31 results about "Gaussian membership function" patented technology

The invention relates to a self-learning mechanism-base fast matching fuzzy reasoning method. The method includes the following steps that: a Gaussian membership degree function method is adopted to construct parameter fuzzification information; a fuzzy rule base is established; external parameters are fuzzificated, so that a fact item can be obtained; the fact item is matched with rules in the fuzzy rule base by adopting a rete algorithm, so that a fuzzy reasoning result can be obtained; the fuzzy reasoning result is subjected to defuzzification, so that a final reasoning result can be obtained; and a sample set is constructed according to the final reasoning result and an actual feedback result, and rule strength self-learning correction is carried out based on the sample set. According to the self-learning mechanism-base fast matching fuzzy reasoning method of the invention, the rete algorithm is adopted, so that the efficiency of fuzzy reasoning can be improved, and the fuzzy reasoning method can be applied to the engineering field with high real-time requirements.

The invention discloses an acoustic-wave-method gas pipeline leakage monitoring method based on ANN (Artificial Neural Network). The acoustic-wave-method gas pipeline leakage monitoring method comprises the following steps that sound wave signal samples on multiple working conditions of a gas pipeline are acquired, the sound wave signal samples on the different working conditions are denoised and characteristic values of the sound wave signal samples on the different working conditions are extracted; the Gaussian membership function is adopted to conduct fuzzy segmentation on the sound wave signal samples on the different working conditions to acquire fuzzy segmentation amount; the F-self-adaption genetic algorithm is used for optimizing the initial training value of the BP (Back Propagation) neural network, the fuzzy segmentation amount is substituted into the BP neural network for training, and a gas pipeline instant working condition judging BP neural network is acquired; gas pipeline working conditions are judged with the output value of the BP neural network according to the principle of proximity and the time-phased statistical method, and the leakage position is confirmed through the cross-correlation function when leakage occurs. The acoustic-wave-method gas pipeline leakage monitoring method has the advantages that the leakage recognition accuracy rate is increased and the leakage judging reliability degree is increased due to the fact that the principle of proximity and the time-phased statistical method are adopted to determine the working condition states of pipelines.

The invention discloses a comprehensive intelligent method for predicting an impact danger level based on micro-seismic fractal, and is applicable to the field of mine pressure impact prediction in the mine safety field. The method comprises three parts of micro-seismic fractal index input, fuzzy intelligent processing analysis and intelligent prediction result output. The method comprises specific steps of calculating micro-seismic time, space and energy fractal indexes, establishing an impact danger level membership matrix, establishing an index abnormity index and a Gaussian membership function, establishing a single-index evaluation matrix R, and calculating and updating each index weight matrix in real time by adopting an F value scoring method in a confusion matrix; calculating the membership probability of each impact danger level, and determining the final impact danger level by combining a maximum membership degree principle (MMDP) and a variable fuzzy feature recognition model (VFPR). The method is clear in mathematical model, high in universality and operability, capable of achieving intelligent probability prediction of different impact danger levels and intelligent recognition of comprehensive results, good in application feasibility and high in intelligent prediction efficiency.

The invention discloses a fuzzy fusion-based multi-sensor crack damage comprehensive diagnosis method. The method comprises the steps of A, collecting a fiber grating spectrum, a Lamb signal and an intelligent coating resistance value; B, extracting characteristic parameters; C, respectively performing fuzzification processing on characteristic parameters of signals collected by fiber grating, piezoelectric and intelligent coating sensors by use of a Gaussian membership function to obtain three membership functions; D, performing fuzzy fusion by use of a Jaeger algorithm to obtain a comprehensive membership function and a fusion factor omega; E, defuzzifying the comprehensive membership function obtained by fuzzy fusion to obtain fused characteristic parameters; and F, performing relationfitting processing on data obtained by fuzzification processing by use of the fusion factor and fiber grating, piezoelectric and intelligent coating sensor data, and obtaining a relation between the fused characteristic parameters and a damage amount by prediction and verification. According to the fuzzy fusion-based multi-sensor crack damage comprehensive diagnosis method, the problem that information judgment and comprehensive diagnosis cannot be directly performed by use of the multi-sensor data at present can be effectively solved.

