81 results about "Fuzzy inference rules" patented technology

The invention relates to a PMSM servo system control method based on fuzzy and active disturbance rejection control. A position signal is given by a differential tracker to arrange a transition process so that the contradiction between rapidness and overshoot of a system is solved, and uncertainty, friction torques and external disturbance due to modeling errors of the system are observed via an expansion state observer; according to the error between the differentials generated by the differential tracker and state variation generated by the expansion state observer, a fuzzy inference rule is obtained with application of experimental experience of technical staff, so that a fuzzy rule control table of an error proportion coefficient, a differential coefficient and an integration coefficient is established; accurate control amount is obtained after de-fuzzification, so that parameter self-adaptive adjustment of a nonlinear error feedback control law is realized; and compensation amounts of the nonlinear error feedback control law and the expansion state observer to the total disturbance forms the control amount, thereby realizing optimal control for an controlled object. The method of the invention improves both tracking precision and disturbance rejection capability of the system.

The invention discloses a GIS fault diagnosis and reliability analysis method based on fuzzy Petri. According to the method, a GIS fault modeling and inference method of a fuzzy Petri net is put forward on the basis of the randomness and diversity of GIS faults and the fuzziness of fault symptoms, the theory of the Petri net and the fuzzy inference rules are organically combined together, and the high-efficiency and accurate parallel inference capacity is achieved. Through the combination of lots of statistics fault cases, the fault diagnosis model of a GIS based on the FPN is established. When the system does not fail, matrix forward reasoning is conducted on the fault state of the GIS by online monitoring and predicting the possible fault symptoms, the credibility of all places can be rapidly and accurately calculated, and the importance degree of the initial place is analyzed; fault reasons are found through backward fuzzy inference rules under the known fault phenomenon.

The invention belongs to the technical field of network security, particularly relates to an attribute-based fuzzy access control calculation method. The method adopts a two-stage weighting method to assign different weights to an object and attributes thereof, then performs comprehensive judgment based on a fuzzy judgment matrix and a weighted average model, calculates a degree of satisfaction of an access request to an authorization policy, and finally obtains an limits of authority policy set based on fuzzy inference rules. Based on different weights of information of three attributes of a main body, resource and environment, the attribute-based fuzzy access control calculation method performs fuzzy processing on a policy decision-making process of access control, obtains the fuzzy authorization policy set, can deal with the problem of access control authorization under the circumstance that access request information partially satisfies judgment conditions, and is of great significance to improving completeness of access control.

The invention relates to a power supply technology, in particular to a method for determining the charge state of a vehicle-mounted storage battery. According to the invention, the method for determining the charge state of the vehicle-mounted storage battery comprises the following steps of: inputting the working voltage, the working current and the working temperature of the storage battery; calculating a corrected working voltage value of the working voltage under the standard working current of the storage battery; determining the corrected working voltage value and a fuzzy value of the working temperature by utilizing respective membership functions; determining a fuzzy value of the charge state of the storage battery by utilizing a fuzzy inference rule according to the corrected working voltage value and the fuzzy value of the working temperature; and calculating the charge state of the storage battery by utilizing an anti-fuzzy algorithm according to the fuzzy value of the charge state of the storage battery. According to the embodiment of the invention, test curves and empirical data can be fully utilized through introducing a fuzzy logic when a SOC (Stress Optical Coefficient) is estimated, so that the result accuracy is improved.

The invention discloses a dissolved oxygen control method based on a fuzzy neural network. The method comprises the following steps: S1, modeling is carried out; S2, fuzzy identification is carried out, and a fuzzy identification method is adopted to identify the structure and parameters of an object model; S3, a clustering method is adopted to determine the number of rules; S4, a neural network is used for learning reasoning data of an expert, a fuzzy reasoning rule is acquired, and each connection weight is adjusted at the same time; S5, after a least square method is used for order identification, a transitive closure method is used for clustering analysis according to a fuzzy similarity relation; S6, a Smith predictor is introduced; and S7, two independent control circuits of blower pressure and oxygen dissolution are adopted. The wastewater treatment effects are good.

