to the most general formulation, in the sense that the system dynamics are described by nonlinear continuous-time differential equations, and the nonlinear measurements are taken at intermittent discrete times randomly spaced. The filtering process includes the procedures of validating the measurement using fuzzy logic, and incorporating factorized forward filtering and backward smoothing to guarantee numerical stability. It provides users a reliable and convenient solution to extracting internal dynamic system state estimates from noisy measurements, with wider applications, better accuracy, better stability, easier design, and easier implementation.

Abstract: The subject of the invention is a process for adjusting a processing unit, whereby at least one set value is varied from a starting value to a final value. In this case, the target value of a correcting value is set based on the starting value and the final value of the adjustment value, and the value of the correcting value is run up from its starting value to its target value using a parameter-dependent transfer function. A control variable is measured, the value of the control variable is compared to a comparison value, and in the case of deviations of the value of the control variable from the comparison value, at least one parameter of the transfer function is corrected.

Abstract: A triggering algorithm contains an accident classification section. The accident classification section senses the type of an accident that occurs. The accident classification section forms probability values for the types of accident in question when an accident cannot be classified unambiguously. By reference to the type of accident or the probabilities that are determined, weighting factors are formed and are taken into account in the calculation of the criterion and/or of one or more comparison thresholds. The triggering algorithm is modular and can thus be easily standardized on a module-specific basis.

Abstract: A method of and system for determining the state-of-health of a proton exchange membrane fuel cell stack connected to a load comprises detecting the real part of the impedance Z1 of the fuel cell stack at a selected frequency; detecting the voltage V1 of the fuel cell stack at open circuit; detecting the voltage V2 of the fuel cell stack when the maximum load current is being drawn from said fuel cell stack; and determining the state of health of said fuel cell stack from a fuzzy system trained in a relationship between said real part of the impedance of the fuel cell stack at the selected frequency, the voltage of the fuel cell stack at open circuit, and the voltage of the fuel cell stack when the maximum load current is being drawn from said fuel cell stack.

Abstract: A method of and system for determining the state-of-health of a lead acid battery of the type commonly used in portable external defibrillators comprises detecting an impedance characteristic of the battery at at least one selected frequency, counting the number of complete charge/discharge cycles that the battery undergoes, and determining the state of health of the battery from a fuzzy system trained in a relationship between the impedance characteristic and the cycle number of the lead acid battery and the state-of-health, wherein the state-of-health is a function of and varies with the battery's ability to deliver power required by the load and the battery's capacity to meet load requirements.

Abstract: A method for diagnosing whether a patient has a disease comprising interpreting the results of an immunoassay, wherein the immunoassay is viewed as a series of separate tests performed in parallel. The immunoassay can be a Western blot and the disease can be Lyme Disease.

Abstract: Embodiments of a fuzzy based thresholding technique for image segmentation are disclosed.

Abstract: Because of the enormous number of neurons and reflex pathways involved in the control of human motion, a comprehensive identification of the structure of the feedback control circuitry is a formidable experimental task. A theoretical approach based on a carefully developed model could shed light on its likely general structure, thereby providing guidance to experimenters. This paper presents a fuzzy logic based control strategy for modeling the human postural dynamics. In particular, we develop a fuzzy controller to coordinate the movement of a three segment sagittal human model to achieve upright posture. The actuation mechanism of the system consists of six muscle like actuators which are co-activated to modulate the stiffness at the joints. Simulation results are presented to illustrate the underlying principles.

Abstract: Embodiments of a fuzzy distinction based thresholding technique for image segmentation are disclosed. In one embodiment, at least one signal value level of the image is determined along which to divide a fuzzy histogram, the histogram being based, at least in part, on the image. The signal value represents a value which produces a divided fuzzy histogram with an extreme value of one of distinctiveness and fuzziness based on a measure of multidimensional distance between measurement distributions and their respective complements. The image is then segmented using the at least one signal value.

Abstract: An integrated circuit comprising a logic processor and a fuzzy logic coprocessor is disclosed which processes a plurality of analog inputs. The logic processor and fuzzy logic processor are combined in the form of a single integrated circuit. The integrated circuit accepts a plurality of analog inputs which are digitized and provided as output to a display peripheral or are used to control an actuator peripheral such as a control unit for a valve. The integrated circuit includes means for loading or exchanging informational elements with other units of an installation.

