An electric vehicle intelligent range-extended battery pack health state detection method and system

Abstract

This invention relates to the field of battery testing technology, specifically to a method and system for detecting the health status of an intelligent range-extended battery pack for electric vehicles. The method includes: collecting the system operating status of the onboard range-extender generator system and electrical data of the battery pack; processing the electrical data to determine each electrical ripple feature related to the operating characteristics of the onboard range-extender generator system; analyzing the cell aging process inside the battery pack using a preset correlation judgment model based on the system operating status and each electrical ripple feature to obtain the overall health coefficient of the battery pack; determining a battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack; and comparing each electrical ripple feature with the battery pack warning threshold to obtain the detection result of the battery pack's health status. This invention addresses the problem that existing methods cannot accurately capture the abnormal temperature generated by the battery pack during charging of the range extender, leading to inaccurate detection of the battery pack's health status.

Description

Technical Field

[0001] This invention relates to the field of battery testing technology, specifically to a method and system for detecting the health status of an intelligent range-extended battery pack for electric vehicles. Background Technology

[0002] Currently, the core powertrain of intelligent range-extended electric vehicles includes a battery pack and a range extender generator system. The battery management system comprehensively monitors and manages the battery pack, collecting real-time data on the total voltage, total current, voltage of each individual cell, and temperature information from multiple temperature probes. This data is used to detect the battery's remaining charge (state of charge) and health status. When the battery is fully charged, the vehicle is powered by the battery pack. When the battery charge drops below a preset threshold, the range extender generator system automatically starts to charge the battery pack, or, under certain conditions, directly powers the drive motor to extend the vehicle's driving range.

[0003] In intelligent range-extended electric vehicles, when the onboard range extender generator system is frequently activated for energy replenishment, the output current and voltage of the generator system exhibit a certain degree of ripple. This makes the electrochemical reaction environment within the battery cells more complex when the battery pack is being charged by the range extender, causing a deviation from the aging path established by the battery management system based on DC charging conditions. When processing rippled current data, the battery management system averages the data, ignoring the additional impact of ripple on battery aging. Existing battery health status detection methods cannot effectively capture the power quality fluctuations caused by range extender charging, the complex superposition effect of current during simultaneous charging and discharging, or the impact of range extender heat on the local temperature field of the battery pack. They also struggle to detect abnormal temperatures generated in the battery pack during charging, leading to inaccurate detection of the battery pack's health status. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting the health status of an intelligent range-extended battery pack for electric vehicles, which solves the problem that existing battery health status assessment methods cannot accurately capture the abnormal temperature generated by the battery pack during charging of the range extender, resulting in inaccurate detection of the battery pack's health status.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for detecting the health status of an intelligent range-extended battery pack for electric vehicles, comprising: Collect system operating status and battery pack electrical data of the on-board range extender generator system, wherein the electrical data includes fluctuation information reflecting the output power characteristics of the on-board range extender generator system; The electrical data is processed to determine each electrical ripple characteristic related to the operating characteristics of the on-board range extender generator system; Based on the system operating status and the characteristics of each power ripple, the aging process of the cells inside the battery pack is analyzed using a preset correlation judgment model to obtain the overall health coefficient of the battery pack. The battery pack warning threshold is determined based on the system operating status and the overall health coefficient of the battery pack. Each power ripple feature is compared with the battery pack warning threshold to obtain the detection result of the battery pack health status.

[0006] Preferably, the step of processing the electrical data to determine each power ripple characteristic related to the operating characteristics of the on-board range extender generator system includes: The electrical data is subjected to instantaneous energy processing to obtain the instantaneous energy value of the electrical data; The instantaneous energy values ​​of the electrical data are subjected to feature extraction processing to obtain the energy envelope features of the instantaneous energy values; Periodic analysis is performed on the energy envelope characteristics of the instantaneous energy value to determine each electrical ripple characteristic related to the operating characteristics of the vehicle-mounted range extender generator system.

[0007] Preferably, the step of periodically analyzing the energy envelope characteristics of the instantaneous energy value to determine each energy ripple characteristic related to the operating characteristics of the on-board range extender generator system includes: The components of the energy envelope feature of the instantaneous energy value are extracted to obtain the harmonic distribution components of the energy envelope feature and the original features of each electrical energy ripple; The harmonic distribution components are matched with a preset vibration mode library to obtain vibration correlation modes; Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics; Based on the original characteristics of each electrical ripple and the characteristics of the periodic interference signal, each electrical ripple characteristic related to the operating characteristics of the vehicle-mounted range extender generator system is determined.

[0008] Preferably, the step of using the vibration correlation mode to perform anomaly analysis on the energy envelope features to obtain the periodic interference signal features includes: Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation. Using the abnormal fluctuation threshold, the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation are compared to obtain the comparison value of each abnormal fluctuation. Based on each abnormal fluctuation comparison value, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics.

[0009] Preferably, the step of using the vibration correlation mode to perform anomaly analysis on the energy envelope characteristics to determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation includes: Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to determine the instantaneous frequency, instantaneous amplitude, and drift rate of the abnormal fluctuations. The instantaneous frequency, instantaneous amplitude, and frequency drift rate of the abnormal fluctuations are analyzed to obtain the abnormal frequency components and abnormal energy distribution characteristics. Based on the abnormal frequency components and abnormal energy distribution characteristics, the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation are determined.

[0010] Preferably, the step of analyzing the cell aging process inside the battery pack using a preset correlation judgment model based on the system operating status and the characteristics of each power ripple, and obtaining the overall health coefficient of the battery pack, includes: Based on the system's operating status, determine the operating speed, output power, and exhaust temperature; Using the operating speed, output power, and exhaust temperature, each electrical ripple characteristic is analyzed to obtain the instantaneous ripple frequency, instantaneous ripple amplitude, and frequency drift rate. Based on the instantaneous frequency, instantaneous amplitude, and frequency drift rate of the ripple, a preset correlation judgment model is used to analyze the cell aging process inside the battery pack, and the overall health coefficient of the battery pack is obtained.

[0011] Preferably, the step of determining the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack includes: Based on the system's operating status, determine the frequency range and modulation characteristics under the operating conditions; The frequency range and modulation characteristics are evaluated using a battery pack estimation model to obtain a battery pack estimation value. The overall health coefficient of the battery pack is used to adjust the estimated value of the battery pack, thereby obtaining the early warning threshold of the battery pack.

