49 results about "Fuzzy inference engine" patented technology

A Soft Computing (SC) optimizer for designing a Knowledge Base (KB) to be used in a control system for controlling a motorcycle is described. In one embodiment, a simulation model of the motorcycle and rider control is used. In one embodiment, the simulation model includes a feedforward rider model. The SC optimizer includes a fuzzy inference engine based on a Fuzzy Neural Network (FNN). The SC Optimizer provides Fuzzy Inference System (FIS) structure selection, FIS structure optimization method selection, and teaching signal selection and generation. The user selects a fuzzy model, including one or more of: the number of input and/or output variables; the type of fuzzy inference; and the preliminary type of membership functions. A Genetic Algorithm (GA) is used to optimize linguistic variable parameters and the input-output training patterns. A GA is also used to optimize the rule base, using the fuzzy model, optimal linguistic variable parameters, and a teaching signal. The GA produces a near-optimal FNN. The near-optimal FNN can be improved using classical derivative-based optimization procedures. The FIS structure found by the GA is optimized with a fitness function based on a response of the actual plant model of the controlled plant. The SC optimizer produces a robust KB that is typically smaller that the KB produced by prior art methods.

A Soft Computing (SC) optimizer for designing a Knowledge Base (KB) to be used in a control system for controlling a suspension system is described. The SC optimizer includes a fuzzy inference engine based on a Fuzzy Neural Network (FNN). The SC Optimizer provides Fuzzy Inference System (FIS) structure selection, FIS structure optimization method selection, and teaching signal selection and generation. The user selects a fuzzy model, including one or more of: the number of input and/or output variables; the type of fuzzy inference model (e.g., Mamdani, Sugeno, Tsukamoto, etc.); and the preliminary type of membership functions. A Genetic Algorithm (GA) is used to optimize linguistic variable parameters and the input-output training patterns. A GA is also used to optimize the rule base, using the fuzzy model, optimal linguistic variable parameters, and a teaching signal. The GA produces a near-optimal FNN. The near-optimal FNN can be improved using classical derivative-based optimization procedures. The FIS structure found by the GA is optimized with a fitness function based on a response of the actual suspension system model of the controlled suspension system. The SC optimizer produces a robust KB that is typically smaller that the KB produced by prior art methods.

The invention relates to an improved storage battery SOC (State Of Charge) estimation method based on the consistency of a unit cell. In the method, the training data set of a storage battery, SOC predicted data corrected on the basis of consistency, and a self-adoption neural network are use for building a fuzzy rule base with an optimized structure and parameters; and then the fuzzy rule base after off-line learning is implanted into the fuzzy inference machine of an embedded controller of BMS (Battery Management System), so as to carry out on-line correction to storage battery SOC estimation. Compared with the prior art, the invention has the advantage that SOC differences in all unit cells inside the battery can be factually reflected on entire SOC estimation.

The invention provides a proportional amplifier PID parameter self-tuning control method comprising the following steps: a real-time parameter value of an output end of the proportional amplifier is detected, and deviation and deviation change rate are calculated according to the real-time parameter value and a given parameter value; the deviation and the deviation change rate are fuzzified into a corresponding deviation fuzzy subset and a deviation change rate fuzzy subset; a PID parameter fuzzy subset is obtained according to the deviation fuzzy subset, the deviation change rate fuzzy subset and a PID parameter fuzzy inference mechanism; after the PID parameter fuzzy subset is de-fuzzified, a PID control parameter is obtained according to a PID parameter initial value and the de-fuzzified change value; output of the proportional amplifier is controlled according to the PID control parameter, the real-time parameter value of the output end of the proportional amplifier is detected again, and the steps are repeated until the real-time parameter value of the output end of the proportional amplifier is equal to the given parameter value. PID parameters are controlled in fuzzy and self-tuning ways so that the proportional amplifier can be self-adaptive to various application occasions. The invention also provides one type of proportional amplifier and a proportional electromagnetic valve.

The invention relates to a power plant fan fault early warning system based on fuzzy reasoning. The system comprises a fan real-time online monitoring module which comprises a fan real-time data acquisition unit, a signal processing unit and a fault feature extraction unit; a fan operation state knowledge base module which comprises a fault rule base unit, a static database unit and a dynamic database unit; and a fuzzy inference engine module which comprises a fuzzy inference unit, a forward inference unit, a reverse inference unit; and an early warning threshold determination module, a fuzzyknowledge acquisition module, an early warning interpretation mechanism module and a human-computer interface interaction module. Compared with the prior art, the problem that early warning is difficult due to strong coupling, nonlinearity, variable working conditions, various fault types and the like of a power plant fan in the traditional sense is solved, real-time monitoring of the state of thefan is achieved, the fault type of the fan can be judged, fault occurrence points can be traced back, and early warning reasons can be analyzed.