The invention discloses a fuzzy logic and evidence reasoning-based transformer insulation stress calculation and evaluation method. The method comprises the following steps of firstly normalizing field data, namely, converting input data for describing a normal value, a limit value and the like of a transformer into dimensionless variables of a [0,1] region; secondly fuzzifying the input data, namely, mapping the input data to a fuzzy membership function, wherein a trapezoidal membership function, a generalized bell-like membership function or a Gaussian membership function can be used; thirdly building a transformer stress evaluation tree model, namely, establishing nodes from the input data of the transformer to an insulation stress of the transformer, and a connection mode of the nodes, and correlating the nodes of upper and lower layers by using a fuzzy correlation matrix; fourthly determining a weight value between the nodes, namely, obtaining a synthesized weight value by using a subjective weight value of an analytic hierarchy process and an objective weight value of eigenvector analysis; and finally obtaining the insulation stress of the transformer, namely, determining a total stress level of the transformer by using a reasoning mechanism and an evidence synthesis rule. The method has the advantages in accuracy, flexibility and capability of processing measurement uncertainty.

The invention provides a robust adaptive fault reconstruction method of a microturbine. The method comprises that a nonlinear part level model of the microturbine is established, a TS fuzzy mathematical method is used to describe a nonlinear system mathematically, and an optimization method is used to optimize a Gaussian-like membership function used in a TS fuzzy system so as to improve the precision of the mathematical model; a state estimation based sliding form observer method is used to reconstruct a fault; and when a nonsingular terminal sliding form observer is designed, linear transformation is introduced into the TS fuzzy system to separate a fault signal from an interference signal, applied interference and a fault signal whose upper bounds of the change rate are not known can be reconstructed simultaneously, and an adaptive law is introduced to update the sliding form gain in real time to eliminate influence of the fault and interference with unknown upper bound of the change rate on sliding form motion. According to experimental results, the method is robust to the fault and interference with unknown upper bound of the change rate, and the unknown fault can be reconstructed in extremely short time.

The invention presents a fuzzy basis function network (FBFN) processing device for the beam pointing error correction of the local multipoint distributed system (LMDS). The beam pointing error caused by wind force can affect the performance of the local multipoint distributed system (LMDS). The correction device uses multi-beam planar array antenna to obtain the signal direction-of-arrival (DOA) of a base station and uses fuzzy basis function network (FBFN) algorithm, which includes thirteen normalized Gaussian membership functions and thirteen rules, to estimate the beam pointing error. The simulation results show that the presented FBFN processing device has better performance in transient response, convergence time and steady state value of the averaged square error than the conventional beam pointing error correction devices.

The invention relates to a time sequence remote sensing image crop classification method combining a TWDTW algorithm and a fuzzy set. The method comprises the following steps: S1, obtaining time sequence remote sensing image data, plot data and crop sample data of a to-be-detected region; S2, preprocessing the time sequence remote sensing image data; S3, constructing an NDVI time sequence data set; S4, respectively constructing standard NDVI time sequence data of different crops and an NDVI time sequence data set of the plot units; S5, constructing a TWDTW algorithm of a non-equal-length time sequence, and obtaining a minimum cumulative distance feature matched with the similarities of different crops; S6, on the basis of the NDVI time sequence data set of the plot unit, phenological characteristics of different crop growth season lengths are calculated; and S7, on the basis of the minimum cumulative distance feature and the growth season length feature, constructing Gaussian membership functions of different crops, and on the basis of a fuzzy set classification rule, realizing refined crop classification on the plot scale. According to the invention, refined classification of crops on the plot scale is realized.

