Membership function data preparation method and an apparatus thereof and an adaption degree operation method and an apparatus thereof

Claim 2

An adaptation degree operation device in which a plurality of membership functions are set within defined sections of an input data range, said sections having boundaries located at peaks of said membership functions, said plurality of membership functions being set by stored data in a memory representing said peaks and, for each section, by stored data in said memory defining the gradient of a pair adjacent and overlapping portions of membership functions associated with said section, said pair of overlapping portions of said membership functions within a section being symmetrical with respect to a center of the section, said device comprising: a judging unit for determining a section to which applied input data belongs; an operation unit for finding in said determined section a first degree of adaptation of said applied input data with respect to a first membership function of the section using peak and gradient data stored in said memory for said determined section and associated with said first membership function, and finding a second degree of adaptation of a second membership function of the determined section based on the found first degree of adaptation; and a weighting operation unit for applying weighting data for said corresponding first and second membership function to said found first and second degrees of adaptation.

Claim 3

A machine implemented method for determining a degree of adaptation of an input variable to membership functions comprising the steps of: setting a plurality of membership functions in an input data range, said set membership functions defining sections having boundaries defined by the peaks of said membership functions, portions of adjacent ones of said membership function residing within said sections and having shapes which are symmetrical with respect to a center line of a section, said membership functions being set by storing data defining the peaks thereof in a memory, and storing data representing the gradient of a portion of one of the membership functions residing within each section in said memory; determining a section to which applied input data belongs; and finding in the determined section a first degree of adaptation of said applied input data with respect to a portion of a first membership function within the determined section using stored data in said memory representing the peak of said first membership function and stored data in said memory defining gradient of said portion of said first membership function for said determined section, and finding a second degree of adaptation of said applied input data with respect to a portion of a second membership function within the determined section based on the found first degree of adaptation.

Claim 4

A machine implemented method for determining a degree of adaptation of an input variable to membership functions comprising the steps of: setting a plurality of membership functions in an input data range, said set membership functions defining sections having boundaries defined by the peaks of said membership functions, portions of adjacent ones of said membership function residing within said sections and having shapes which are symmetrical with respect to a center line of a section, said membership functions being set by storing data defining the peaks thereof in a memory, and storing data representing the gradient of a portion of one of the membership functions residing within each section in said memory; determining a section to which applied input data belongs; and finding in the determined section a first degree of adaptation of said applied input data with respect to a portion of a first membership function within the determined section using stored data in said memory representing the peak of said first membership function and stored data in said memory defining gradient of said portion of said first membership function for said determined section, and finding a second degree of adaptation of said applied input data with respect to a portion of a second membership function within the determined section based on the found first degree of adaptation.

Claim 5

A machine implemented method for determining a degree of adaptation of an input variable to stored membership functions, said stored membership functions residing within an input data range divided into sections by the peaks of said membership functions, portions of adjacent ones of said membership functions residing in overlapping fashion within a section, the shapes of the portions of said adjacent membership functions within a section being symmetrical with respect to a center line of a section, said membership functions being stored in a memory in the form of data defining the peaks of the membership functions and data representing the gradient of a portion of one of the overlapping membership functions residing in each section, said method comprising the steps of: determining a section to which applied input data belongs; and finding in the determined section a first degree of adaptation of said applied input data with respect to a first membership function partially within the determined section using stored data in said memory of the peak of said first membership function and stored data in said memory representing the gradient for said determined section, and finding a second degree of adaptation of said applied input data with respect to a second membership function partially within the determined section based on the found first degree of adaptation.

Claim 6

A machine implemented method for determining a degree of adaptation of an input variable to stored membership functions, said stored membership functions residing within an input data range divided into sections by the peaks of said membership functions, portions of adjacent ones of said membership functions residing in overlapping fashion within a section, the shapes of the portions of said adjacent membership functions within a section being symmetrical with respect to a center line of a section, said membership functions being stored in a memory in the form of data defining the peaks of the membership functions and data representing the gradient of a portion of one of the overlapping membership functions residing in each section, said method comprising the steps of: determining a section to which applied input data belongs; and finding in the determined section a first degree of adaptation of said applied input data with respect to a first membership function partially within the determined section using stored data in said memory of the peak of said first membership function and stored data in said memory representing the gradient for said determined section, and finding a second degree of adaptation of said applied input data with respect to a second membership function partially within the determined section based on the found first degree of adaptation.

Claim 7

A method for operating a degree of adaptation according to claim 3, wherein said second degree of adaptation is obtained by subtracting said first degree of adaptation from "1".

Claim 8

A method for operating a degree of adaptation where weight data is applied to each membership function according to claim 20, wherein a weighting operation is executed by applying weighting data of corresponding membership functions to said first and second degrees of adaptation.

Claim 9

An adaptation degree operation device comprising: a memory in which a plurality of membership functions are set to define sections of an input data range, said sections having boundaries at peaks of said overlapping membership functions, shapes of a pair of membership functions associated with a said section being symmetrical with respect to a center of the section, said memory storing data representing the peaks of said membership functions and data representing a gradient of the membership functions for each of said sections; a judging unit for determining a section to which applied input data belongs; an operation unit for finding in said determined section a first degree of adaptation of said applied input data with respect to a first membership function of the section using peak and gradient data stored in said memory for said determined section and associated with said first membership function, and finding a second degree of adaptation of a second membership function of the determined section based on the found first degree of adaptation.

Claim 10

An adaptation degree operation device comprising: a memory in which a plurality of membership functions are set to define sections of an input data range, said sections having boundaries at peaks of said overlapping membership functions, shapes of a pair of membership functions associated with a said section being symmetrical with respect to a center of the section, said memory storing data representing the peaks of said membership functions and data representing a gradient of the membership functions for each of said sections; a judging unit for determining a section to which applied input data belongs; an operation unit for finding in said determined section a first degree of adaptation of said applied input data with respect to a first membership function of the section using peak and gradient data stored in said memory for said determined section and associated with said first membership function, and finding a second degree of adaptation of a second membership function of the determined section based on the found first degree of adaptation.

Claim 11

A device according to claim 6, wherein said operating unit finds said second degree of adaptation by subtracting said first degree of adaptation from "1".