109 results about "Fault severity" patented technology

The invention discloses a network fault diagnosis method based on deep learning in a network virtualization environment. The network fault diagnosis method comprises the steps of: dividing a network into a physical network and a virtual network, combining the characteristics of occurrence of network faults, considering the time influencing factor, network topological connection characteristics and a mapping relation between the virtual network and the physical network, and comprehensively evaluating the network faults by means of a fault severity grading probability; regarding network characteristic parameters with influence degrees as a model learning resource, paying attention to the correspondence between variation trend of network historical data and fault tags, establishing a network fault diagnosis model with multiple fault grading probabilities in the network virtualization environment based on a viewing angle of deep learning, and training network parameters by using the network fault diagnosis model; and adjusting a fault prediction model in the training process, and utilizing an optimized and adjusted deep learning network to realize fault diagnosis in the network virtualization environment. The network fault diagnosis method can carry out deep analysis on the network parameters in the network virtualization environment, therefore the network fault diagnosis method has higher precision in predicting the network faults.

A method and apparatus is provided for fault notification based on a severity level. The method comprises detecting a fault associated with a processing tool that is adapted to process one or more workpieces, determining a fault severity level of the detected fault and selecting at least one user to notify of the fault based on the severity level of the fault.

A system and method for detecting a rotor fault condition in an AC induction machine is disclosed. The system includes a processor programmed to receive voltage and current data from an AC induction machine, generate a current frequency spectrum from the current data, and identify rotor-fault related harmonics in the current frequency spectrum. The processor is also programmed to calculate a fault severity indicator using the voltage and current data, identified rotor-fault related harmonics, and motor specifications, analyze the fault severity indicator to determine a possibility of rotor fault. The processor generates an alert based on the possibility of rotor fault.

The invention relates to the technical field of satellite fault diagnosis and fault-tolerant control, and proposes a fault diagnosis method, which has very good robustness to external interference, isfree of influence of external interference in a fault diagnosis process, prevents generation of misinformation and missing report, and can effectively estimate fault severity. According to the satellite fault diagnosis and fault-tolerant control method based on a self-adaptive observer, first a satellite kinematics and dynamics model under an actuator fault and external interference influence isestablished to analyze influence of fault effective factors on a system; secondly, an observer is designed for a satellite fault system, a residual error evaluation function is established based on observer output and system actual output, an evaluation threshold value is obtained, and the purpose of fault detection is achieved; then, when occurrence of a fault is detected, a self-adaptive observer is designed to complete a fault diagnosis task; and finally, based on an estimated value of the fault, a finite time self-adaptive fault-tolerant controller is designed. The satellite fault diagnosis and fault-tolerant control method based on a self-adaptive observer is mainly applied to fault diagnosis and fault-tolerant control occasions.

The invention discloses a control method of an electronic automobile battery management system (BMS). The control method comprises a charge control method; the BMS has a function of receiving a charge connection confirmation signal on a charge plug, and is capable of adopting different control strategies according to the situation that whether an overall controller is in an idle mode or a powered or driving mode. By adopting the system, alarms of two grades can be made to a whole automobile according to abnormal states of a battery pack on the basis of fault severity; in the situation of a second-grade alarm, a motor controller can be operated in a power reduction manner according to available charge/discharge current; in the situation of a first-grade alarm, charging/discharging of a battery is stopped, so that safe use of the power battery can be ensured, and the service life of the power battery can be effectively prolonged.

The invention relates to a method for online monitoring the content of gas in transformer oil and a system thereof. A gas optoacoustic spectroscopy is adopted to measure the gas dissolved in the transformer oil, namely, after the gas is sent into an opto-acoustic pond, a gas circuit valve is closed, laser sent by a laser is used for radiating sample gas in the opto-acoustic pond, sound wave is generated by the gas which passes through light-absorbing pulse, and the intensity of the sound wave is detected by a rheomicrophone; then, a signal is amplified by preposing and phase locking, and the data of the signal is collected by a high-accuracy digital-analog converter and a DSP data transmitter and stored; at the same time, opto-acoustic phase drift analysis is carried out by a lock-in amplifier, and the running quality or fault category and fault severity of an electrical power unit is exactly judged according to the sorts and the concentration of the gas, thus judging the running state or the property and degree of fault of the electrical power unit with high accuracy. The invention has simple structure and stable running, and can realize online monitoring for a long time.