The invention belongs to the technical field of greenhouse intelligent regulation and control, and relates to a greenhouse intelligent regulation and control method based on agricultural solar term experience data. According to the method, by importing the agricultural solar term experience data and using a fuzzy neural network strategy as the basis, a fuzzy neural network controller in which a greenhouse environmental factor is coupled with a greenhouse regulation and control mode is constructed, the control precision of the fuzzy neural network controller is improved on the basis by using a neural network online backpropagation learning algorithm, auxiliary optimization is performed on a topological structure, a connecting weight, a membership function parameter or a fuzzy inference rule of the fuzzy neural network controller by using a genetic optimization algorithm, so that a genetically optimized fuzzy neural network controller is formed. According to the greenhouse intelligent regulation and control method provided by the invention, the fuzzy neural network controller is constructed based on the agricultural solar term experience data, a new idea is provided for the greenhouse regulation and control method, the guiding role of the agricultural solar term experience data on agricultural production is fully exerted, and meanwhile, the agricultural production cost is reduced.

The invention discloses a fuzzy active disturbance rejection control method of a permanent magnet synchronous motor used for industrial robots. The method comprises the following steps: collecting the rotor position signals of a motor servo system, then using the rotor position signals as the feedback signals of an active disturbance rejection controller, using a fuzzy inference rule, establishing a fuzzy rule control table, and realizing high precision control for the motor servo system through automatic adjustment of parameters and compensation for system disturbance. A traditional vector-controlled PI adjustment method is replaced by using the fuzzy active disturbance rejection control method, the rapidity and stability of a driving system of the permanent magnet synchronous motor used for industrial robots are improved, and the precision and disturbance rejection capacity of the system are improved.

The invention provides an FHM-based cold-rolling mill working roll emulsion sectional cooling control method. The method comprises determining the amount of nozzles covering strip steel and sequentially numbering the controllable nozzles; establishing fuzzy inference rules of an FHM-based cold-rolling mill working roll emulsion sectional cooling system; defining generalized input variables for the fuzzy inference rules and obtaining the fuzzy membership functions of fuzzy sets corresponding to the generalized input variables; by means of the obtained fuzzy inference rules, the obtained generalized input variables and the obtained fuzzy membership functions of the fuzzy sets, calculating out the spraying flow adjusting amount of an i-th nozzle at a current time to obtain an emulsion spraying flow adjusting amount calculation model of a controllable nozzle to achieve the flow control valve online control of the sectional cooling controllable nozzle. According to the FHM-based cold-rolling mill working roll emulsion sectional cooling control method, by establishing the physical relationship between plate shape deviation and the emulsion spraying flow adjusting amount of the controllable nozzle at the current time, the emulsion spraying flow adjusting amount calculation model of the controllable nozzle can be obtained and meanwhile the precise online control of cold-rolling mill working roll emulsion sectional cooling can be achieved.

The invention discloses a method for controlling an oxidation potential water generator by establishing a mathematical model. Through a large number of experiments, this method obtains the data relationship between pH and available chlorine, pulse salt pump frequency, electrolysis voltage, electrolysis current, water inflow flow, ORP and temperature, and establishes The fuzzy inference rules between pH and available chlorine and other factors, and then according to the deviation between ORP, pH and available chlorine and their respective reference values, adjust the pulse salt pump frequency, electrolysis voltage and current through the controller, so as to ensure the water output index of stability.

The invention provides a system and a method for optimizing a cutting process, relating to the technical field of the cutting process. The system comprises an evaluation index selector, a data quantitative processor, a sample classifier, a process scheme evaluator and a process scheme selector. The method for realizing the optimization for the cutting process by adopting the system and on the basis of a particle swarm classifying and self-adaption nerve fuzzy inference system comprises the following steps: generating a training sample set by adopting a particle swarm optimization algorithm according to the collected process index data; establishing a self-adaption nerve fuzzy inference system model by learning the training sample set; autonomously generating a fuzzy inference rule, thereby acquiring an evaluation value and confirming the optimal cutting process scheme. According to the method provided by the invention, the intelligent evaluation and the prior evaluation can be realized; the evaluation process is simple and clear; through the method and the system, different cutting process schemes of a same part can be compared and then the optimal process route can be confirmed; under the condition of guaranteeing the emission reduction and environment protection, the cutting process can be more effectively realized.