[0012] Preferably, the step of adjusting the estimated value of the battery pack using the overall health coefficient of the battery pack to obtain the battery pack early warning threshold includes: The estimated value of the battery pack is verified to obtain the estimated value of the battery pack after verification. Using the overall health coefficient of the battery pack, the estimated value of the battery pack after passing the verification is adjusted to obtain the preliminary adjusted value of the battery pack; The battery pack adjustment threshold is used to evaluate the initial adjustment value of the battery pack, and an adjustment value evaluation coefficient is obtained. Based on the adjustment value evaluation coefficient and the initial adjustment value of the battery pack, the battery pack warning threshold is obtained.

[0013] Preferably, the steps for collecting system operating status of the on-board range extender generator system and electrical data of the battery pack include: Collect raw operating status data of the vehicle-mounted range extender generator system and raw electrical data of the battery pack; The original operating state of the system and the original electrical data of the battery pack are preprocessed to obtain the preprocessed original operating state of the system and the original electrical data of the battery pack. The preprocessed original operating status of the system and the original electrical data of the battery pack are verified to obtain the system operating status of the on-board range extender generator system and the electrical data of the battery pack.

[0014] The present invention also provides a health status detection system for an intelligent range-extended battery pack for electric vehicles, the system comprising: The data acquisition module is used to collect the system operating status of the vehicle-mounted range extender generator system and the electrical data of the battery pack. The electrical data includes fluctuation information reflecting the output power characteristics of the vehicle-mounted range extender generator system. The feature processing module is used to process the electrical data and determine each power ripple feature related to the operating characteristics of the vehicle-mounted range extender generator system. The coefficient analysis module is used to analyze the aging process of the cells inside the battery pack based on the system operating status and the characteristics of each power ripple, using a preset correlation judgment model to obtain the overall health coefficient of the battery pack. The threshold determination module is used to determine the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack. The status detection module is used to compare each power ripple feature with the battery pack warning threshold to obtain the detection result of the battery pack health status.

[0015] Compared with the prior art, the method and system for detecting the health status of intelligent range-extended battery packs for electric vehicles of the present invention have the following advantages: This invention collects system operating status data of the on-board range extender generator system and electrical data of the battery pack. By processing this data, it determines the electrical ripple characteristics related to the operating characteristics of the range extender generator system, thereby quantifying the non-ideal characteristics of the range extender's output power. Combining the system operating status and electrical ripple characteristics, a pre-defined correlation judgment model is used to conduct an in-depth analysis of the cell aging process within the battery pack, obtaining the overall health coefficient of the battery pack. Then, based on the system operating status and the overall health coefficient of the battery pack, a warning threshold for the battery pack is determined. Finally, the electrical ripple characteristics are compared with this warning threshold to obtain the detection result of the battery pack's health status. By identifying and quantifying the electrical ripple caused by the range extender, battery capacity decay and abnormal increases in internal resistance can be detected earlier and more accurately, effectively avoiding sudden performance drops and safety hazards in the battery pack, and significantly improving the reliability and safety of intelligent range-extended electric vehicles in specific operating scenarios. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly described below. In all the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a flowchart of a method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to the present invention.

[0018] Figure 2 This is a structural block diagram of a health status detection system for an intelligent range-extended battery pack for electric vehicles according to the present invention.

[0019] In the diagram: 210, Data Acquisition Module; 220, Feature Processing Module; 230, Coefficient Analysis Module; 240, Threshold Determination Module; 250, State Detection Module.

[0020] The implementation and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] The following drawings disclose several embodiments of the present invention. For clarity, many practical details will be described in the following description. However, it should be understood that these practical details are not intended to limit the invention. That is, in some embodiments of the invention, these practical details are not essential. Furthermore, for the sake of simplicity, some conventional structures and components will be shown in the drawings in a simple schematic manner.

[0022] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0023] Furthermore, in this invention, the use of terms such as "first" and "second" is for descriptive purposes only and does not specifically refer to any order or sequence, nor is it intended to limit the invention. They are merely used to distinguish components or operations described using the same technical terms, and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but only if they are feasible for those skilled in the art. If a combination of technical solutions is contradictory or impossible to implement, such a combination should be considered nonexistent and not within the scope of protection claimed by this invention.

[0024] To further understand the content, features, and effects of this invention, the following embodiments are provided, and detailed descriptions are given below in conjunction with the accompanying drawings: Please see Figure 1 This invention provides a method for detecting the health status of an intelligent range-extended battery pack for electric vehicles, comprising the following steps: S100. Collect the system operating status of the on-board range extender generator system and the electrical data of the battery pack. The electrical data includes fluctuation information reflecting the output power characteristics of the on-board range extender generator system. The system operating status refers to various parameters of the on-board range extender generator system during operation, such as operating speed, output power, and exhaust temperature, reflecting the workload and efficiency of the range extender. The electrical data refers to electrical quantities such as voltage and current collected from the battery pack, and the collected electrical data includes fluctuation information reflecting the output power characteristics of the on-board range extender generator system. The collected electrical data also includes instantaneous fluctuations or ripple components caused by the operating characteristics of the range extender. In this step, the data acquisition module monitors the operating parameters of the on-board range extender generator system in real time, for example, obtaining information such as engine speed, generator output power, and exhaust temperature through the CAN bus interface. Simultaneously, this module is also responsible for collecting the electrical data of the battery pack, including the total voltage and current of the battery pack, and the voltage and temperature of each individual battery cell. High-sampling-rate data acquisition equipment, such as analog-to-digital converters with sampling frequencies reaching several kilohertz, is employed to capture instantaneous fluctuations in current and voltage. The data acquisition module is equipped with multiple sensors, such as current sensors, voltage sensors, and temperature sensors, positioned at key locations within the battery pack and onboard range extender system to ensure comprehensive and accurate data. For example, the current sensor can employ a Hall effect sensor to effectively capture the ripple component of the range extender's output current.

[0025] S200. The electrical data is processed to determine each electrical ripple feature related to the operating characteristics of the on-board range extender generator system. The electrical ripple feature refers to the periodic or non-periodic fluctuation patterns extracted from the electrical data that are related to the operating characteristics of the on-board range extender generator system, reflecting the quality of the range extender's output power and its impact on the battery pack. Specifically, the time-domain signal is converted into a frequency-domain signal using methods such as Fourier transform or wavelet transform, thereby identifying ripple components of different frequencies. For example, a fast Fourier transform is used to perform spectral analysis on the collected current data to identify characteristic frequencies related to the range extender engine speed and the number of pole pairs of the generator. Alternatively, methods based on empirical mode decomposition or variational mode decomposition are used to decompose the electrical data into eigenmode functions, thereby extracting fluctuation features at different scales. The fluctuation features are further quantified by calculating parameters such as amplitude, frequency, and phase.