The embodiment of the present invention provides a diagnostic method and device of a train bearing failure. The method mainly comprises a step of using a vibration sensor to collect the vibration data of a train bearing and a framework and obtaining a bearing signal according to the train bearing and the framework, a step of carrying out analysis processing on the bearing signal and obtaining the time-domain characteristic parameter and the frequency domain characteristic parameter of the bearing signal, and a step of using a failure classification algorithm to solve the failure state information of the bearing based on the time domain characteristic parameter and the frequency domain characteristic parameter of the bearing signal and the data of the bearing of a training sample under various states. Through collecting the vibration data of the train bearing and the framework by using the vibration sensor, the time domain characteristic parameter of the bearing signal is obtained, whether the bearing corresponding to the sensor is failed or not is judged according to the time domain and frequency domain characteristic parameters, the position of a failure is positioned at the same time, and the severity degree of the failure is identified. According to the embodiment of the present invention, the fault accurate diagnosis, fault positioning and fault severity degree discrimination of a subway train bearing are realized.

The invention provides a rolling bearing fault diagnosis method based on improved multi-scale amplitude perceived permutation entropy, relates to the field of digital signal processing, and aims at solving the problems of low separability of feature extraction, low accuracy of fault recognition and insufficient analysis of fault severity in the existing fault diagnosis method of the rolling bearing vibration signal. The method comprises the following steps: step 1. obtaining vibration signal sample sets of the rolling bearing under different fault types and different fault degrees; step 2. obtaining the optimal PR component for subsequent feature extraction; step 3. obtaining fault feature vectors of different fault types and different fault degrees; step 4. inputting the feature vector into the random forest classifier; and step 5. obtaining the fault type and the fault severity of the rolling bear. The extracted feature vector has good separability and high fault description ability,and the average recognition accuracy rate reaches 99.25%. The method can be widely applied to the field of bearing fault diagnosis.

The invention discloses a method and a system for detecting a device fault. The method includes: 1, acquiring model parameters and a rule base, wherein the model parameters include a device type, a device structure, the number of measuring points, the distribution of the measuring points and the attributes of the measuring points, and the rule base includes fault types, corresponding fault characteristic parameters and set thresholds; 2, constructing the fault models of a device according to the model parameters and the rule base, wherein every device type corresponds to multiple fault types, and every fault model comprises the fault characteristic parameters, the set thresholds of the fault characteristic parameters, and the corresponding fault types; 3, acquiring the infrared thermogram of every measuring point of the device; 4, extracting the device characteristic parameters from the infrared thermogram according to the model parameters; and 5, acquiring the fault detection result of the device according to the fault models and the device characteristic parameters, wherein the fault detection result comprises the fault location and the fault severity level. The method and the system have the advantages of realization of the analysis of the running state of the device, realization of online detection, and improvement of the detection accuracy of the device faults.

The invention provides a self-adaptive current differential motion protection method for direct-current lines. The method is suitable for direct-current power transmission lines in two-end and multi-end direct-current power transmission systems. To the two-end system, setting-point currents are respectively calculated through electric quantities at two ends so as to realize the current differential protection; and to the multi-end system, branch-point currents are respectively calculated through the voltage and the current of a converter station line end so as to realize the current differential protection. A self-adaptive fault severity setting method is brought forward, and the self-adaptive adjustment of a differential current threshold is realized through the change of the voltage, so that the rapidity and the sensitivity of the protection are taken into account. The self-adaptive current differential protection method for the direct-current lines can improve the sensitivity and the reliability of the protection of direct-current lines and is suitable for the two-end and multi-end direct-current power transmission systems and a voltage source converter type direct-current power transmission system.

