41 results about "Faulty cell" patented technology

This application relates to the field of new energy technology, specifically providing a battery fault analysis method, system, control device, and storage medium, aiming to solve the technical problem of low efficiency in existing battery fault analysis technologies. To this end, the battery fault analysis method of this application includes: acquiring relevant parameters of the battery under test; determining the fault type of the battery under test based on the parameters; selecting at least one corresponding model according to the fault type; analyzing the fault of the battery under test based on the model; and obtaining management decisions for the battery under test based on the analysis results. This application initially determines the fault type of the battery, and then selects the corresponding model for analysis according to the different types of faults. Through this application, different types of models can be called for analysis of different types of battery faults, improving the accuracy of the analysis results and thus improving the analysis efficiency.

The application relates to the technical field of battery fault detection, in particular to a lithium ion battery internal short circuit fault detection method; in the method, first, constant voltage testing is carried out at a high SOC, and the short circuit resistance of the battery is calculated; whether the battery is short-circuited is judged according to the short circuit resistance data; if yes, then constant voltage testing is carried out at a low SOC, and the short circuit resistance is calculated; whether the battery is internally short-circuited or externally short-circuited is judged according to whether the short circuit resistance data is the same as the previous short circuit resistance data; if the battery is internally short-circuited and is a single-layer single-body structure battery, then the specific internal short circuit type is judged according to the two short circuit resistance data; through the fault detection method, the internal short circuit type of the fault resistance can be effectively judged.

This application relates to a method, apparatus, computer device, and storage medium for diagnosing battery faults. The method acquires the operating signal of the battery under test and then inputs this signal into a fault diagnosis model for fault diagnosis, obtaining a diagnostic result. The operating signal includes the operating voltage and / or polarization voltage of the battery under test. The fault diagnosis model is trained based on a sample dataset, which is generated from a battery model corresponding to the battery under test. This method generates a large number of sample datasets using the battery model and then uses these datasets to train a high-accuracy fault diagnosis model, thus improving the accuracy of fault diagnosis.

The application relates to the technical field of battery monitoring, in particular to a construction method of a battery fault early warning model, which comprises the following steps: S1, collecting new energy automobile lithium battery full life cycle operation data; S2, screening actual characteristic segments of internal short circuit fault precursors; S3, screening simulation characteristic segments of internal short circuit fault precursors; S4, completing construction of a high-quality fault database; S5, generating a composite input sequence of a Transform encoder; S6, processing through the Transform encoder and outputting a fault determination result, and training and optimizing to obtain the battery fault early warning model. The battery fault early warning model constructed by the application solves the problems of fault sample scarcity and early warning lag in the prior art.

This disclosure relates to a battery fault early warning method, device, controller, and vehicle, belonging to the field of power battery technology. It enables second-level fault reasoning, timely detection of sudden battery anomalies, and improved accuracy of battery fault early warning. The battery fault early warning method is applied to a web-based application and includes: acquiring vehicle battery data; determining multi-dimensional features of each individual battery cell in the vehicle based on the battery data, the multi-dimensional features including real-time features, historical statistical features, and correlation features; inputting the multi-dimensional features into a fault early warning model to obtain the fault risk probability of each individual battery cell in the vehicle output by the fault early warning model, the fault early warning model being constructed based on a gradient boosting tree and a long short-term memory network; and performing battery fault early warning based on the fault risk probability of each individual battery cell.

The present invention relates to a battery fault diagnosing device and a method of determining a fault detection counter (FDC) value by applying a weighting, and the battery fault diagnosing device of the present invention may include a measuring unit that generates measurement data on the cell voltage of each of a plurality of battery cells included in the battery pack, an FDC determining unit that increases an FDC value with an increase inclination or decreases the FDC value with a decrease inclination on the basis of the result of comparison between the measurement data and a predetermined threshold voltage value, a weighting determining unit that determines a target weighting to be applied to the increase inclination or the decrease inclination, on the basis of the FDC value at the time of applying the increase inclination or the decrease inclination and the magnitude of a voltage deviation obtained by subtracting the threshold voltage value from the measurement data, and a DTC setting unit that sets a diagnostic trouble code (DTC) value corresponding to the FDC value to a first value indicating that the battery pack is abnormal, when the FDC value reaches a predetermined upper limit threshold, and sets the DTC value corresponding to the FDC value to a second value indicating that the battery pack is normal, when the FDC value reaches a predetermined lower limit threshold.