[0026] S300. Based on the system operating status and each energy ripple characteristic, a preset correlation judgment model is used to analyze the cell aging process inside the battery pack to obtain the overall health coefficient of the battery pack. The preset correlation judgment model is a pre-established mathematical model or algorithm used to analyze the relationship between the system operating status, energy ripple characteristics, and the cell aging process inside the battery pack. This model can be constructed using machine learning, expert systems, or physical models, aiming to reveal the impact mechanism of the range extender's operating conditions on battery aging. Specifically, a machine learning-based correlation judgment model is first established, such as a support vector machine, neural network, or random forest model. This model is trained using historical data to learn the complex nonlinear relationship between system operating status (such as range extender operating time and load changes) and energy ripple characteristics (such as ripple amplitude and frequency) and battery pack aging indicators (such as capacity decay rate and internal resistance growth rate). In practical applications, when new system operating status and energy ripple characteristics are input into the model, the model outputs a predicted value, i.e., the overall health coefficient of the battery pack. For example, the model can be trained to accelerate the decline in the health of the battery pack when the range extender is running under specific high-load conditions and the amplitude of the frequency component of the power ripple exceeds a threshold.

[0027] S400. Determine the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack. The overall health coefficient of the battery pack is a quantitative indicator used to comprehensively assess the current health status of the battery pack, usually expressed as a percentage, reflecting the degree of aging phenomena such as battery capacity decay and increased internal resistance. The battery pack warning threshold is a critical value dynamically determined based on the battery pack's health coefficient and the system operating status. When the energy ripple characteristic exceeds this threshold, it indicates that the battery pack has abnormal aging or potential failure risk, requiring a warning. This step can dynamically adjust the warning threshold based on the current system operating status, such as the vehicle's driving mode (urban commuting and highway cruising) or the range extender's activation frequency. Simultaneously, the overall health coefficient of the battery pack also affects the setting of the warning threshold. For example, when the battery pack's health coefficient is high, the warning threshold can be set relatively leniently; while when the health coefficient is low, the warning threshold can be set more strictly to improve safety. Alternatively, the threshold determination module can use a rule-based expert system to calculate an appropriate warning threshold based on a preset rule base and current state parameters. Alternatively, an adaptive algorithm can be used to dynamically adjust the warning threshold based on the battery pack's historical aging trend and current health status, in order to adapt to the actual aging of the battery pack.

[0028] S500: Compare each energy ripple feature with a battery pack warning threshold to obtain the detection result of the battery pack health status. Specifically, if a certain energy ripple feature (e.g., the ripple amplitude at a specific frequency) exceeds a preset battery pack warning threshold, the system will determine that the battery pack is abnormal and issue a warning signal. For example, when a high-frequency ripple amplitude related to the operation of the range extender is detected to continuously exceed the warning threshold, the system can determine that there may be local overheating or uneven aging of the cells inside the battery pack and issue a warning to the driver or operator. The comparison can be a simple numerical comparison or a more complex statistical method, such as calculating the degree of deviation between the ripple feature and the threshold to assess the severity of the abnormality.

[0029] This invention actively captures and analyzes the fluctuation information of the output electrical energy of the on-board range extender generator system, i.e., the electrical energy ripple characteristics. This allows the system to identify factors that accelerate battery aging caused by specific operating conditions of the range extender, which are overlooked by traditional methods. For example, when the range extender frequently starts and the load changes, its output current and voltage may exhibit significant ripple, which can cause additional stress on the battery cells and accelerate their aging. By extracting ripple characteristics, potential problems can be detected earlier and more accurately. Secondly, by introducing a preset correlation judgment model, this model can correlate the system operating state and electrical energy ripple characteristics with the aging process of the cells inside the battery pack. Under specific range extender operating conditions, this model can identify the accelerated aging effect of ripple of a specific frequency and amplitude on a specific cell in the battery pack, thus avoiding the limitation of attributing all aging phenomena to normal charge-discharge cycles. Furthermore, the battery pack warning threshold is dynamically determined based on the system operating state and the overall health coefficient of the battery pack. This dynamic adjustment mechanism allows the warning threshold to better adapt to the actual aging condition of the battery pack and the vehicle's operating conditions. For example, when the battery pack health coefficient is low, the warning threshold will be tightened accordingly to detect potential risks earlier. Under high load or specific operating modes, the warning threshold will also be adjusted to avoid false alarms or missed alarms. Finally, by comparing each power ripple characteristic with the dynamically determined battery pack warning threshold, the detection result of the battery pack health status can be obtained. This allows for the timely detection of abnormal aging or potential fault risks within the battery pack, thereby significantly improving the accuracy of battery pack detection, operational reliability, and safety.

[0030] In some embodiments of this application described above, the step of processing the electrical data to determine each electrical ripple characteristic related to the operating characteristics of the on-board range extender generator system includes: The electrical data undergoes instantaneous energy processing to obtain its instantaneous energy value. This instantaneous energy processing refers to converting the original electrical data into its instantaneous energy value using specific signal processing algorithms, such as Hilbert transform or wavelet transform. This processing aims to reveal the time-varying energy intensity within the electrical data, thereby better capturing the inherent fluctuation information.

[0031] The instantaneous energy values ​​of the electrical data are processed by feature extraction to obtain the energy envelope features of the instantaneous energy values. This step extracts the overall fluctuation trend or profile from the instantaneous energy values. The energy envelope features can reflect the intensity changes of electrical energy fluctuations at different time scales, which helps to filter out high-frequency noise and highlight low-frequency or mid-frequency fluctuation patterns related to the operating characteristics of the generator system. For example, this can be achieved through techniques such as envelope demodulation and low-pass filtering.

[0032] Periodic analysis is performed on the energy envelope characteristics of the instantaneous energy value to determine each electrical ripple feature related to the operating characteristics of the on-board range extender generator system. This step involves identifying periodic components in the energy envelope characteristics using methods such as Fourier transform, power spectral density analysis, or autocorrelation analysis. These periodic components are correlated with specific operating frequencies or harmonic frequencies of the on-board range extender generator system (such as engine speed and generator frequency), thereby enabling the determination of each electrical ripple feature related to the operating characteristics of the on-board range extender generator system.

[0033] This invention, by first performing instantaneous energy processing on the raw electrical data, effectively makes the instantaneous power or energy changes in the electrical signal explicit, thus laying the foundation for subsequent feature extraction. Since the electrical data contains fluctuation information reflecting the output electrical energy characteristics of the on-board range extender generator system, this fluctuation information is more intuitive in the instantaneous energy domain. By extracting features from the instantaneous energy values, energy envelope features are obtained, simplifying complex instantaneous energy changes into more representative envelope curves, thereby focusing on the overall trend of energy fluctuations rather than subtle high-frequency noise. Finally, by performing periodic analysis on the energy envelope features, periodic components related to the operating frequency and harmonic frequencies of the on-board range extender generator system can be accurately identified. These components are the key electrical ripple features characterizing the generator system's operating characteristics.