The invention discloses a multi-parameter monitoring device and method suitable for the abnormal state of a mechanical device. The device comprises a data analysis and fault diagnosis server, and a plurality of early warning monitoring terminals and at least two data information acquisition terminals which are in signal connection with the data analysis and fault diagnosis server. During monitoring, the data analysis and fault diagnosis server reads various sensing data, preprocesses the sensing data, performs abnormal vibration analysis and mechanical fault diagnosis on the preprocessed vibration data, obtains a sensing data fusion estimation result by adopting a fuzzy reasoning mechanism, and outputs a result of the overall operation state of the mechanical device. According to the monitoring method, an abnormal vibration analysis and mechanical fault diagnosis method is adopted to judge the abnormal vibration type, the fuzzy reasoning is conducted on the mechanical device operationstate according to a fuzzy reasoning rule knowledge base, and the overall operation state of mechanical device is evaluated, so that the long-term online unattended abnormal state monitoring and faultearly warning of multiple devices are realized.

The invention discloses a roaming switching control method and device and electronic equipment, and the method comprises the steps: carrying out the sampling according to a preset sampling time interval, and obtaining the current state information of a network system where a current wireless AP connected with a to-be-roamed client is located; inputting the current state information into a preset fuzzy inference system, and according to a preset fuzzy inference mechanism in the fuzzy inference system and a rule set containing a plurality of fuzzy logic rules, performing inference to obtain roaming quality determined by downlink signal intensity corresponding to each wireless AP and the load quantity of the client; determining the optimal roaming quality from all the roaming qualities, and taking the wireless AP corresponding to the optimal roaming quality as a target wireless AP; and controlling the to-be-roamed client to roam from the current wireless AP to the target wireless AP. According to the invention, the roaming switching control can be carried out by adopting the rule set suitable for the wireless network environment, and the most suitable target wireless AP can be accurately determined.

The invention discloses a weld tracking application adaptive fuzzy control method comprising the following steps: 1, fuzzifying an obtained discrepancy e and a deviation change rate (e<.>=de/dt) so as to obtain a fuzzy variable deviation E and a deviation change rate EC; 2, combining with expert knowledge and operation experiences, and building fuzzy rules according to the deviation e and deviation change rate ec; 3, inputting the inputted fuzzy variable deviation E and the deviation change rate EC into a fuzzy inference engine so as to obtain a fuzzy output value U; 4, using a maximum value method to defuzzify, and converting the fuzzy output value U into an accurate value u; 5, introducing a controller scale factor ki (i=1, 2,..n) and a controller quantification factor ku, and setting input/output fuzzy universe and a fuzzy rule function; 6, real time updating the fuzzy rule function according to the input/output value changes, and calculating and outputting a offset voltage value of a controlled object. The method can adjust the fuzzy controller parameters in a control process, thus solving the problems that the control process is unstable, poor in dynamic performance and rough in the process.

The invention discloses a high-conflict evidence fusion method based on fuzzy reasoning, relates to a road condition assessment method based on an automobile physical information fusion system, and belongs to the technical field of multi-sensor data fusion. The method is based on the VCPS technology. The method comprises the steps of acquiring real-time operation information of an automobile, forming evidence for judging the current road condition by multiple automobiles, calculating support probability distribution of an evidence source through a support probability function, further obtaining a maximum distance and an average distance of probability distribution between evidences, and performing effective measurement on a conflict degree between the evidences through a fuzzy reasoning mechanism. And on the basis, the support degree and the credibility of the evidence can be obtained. The method also considers the uncertain information of the evidence at the same time, and calculatesthe relative importance of the evidence through Dane entropy. And the weight of the evidence is obtained by combining the credibility and the importance, and finally weighted average is performed on the evidence to obtain an average evidence. And finally, the average evidence is integrated for multiple times by utilizing a DS combination rule to obtain a reliable assessment evidence.

The present invention discloses a landslide hazard degree evaluation method under support of a GIS (Geographic Information System) and an artificial intelligence technology. The method comprises: acquiring knowledge about a relation between a landslide and a geographical environment by interviewing a landslide expert, and screening geographical environment factors according to conditions, such as a degree of influence of the geographical environment factors on the landslide, data source quality and an acquisition difficulty degree, in expert knowledge; extracting quantitative geographical environment factors by use of the GIS; expressing the expert knowledge about the relation between the landslide and the geographical environment factors by use of a fuzzy membership degree function, and constructing a fuzzy inference engine capable of calculating a landslide hazard degree value based on the geographical environment factors; and inputting the quantitative geographical environment factors into the fuzzy inference engine, and calculating spatial distribution of a regional landslide hazard degree. The method disclosed by the prevent invention effectively overcomes the defects such as high sample quality and quantity requirements, and poor interpretability and poor portability of a statistical method, improves precision and the detail level of landslide hazard degree spatial distribution speculation, and has extensive application prospects in the aspects of regional landslide hazard degree analysis and landslide disaster prevention and reduction.