The embodiment of the invention provides a satellite network switching method and device. The method comprises the following steps: determining at least one parameter and a target fuzzy logic switching model corresponding to the at least one parameter; acquiring a user terminal and covering at least one input of at least one satellite of the user terminal; performing fuzzy processing on the at least one input according to at least one subordinating degree function corresponding to the at least one input and a preset linguistic fuzzy set of the target fuzzy logic switching model so as to obtain a fuzzy reasoning result corresponding to each input of each satellite; correspondingly determining a switching factor of each satellite in the at least one satellite according to a fuzzy reasoning rule of the target fuzzy logic switching model; and selecting at least one to-be-switched satellite in the at least one satellite greater than the preset duration and the biggest switching factor in the corresponding switching factors, determining the to-be-switched satellite corresponding to the switching factor as a target satellite, and switching the user terminal to the target satellite.

The invention relates to a speed control method for car-following motions, which is characterized in that the acceleration lva is acquired by means of three times of the relative distances between a front car and a following car, the speeds of both the front and following cars, and the time interval Delta t in which the relative distances between the two cars, and used as a second input variable. Ds and lva act as two input variables of a fuzzy controller, and the acceleration of the following car is used as the output variable fva of the fuzzy controller; ds, lva and fva are respectively divided into seven fuzzy word sets, namely, minus large, minus middle, minus small, zero, plus small, plus middle, and plus large, which are respectively represented as nb, nm, ns, ze, ps, pm, and pb; the membership functions of the seven fuzzy word sets adopt the common triangle functions. The fuzzy inference rules of fuzzy controller adopt the 49 rules inducted in a rule table; the fuzzy set of the variable fva is calculated by use of the maximum-minimum approach; finally, the method of mass center is adopted to deblur the fuzzy sets of the output variable fva, thus obtaining the acceleration of the following car.

The invention discloses a coach fire detection, extinguishing and pre-warning system and a method thereof. A lower computer device is arranged on each cabin of a coach; parameter information in the cabins is monitored by the lower computer devices and fire pre-warning grades are generated on the basis of a fuzzy inference rule table; the lower computer devices are in a communication connection with an upper computer by utilizing buses; when the fire pre-warning grade in any cabin reaches the alarm threshold value, the lower computer device in the cabin transfers the pre-warning grade, space information and fault parameter information to the upper computer device; and the upper computer device starts an alarm or fire extinguishing equipment to realize active fire control. The all-dimensional fire monitoring of the whole coach can be realized, the identification scope of fire faults of coaches can be expanded, people' safety can be ensured and the property loss can be reduced. Meanwhile, the fire is judged by the multi-information fusion technology of the fuzzy theory, so that the uncertainity and limitation of fire judgment with a single parameter is avoided and the fire early-warning accuracy of the coaches is improved.

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 relates to the technical field of unmanned aerial vehicles, particularly to a self-adaptive fuzzy active disturbance rejection control method of a variable-load quadrotor unmanned aerialvehicle. The method comprises the following steps: establishing a variable-load quadrotor unmanned aerial vehicle system model; designing a self-adaptive control system for the variable-load quadrotor unmanned aerial vehicle system model based on a self-adaptive variable-load compensation method; and introducing a fuzzy control method into the self-adaptive control system, and optimizing the self-adaptive control system to obtain the fuzzy active disturbance rejection control system. According to the invention, the adaptive control system provided by the invention can quickly and accurately track and estimate the load variation, so that the external disturbance compensation coefficient of the active disturbance rejection controller can be corrected according to the dynamic change of the load, and the robustness and stability of load disturbance resistance of a nonlinear time-varying unmanned aerial vehicle system are enhanced; and fuzzy control is applied to the active disturbance rejection controller, online self-tuning adjustment is carried out on control parameters by formulating a reasonable fuzzy reasoning rule, and the response speed of the system can be effectively increased.

