34results about How to "Accurate Fault Diagnosis Results" patented technology

The invention discloses a state monitoring and fault diagnosis system of a gas turbine generating unit and a use method of the state monitoring and fault diagnosis system. The state monitoring and fault diagnosis system comprises a vibration monitoring module and a network module of a control system of the unit, a signal acquisition module and a computer which is provided with a LabVIEW platform, wherein a state performance signal is transmitted to the computer by the network module; the signal acquisition device is provided with a signal acquisition circuit and a signal processing circuit; a vibration signal output by the vibration monitoring module is acquired by the signal acquisition circuit, is processed by the signal processing circuit and is then transmitted to the computer; data acquisition is performed after an acquisition channel of the signal acquisition circuit is set by the LabVIEW platform; the LabVIEW platform comprises a real-time analysis module, an off-line analysis module and a fault diagnosis module; on-line monitoring and analysis are performed on the state performance signal and the vibration signal by the real-time analysis module; off-line analysis is performed on recorded data by the off-line analysis module; and fault diagnosis is performed on the unit by the fault diagnosis module. The state monitoring and fault diagnosis system is easy to operate and high in diagnosis accuracy and efficiency and can be applied to state monitoring and diagnosis of each turbine generating unit.

The invention relates to online monitoring technology, in particular to an online monitoring system and an online monitoring method for a motor of an electric automobile. The online monitoring system comprises a motor controller arranged on the electric automobile, a motor serving as a power driving unit of the electric automobile, and a remote monitoring management center; and the motor controller of the electric automobile directly monitors data such as various parameters of the motor controller and the motor, the running state of the motor and the like, and sends the data to the remote monitoring management center for monitoring management. The online monitoring system can directly monitor the actual running state of the motor of the electric automobile, reflects the running state of the motor of the electric automobile in real time, and realizes control of the motor state such as online tracking, monitoring, fault early warning and the like so as to discover the potential safety hazard of the motor, stop the ineffective control of the motor controller on the motor, prevent accidents and provide important safety technical guarantee for the personal safety of a driver and passengers.

The invention provides a sparse feature extraction method for a fault signal of a rotating machine. The method includes the steps of constructing a fault overcomplete atom library according to a faultgeneration mechanism of an impact fault of the rotating machine; calculating a multi-resolution time-frequency spectrum of a to-be-analyzed signal through multi-resolution generalized S-transform resolution and normalization processing according to a mathematical model of a to-be-observed signal and the fault overcomplete atom library; calculating a maximum value of the normalized multi-resolution time-frequency spectrum, and calculating an atom set capable of best performing sparse representation on the to-be-observed signal by combination with an orthogonal matching tracking algorithm; andaccording to the fault sparse representation signal and fault feature frequency of a device and rotating speed information, determining a fault type, to achieve a rapid fault diagnosis of the mechanical device. The sparse feature extraction method has the advantages that the complexity of the atomic search and the search time are greatly reduced, the efficiency of sparse decomposition is improved,and the efficiency of fault diagnosis of impact faults such as crack, pitting or spalling is improved.

The invention relates to a state maintenance and fault diagnosis method for power transformation equipment. The method is mainly and technically characterized in that an optimized adaptive adjustment-based learning rate model is established through improving typical and differential samples. According to the optimized adaptive adjustment-based learning rate model, the state maintenance and the fault diagnosis for power transformation equipment are carried out. The method is reasonable in design, and the state maintenance and the fault diagnosis for power transformation equipment are carried out through the optimized adaptive adjustment-based learning rate model. The redundant diagnosis information can be effectively reduced. Meanwhile, the complicated calculation and derivation is not required, so that a fault recognition neural network model is more practical. The model is better in line with the actual situation and the maintenance efficiency is improved. The fault maintenance cost and the power shortage loss cost of users are controlled. The problems of blind maintenance and insufficient maintenance can be avoided. The balance among the safety, the efficiency and the economy isfully realized. The replacement, the transformation, the operation maintenance and the maintenance test of power transformation equipment are guided by the method. Meanwhile, the method provides a direct basis for the management decisions, the equipment maintenance and the like of the power transformation equipment.