The invention provides a method for detecting fiber bragg grating acoustic emission of a helicopter bearing fault in order to detect an early-time state of the fault of rolling bearings on a helicopter, enhance the capacity for recognizing the fault of the bearings of the helicopter and prolong the service life of the bearings of the helicopter. The detecting method includes the steps of sensor installation, signal sensing and demodulation, wherein the sensor installation includes pasting-installation of a grating sensor (packaged or non-packaged), the single-fiber-bragg-grating sensing method is adopted for the sensing, and the demodulation includes photoelectric conversion, amplification, acquiring, resonance demodulation processing and the like. The method for detecting fiber bragg grating acoustic emission of the helicopter bearing fault is mainly used for detecting the fault of the bearings of the helicopter and faults of bearings of other rotating machines at the home position and on line. Compared with traditional detecting methods (eye observing, vibration, magnetic powder and the like), the method for detecting fiber bragg grating acoustic emission of the helicopter bearing fault can effectively detect the early-time state of the faults and accurately judge the fault type and the fault severity order, and the method has great significance for on-condition maintenance and service life prolonging of the helicopter.

The invention provides a method for monitoring the state of an underwater vehicle based on fuzzy support vector domain description, and belongs to the technical field of fault diagnosis for the underwater vehicle. The method comprises the steps of respectively extracting feature information from AUV (Autonomous Underwater Vehicle) longitudinal velocity signal wavelet approximate components and propeller control signals when the operating state of a propeller is to be monitored based on a modified Bayesian classification algorithm, constructing a sample to be monitored based on the extracted feature information, calculating a fuzzy membership coefficient, substituting the fuzzy membership coefficient into a fuzzy support vector domain description monitoring model so as to acquire a monitoring coefficient, and judging the operating state of an AUV propeller based on the monitoring coefficient. The method provided by the invention not only can judge whether the AUV propeller has a fault or not, but also can judge the fault severity of the AUV propeller, and is especially suitable for being applied to state monitoring for the autonomous underwater vehicle propeller.

The invention provides a fault statistic analysis based motor train unit repair decision method and apparatus. The method comprises the steps of calculating an instantaneous fault rate, fault severity level magnitude and fault probability level magnitude of a fault according to fault information of a fault mode of a part of a motor train unit; calculating fault criticality level magnitude according to the fault severity level magnitude and the fault probability level magnitude; determining a fault repair mode according to the fault criticality level magnitude and the fault severity level magnitude; calculating running mileage or running time, corresponding to a loss critical point of the part, of the motor train unit according to the instantaneous fault rate; and determining a repair cycle of the part with the corresponding fault mode according to the fault repair mode and the running mileage or the running time, corresponding to the loss critical point of the part, of the motor train unit. According to the motor train unit repair decision method and apparatus, the fault can be quantitatively analyzed, the judgment efficiency of a repair policy for important parts of the motor train unit can be improved, and judgment standards of repair decision are unified.

A method of health management of a monitored system includes collecting component condition indicator data used to calculate a plurality of component health indicators. Component fault severity and potential failure modes are determined utilizing the component condition indicator data. The potential failure modes are ranked in order of likelihood to isolate the failure mode. A system of health management for monitored apparatus includes a fault severity module to derive a plurality of component health indicators from collected component condition indicator data, the plurality of component health indicators indicative of fault severity of a plurality of components. A fault isolation module separately derives a ranked listing of potential fault/failure modes utilizing the component condition indicator data. The system further includes a communication portion to which the plurality of component health indicators and the ranked listing of fault/failure modes are transmitted for review and/or action by cognizant personnel.

The invention provides a virtual network mapping method and device based on regional fault prediction. The method comprises the following steps: constructing a virtual request queue by received virtual network requests; determining regional fault severity according to the statistical information of an initialized regional fault model; selecting a virtual network request having a maximum average profit; executing a node mapping scheme based on a greedy strategy on the virtual network request having the maximum average profit; judging whether the virtual network request having the maximum average profit permits link splitting or not; when link splitting is permitted, allocating a corresponding bandwidth for each branch in a candidate mapping path according to the regional fault severity; and executing virtual link mapping on each branch till the execution of virtual link mapping of all virtual links is completed. The probability of potential regional fault occurrence is predicted and mapping of a virtual network is guided by using the regional fault model and a risk assessment method, so that fault regions are avoided, or the performance loss expectancy of a regional fault is reduced.

