Using fuzzy logic (G06N7/02)

Modified fuzzy rule intelligent controller

Device has a control object, an efficiency coefficient unit, a control neural network self-training rule unit, a system operation history unit, a control neural network, a fuzzification unit, a fuzzy output unit and a defuzzification unit.

Method of registering unit element using fuzzy logic methods

In the method of recording a unit element, readings obtained after signal discretisation are associated with membership functions which characterise "good" reception of logic "1" or logic "0". Further, a procedure for resolving a fuzzy integral is carried out, after which a decision on receiving a binary unit "1" or zero "0" is made. To increase reliability of receiving a unit element on the decision boundary, presence of a zone of uncertain registration, wherein a decision on reception of a "delete" signal is made, is taken into account.

Optoelectronic defuzzification apparatus

Apparatus has a source of coherent radiation, an optical n-output splitter, two transparency filters, n optical Y-splitters, two optical integrators with a definite integration function, two photodiodes, a field-effect transistor with a controlled p-n-junction connected in a common-source circuit.

Optical boundary disjunctor for fuzzy sets

Device has a radiation source, two linear transparency filters, three optical n-output splitters, an optical three-output splitter, a group of n units for calculating the result, each having two optical Y-couplers, two pairs of optically connected waveguides, a photodetector and a piezoelectric element in which a second pair of optically connected waveguide is integrated.

Optical disjunctor for continuous (fuzzy) sets

Device contains m groups of k units for spatial distribution of optical flux, each consisting of a photodetector, a radiation source, an electrooptic deflector, a group of n optical waveguides, a linear transparency filter, a group of n optical j-output splitters and a group of n optical (n-j+1)-input couplers, k groups of n optical m-input couplers, k groups of n intensity normalisation units, each consisting of m-1 pairs of optically connected waveguides, m-1 transparency filters and an optical m-input coupler and k optical n-input couplers.

Optical or gate for fuzzy sets

Device has m groups of k photodetectors, m coherent radiation sources, m 2k-output optical splitters, m groups of k optical amplitude modulators, m groups of k optical phase modulators, m groups of k optical Y-couplers, k minimum signal selectors, k square-root extractors and k subtracters.

Optical and gate for continuous sets

Device has a radiation source, an optical Y-splitter, two optical k×n output splitters, two matrix transparency filters, k groups on n optical Y-couplers, k groups on n pairs of optically connected waveguides, k groups on n transparency filters, k optical n-input couplers.

Optical or gate for continuous sets

Device has a radiation source, an optical Y-splitter, two optical k×n output splitters, two matrix transparency filters, k groups on n optical Y-couplers, k groups on n intensity normalisation units, k optical n-input couplers.

Method of constructing fuzzy logic systems and device for implementing said method

Method of constructing fuzzy logic systems, where a sequence of fuzzy logic rules is created first. From each of these rules, a numerical characteristic is assigned - control quality index, where the fuzzy logic rules are enforced based on a trained neural network. Information signals or signals from the control object are transmitted to the inputs of the neural network. A sequence of output signals or a sequence of instructions and recommendations is formed at the output, where the trained neural network is a large trained artificial neural network. Each of the fuzzy logic rules is enforced by a separate fragment of the large trained artificial neural network (domain), where the number of domains corresponds to the number of fuzzy logic rules and also contains a certain excess number of backup domains. One of the domains functions of an arbitrator and switches outputs of the domains with outputs of the neural network based on the control quality index.

Multipurpose module

Invention concerns field of computer facilities and application in computing systems with parallel processing of the information and high speed can find. The device contains blocks of devices "AND", (n+1)-input adders, the logic block consisting from r of chains from consistently included frequency-modulated generator of start, the high-frequency self-oscillator with the plan of self-clearing which is delivery-radiating systems of the high-frequency self-oscillator.

Neural fuzzy net recognition device

Invention refers to data processing of special application, specifically for nonweighted digital codes driven signal and image transformation to weighed codes, and can be used for signal and image processing and recognition. Device contains two membership function drivers, overall similarity index driver and two fuzzy value shapers.

Device for servicing of various priority calls of computer system users

Device consists of N ≥ 2 users blocks, timing pulse generator, OR element, NAND element, selector-multiplexer, N input NAND elements, priorities encipherer, priorities analyzers, priorities authenticator.