The invention discloses a vibration equipment fault diagnosis method and system, and relates to the field of equipment fault diagnosis. Vibration equipment vibrates non-periodically and/or is asymmetric in the aspect of appearance. The method comprises the following steps of: arranging two groups of vibration sensors on a vibration equipment shell at two sides of a centroid plane, wherein each group of vibration sensors comprises a plurality of vibration sensors, each vibration sensor is used for acquiring dynamic vibration signal of the equipment, and the centroid plane is parallel to a planewhere a vibration equipment base is located; analyzing the dynamic vibration signal acquired by each vibration sensor to obtain a feature weighting signal, and generating a contour feature signal ofthe vibration equipment according to all the feature weighting signals; carrying out dimensionality reduction on the contour feature signal on the basis of a nonlinear manifold learning method so as to obtain low-dimensional feature descriptions of the vibration equipment; and classifying the low-dimensional feature descriptions by adoption of a classifier so as to obtain a fault diagnosis result.The method and system are capable of effectively reducing minor and interference information in original vibration data so as to obtain stable and correct fault diagnosis results.

The invention discloses a linear analog circuit fault diagnosis method. The linear analog circuit fault diagnosis method comprises the steps of firstly simulating an analog circuit to obtain fault feature column vectors corresponding to fault source elements, dividing the corresponding fault source elements into an ambiguity set when the fault feature column vectors are identical within an error range, selecting a representative fault source element from each ambiguity set, respectively performing Q-time simulation on each representative fault source element to obtain Q fault feature column vectors, using each representative fault source element as a category, obtaining classification model data according to the corresponding fault feature column vectors, obtaining fault column vectors according to normally-operating output voltage phasor quantity vectors and fault output voltage phasor quantity vectors when the analog circuit breaks down, and classifying the fault column vectors according to the classification model data and corresponding classification methods to obtain fault diagnosis results. The linear analog circuit fault diagnosis method is simple and easy to operate and is based on multiple-time simulation, and the influence of tolerance on fault diagnosis can be effectively reduced.

The invention relates to a wind turbine generating set fault diagnosis method based on bispectral entropy. The method includes the steps: collecting vibration signals of a wind turbine generating set in a normal running state, a minor faulty state, a medium faulty state and a severe faulty state; calculating fault characteristic bands of all the vibration signals; establishing a fault deviation degree detection model; collecting vibration signals of a to-be-detected wind turbine generating set, carrying out fault characteristic band calculation on the to-be-detected vibration signals, obtained fault characteristics are input to the fault deviation degree detection model, calculating the fault deviation degree between the faulty state and the above four faulty states, and finally obtaining the minimal deviation degree which is the faulty state of the wind turbine generating set. The fault characteristics of non-stationary signals can be effectively extracted, fault diagnosis for the wind turbine generating set can be carried out, and the fault diagnosis precision can be improved. The method can be widely applied to fault diagnosis for wind turbine generating set equipment.

The invention discloses an equipment fault reason determination method, a storage medium and electronic equipment, and relates to the technical field of knowledge maps. The method comprises the steps:obtaining text information used for describing an equipment fault; extracting keywords from the text information and determining corresponding part-of-speech of the keywords in the text information;determining an association relationship between the extracted keywords according to the text information; searching a fault description path matched with the keyword, the part-of-speech correspondingto the keyword and the association relationship in a pre-constructed fault knowledge graph; searching a fault type node associated with the fault description path in the fault knowledge graph; and outputting an equipment fault reason corresponding to the fault type node. The invention has the beneficial effects that the literal semantics and the implicit semantics in the text information can be fully mined, so that accurate search is realized. By utilizing a knowledge graph technology, a knowledge graph with semantic comprehension is constructed, and a simple, efficient and accurate fault search technology is realized.

The invention discloses a satellite sensor fault detection and positioning method based on a self-encoding observer. According to the method, fault detection and positioning are carried out on a satellite attitude sensor based on a self-encoding observer. To be specific, according to an observer design based on an auto-encoder and a support vector machine system, self-coding observers are built according to a satellite attitude measurement method, inputs and outputs of the observers are compared to generate residual errors, the support vector machine is used for detecting faults according to mean values, mean square errors and information entropies of the residual errors, and detection results of the multiple self-coding observers are combined to locate fault sensors. According to the invention, fault detection and fault positioning can be carried out in combination with signals of a plurality of satellite attitude sensors, and the reliability and ground monitoring capability of a satellite attitude control system are improved.