The invention discloses a hybrid electric vehicle power battery initial electric quantity algorithm, comprising: ANFIS construction, ANFIS network structure determination, selection of a Gaussian membership function as an input and output membership function, division of an input variable space, and calculation of battery initial electric quantity Q1 by using a BP neural network and an ANFIS model. According to the scheme, the algorithm can monitor the electric quantity of the power battery in real time. The service life of the power battery is prevented from being influenced by overcharging or overdischarging of the power battery. Through analyzing the charging and discharging process of the battery is analyzed, key parameters of SOC is determined, the test model is corrected on the MatLab platform. Comparison by experimental simulation shows that ANFIS has good adaptive ability and generalization ability. The initial electric quantity estimation error of the battery is reduced to belower than 3%. The method can be used for an intelligent monitoring system of the hybrid electric vehicle, a temperature compensation coefficient is added, the temperature influence is corrected through the temperature compensation coefficient, and the estimation precision can be better improved.

The invention discloses a human body-behind-wall multi-state target detection method based on fuzzy pattern recognition and a genetic algorithm. The human body-behind-wall multi-state target detection method comprises processing a received signal of a P410 radar, and extracting a characteristic parameter of the received signal; and forming a membership function set by means of the extracted characteristic parameter and multiple states behind a wall. A gaussian function is selected as a sub-membership function, the mean value and variance in the sub-membership function are optimized by means of the genetic algorithm, and the membership function set is constructed; a human body-behind-wall target prediction function is established in dependence on a fuzzy pattern recognition theory, and through calculation, which kind of state the detected data belongs to can be obviously recognized on the basis of a maximum membership degree principle. The human body-behind-wall multi-state target detection method based on the fuzzy pattern recognition and the genetic algorithm is mainly applied to the disaster rescue field and the anti-terrorism criminal investigation field in order to guarantee that a target under which a living body is buried is detected and rescued and the personal safety of a hostage is guaranteed when the hostage is seized in the anti-terrorism action.

The invention discloses a driving state detecting device and method based on reverse binoculus. Determination of a current driving state of a driver is conducted according to the yaw rate of a vehicleand the postures of the head of the driver. The method comprises the steps that firstly, detection and recognition of traffic lanes are conducted according to a Hough algorithm to calculate the yaw rate of the vehicle; meanwhile estimation on the postures of the head of the driver is conducted by adopting a multi-point perspective algorithm; and then fuzzy judgement rules based on a Gause membership function are established, recognition of the driving state of the driver is conducted according to the yaw rate of the vehicle and the postures of the head of the driver, and a complete driver state detection model is established.

The invention relates to a novel data-driven model predictive control method, which comprises the following steps of: giving correlation weight between each input variable and output from input variables obtained from a wind power plant by adopting a two-stage fuzzy curve method, and quickly selecting important input variables according to input variable indexes; then, determining prerequisite parameters of a fuzzy model by adopting fuzzy clustering and a Gaussian membership function, and identifying conclusion parameters of the fuzzy model by adopting recursive least squares; and the identified model is used for participating in data-driven model prediction control, so that the wind power plant can smoothly perform primary frequency modulation, the MPC problem for wind power plant frequency control is completed, and a relatively accurate solution can be quickly obtained.

The invention relates to a cold-rolling mill second flow thickness control method and device based on fuzzy control. The method comprises the steps that the thickness of a strip steel outlet is pre-calculated according to a second flow equation, and an outlet thickness gauge is used for measuring the thickness and correcting the thickness to obtain an outlet thickness difference; the outlet thickness difference is divided into a plurality of fuzzy grades, and a subordinating degree value of the outlet thickness difference is obtained; and a fuzzy rule is determined, proportional and integral control output control amounts are calculated according to the fuzzy rule and the membership degree value, and the total output control amount is calculated according to the proportional and integral control output control amounts. According to the cold-rolling mill second flow thickness control method and device, the thickness of the strip steel outlet is firstly pre-calculated according to the second flow equation, and the outlet thickness gauge is used for measuring the thickness and correcting the thickness to obtain the outlet thickness difference; then the outlet thickness difference is divided into the multiple fuzzy grades, and a Gaussian subordinating degree function is used for obtaining the subordinating degree value; the fuzzy rule and the adaptive coefficient are set, and the control amounts of proportional control and integral control are respectively obtained; and finally the second flow AGC control amount is synthesized.