The invention discloses a method and device for programming the track of a tour inspection robot, and belongs to the technical field of robot motion control. The method comprises the steps that 1) the position error E between a driving track and a magnetic track of the tour inspection robot and the change rate EC of the position error are obtained; 2) a fuzzy input is provided for a fuzzy controller which is set in advance; 3) the fuzzy controller obtains a fuzzy output U according to the preset fuzzy interference rule and program; 4) voting is carried out via the preselected membership function of the fuzzy output and the gravity center of the area enclosed by abscissa, and a clear output is obtained; and 5) the obtained clear output is converted into a practical output in a linear scale transformation method, and the practical output is transmitted to a driving motor of the tour inspection robot to adjust the rotation angle of the robot. Compared with the prior art, the method and device have the advantages that adjustment is appropriate and swinging tends not to occur in the operation process of the robot.

The present invention discloses a hierarchy fuzzy system based on a unified model. First, a hierarchy fuzzy system with an increasing shape structure is established, namely, each basic hierarchy unit only has two physical input quantities collected from an industry production field. Through defining a uniform universe of discourse interval, the variation range of the actual physical input quantity is analyzed, the input quality is mapped uniformly on the defined interval, namely, the interval is used as the basic universe of discourse of the fuzzy system; each basic fuzzy unit adopts the same fuzzy reasoning way, namely, the same universe of discourse, the same fuzzy way and the uniform fuzzy reasoning rule, the fuzzy reasoning adopts the Mamdani composition arithmetic which is simple and convenient to calculate, based on which, a vacancy replacement processing mode is adopted for the unsure physical input quantity, the output weight sum of each layer of fuzzy basic unit is adopted for system output, and the feedback and adjusting for the system result is realized through an interval mapping adjusting way. The present invention solves the actual problems of the complexity of design and the real-time property of operation, thereby guaranteeing the practicability of the hierarchy fuzzy system.

The invention discloses a driving intention identifying method suitable for multi-performance comprehensive optimization of a gear shift schedule, and belongs to the technical field of step type automatic transmissions of cars. The driving intention identifying method comprises the following steps of: adopting a dynamic property and economical efficiency expectation fuzzy inferer to perform quantization on dynamic property and economical efficiency expectation of a driver separately, and determining emission performance expectation on this basis; respectively determining input parameters and output parameters of the fuzzy inferer according to the construction of the dynamic property and economical efficiency expectation fuzzy inferer, and determining subordinate functions of the input parameters and the output parameters; and finally, establishing a dynamic property and economical efficiency fuzzy inference rule base, and taking each performance expectation quantized value as a weighting coefficient to construct a gear shift performance comprehensive evaluative function which is used as a gear shift rule multi-performance index comprehensive optimization target function. On the basis of reflecting the driving intention of the driver, the driving intention identifying method enables dynamic property, economical efficiency and emission comprehensive performance of a whole vehicle to be optimal, and also enables automatic gear shift to be more personalized, so that the defect of simply sorting driving intentions in the prior art is made up.

The invention discloses a control system and method of floor heating temperature based on a particle swarm optimization algorithm. The method comprises the following steps of: establishing a fuzzy reasoner by using the particle swarm optimization algorithm; calculating the temperature deviation and the variation rate of the temperature deviation after collecting the temperature in the room according to the preset temperature and the collected temperature; calculating parameters of a PID controller according to the temperature deviation, variation rate of the temperature deviation, and fuzzy reasoner; controlling the heating time of a regulating valve by the PID controller according to the temperature deviation and the parameters of the PID controller. The control system and method of floorheating temperature adopts the particle swarm optimization algorithm to establish the fuzzy reasoner. Therefore, the obtained fuzzy reasoning table is more accurate. The control system can better reflect the relationship between temperature changes and control time in different rooms. Therefore, the control system can improve the accuracy of the floor heating temperature control, and make the user more comfortable and more energy efficient. At the same time, the control system reduces the difficulty of parameters adjustment, and reduces the influence of subjective factors. The control systemcan be widely used in the field of floor heating temperature control.

The invention discloses a method of micro grid energy scheduling based on dynamic adaptive fuzzy logic. The method includes the steps of taking one-day prediction information of a micro grid as a training set for generating an optimal fuzzy controller; taking the obtained prediction information as an analog input, and with the dual optimization goal of minimizing the operation cost of the micro grid and minimizing the power fluctuation of a public grid, performing dynamic adaptive design on two aspects, including the shape of a membership function and a fuzzy inference rule, of a fuzzy interference system through a new heuristic optimization algorithm; and further using the obtained optimal fuzzy controller for real-time control of the micro-grid energy management. For one thing, the adaptive optimal controller considers the running information for a period of time and adjusts its own fuzzy inference system according to different environments so that the optimal scheduling of the global time can be realized; and for another, since the fuzzy logic inference has good fault tolerance towards its own imprecise information, the method can effectively avoid the interference of the uncertainty caused by a prediction error.