The invention relates to the technical field of fault diagnosis, in particular to a motor fault diagnosis method and device, a storage medium and electronic equipment, and solves the problem that a fault diagnosis result is inaccurate due to the adoption of single-domain feature extraction in related technologies. The method comprises the following steps: collecting a vibration signal of a to-be-tested motor; sampling the vibration signal to obtain a vibration signal sample set; extracting vibration features in the vibration signal sample set to obtain a vibration feature sample set, wherein the vibration features comprise a time domain feature, a frequency domain feature and a time frequency feature; and inputting the vibration characteristic sample set into a pre-trained diagnosis model to obtain a diagnosis result. According to the invention, fault diagnosis is carried out in a mode of extracting the multi-domain features, the performance degradation degree of the motor equipment can be accurately described, and an accurate fault diagnosis result can be efficiently obtained.

The invention discloses an ANN-based ACS closed loop system fault location method. The method locates a satellite actuator fault and a sensor fault based on a neural network. The method comprises the steps that an artificial neural network and a support vector machine system are constructed; ANN is used to carry out modeling on a satellite ACS closed loop system; comparing with an actual satellite signal is carried out to generate residuals; SVM is used to detect the faults based on the mean, mean square error and information entropy of the residuals; and according to the influence of an actuator and a sensor on an attitude control closed loop and the transfer relationship of the faults in the loop, attitude kinematics is combined to locate fault points. According to the invention, on-line fault diagnosis and locating are carried out on the actuator fault and the sensor fault of the satellite ACS closed loop system; and the control efficiency and reliability of fault-tolerant control are improved.

The invention provides an oil-immersed transformer fault diagnosis method and system and a readable storage medium. The oil-immersed transformer fault diagnosis method comprises the following steps: S1, acquiring corresponding gas contents in different fault states and normal states; s2, screening the gas in the S1, selecting the gas with centrality and integrity as a characteristic gas input model, and performing integer coding on the transformer state corresponding to the characteristic gas as an output result of model training; s3, preprocessing is carried out, and effective data is reserved as a first training set; s4, carrying out layered division, sample synthesis balance and standardization treatment; s5, training and testing are carried out; and S6, carrying out transformer fault diagnosis. According to the invention, the defects of the existing oil-immersed transformer fault diagnosis algorithm based on the traditional artificial intelligence algorithm can be made up, the fault diagnosis accuracy of the oil-immersed transformer is improved, and the power supply reliability is improved.

The invention relates to a method and device for monitoring the state of an AGV, computing equipment and a storage medium. The method comprises the following steps: a first signal acquisition step: acquiring first signal data from an AGV at a first sampling rate, the first signal data at least comprising a vibration signal; a feature extraction step: extracting features from the first signal data;an abnormal state judgment step: determining whether the AGV is in an abnormal state or not based on the extracted features, executing a storage step under the condition that the AGV is determined tobe in a normal state, and executing a second signal acquisition step under the condition that the AGV is determined to be in an abnormal state; a second signal acquisition step: acquiring second signal data from the AGV at a second sampling rate, the second sampling rate being greater than the first sampling rate, and the number of types of the second signal data being equal to or greater than the number of types of the first signal data; and a storage step: storing the collected first signal data or second signal data.

The invention discloses a fault detection method and device for a special motor. The method comprises the steps of obtaining first rotating speed information of a first special motor; obtaining first volume information of the first special motor; inputting the first rotating speed information and the first volume information into a first fault detection model to obtain first fault information; obtaining first image information; if the first special motor has the first smoke point, obtaining the position information of the first smoke point; obtaining first smoke amount information according to a smoke sensor; inputting the position information of the first smoke point and the first smoke amount information into a second fault detection model to obtain second fault information; obtaining third fault information; obtaining first reminding information according to the third fault information; and according to the first reminding information, reminding that the first special motor breaks down. The technical problem that in the prior art, fault diagnosis in the operation process of the special motor is not accurate enough is solved.

The invention relates to a fuzzy diagnosis method for the water vapor chemical fault of a thermal power generating unit. The fuzzy diagnosis method comprises: according to the type of a thermal power generating unit water vapor chemical fault to be diagnosed, determining fuzzy diagnosis monitoring parameters so as to be used as fuzzy variables; according to the determined fuzzy variables, constructing a fuzzy rule for the thermal power generating unit water vapor chemical fault; according to the constructed fuzzy rule, determining fuzzy relation matrixes corresponding to each fuzzy variable; and recording the specific values of the fuzzy variables, and carrying out fuzzy diagnosis inference according to the fuzzy relation matrixes so as to obtain the diagnosis result. Compared to the method in the prior art, the method of the present invention has advantages of fast diagnosis, accurate diagnosis result, easy achieving, and the like.