Abstract: A system includes a sensor section which receives fuzzy information inputs X containing a plurality of components and converts the plurality of components into a plurality of measurable input physical quantities, a converter which receives a plurality of input physical quantities and converts the input physical quantities into a plurality of pulses having pulse widths corresponding to the magnitudes thereof, and a feature extraction section (NF) which receives a plurality of pulses, selects the plurality of pulses by using a set pulse width as a reference, and extracts feature information items y0, y1, y2, . . . which express the features of the fuzzy information inputs X from the fuzzy information inputs X according to the number of selected pulses.

Abstract: A fuzzy logic based digital color transition improvement method and apparatus for increasing color sharpness by replacing the slow transition color edges with edges that have steeper rising and falling times. Fuzzy logic used here to decide where the transition happens and how to enhance the signal when transition happens. Based on the results of fuzzy logic inference, through weighting among input digital color signal, its N-pixel delayed signal and 2N-pixel delayed signal, the output signal has steep and smooth color edges without ringing.

Abstract: A method is provided for deriving a near-optimal fuzzy automaton for a given separation problem. The method includes the steps of: forming a first generation population (24) of fuzzy automata, where the first generation population of fuzzy automata includes a plurality of fuzzy automata; performing a mutation operation (28) on each fuzzy automaton in the first generation population of fuzzy automata; reproducing the first generation population of fuzzy automata using a survival of the fittest operation (30, 32, 34); and applying a cross-over operator (36) to the reproduced first generation population of fuzzy automata, thereby yielding a next-generation population of fuzzy automata. A near-optimal fuzzy automaton is identified by evaluating the performance (38) of each fuzzy automaton in the next-generation population; otherwise the methodology is repeated until a near-optimal fuzzy automaton is derived for the given separation problem.

Abstract: An industrial process material flow within an industrial process is controlled by a process control loop that includes a process controller, a field device controller and a field device in the process as well as process variable feedback from the process to the process controller. Both the process controller and the field device controller, such as a valve controller, are fine-tuned so that their interaction is taken into account and controller parameters, which provide an optimal control loop performance are found. This results in a significant improvement in the control loop performance, i.e., in a lower process variability, as compared with the traditional tuning method, which does not take into account the interaction between the two controllers.

Abstract: A neural network construct is trained according to sets of input signals (descriptors) generated by conducting a first experiment. A genetic algorithm is applied to the construct to provide an optimized construct and a CHTS experiment is conducted on sets of factor levels proscribed by the optimized construct.

Abstract: A universal robust filtering process is adapted for eliminating the need of prediscretizing a continuous-time differential model into a discrete-time difference model. It provides a universal robust solution to the most general formulation, in the sense that the system dynamics are described by nonlinear continuous-time differential equations, and the nonlinear measurements are taken at intermittent discrete times randomly spaced. The universal robust filtering process includes the procedures of validating the measurement using fuzzy logic, and incorporating factorized forward filtering and backward smoothing to guarantee numerical stability. It provides users a reliable and convenient solution to extracting internal dynamic system state estimates from noisy measurements, with wider applications, better accuracy, better stability, easier design, and easier implementation.

Abstract: A system is controlled by an actor-critic based fuzzy reinforcement learning algorithm that provides instructions to a processor of the system for applying actor-critic based fuzzy reinforcement learning. The system includes a database of fuzzy-logic rules for mapping input data to output commands for modifying a system state, and a reinforcement learning algorithm for updating the fuzzy-logic rules database based on effects on the system state of the output commands mapped from the input data. The reinforcement learning algorithm is configured to converge at least one parameter of the system state to at least approximately an optimum value following multiple mapping and updating iterations. The reinforcement learning algorithm may be based on an update equation including a derivative with respect to said at least one parameter of a logarithm of a probability function for taking a selected action when a selected state is encountered.

Abstract: A method for determining state of health (SOH) of an electrochemical device using fuzzy logic (i.e., an intelligent system) is presented. State of health of an electrochemical device is determined by an internal characteristic parameter (or external operating and environmental conditions) of the electrochemical device and a characteristic parameter of a load with an intelligent system. The electrochemical device comprises such devices as primary (“throwaway”) batteries, rechargeable batteries, fuel cells, hybrid batteries containing a fuel cell electrode or electrochemical supercapacitors. The intelligent system is trained in the relationship between the characteristic parameters of the electrochemical device, the characteristic parameters of the load and the SOH of the electrochemical device.

Abstract: Automatic detection of defects during the fabrication of semiconductor wafers is largely automated, but the classification of those defects is still performed manually by technicians. This invention includes novel digital image analysis techniques that generate unique feature vector descriptions of semiconductor defects as well as classifiers that use these descriptions to automatically categorize the defects into one of a set of pre-defined classes. Feature extraction techniques based on multiple-focus images, multiple-defect mask images, and segmented semiconductor wafer images are used to create unique feature-based descriptions of the semiconductor defects. These feature-based defect descriptions are subsequently classified by a defect classifier into categories that depend on defect characteristics and defect contextual information, that is, the semiconductor process layer(s) with which the defect comes in contact.