The invention provides a wind speed accuracy controlling method and system of a transonic-speed wind tunnel. The method comprises the following steps: S1, performing data normalization type fuzzy processing on variables inputted by a wind tunnel flow field prediction model under different preset conditions; S2, establishing a fuzzy model rule library of the wind tunnel flow field prediction modelbased on preset working conditions, acquiring a zero-order TS fuzzy model with multiple rules according to the variables after fuzzy processing, and establishing a fuzzy inference rule of a TS fuzzy system; S3, acquiring a matching degree of meeting an Mth fuzzy rule by an input variable according to the fuzzy inference rule; and S4, substituting the calculated matching degree into the TS fuzzy model, and weighting the outputs of all rules according to respectively occupied weights to obtain a Mach number model conforming to the blowing test accuracy. According to the invention, on the basis of the mode of combination of fuzzy control with prediction condition, the accuracy of the Mach number is controlled; the blowing test accuracy is realized well; and the wind tunnel flow field prediction model based on the fuzzy model prediction meets the blowing test requirement well.

A fuzzy speed controller control method for an asynchronous motor of a diesel locomotive. Because the fuzzy control technology has the function of nonlinear structure self-seeking, it can select suitable PI parameters according to different speed errors and the speed error change rates corresponding to different speed errors, so that To achieve the purpose of quickly following the given torque and given speed in the full speed range, and to improve the static and dynamic characteristics of the system. In the fuzzy control, the fuzzy controller takes the speed error signal and the differential of the speed error signal as input, input the input signal into the fuzzy controller, and then realize fuzzy control through fuzzy reasoning and fuzzy reasoning rules. According to different speed errors, different speed The rate of change of the speed error corresponding to the error, select the appropriate PI parameter. The present invention uses a fuzzy controller to select more suitable PI parameters according to fuzzy reasoning and fuzzy rules in the full speed range and according to the input, so that the system can respond faster when the load or the rotating speed changes suddenly.

The invention relates to an Internet of Vehicles relay node selection method based on fuzzy logic, and belongs to the field of information communication in a Vehicular Ad-hoc Network (VANET). An emergency safety message in the vehicle-mounted ad hoc network is transmitted between vehicles by means of a broadcast technology, wherein the key point is selection of a relay node. The core idea of the method is to comprehensively consider the coverage area, the channel busy / idle ratio and the moving factor of the vehicle based on the fuzzy logic; the forwarding priority of the vehicle serving as therelay node is obtained by establishing an appropriate fuzzy reasoning rule, and the vehicle takes the neighbor vehicle node with the maximum forwarding priority as the relay node of the next hop, sothat the number of the relay nodes is effectively reduced. The invention aims to solve the problems of broadcast message delay and redundancy caused by participation of a plurality of vehicle nodes inmessage broadcast.

The invention discloses a power grid fault diagnosis method, including the steps of: topologically analyzing a power grid with a fault according to a preset topological analysis rule to obtain a power-losing element; establishing a Petri network model of the power-losing element according to the pre-configured protection information and the action information of a circuit breaker corresponding to protection information of the power-losing element, wherein the Petri network model is composed of a library, a transition and a connecting arc between the library and the transition, and the library includes a protection library, a circuit breaker library and a element library; valuating the parameters of the Petri network model to obtain an intuitionistic fuzzy Petri network model; the parameters include a library identification value, a transition threshold value and a connecting arc weight; the library identification value includes a protection library identification value, a circuit breaker library identification value and an element library identification value; obtaining the element library identification value according to the intuitionistic fuzzy Petri network model and a preset fuzzy reasoning rule; obtaining the fault probability of the power-losing element and determining the fault element according to the element library identification value and a preset measure function.