This invention relates to the field of battery fault diagnosis technology, and discloses an intelligent diagnostic method, system, device, and medium for battery short-circuit faults. The method includes: acquiring a set of timestamped battery cell voltage data and generating a diagnostic sequence containing detection nodes; based on the detection node set, controlling a monitoring terminal to collect voltage data, capture current fluctuation signals and extract feature vectors, determine anomaly marker information, and control the diagnostic process according to the status of the detection node set and preset conditions; after diagnosis, sending the feature vectors to an analysis server to obtain fault location identifiers, storing the anomaly marker information, analyzing the differences between the marker information log and the historical log database to determine the fault evolution path, and generating and archiving a diagnostic report when matching a preset path model. The system includes acquisition, sequence construction, and diagnostic execution units. The device includes a processor and storage device, and the medium stores the corresponding program. This invention improves the accuracy, efficiency, and intelligence level of fault diagnosis.

The present application belongs to the field of energy storage battery fault detection and early warning, and discloses a battery fault early warning device and method, which is used for identifying static failure and predicting working condition hidden danger in static state, and outputting early warning after excitation confirmation when necessary. The device comprises data acquisition, data processing, fault analysis, excitation confirmation and early warning output modules. The acquisition module obtains data such as voltage, current, temperature, insulation and sampling / communication state and organizes them according to short window, long window and event segment; the processing module completes alignment, calibration, cleaning and consistency check. The analysis module calculates data quality weight, evaluates deviation degree of the marker quantity in combination with the health baseline library, and fuses evidences to obtain risk and output static prediction result; when the threshold value is reached and the boundary is allowed, the minimum identifiable excitation is executed to obtain confirmation score C, and the final risk is fused to output alarm and evidence chain in stages, so as to realize closed-loop early warning.

This application discloses a fault processing method and apparatus, and a computer-readable storage medium, and relates to the field of electric power technologies, so as to improve the flexibility of battery fault processing. The method includes: obtaining a first parameter value of a target parameter of a battery at a first moment; determining a target fault level among multiple fault levels of the battery based on the first parameter value, where the target fault level is used to indicate a risk level of the battery at the first moment; and executing a target fault processing policy corresponding to the target fault level.

The application discloses a battery pack, a battery fault detection method and a battery management system. The battery pack comprises a plurality of battery modules. A first diode, a second diode, a first battery cell and a second battery cell are arranged in each battery module, and the battery modules are connected in the following manner: the anodes of the first diodes are electrically connected with the positive poles of the first battery cells and the charging positive terminals of the battery modules respectively, and the cathodes of the first diodes are electrically connected with the positive poles of the second battery cells and the discharging positive terminals of the battery modules respectively; the anodes of the second diodes are electrically connected with the negative poles of the first battery cells and the discharging negative terminals of the battery modules respectively, and the cathodes of the second diodes are electrically connected with the negative poles of the second battery cells and the charging negative terminals of the battery modules respectively. The technical scheme provided by the application can remove the faulty battery cell from work and maintain the normal battery cells to continue normal work, thereby reducing the overall failure rate of the battery pack.

The application belongs to the field of electric power, and discloses a risk early warning method and device for an energy storage power station, which comprises the following steps: collecting multi-source heterogeneous operation data of battery monomer level, battery cluster level, battery cabin level and system level in the energy storage power station, and preprocessing the multi-source heterogeneous operation data; extracting a dynamic feature set related to battery fault evolution from the preprocessed data, and constructing a state vector for representing the safety state of the system based on the dynamic feature set; constructing a system state space model based on the dynamic fault evolution process, and initializing model parameters based on test data; predicting the future evolution path of the state vector based on the system state space model and real-time operation data, and calculating the probability and early warning time of the evolution path entering a predefined fault region; determining the current early warning level according to the probability and early warning time, and triggering a linkage control strategy corresponding to the early warning level.