[0034] In some embodiments of this application described above, the step of periodically analyzing the energy envelope characteristics of the instantaneous energy value to determine each energy ripple characteristic related to the operating characteristics of the on-board range extender generator system includes: The energy envelope characteristics of the instantaneous energy value are extracted to obtain the harmonic distribution components and the original characteristics of each electrical ripple. This step involves using signal decomposition techniques, such as empirical mode decomposition, variational mode decomposition, or wavelet decomposition, to decompose the complex energy envelope characteristics into simpler components with different frequencies and amplitudes. The harmonic distribution components are periodic components exhibiting specific frequency multiples in the energy envelope characteristics. These components may originate from the generator's own harmonic output or from other periodic interference sources. The original characteristics of each electrical ripple refer to the uncorrected periodic fluctuation information directly related to the output electrical characteristics of the on-board range extender generator system after the initial decomposition. The purpose is to decompose the original signal to distinguish periodic components from different sources.

[0035] The harmonic distribution components are matched with a preset vibration mode library to obtain vibration correlation patterns. Specifically, this step involves comparing the frequency, amplitude, and phase characteristics of the extracted harmonic distribution components with a pre-established database containing characteristics of known vibration sources (such as engines and transmission systems). The preset vibration mode library can store typical vibration frequencies and harmonic characteristics of various mechanical components under different operating conditions. Through matching, it is possible to identify which periodic components in the energy envelope characteristics may be related to non-electrical ripple factors such as mechanical vibration, thereby obtaining vibration correlation patterns. Vibration correlation patterns refer to the periodic characteristic patterns identified that are related to non-electrical ripple factors such as mechanical vibration. Their purpose is to identify and isolate periodic interference caused by non-electrical factors such as mechanical vibration.

[0036] Using the vibration correlation pattern, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics. This step refers to conducting in-depth analysis of periodic fluctuations in the energy envelope characteristics that match the identified vibration correlation pattern to determine whether they are abnormal interferences. For example, a threshold can be set; when the amplitude, frequency, or duration of a periodic component in the energy envelope characteristics highly matches the vibration correlation pattern and exceeds the range of normal electrical ripple, it is identified as a periodic interference signal characteristic. Periodic interference signal characteristics refer to periodic fluctuations determined to be unrelated to the operating characteristics of the on-board range extender generator system and caused by external or internal interference sources. The purpose is to accurately identify and quantify interference signals.

[0037] Based on the original characteristics of each power ripple and the characteristics of the periodic interference signal, each power ripple feature related to the operating characteristics of the on-board range extender generator system is determined. This step involves using signal processing techniques (such as filtering, signal reconstruction, or component removal) to remove or weaken the periodic interference signal features from the original power ripple characteristics after obtaining them, thereby obtaining purer and more accurate power ripple features that reflect the output power characteristics of the on-board range extender generator system. The purpose is to eliminate interference and obtain accurate power ripple features.

[0038] Specifically, during the operation of an electric vehicle, the onboard range extender generator system is working, and the instantaneous energy value of its output electrical energy is collected and subjected to feature extraction to obtain energy envelope characteristics. To accurately identify the electrical ripple characteristics, firstly, wavelet decomposition can be used to extract components from the energy envelope characteristics, resulting in multiple wavelet coefficient layers. Some layers represent harmonic distribution components, while others contain the original electrical ripple characteristics. Next, the frequency and amplitude information of the harmonic distribution components are matched with a pre-established vibration mode library. This vibration mode library may contain typical vibration frequencies of the engine at different speeds (e.g., the second and fourth harmonic frequencies of the engine). If the matching result shows that a certain harmonic component is highly correlated with a certain vibration mode of the engine, the corresponding vibration correlation mode is obtained. Subsequently, using this vibration correlation mode, anomaly analysis is performed on the part of the energy envelope characteristics related to that vibration mode. For example, by deriving the energy entropy or peak factor of this part and comparing it with a preset threshold, the periodic interference signal characteristics caused by engine vibration are identified. Finally, the identified periodic interference signal features are removed from the original power ripple features (e.g., through band-stop filtering or signal reconstruction) to obtain each power ripple feature that is more accurate in relation to the operating characteristics of the on-board range extender generator system, which is then used for subsequent battery pack health status detection.

[0039] This embodiment extracts components from the energy envelope features of instantaneous energy values, decomposing complex signals into more easily analyzable harmonic distribution components and original energy ripple features, laying the foundation for subsequent interference identification. Secondly, by matching the harmonic distribution components with a preset vibration mode library, periodic interference patterns related to non-electrical factors such as mechanical vibration can be effectively identified, avoiding misjudgment of interference as energy ripple. Anomaly analysis of the energy envelope features using vibration correlation patterns accurately quantifies and identifies the characteristics of periodic interference signals. Finally, by combining the original energy ripple features and the periodic interference signal features, interference signals can be separated or eliminated from the original ripple, resulting in purer and more accurate energy ripple features that reflect the operating characteristics of the on-board range extender generator system. This makes subsequent analysis of the aging process of the battery pack's internal cells more reliable and avoids misjudgments caused by interference signals.

[0040] In some embodiments of this application described above, the step of performing anomaly analysis on the energy envelope characteristics using the vibration correlation mode to obtain periodic interference signal characteristics includes: Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation. The centroid of the abnormal fluctuation spectrum refers to the center frequency of the abnormal fluctuation energy distribution in the spectral analysis of the energy envelope characteristics, reflecting the main frequency components of the abnormal fluctuation. The energy entropy of the abnormal fluctuation is used to quantify the uniformity or complexity of the distribution of abnormal fluctuation energy in the spectrum; a higher entropy value generally indicates a more complex fluctuation component or a more dispersed distribution. The peak factor of the abnormal fluctuation is an indicator of the instantaneous impact characteristics of the abnormal fluctuation signal; a high peak factor usually indicates the presence of a sharp instantaneous impact or pulse.

[0041] By using abnormal fluctuation thresholds, the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation are compared to obtain a comparison value for each abnormal fluctuation. Specifically, the abnormal fluctuation threshold is a pre-set reference limit used to determine whether a fluctuation is abnormal. This threshold is set based on historical data, expert experience, or statistical analysis methods; for example, it is set as the range of the mean of the corresponding parameter under normal operating conditions plus or minus a certain standard deviation. By comparing the determined centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation with their respective abnormal fluctuation thresholds, a comparison value for each abnormal fluctuation can be obtained, thereby reflecting the degree of deviation of the current fluctuation from the normal state.