A health monitoring system for complex networked apparatus includes a number of neuro-fuzzy inference apparatuses feeding inference results into a data fusion hierarchy. At each level in the hierarchy, fuzzy inference is applied to generate a desired output signal by processing selected input signals in accordance with a knowledge base defining fuzzy membership functions and fuzzy inference rules defined in advance. The knowledge base includes alternative definitions of membership functions and / or inference rules. The apparatus selects which definition to use according to environmental or other conditions, and predetermined selection criteria.

The invention belongs to the technical filed of robot trial planning, in particular to a navigation trail planning method of a mobile robot based on distance-type fuzzy reasoning. According to the navigation trail planning method, a fuzzy reasoning rule is adopted, knowledge of navigation, movement toward a destination and obstacle avoidance of the mobile robot is formulated, linguistic variablesin the fuzzy reasoning rule are quantified through a fuzzy set membership function, and based on distance-type fuzzy reasoning, a walking speed target value and a rotation speed target value of a given path of the mobile robot are calculated in real time by adopting information which is detected by a sensor on the mobile robot, so that the mobile robot achieves highly intelligent behavior like human navigation.

A health monitoring system for complex networked apparatus includes a number of neuro-fuzzy inference apparatuses feeding inference results into a data fusion hierarchy. At each level in the hierarchy, fuzzy inference is applied to generate a desired output signal by processing selected input signals in accordance with a knowledge base defining fuzzy membership functions and fuzzy inference rules defined in advance. The knowledge base includes alternative definitions of membership functions and / or inference rules. The apparatus selects which definition to use according to environmental or other conditions, and predetermined selection criteria.

The invention provides a cold rolling band steel plate shape intelligent optimization control method. A plate shape deviation signal is divided into a left wave and right wave component, a middle wave and two-side wave component and a quarter wave and side middle wave component through model recognition. A fuzzy inference rule for cold rolling mill plate shape control is built. A generalized input variable and a generalized fuzzy subset of the variable are defined for the obtained fuzzy inference rule, and then a fuzzy membership function of a fuzzy set that the generalized input variable corresponds to is taken. Online adjusting quantity of a cold rolling mill roll inclining device, a working roll bending device and a middle roll bending device is calculated to obtain an online adjusting quantity calculation model of the cold rolling mill plate shape intelligent optimization control. Online adjustment is conducted according to the obtained online adjusting quantity calculation model. The problem of poor cold rolling band steel product plate shape control quality caused by the fact that transmission lag exists between a mill body and a plate shape instrument and affecting factors of the plate shape are not considered comprehensively is solved.

The invention discloses a real-time detection method and system for solder skips based on fuzzy reasoning and edge calculation, and the method comprises the steps: firstly collecting normal welding time sequence data and solder skips abnormal time sequence data, and then carrying out the feature extraction and PCA dimension reduction; constructing a fuzzy reasoning system by using the sample features after dimension reduction to obtain the welding missing credibility; according to the relevance between the sample label and the welding leakage credibility, an abnormal threshold value of the welding leakage credibility is determined; and for new welding time sequence data, obtaining new solder skips credibility through fuzzy reasoning, and comparing the new solder skips credibility with an abnormal threshold value to realize real-time detection of solder skips defects. The real-time intelligent detection method for the solder skipping defect is constructed by combining PCA dimension reduction with fuzzy reasoning, the writing complexity of the fuzzy reasoning rule is reduced, meanwhile, the high recognition precision is kept, the hardware resource requirement is greatly reduced, meanwhile, the cost performance and the real-time performance are high, and the practical value is remarkable.

The invention brings forward a high-voltage cable joint state detection method. The method comprises the following steps: acquiring a cable joint temperature, a cable current and an environment temperature; respectively establishing membership functions; converting the environment temperature, the cable joint temperature and the cable current into fuzzy values respectively according to the membership functions; determining fuzzy subclasses which the fuzzy values of the environment temperature, the cable joint temperature and the cable current belong to in corresponding domain spaces; and establishing fuzzy inference rules so as to determine cable joint states and generation factors for the over-high cable joint temperature. According to the detection method provided by the embodiment of the invention, whether cable joints are abnormal and reasons why problems occur can be accurately and reliably analyzed, and alarms can be given timely, such that risks caused by abnormities of the cable joints can be avoided, and life and property safety of passengers are guaranteed. The invention also brings forward an automobile high-voltage cable joint state detection system.