Various systems, methods, and apparatuses for operating a wireless charging device in an electric vehicle are disclosed. One method includes detecting a system fault indicative of one or more faults in the wireless charging device in the electric vehicle or in the transmitter. The method further includes determining a fault severity level from a plurality of fault severity levels based on a type of the system fault detected. A total number of types of system faults can be greater than a total number of the plurality of fault severity levels. The method further includes performing one or more system fault response operation based on the determined fault severity level. Each of the plurality of fault severity levels can be associated with a different set of system fault response operations.

A power grid risk comprehensive assessment method based on improved power betweenness is used for objectively evaluating power grid risks. The technical scheme is that the method first calculates improved power betweenness of each node and line; then the improved power bewteenness is introduced into a power grid fault consequence severity index, thereby obtaining a power grid fault consequence severity index based on improved power betweenness; then an intuition fuzzy analytic hierarchy process is adopted to make a comprehensive assessment of a fault severity index, and an evaluation result and a power grid fault probability are combined, thereby obtaining a final power grid risk value; and finally, power grid risks are accurately assessed according to the final power grid risk value. The power grid risk comprehensive assessment method based improved power betweenness introduces the improved power betweenness into the power grid fault consequence severity index, so that each index contains consideration of influence on a load side in a calculation process, and the overall fault severity evaluation result is relatively reasonable. The power grid risk comprehensive assessment method based improved power betweenness can comprehensively, objectively and accurately make an evaluation of power grid risks.

The invention discloses a multi-source automobile security information processing method. According to the method, automobile security information from multiple sources is acquired, including owner complaint information and vehicle security correlation information; according to a preset automobile fault expert knowledge base, an automobile fault tag and a fault severity level of each piece of automobile security information are determined; according to all the owner complaint information and corresponding tags, clustering is performed on all the owner complaint information to form automobile fault cases; and according to all the vehicle security correlation information and corresponding tags, the vehicle security correlation information and the automobile fault cases are correlated to forman automobile fault case knowledge graph, and the automobile fault case knowledge graph can be used for automobile security risk level assessment. Thus it can be known that sources of reference information for automobile security risk assessment are enriched, and the reference value of the reference information for automobile security risk assessment is greatly raised.

The invention provides a motor fault early warning method and system. The method comprises the following steps of respectively calculating a rotor broken bar generalization stage number, a stator winding turn-to-turn short circuit generalization stage number, a bearing outer raceway fault generalization stage number, a bearing inner raceway fault generalization stage number, a bearing rolling bodyfault generalization stage number and a bearing retainer fault generalization stage number according to three-phase voltage and three-phase signal of a motor and a preset rotor broken bar fault diagnosis model, a stator winding fault diagnosis model and a bearing fault diagnosis model; and evaluating the health state of the motor according to the generalization series. By adopting the motor faultearly warning method and system, the motor fault severity can be judged, a shutdown maintenance plan is arranged in advance, the production efficiency is improved, the equipment operation and maintenance cost is reduced, and the economic and safety benefits of enterprises are increased.

The invention discloses a transient stability risk assessment method of an electric power system containing a large-scale wind power station. The method comprises the following steps: establishing a power grid fault uncertainty model; obtaining a predicated fault set according to a power grid fault; calculating the fault occurrence probability of the predicated fault set; simulating a typical wind speed scene, and carrying out interval division on a wind speed sample; calculating wind speed interval probability; carrying out time-domain simulation analysis on each wind speed scene and a fault scene and constructing a non-linear utility function; calculating to obtain a fault severity degree; calculating a risk index under each scene and obtaining a general transient stability risk index according to the risk index of each scene. The assessment method disclosed by the invention considers the uncertainty of wind speeds and the uncertainty of a system fault state, and a scene simulation and interval probability analysis method is adopted and is used for researching system transient stability under each wind speed scene; the defect of a traditional method that the time consumption is great is overcome; the method for measuring the fault severity degree meets the objective reality better.