[0042] Based on each abnormal fluctuation comparison value, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics.

[0043] This embodiment, when performing anomaly analysis on energy envelope characteristics, does not rely on a single or vague anomaly judgment, but rather refines it into a quantitative analysis of key indicators such as the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation. These indicators comprehensively capture anomalous information from multiple dimensions, including frequency distribution, energy complexity, and instantaneous impact. By comparing the quantified indicators with preset abnormal fluctuation thresholds, a comparison value for each abnormal fluctuation can be obtained, thereby achieving accurate identification and quantification of abnormal fluctuations. This makes the extraction of periodic interference signal characteristics more accurate and reliable, avoiding misjudgments or omissions that might result from judging based on a single indicator.

[0044] Preferably, the step of using the vibration correlation mode to perform anomaly analysis on the energy envelope characteristics to determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation includes: Using the vibration correlation model, anomaly analysis is performed on the energy envelope characteristics to determine the instantaneous frequency, instantaneous amplitude, and frequency drift rate of the abnormal fluctuations. This step refers to a refined analysis of the fluctuation components in the energy envelope characteristics under the guidance of the matched vibration correlation model. Instantaneous frequency refers to the frequency of the signal at a certain moment, instantaneous amplitude refers to the amplitude of the signal at a certain moment, and frequency drift rate reflects how quickly the frequency changes over time. This reveals the dynamic characteristics of abnormal fluctuations in the energy envelope characteristics.

[0045] The instantaneous frequency, instantaneous amplitude, and frequency drift rate of the abnormal fluctuations are analyzed to obtain the abnormal frequency components and abnormal energy distribution characteristics. This step can be further processed by statistical or signal processing methods (e.g., wavelet analysis and Hilbert transform) on the instantaneous parameters to extract the main frequency components and their corresponding energy distributions in the abnormal fluctuations. Abnormal frequency components refer to frequency components that should not appear or are significantly enhanced under normal operating conditions, while abnormal energy distribution characteristics describe the distribution and intensity of abnormal frequency components throughout the energy spectrum.

[0046] Based on the abnormal frequency components and abnormal energy distribution characteristics, the spectral centroid, energy entropy, and peak factor of the abnormal fluctuation are determined. The spectral centroid measures the degree of concentration in the spectral distribution, the energy entropy reflects the uniformity or complexity of the spectral energy distribution, and the peak factor characterizes the sharpness or impact of the signal waveform. These characteristic parameters can quantify the spectral characteristics and energy distribution features of abnormal fluctuations from different dimensions, providing a quantitative basis for subsequent abnormal fluctuation judgment.

[0047] This embodiment utilizes vibration correlation modes to perform anomaly analysis on energy envelope characteristics, thereby accurately determining the instantaneous frequency, instantaneous amplitude, and frequency drift rate of anomalous fluctuations. Instantaneous parameters directly capture the dynamic changes of anomalous fluctuations within the energy envelope characteristics, laying the foundation for subsequent in-depth analysis. Further analysis of these instantaneous parameters effectively extracts anomalous frequency components and anomalous energy distribution characteristics. These characteristics are the essential manifestations of anomalous fluctuations in the frequency and energy domains, clearly revealing their inherent laws. Ultimately, based on the anomalous frequency components and anomalous energy distribution characteristics, the spectral centroid, energy entropy, and peak factor of anomalous fluctuations can be accurately derived. These quantitative indicators describe the spectral characteristics and energy distribution of anomalous fluctuations from multiple perspectives, providing a comprehensive and accurate basis for subsequent anomalous fluctuation judgments, avoiding misjudgments or omissions caused by relying on a single indicator.

[0048] In some embodiments of this application, the step of analyzing the cell aging process inside the battery pack and obtaining the overall health coefficient of the battery pack by using a preset correlation judgment model based on the system operating status and the characteristics of each power ripple includes: Based on the system's operating status, the operating speed, output power, and exhaust temperature are determined. Operating speed refers to the number of revolutions the range extender generator makes per unit time, reflecting the generator's mechanical motion intensity; output power refers to the actual electrical energy output capacity of the generator, directly related to the battery pack's charging load; exhaust temperature reflects the generator's thermal state, indirectly affecting the battery pack's heat dissipation environment and temperature stress. This provides crucial operating condition background information for subsequent power ripple characteristic analysis.

[0049] By analyzing each electrical ripple characteristic using the aforementioned operating speed, output power, and exhaust temperature, the instantaneous ripple frequency, instantaneous ripple amplitude, and frequency drift rate are obtained. The instantaneous ripple frequency is the instantaneous oscillation frequency of the electrical ripple at a given moment, which may be related to the generator's speed or specific harmonic components. The instantaneous ripple amplitude refers to the instantaneous amplitude of the electrical ripple at a given moment, reflecting the intensity of the electrical energy fluctuation. The frequency drift rate indicates the rate of change of the ripple frequency over time, predicting abnormalities or aging of certain components within the generator or battery pack. By extracting refined ripple characteristic parameters, a more comprehensive and in-depth understanding of the complex relationship between electrical energy fluctuations and system operating status can be achieved.

[0050] Based on the instantaneous frequency, instantaneous amplitude, and frequency drift rate of the ripple, a preset correlation judgment model is used to analyze the cell aging process inside the battery pack, and the overall health coefficient of the battery pack is obtained.

[0051] Specifically, during high-speed driving, the onboard range extender generator system of the electric vehicle operates continuously. The data acquisition module collects real-time data on the generator's operating speed (3000 rpm), output power (15 kW), and exhaust temperature (250°C). Simultaneously, the battery pack's electrical data is collected and analyzed to obtain the electrical ripple characteristics. When analyzing the aging process of the battery cells within the battery pack, each electrical ripple characteristic is analyzed in depth based on the operating speed, output power, and exhaust temperature. For example, through signal processing algorithms, the instantaneous ripple frequency under the current operating condition is extracted as 120 Hz, the instantaneous ripple amplitude as 0.5 V, and the frequency drift rate as 0.01 Hz / s is observed. Subsequently, refined parameters (operating speed, output power, exhaust temperature, instantaneous ripple frequency, instantaneous ripple amplitude, and frequency drift rate) are input into a preset correlation judgment model. This model is based on machine learning or an expert system and has internally learned the complex relationship between parameters and the degree of cell aging under different operating conditions. The resulting overall health coefficient for the battery pack was 0.85. This more accurately reflects the current aging level of the battery pack; for example, 0.85 indicates that the battery pack is in a mild aging stage and requires continuous monitoring. Through meticulous parameter extraction and analysis, the assessment of the battery pack's health status is more precise, thus providing users with more reliable early warning information.