This invention discloses a battery fault handling method, apparatus, and electronic device. The method includes: upon detecting a target fault in a first battery cluster within a target battery pack, reducing the current flowing through the first battery cluster's circuit to below a target safety threshold, disabling the first battery cluster, and rearranging the logical addresses of multiple battery clusters in a battery topology array to adjust the usage priority of the first battery cluster from a first priority to a second priority; wherein the battery topology array stores the logical addresses of multiple battery clusters in the target battery pack; in response to the first battery cluster meeting the recovery conditions, enabling the first battery cluster and rearranging the logical addresses of multiple battery clusters in the battery topology array to adjust the usage priority of the first battery cluster from a second priority to a first priority. This achieves safe current limiting and priority reduction isolation during faults, and automatic priority reset after recovery, ensuring system safety and stability and efficient power supply recovery.

The embodiment of the application discloses a battery fault detection method, system, device and electric vehicle, the temperature of each battery monomer in the battery pack is detected through at least one temperature sensor, the temperature group of each battery monomer is obtained, the target temperature of each battery monomer is obtained from each temperature group, the target temperature threshold corresponding to each target temperature is determined from the temperature threshold parameter group corresponding to each temperature group, and the failure state of the battery pack is determined according to the comparison result of each target temperature and the target temperature threshold corresponding to each target temperature. Therefore, the embodiment can perform temperature anomaly judgment by selecting the temperature detected by the available temperature sensor and the reasonable temperature threshold, which improves the accuracy of temperature anomaly judgment and battery fault judgment of each battery monomer, thereby ensuring the use safety of the electric vehicle, and avoiding the situation that the number of fault reports is relatively large due to temperature monitoring anomaly caused by failure of a certain temperature sensor.

The application discloses a proton exchange membrane fuel cell fault diagnosis method based on physical information guidance and related equipment, which can be applied to the technical field of battery fault diagnosis. Through the physical prior feature screening mechanism, the running state monitoring data is processed, the influence of redundant variables and working condition noise on the diagnosis process is reduced from the source, then the key variables are input into the neural network model after pretreatment, the neural network model is semi-supervised trained based on the physical consistency constraint and the corresponding fault category probability to output the target bottleneck representation, the consensus clustering result is obtained after the target bottleneck representation is subjected to consensus clustering, the consensus clustering result is inferred and corrected based on the physical post-correction rule, then the proton exchange membrane fuel cell is subjected to fault diagnosis combined with real-time state monitoring data, so that the accuracy of the fault diagnosis result in the proton exchange membrane fuel cell can be effectively improved.

This invention belongs to the field of battery fault diagnosis, and particularly relates to a fault diagnosis method for lithium battery voltage and current sensors. The method includes: constructing an equivalent circuit model of a lithium-ion battery and identifying its parameters; constructing an augmented system model with a singular structure, treating the faults of the voltage and current sensors as augmented states; constructing an augmented state observer; designing the gain of the augmented state observer so that the poles of its error dynamic system are located within a preset disk domain; running the augmented state observer to simultaneously output a first estimate of the battery system state and a second estimate of the sensor fault; and achieving fault diagnosis for the sensor. This invention solves the problems of unstable and lagging sensor fault identification and inability to accurately estimate the specific shape of faults under dynamic operating conditions using traditional methods, and can achieve high-precision real-time estimation of various complex faults such as bias, periodicity, and intermittency.

The application discloses a power battery fault detection and single body positioning method. The method obtains single body voltage, temperature, time stamp and running state data, completes structure identification, unified alignment and effective mask construction; uses normal samples to train an unsupervised reconstruction model, extracts reconstruction anomaly and physical offset features, and realizes file-level fault detection in combination with dynamic threshold calibration; further executes single body positioning on the fault file, and outputs the fault single body and early warning time. The method does not require a large number of fault labels, and is suitable for early fault detection and fine positioning of power batteries with different structures.