The invention belongs to the field of fuzzy control, and particularly relates to a single intersection traffic signal fuzzy controller which comprises a knowledge base (A*oR), a fuzzy inference engine (oR), quantification factors (Kx, Ky and Kt), a fuzzy input module (D/F) and a clearness output module (F/D), two-input signals are traffic flows x1 and y1 in an annular coil vehicle detector acquisition road network, a single-output single is real-time phase green light extension time u, the traffic flows x1 and y1 are converted into fuzzy input variables X and Y through the fuzzy input module (D/F), the fuzzy input variables X and Y are inputted to the fuzzy inference engine (oR) and ratiocinated and analyzed by the aid of the knowledge base (A*oR) to obtain a fuzzy control green light extension time output variable U, and the real-time phase green light extension time u can be acquired through the clearness output module (F/D). Current phase traffic flow information can be fuzzily controlled, a green light signal is delayed when the traffic flow is large, so that the traffic flow is evacuated, and smooth traffic is realized.

The invention discloses an adaptive color image steganography method based on a fuzzy inference system, which comprises the following steps: step 1, separating color channels of a color carrier image,and extracting features of each color channel of the carrier image; 2, determining a membership function, giving a fuzzy rule, taking the features extracted in the step 1 as input, obtaining a fuzzyoutput through a fuzzifier and a fuzzy inference engine, and obtaining a clear output value through a defuzzifier; and step 3, judging each pixel type, estimating the embedding capacity of each pixeltype according to the complexity of the color channel, encrypting information to be hidden, adaptively embedding secret information with different digits for different pixel types by utilizing an LSBalgorithm, and combining image color channels to obtain a secret-loaded image. According to the method, the color space complexity of the color image is fully considered, different capacities are adaptively embedded according to the texture complexity and visual characteristics of the image, and the image embedding capacity, imperceptibility and secret information security can be improved.

The invention discloses an energy-saving fuzzy controller method for a synchronous motor, which comprises a fuzzy interface, a knowledge base, a fuzzy inference machine and a defuzzy interface. The method comprises the following steps of step 1, constructing a fuzzy controller in a Matlab simulation system; Step 2, combining the fuzzy controller constructed in the step 1 with a traditional PID controller to form a fuzzy PID controller. According to the invention, a composite controller is designed to improve the speed response, so that the system is good in dynamic and static performance. Therequirements of a high-precision system are met.

The invention belongs to the technical field of motor control methods, and particularly relates to a novel electric loading system control method. For solving the problem that loading accuracy and stability of an existing intelligent electric loading system control method are difficult to ensure, the compound control method based on fuzzy CMAC and an indistinct logic computer is provided. The control method includes the steps that in fuzzy CMAC control, a loading instruction signal is mapped to all units in space, then membership degrees of an input signal to all fuzzy sets serve as weight information of corresponding weight values, and weighting summation is carried out on all the weight values to obtain preset output approximate values. Meanwhile, activated input fuzzy sets in the indistinct logic computer are mapped into all output fuzzy sets, and an improved high defuzzifier is used for calculation to reach increment needed by preset output; finally the two are added to obtain a final output signal. The electric loading system control method based on fuzzy CMAC and the indistinct logic computer can improve system control precision and ensures system stability.

The invention provides a method for verifying the integrity of investigation information. The method includes the steps that investigation information is read from a database; the investigation information is preprocessed based on a preprocessing rule base, and feature extraction is carried out to obtain feature vectors; the obtained feature vectors serve as input vectors of a fuzzy inference machine, and a feature-level integrity verification result of the feature vectors is obtained; comprehensive decision-making processing is carried out on the feature-level integrity verification result, and an investigation information integrity verification result is obtained. The method suitable for verifying the integrity of the investigation information is designed, and the investigation information is preprocessed three times in a progressive mode; besides, the processing process can be backtracked, so that processing rules are easy to modify, and the investigation information processing efficiency and processing precision are improved; moreover, integrity verification is carried out based on the feature level first, and then integrity verification is carried out after feature-level verification is modified according to conflicts and the Euclidean distance, so that the precision of investigation information integrity verification is improved, and the processing speed of investigation information integrity verification is improved.

The invention discloses a solid-liquid phase change material surface heat flow measurement method and system based on a dispersion fuzzy reasoning mechanism. The method comprises the steps of: S1, building a solid-liquid phase change material internal radiation heat conduction coupling heat exchange model, and giving an initial guess value of material surface heat flow distribution; S2, selecting M temperature measuring points, and recording temperature information; S3, calculating the temperatures of the M measuring point positions along with the time change; S4, establishing an input signal of surface heat flow measurement according to the measured values and the calculated values of the temperatures of the M measuring points; S5, constructing a fuzzy reasoning unit for measuring the surface heat flow of the solid-liquid phase change material, and obtaining a reasoning result; S6, constructing a dispersive fuzzy reasoning module based on the fuzzy reasoning unit and a comprehensive weighting coefficient; and S7, updating the heat flow distribution according to an output result of the dispersive fuzzy reasoning module, and checking iteration. The system is used for executing the method. The method and system have the advantages of simple principle, easiness in implementation, simplicity and convenience in operation, high measurement precision and the like.