[0052] This embodiment refines the system's operating state into operating speed, output power, and exhaust temperature, and uses these parameters to conduct in-depth analysis of the electrical ripple characteristics. This extracts the instantaneous ripple frequency, instantaneous ripple amplitude, and frequency drift rate, avoiding potentially insufficient refinement in the correlation analysis between system operating state and electrical ripple characteristics. Specifically, the ripple characteristics of the output electrical energy of a range extender generator exhibit different patterns under different operating speeds, output power, and exhaust temperatures. For example, fluctuations in speed lead to changes in ripple frequency, while increases or decreases in output power affect the ripple amplitude. Exhaust temperature, as an environmental factor, indirectly alters the battery pack's response to ripple by influencing its temperature. By correlating operating parameters with the instantaneous frequency, instantaneous amplitude, and frequency drift rate of the ripple, a pre-defined correlation judgment model can more accurately capture the aging characteristics of the battery cells within the battery pack under specific operating conditions. Through multi-dimensional and refined parameter extraction and analysis, subtle aging signs that are often overlooked can be identified, improving the accuracy and sensitivity of the cell aging process analysis.

[0053] In some embodiments of this application described above, the step of determining the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack includes: Based on the system's operating status, the frequency range and modulation characteristics under operating conditions are determined. This step involves monitoring real-time operating parameters of the onboard range extender generator system, such as engine speed and load, during electric vehicle operation to identify specific frequency ranges and signal modulation characteristics that may affect the battery pack's electrical data under the current operating conditions. The purpose is to provide accurate operating condition background information for subsequent battery pack prediction.

[0054] A battery pack prediction model is used to evaluate the frequency range and modulation characteristics, yielding a battery pack prediction value. Specifically, a pre-established battery pack performance model is used, combined with the frequency range and modulation characteristics under current operating conditions, to make a preliminary prediction of the battery pack's health status under ideal or specific conditions. This battery pack prediction model is a machine learning model trained on historical data, such as a support vector machine, neural network, or regression model, or it can be an electrochemical model based on physicochemical principles. Its purpose is to provide a baseline prediction value consistent with current operating conditions.

[0055] The overall health coefficient of the battery pack is used to adjust the estimated value of the battery pack, resulting in a battery pack warning threshold. This step involves using the overall health coefficient of the battery pack, obtained through analysis using a preset correlation judgment model, as a correction factor to adjust the initially obtained battery pack estimate. For example, when the overall health coefficient of the battery pack is low, the warning threshold can be appropriately lowered to make it more sensitive to anomalies; conversely, when the health coefficient is high, the threshold can be appropriately relaxed. The purpose is to ensure that the finally determined battery pack warning threshold can more accurately reflect the actual aging degree and health status of the battery pack, improving the personalization and accuracy of the warning.

[0056] Specifically, under high-speed driving conditions, the onboard range extender generator system of an electric vehicle operates at a higher speed. First, based on the current high-speed driving conditions, a specific frequency range (e.g., 100Hz-500Hz) and modulation characteristics, such as significant high-frequency harmonic components, are determined. Next, using a pre-trained battery pack prediction model and considering the operating condition characteristics, a predicted battery pack health value is obtained; for example, under the current conditions, the battery pack's health status should be 90%. However, if analysis of electrical data and the cell aging process reveals that the overall health coefficient of the battery pack is actually 85%, lower than the predicted value, then the predicted 90% battery pack health value is adjusted using this 85% overall health coefficient. For example, an adjustment function is used to derive a warning threshold of 88% to more accurately reflect the actual aging degree of the battery pack, ensuring timely warnings when the battery pack's health status slightly declines. This warning threshold can be intelligently adapted to the actual health status of the battery pack.

[0057] This embodiment first precisely determines the frequency range and modulation characteristics under operating conditions based on the system's operating status, providing a more refined input for the battery pack prediction model. This allows the battery pack prediction value to more accurately reflect the battery performance under current operating conditions. Furthermore, by incorporating the overall health coefficient of the battery pack into the adjustment process of the battery pack prediction value, the actual aging degree and health status of the battery pack can be integrated into the warning threshold. Dynamic adjustments based on the actual health status enable the determined battery pack warning threshold to be more personalized and accurately adapted to the current state of the battery pack, effectively avoiding discrepancies between the warning threshold and the actual health status of the battery pack, and significantly improving the accuracy and reliability of the warning.

[0058] In some embodiments of this application described above, the step of adjusting the estimated value of the battery pack using the overall health coefficient of the battery pack to obtain the battery pack early warning threshold includes: The estimated value of the battery pack is verified to obtain a qualified estimated value. Specifically, the estimated value verification refers to the process of checking the validity or reliability of the initially obtained estimated value of the battery pack. Its purpose is to eliminate abnormal data or estimated values ​​that do not meet preset conditions, ensuring that the baseline data for subsequent adjustments is accurate and reliable. For example, this can be done by comparing with historical data, verifying with the model's prediction range, or by setting reasonable upper and lower limits to determine whether the estimated value of the battery pack is within the normal range. The qualified estimated value of the battery pack is data whose validity has been verified, laying the foundation for subsequent precise adjustments.

[0059] Using the overall health coefficient of the battery pack, the estimated value of the calibrated battery pack is adjusted to obtain a preliminary adjusted value. This preliminary adjusted value is the result of a preliminary correction based on the estimated value of the calibrated battery pack, combined with the overall health coefficient. This aims to initially reflect the actual health condition of the battery pack in the estimated value, making it closer to the current true state of the battery pack.

[0060] The initial adjustment value of the battery pack is evaluated using a battery pack adjustment threshold to obtain an adjustment value evaluation coefficient. Specifically, the battery pack adjustment threshold serves as a reference standard for measuring the reasonableness or adjustment range of the initial battery pack adjustment value. This threshold is preset based on factors such as empirical data, battery type, and operating conditions. The adjustment value evaluation coefficient is obtained by evaluating the initial battery pack adjustment value against the battery pack adjustment threshold. This evaluation coefficient can be a weighting factor, confidence score, or correction parameter to quantify the effectiveness of the initial adjustment or the degree to which further correction is needed.

[0061] Based on the adjustment value evaluation coefficient and the initial adjustment value of the battery pack, the battery pack warning threshold is obtained.