The application relates to a power battery abnormality detection and probability quantification method based on kernel density estimation and extreme value theory, and belongs to the battery fault detection field, and comprises the following steps: S1, collecting electric vehicle operation data, segmenting and preprocessing; S2, constructing a battery feature vector on historical fault-free samples, forming a fault-free reference feature set after preprocessing; S3, using a KDE model to perform abnormality detection on an evaluated sample, obtaining the probability density of the evaluated sample, taking logarithm to obtain a logarithmic density value, and statistically obtaining a tail threshold value of the fault-free logarithmic density value in the training stage; S4, in the abnormality detection process, defining the deviation degree of the evaluated sample below the threshold value as abnormality strength gap, performing GPD fitting on the tail distribution of the gap corresponding to the historical fault-free sample, and obtaining model parameters; and S5, inputting the gap of the evaluated sample into the GPD model, calculating an abnormality probability value, and realizing probability quantification output of the abnormality degree of the battery.

The application discloses a power battery fault analysis method based on weighted fusion, comprising the following steps: S1, collecting operation data of the power battery under multiple preset working conditions; S2, preprocessing the operation data and extracting key data features; S3, constructing a fault type identification model and outputting a fault type, if the fault type comprises a sub-fault type, executing step S4, otherwise, executing step S5; S4, constructing a sub-fault type identification model and outputting a sub-fault type; S5, verifying the fault type or the fault sub-type and outputting a verification conclusion and a maintenance suggestion. Through the two progressive models and the dynamically adjusted weights, the application realizes coarse positioning to accurate positioning of the battery fault, and after the working condition reproduction verification, the core fault type is output through the unique fault determination mechanism, so that the problems of qualitative ambiguity of the traditional diagnosis method, low efficiency of the maintenance decision and inaccurate fault positioning are solved.

A battery detection method, apparatus (200), device, storage medium, and computer program product are disclosed. The battery detection method includes: acquiring first operating data of a first battery (110); determining a first detection result of the first battery (120) based on the fault detection model corresponding to the first battery and the first operating data, wherein the first detection result includes a first score for the first battery to experience a target fault; and determining a second detection result of the first battery (130) based on the first score and a time influence factor for the first battery to experience the target fault, wherein the time influence factor is determined based on the time information of the first battery experiencing the target fault within a first preset time period. This method can solve the problems of low detection accuracy and high false alarm rate in battery fault detection results.

This invention discloses a battery fault detection method, system, device, and medium based on multivariate information entropy, relating to the field of battery fault detection technology for electrochemical energy storage power stations. The method includes: using the highest single-cell temperature within a battery module as a representative temperature, calculating the temperature change of each module and the comprehensive average and sample standard deviation of all module temperature changes to construct an adaptive confidence interval; identifying modules whose temperature changes do not fall within this interval as target faulty battery modules; then collecting voltage and temperature time-series data of all cells within the target module to calculate the multivariate information entropy; and selecting candidate faulty cells based on the multivariate information entropy of all cells. Finally, calculating the local relative entropy deviation of the candidate cells and comparing it with the upper limit of the statistical distribution of deviation values ​​for all cells; if the deviation exceeds the upper limit, it is determined to be a thermal runaway fault. This fault detection method can effectively reduce the computational burden, achieve threshold-free adaptive detection, and thus accurately locate battery faults.

The application discloses a battery early internal short circuit fault identification method based on capacity increment analysis and belongs to the technical field of battery fault identification; the method comprises the following steps: extracting an incremental capacity curve of a constant current charging segment of a battery; dividing a peak value interval of the incremental capacity curve, extracting feature data, and identifying fault features of the feature data by calculating a Pearson correlation coefficient value between the feature data and a fault label; constructing a fault classification model, inputting the fault features into an input vector, inputting the fault label at the same time, training to obtain a decision function, and calculating an output classification result by using the decision function; and the method has the beneficial effects that the battery early internal short circuit fault is identified based on the capacity increment analysis method, the fault features are screened by dividing the peak value interval and calculating the Pearson correlation coefficient of the feature data and the fault label, a classification model is constructed for diagnosis, and the effective identification of the battery early internal short circuit fault can be realized.

A method for early detection of a thermal runaway fault for a high voltage battery system of a vehicle is disclosed. The method may include receiving, by a battery fault detection system, historical operational data associated with the high voltage battery system of the vehicle, preparing the historical operational data for input into a trained predicted operation model, modeling predicted operational data of the high voltage battery system based on the historical operational data, comparing the historical operational data and the predicted operational data, and determining the thermal runaway fault based on the comparison between the historical operational data and the predicted operational data.