Abstract: A fuzzy logic controller system comprising a plurality of fuzzy logic controller units cascadably coupled to one another, each controller having a membership processor for receiving a single input signal corresponding to a sensed parameter from a controlled system and performing fuzzy inference operations based on the input signal for outputting an incremental output signal indicative of a requested change in a controlled output from a current level, a permissions processor for scaling the incremental output signal in response to a fulfillment signal from another controller, a processor responsive to the membership processor and to the permissions processor for providing a fulfillment signal to another controller indicative of a permissible range of freedom for scaling the incremental output signal of the another controller in the cascading sequence, an adder for summing each of the incremental output signals from the plurality of controllers and applying the sum to the controlled output for providing a corre

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: In a process for designing a fuzzy regulator with at least one input parameter and at least one output parameter (x1, x2, y), where an incomplete rule base with rules that are defined explicitly is supplemented by an implicit rule whose conditional part yields a truth value that has its greatest value when the explicit rules yield a truth value of 0 and is the smallest when the explicit rules have a truth value of 1, the truth value of the implicit rule may be determined by forming the maximum from the truth values of the explicit rules and calculating the complement to 1. This leads to a great simplification of the design of the regulator and prevents instabilities that can occur with an incomplete rule base.

Abstract: In a system of information processing by fuzzy logic, force coefficients are associated with values of variables processed by a fuzzy logic processor. The force coefficients show the degree of urgency with which information sent has to be taken into consideration or indicating the imperative nature of this information. In a network structure, this force coefficient is incorporated into address signals conveyed on the network. Then, in the membership function memories of the fuzzy logic processors, there are memories that take account of the classes of addresses for the extraction therefrom of the character of urgency of the information elements concerned.

Abstract: Control of an interconnection process for optical fibers involves evaluating the status of all modules used in the process and moving an optical fiber through the process in accordance with the status of the modules. Such control allows more than one fiber, and more than one fiber type, to be processed simultaneously. The processing may be optimized using fuzzy logic to track when operations performed by a module are near completion. A graphical user interface displaying the movement of the optical fibers through the process may also be provided.

Abstract: A system and method for automatically tuning a fuzzy logic process controller is presented. A process under control is automatically assessed to determined process control characteristics including, for example, ultimate gain, ultimate period and time delay. Then, these quantities are used to calculate tunable control parameters which include scaling factors, which are applied to the fuzzy logic controller in order to optimize the performance of the fuzzy logic controller when connected to control the process.

Abstract: An autonomic system for updating a fuzzy neural network includes a process of calculating an estimated value based on fuzzy inference by using a neural network structure, wherein a parameter to be adjusted or identified by fuzzy inference and outputted from the neural network is made to correspond to coupling loads which are updated by learning, i.e., fuzzy rules and membership functions are adjusted by learning. This system is characterized in that the addition and deletion of fuzzy rules are conducted based on changes in output errors in an autonomic manner, thereby effectively obtaining appropriate numbers of fuzzy rules optimal for an object such as a vehicle engine having strong non-linearity. Fuzzy rules are formed by a combination of membership functions representing variables such as an engine speed and a throttle angle.

Abstract: A PID control system and method which maintains continuity of the control output in spite of changes in controller parameters. The controller parameters include a proportional gain, an integral gain, a derivative gain, and a manual/automatic mode parameter. The control output at the time of a parameter change is set equal to the previous controller output, and the integrated error is set to a value consistent with the previous controller output according to a PID controller equation. This value for the integrated error serves to ensure the sustained continuity of the control output after the controller parameter change. Alternatively, the control output at the time of a parameter change is set equal to a first value continuous with respect to one or more previous control output values, and the integrated error is set to a value consistent with the first value according to the PID controller equation. Furthermore, the system and method employs integrator anti-windup. If the control output saturates, i.e.

Abstract: A fuzzy control apparatus for a jet loom detects a flying state of a weft yarn, performs, based on a detected value thus detected, fuzzy inference to determine a modification amount to a rotational frequency of a motor, drives the motor to rotate at a predetermined rotational frequency, and corrects the rotational frequency of the motor based on the correction amount.