[0062] Specifically, the estimated battery pack value is compared with a preset normal operating range. If it exceeds this range, it is marked as abnormal and corrected or discarded. Alternatively, a predictive model is built using historical data, and the current estimated battery pack value is compared with the model's predicted value. If the deviation is too large, verification is performed. For example, if the estimated value is not within the expected range, it needs to be corrected or re-collected. After successful verification, the verified estimated battery pack value is obtained. Subsequently, the overall health coefficient of the battery pack is used to make a preliminary adjustment to the verified estimated battery pack value. For example, a linear or non-linear function relationship can be used for adjustment to obtain a preliminary adjusted value. Next, the preliminary adjusted value of the battery pack is evaluated using a battery pack adjustment threshold. For example, the preliminary adjusted value of the battery pack can be a percentage, representing the allowable fluctuation range of the preliminary adjusted value relative to a certain benchmark value. If the deviation between the preliminary adjusted value of the battery pack and the benchmark value (such as the ideal warning value) is within the deviation range, the adjustment value evaluation coefficient can be set to 1; if the deviation exceeds the deviation range, the adjustment value evaluation coefficient is reduced according to the magnitude of the deviation. Finally, based on the adjustment value evaluation coefficient and the initial adjustment value of the battery pack, the battery pack warning threshold is determined. This yields a more accurate and reliable battery pack warning threshold after multiple verifications and evaluations.

[0063] This embodiment introduces a pre-estimation verification step, first verifying the reliability of the battery pack pre-estimation, effectively avoiding subsequent adjustment deviations caused by inaccurate initial pre-estimations. Because the pre-estimation is rigorously verified, subsequent adjustments based on the overall health coefficient can be made on a more solid and reliable data foundation. Using the overall health coefficient of the battery pack, the verified battery pack pre-estimation is initially adjusted, allowing the pre-estimation to initially reflect the actual health status of the battery pack. By introducing a battery pack adjustment threshold to evaluate the initial adjustment value and obtain an adjustment value evaluation coefficient, the effectiveness and rationality of the initial adjustment can be quantitatively analyzed. This enables the final battery pack warning threshold to comprehensively consider the initial adjustment results and its evaluation, thereby achieving refined and intelligent determination of the warning threshold.

[0064] In some embodiments of this application described above, the steps for collecting the system operating status of the on-board range extender generator system and the electrical data of the battery pack include: This step involves acquiring the raw operating status of the onboard range extender generator system and the raw electrical data of the battery pack. This means directly obtaining unprocessed or unfiltered initial data from onboard sensors, controllers, or the data bus. The raw operating status of the system may include real-time parameters such as engine speed, load current, output voltage, and exhaust temperature, while the raw electrical data may include individual battery cell voltage, total voltage, charging / discharging current, battery temperature, and internal resistance.

[0065] The original operating status of the system and the electrical raw data of the battery pack are preprocessed to obtain preprocessed original operating status and electrical raw data of the battery pack. This step refers to the operations performed on the collected raw data to eliminate noise, fill missing values, correct outliers, or perform data format conversion, thereby improving the quality and usability of the data. The preprocessed original operating status and electrical raw data of the battery pack are a dataset that has undergone preliminary cleaning and normalization.

[0066] The preprocessed original system operating status and battery pack electrical data are verified to obtain the system operating status and battery pack electrical data of the on-board range extender generator system. This step involves checking the validity, completeness, and consistency of the preprocessed data to ensure that the data meets preset quality standards and logical rules, ultimately yielding reliable system operating status and battery pack electrical data for subsequent health status detection.

[0067] Specifically, the raw operating status of the on-board range extender generator system can include real-time sensor data such as engine speed, load current, output voltage, and exhaust temperature, while the raw electrical data of the battery pack can include battery voltage, current, temperature, and internal resistance. During preprocessing, moving average filtering can be used to remove instantaneous noise from sensor data, Lagrange interpolation can be used to fill in minor data gaps caused by communication interruptions, and voltage and current data can be normalized to eliminate dimensional interference. During the verification phase, reasonable threshold ranges can be set; for example, if the battery voltage exceeds the normal operating range (e.g., 2.5V-4.2V), it is marked as abnormal data and corrected or removed. Simultaneously, the continuity of the data sequence can be checked to ensure the integrity of the data stream. Through this refined processing, the high quality and reliability of the system operating status and electrical data ultimately used for health status detection are ensured.

[0068] This embodiment introduces raw data acquisition, preprocessing, and verification steps to ensure higher quality and reliability of system operating status and electrical data input into subsequent analysis modules. Specifically, acquiring raw data is the foundation for obtaining all information; the preprocessing step effectively removes noise and anomalies from the data, fills in missing data, and makes the data more regular and consistent; the verification step further verifies the validity and accuracy of the data, avoiding misjudgments or omissions caused by data quality issues. Because of the strict quality control at the data source, subsequent determination of power ripple characteristics, cell aging analysis, and the determination of health coefficients and warning thresholds can be based on solid and reliable data, significantly improving the accuracy and robustness of the entire health status detection method.

[0069] Based on any of the above embodiments, a method for detecting the health status of an intelligent range-extended battery pack for electric vehicles is provided. Figure 2 The present invention also provides a health status detection system for an intelligent range-extended battery pack for electric vehicles, the system comprising a data acquisition module 210, a feature processing module 220, a coefficient analysis module 230, a threshold determination module 240, and a status detection module 250.

[0070] The data acquisition module 210 is used to collect the system operating status of the vehicle-mounted range extender generator system and the electrical data of the battery pack. The electrical data includes fluctuation information reflecting the output power characteristics of the vehicle-mounted range extender generator system.

[0071] The feature processing module 220 is used to process the electrical data and determine each power ripple feature related to the operating characteristics of the on-board range extender generator system.

[0072] The coefficient analysis module 230 is used to analyze the aging process of the cells inside the battery pack based on the system operating status and the characteristics of each power ripple, using a preset correlation judgment model, to obtain the overall health coefficient of the battery pack.

[0073] The threshold determination module 240 is used to determine the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack.

[0074] The status detection module 250 is used to compare each power ripple feature with the battery pack warning threshold to obtain the detection result of the battery pack health status.

[0075] In this embodiment, the data acquisition module 210 acquires electrical data containing fluctuation information and the system operating status, and the feature processing module 220 performs in-depth analysis of the power ripple characteristics. Subsequently, the coefficient analysis module 230 combines the system operating status and power ripple characteristics, and uses a preset correlation judgment model to accurately analyze the cell aging process inside the battery pack, thereby obtaining the overall health coefficient of the battery pack. Further, the threshold determination module 240 dynamically determines the battery pack warning threshold based on the system operating status and the overall health coefficient. Finally, the status detection module 250 outputs the detection result of the battery pack health status by comparing the power ripple characteristics with the warning threshold. This system can effectively identify accelerated aging and local uneven aging problems of the battery pack under special operating conditions of the range extender, thereby significantly improving the accuracy and reliability of battery health status assessment.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the present invention specification.