This invention discloses a battery system fault detection method and system, relating to the field of battery fault detection technology. The method includes the following steps: collecting performance data during the stable operation phase of the battery and generating a degradation evolution baseline trajectory; collecting real-time performance data during the continuous operation phase of the battery and generating the current degradation evolution trajectory; performing trend decoupling analysis on the current degradation evolution trajectory using a variational mode decomposition algorithm and determining its consistency with the degradation evolution baseline trajectory to obtain potential abnormal evolution behavior; analyzing the potential abnormal evolution behavior using a deviation algorithm to obtain anomalous evolution characteristics and anomalous evolution trends; determining the existence of a fault based on the anomalous evolution characteristics, determining the fault evolution stage based on the anomalous evolution trend, and obtaining the fault detection result based on the fault existence and fault evolution stage. This invention generates fault detection results through decoupling analysis, thereby achieving fault detection in battery systems.

The application discloses a kind of multi-dimensional residual fusion battery fault detection methods based on multiple adversarial generation network, it is related to the technical field of new energy automobile and energy storage system safety management, the method constructs multiple adversarial generation network including generator and triple discriminator, and through multidimensional constraint mechanism, network is jointly trained, so that generator can be high-fidelity reconstruction to normal battery data, in online detection stage, real-time operation data are input into generator to obtain reconstruction sequence, and calculate morphological difference, distribution difference and point-to-point difference, are fused into multi-view reconstruction residual, and then based on historical health data, dynamic threshold is calculated, in continuous multiple time windows, multi-view residual and dynamic threshold are compared, and finally the battery state is determined;The application captures amplitude anomaly, waveform morphological distortion and statistical distribution deviation by using the multi-view residual fusion method, to early perceive weak faults, realize high sensitivity, high reliability battery fault diagnosis.

This invention discloses a fault early warning device and method for fuel cells in hydrogen-powered ships, relating to the field of fuel cell technology. The device includes a multi-dimensional parameter sensor group, a processor, a memory, and an actuator. The sensor group collects multi-dimensional operating parameters, the processor calls an improved support vector machine voltage prediction model from the memory to process the data, and the actuator outputs graded indication signals. The method constructs vectors from the collected data, performs data transformation and normalization, solves model coefficients through matrix operations, performs new data transformation and voltage prediction, and eliminates extreme errors to achieve graded early warning based on the error sum. This invention designs a new kernel function calculation mode to reduce noise interference, solve the problems of insufficient accuracy and sensitivity and complex parameter tuning in traditional early warning systems, improve system reliability and safety, reduce maintenance costs, extend equipment life, and contribute to the environmentally friendly and sustainable development of hydrogen-powered ships.

The utility model discloses a kind of multi-redundancy whole machine power supply device for unmanned plane and unmanned plane, including four independent battery unit B1, B2, B3, B4 and parallel bus, the positive pole of battery unit B1 is connected with the negative pole of battery unit B2, forms first series branch, the positive pole of battery unit B3 is connected with the negative pole of battery unit B4, forms second series branch, the positive pole input end of parallel bus is simultaneously connected with the positive pole of B1 in first series branch and the positive pole of B3 in second series branch, the negative pole input end of parallel bus is simultaneously connected with the negative pole of B2 in first series branch and the negative pole of B4 in second series branch;Parallel bus is also connected with the main power output port for supplying power to unmanned plane whole machine. The utility model solves the problem that existing technology exists electric unmanned plane and causes whole machine power failure crash risk due to single battery fault.

A kind of power battery failure cell thermal runaway accident cause research method, by comparing the characteristics of voltage drop curve and temperature change curve when accident occurs, and combining with the appearance observation and analysis of failure cell after fire accident, the effective and rapid judgment of the cause of thermal runaway accident of failure cell in new energy vehicle fire accident investigation is realized.In the present application, the power battery failure cell thermal runaway accident cause research method starts from the voltage and temperature curve characteristics of cell level, and combines with the state of cell sample after accident, which is suitable for distinguishing and judging the cause of the accident.The method is easy to operate, and is convenient for relevant technical researchers to carry out new energy vehicle fire accident investigation and rapid judgment of the cause of the accident.