Abstract: An apparatus for regulating welding parameters during laser beam welding includes a camera for detecting the geometry of a melt bath formed during the welding process. The camera is connected with an image-data-processing unit. The welding depth is regulated by the processing unit as a function of the detected melt bath length or melt bath area. The focus position of the laser beam is regulated as a function of the geometric similarity factor, which is calculated as the quotient of the melt bath area and the distance between the geometric centers of gravity of the laser beam keyhole and the melt bath area. Detection and regulation of gaps between the workpieces of height offset between the workpieces and/or lateral offset of the laser beam is also provided for.

Abstract: A method of growing turfgrass wherein a relatively large amount of precipitator coal fly ash is applied to an existing soil and a turfgrass is established thereon. The coal fly ash is weathered before or after application to the existing soil. A field employing a treated soil layer is also provided having improved soil characteristics which are particularly beneficial in the growth, harvest, and shelf life of turfgrasses.

Abstract: A processor operating in a fizzy logic's mode and including a fuzzyfication unit receiving a plurality of input variables on its input and being adapted to compute a membership value of such variables in a membership function, a processing unit connected downstream of the fuzzyfication unit to produce a fuzzy set, i.e. the results of fuzzy logic inference operations performed on the plurality of variables, and a defuzzyfication unit operative to translate the inference results into a so-called crisp value. The processor further includes a first memory device connected to the fuzzyfication unit and containing the set of membership functions, a second memory device connected to the defuzzyfication unit and containing the set of crisp value which appear in the THEN part of the fizzy rules, and a further memory device connected to the defuzzyfication unit.

Abstract: An adaptive fuzzy feature mapping (AFFM) technique provides a method for identifying and matching a new data pattern against a set of known data patterns using a combination of distance measurements and fuzzy logic functions. Known data patterns are stored as organized nodes in a pattern map wherein each organized node is defined by one or more attribute coefficients. As distance measurement is computed between a new data pattern and each organized node of the pattern map using distance measurement wherein the organized node having the smallest distance measurement to the new data pattern receives the highest ranking. Traversing the organized nodes according to the ranking, the new data pattern is compared to each organized node using fuzzy logic functions. If the new data pattern matches an organized node based on an acceptable degree of fuzziness, the attribute coefficients of the organized node are updated to reflect those coefficients of the new data pattern.

Abstract: The fuzzy inference circuit includes a sensor sensing a state of a control object, and a charge coupled device storing fuzzy rule function values for at least one fuzzy rule function. The charge coupled device further performs a fuzzy inference on output of the sensor using the fuzzy rule function values to generate a control signal for controlling the control object.

Abstract: A vehicle route guidance system considers driver preferences, vehicle parameters such as speed and performance capabilities in a navigation computer, and outputs flexible guidance instructions based on these considerations. The system monitors vehicle parameters such as current location and speed with sensors such as an odometer (109), a compass (115), and a GPS receiver (113). The system also determines the location of a vehicle maneuver and then, considering the former attributes with a model free mechanism, in this case a navigation computer with fuzzy inferencing (101), communicates to the driver instructing him how to manoeuver.

Abstract: An adaptive run-time controller is used to adjust the run-time of a fabrication and assembly process simulation model based on inputs generated by outside forces. The run-time controller utilizes fuzzy logic to evaluate the outside forces against existing control rules to determine new simulation run-times based upon this evaluation. The run-time controller then analyzes and adjusts the existing control rules based on current, historical and trend data. The new rules are integrated into the run-time controller to accurately reflect the current status of the fabrication and assembly process for subsequent simulation runs. As such, the run-time controller inputs and evaluates outside forces that were impossible or impracticable to integrate in the simulation model.

Abstract: An intelligent suction device, particularly for vacuum cleaners and the like, includes a fuzzy-logic controller for controlling the motor of a fan or turbine. The suction pressure is detected by at least one pressure sensor which feeds back its measurement, in a closed loop, to the controller.

Abstract: A simplified fault logic accomodation control which utilizes fuzzy logic to assist in calculating a preselected weighted average of the values of two redundant sensors that are sensing the same parameter. A synthesized, third, predicted value of the parameter is used in the fuzzy logic to arbitrate between the two sensed values.

Abstract: An inference rule determining process according to the present invention sequentially determines, using a learning function of a neural network model, a membership function representing a degree which the conditions of the IF part of each inference rule is satisfied when input data is received to thereby obtain an optimal inference result without using experience rules. The inventive inference device uses an inference rule of the type "IF . . . THEN . . ." and includes a membership value determiner (1) which includes all of IF part and has a neural network; individual inference quantity determiners (21)-(2r) which correspond to the respective THEN parts of the inference rules and determine the corresponding inference quantities for the inference rules; and a final inference quantity determiner which determines these inference quantities synthetically to obtain the final results of the inference.