Claims

1. A method for detecting the health status of an intelligent range-extended battery pack for electric vehicles, characterized in that, include: Collect system operating status and battery pack electrical data of the on-board range extender generator system, wherein the electrical data includes fluctuation information reflecting the output power characteristics of the on-board range extender generator system; The electrical data is processed to determine each electrical ripple characteristic related to the operating characteristics of the on-board range extender generator system; Based on the system operating status and the characteristics of each power ripple, the aging process of the cells inside the battery pack is analyzed using a preset correlation judgment model to obtain the overall health coefficient of the battery pack. The battery pack warning threshold is determined based on the system operating status and the overall health coefficient of the battery pack. Each power ripple feature is compared with the battery pack warning threshold to obtain the detection result of the battery pack health status.

2. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 1, characterized in that, The steps of processing the electrical data to determine each electrical ripple characteristic related to the operating characteristics of the on-board range extender generator system include: The electrical data is subjected to instantaneous energy processing to obtain the instantaneous energy value of the electrical data; The instantaneous energy values ​​of the electrical data are subjected to feature extraction processing to obtain the energy envelope features of the instantaneous energy values; Periodic analysis is performed on the energy envelope characteristics of the instantaneous energy value to determine each electrical ripple characteristic related to the operating characteristics of the vehicle-mounted range extender generator system.

3. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 2, characterized in that, The steps of periodically analyzing the energy envelope characteristics of the instantaneous energy value to determine each energy ripple characteristic related to the operating characteristics of the on-board range extender generator system include: The components of the energy envelope feature of the instantaneous energy value are extracted to obtain the harmonic distribution components of the energy envelope feature and the original features of each electrical energy ripple; The harmonic distribution components are matched with a preset vibration mode library to obtain vibration correlation modes; Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics; Based on the original characteristics of each electrical ripple and the characteristics of the periodic interference signal, each electrical ripple characteristic related to the operating characteristics of the vehicle-mounted range extender generator system is determined.

4. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 3, characterized in that, The steps for performing anomaly analysis on the energy envelope features using the vibration correlation mode to obtain the periodic interference signal features include: Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation. Using the abnormal fluctuation threshold, the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation are compared to obtain the comparison value of each abnormal fluctuation. Based on each abnormal fluctuation comparison value, anomaly analysis is performed on the energy envelope characteristics to obtain periodic interference signal characteristics.

5. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 4, characterized in that, The steps of using the vibration correlation mode to perform anomaly analysis on the energy envelope characteristics and determine the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation include: Using the vibration correlation mode, anomaly analysis is performed on the energy envelope characteristics to determine the instantaneous frequency, instantaneous amplitude, and drift rate of the abnormal fluctuations. The instantaneous frequency, instantaneous amplitude, and frequency drift rate of the abnormal fluctuations are analyzed to obtain the abnormal frequency components and abnormal energy distribution characteristics. Based on the abnormal frequency components and abnormal energy distribution characteristics, the centroid of the abnormal fluctuation spectrum, the energy entropy of the abnormal fluctuation, and the peak factor of the abnormal fluctuation are determined.

6. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 1, characterized in that, Based on the system operating status and the characteristics of each power ripple, the steps for analyzing the cell aging process inside the battery pack using a preset correlation judgment model to obtain the overall health coefficient of the battery pack include: Based on the system's operating status, determine the operating speed, output power, and exhaust temperature; Using the operating speed, output power, and exhaust temperature, each electrical ripple characteristic is analyzed to obtain the instantaneous ripple frequency, instantaneous ripple amplitude, and frequency drift rate. Based on the instantaneous frequency, instantaneous amplitude, and frequency drift rate of the ripple, a preset correlation judgment model is used to analyze the cell aging process inside the battery pack, and the overall health coefficient of the battery pack is obtained.

7. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 1, characterized in that, The steps for determining the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack include: Based on the system's operating status, determine the frequency range and modulation characteristics under the operating conditions; The frequency range and modulation characteristics are evaluated using a battery pack estimation model to obtain a battery pack estimation value. The overall health coefficient of the battery pack is used to adjust the estimated value of the battery pack, thereby obtaining the early warning threshold of the battery pack.

8. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 7, characterized in that, The steps for adjusting the estimated value of the battery pack using the overall health coefficient of the battery pack to obtain the battery pack early warning threshold include: The estimated value of the battery pack is verified to obtain the estimated value of the battery pack after verification. Using the overall health coefficient of the battery pack, the estimated value of the battery pack after passing the verification is adjusted to obtain the preliminary adjusted value of the battery pack; The battery pack adjustment threshold is used to evaluate the initial adjustment value of the battery pack, and an adjustment value evaluation coefficient is obtained. Based on the adjustment value evaluation coefficient and the initial adjustment value of the battery pack, the battery pack warning threshold is obtained.

9. The method for detecting the health status of an intelligent range-extended battery pack for electric vehicles according to claim 1, characterized in that, The steps for collecting system operating status data of the on-board range extender generator system and electrical data of the battery pack include: Collect raw operating status data of the vehicle-mounted range extender generator system and raw electrical data of the battery pack; The original operating state of the system and the original electrical data of the battery pack are preprocessed to obtain the preprocessed original operating state of the system and the original electrical data of the battery pack. The preprocessed original operating status of the system and the original electrical data of the battery pack are verified to obtain the system operating status of the on-board range extender generator system and the electrical data of the battery pack.

10. A health status detection system for an intelligent range-extended battery pack for electric vehicles, characterized in that, The system includes: The data acquisition module is used to collect the system operating status of the vehicle-mounted range extender generator system and the electrical data of the battery pack. The electrical data includes fluctuation information reflecting the output power characteristics of the vehicle-mounted range extender generator system. The feature processing module is used to process the electrical data and determine each power ripple feature related to the operating characteristics of the vehicle-mounted range extender generator system. The coefficient analysis module is used to analyze the aging process of the cells inside the battery pack based on the system operating status and the characteristics of each power ripple, using a preset correlation judgment model to obtain the overall health coefficient of the battery pack. The threshold determination module is used to determine the battery pack warning threshold based on the system operating status and the overall health coefficient of the battery pack. The status detection module is used to compare each power ripple feature with the battery pack warning threshold to obtain the detection result of the battery pack health status.

Images