A method and system for monitoring the health of batteries throughout their entire lifecycle at battery swapping stations

Abstract

This invention relates to the field of new energy vehicle technology and discloses a method and system for monitoring the full life cycle health of batteries in battery swapping stations. The method includes: applying a small sinusoidal current within a preset charging window and collecting infrared temperature sequences during a historical period; calculating the regional average temperature; extracting the amplitude and phase of the first and second harmonics; establishing a phase-penetration mapping; forming a thermal state index; and determining the crossover frequency of the two harmonic phases. A lifespan function is constructed by combining the battery's operating data. During the monitoring phase, the predicted crossover frequency value of the target battery is obtained to estimate the lifespan, and real-time visual monitoring is achieved through cloud collaboration, improving prediction accuracy and battery swapping safety.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle technology, and more specifically, to a method and system for monitoring the health of batteries throughout their entire life cycle at battery swapping stations. Background Technology

[0002] With the widespread adoption of battery swapping stations in the electric vehicle industry, power batteries are frequently transferred between different vehicles, operating in complex and variable environments, including rapid temperature changes, humidity and salt spray effects, and different heat dissipation conditions. This diverse operating environment makes battery health monitoring a core issue in full lifecycle management. Existing monitoring methods mostly rely on electrical data such as voltage, current, and capacity, or obtain information through limited temperature measurements. However, these methods are insufficient to reflect the true thermal changes inside the battery, let alone intuitively reveal the internal loss process that occurs as the battery's lifespan degrades.

[0003] Currently, common methods for assessing battery health mainly include capacity degradation estimation and impedance testing. Capacity estimation requires long-term accumulation of charge and discharge data, which is difficult to achieve in battery swapping scenarios. Impedance testing often requires additional specialized equipment or complex algorithm calculations, which cannot be completed quickly during the battery swapping process. Furthermore, voltage and current signals are affected by vehicle operation and environmental fluctuations, exhibiting a certain degree of lag and failing to quickly reflect early changes in battery health. This makes existing full lifecycle health monitoring methods significantly inadequate for battery swapping station scenarios.

[0004] The heat generated by a power battery includes irreversible heat related to the square of the current and internal resistance, as well as reversible heat related to the direction and state of charge and discharge. As the battery is cycled, its internal resistance gradually increases, and the proportion of irreversible heat generation continuously rises. This change often precedes the manifestation of electrical indicators such as capacity decay, thus possessing the potential to serve as a predictive signal of health status. However, due to differences in heat dissipation conditions, environment, and surface characteristics, conventional temperature point measurements or passive infrared imaging are easily interfered with, making it impossible to maintain comparability between different battery swapping stations. This hinders the establishment of a reliable full-lifecycle health monitoring and prediction mechanism based on these thermal changes. Summary of the Invention

[0005] This invention provides a method and system for monitoring the health of batteries throughout their entire lifecycle at battery swapping stations, addressing the technical problems mentioned in the background section.

[0006] This invention provides a method for monitoring the health of batteries throughout their entire lifecycle at battery swapping stations, comprising:

[0007] Within a historical period:

[0008] Within a preset charging window, small sinusoidal currents are applied to the battery sequentially according to a preset frequency set to collect the corresponding infrared temperature sequence of the battery, and further obtain the regional average temperature.

[0009] For the average temperature of the region corresponding to each frequency in the preset frequency set, the amplitude and phase of the first harmonic and the amplitude and phase of the second harmonic are extracted respectively.

[0010] A phase-equal penetration mapping is established on the dimension of the first harmonic phase using the second harmonic phase;

[0011] The equal-phase first harmonic frequency corresponding to each second harmonic frequency is determined based on the phase equal-penetration mapping; the thermal state index sequence is constructed by the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency to further determine the double harmonic phase crossover frequency.

[0012] The battery's operating volume is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle.

[0013] During the monitoring period:

[0014] Obtain the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted lifetime of the target battery;

[0015] Through collaborative interaction between the cloud and the battery swapping station, the battery can be visualized and monitored based on the predicted lifespan of the target battery.

[0016] Furthermore, within a preset charging window, small sinusoidal currents are applied to the battery sequentially according to a preset frequency set to collect the corresponding infrared temperature sequence of the battery, thereby obtaining the regional average temperature, including:

[0017] Within a preset charging window, a small sinusoidal current of the corresponding frequency is applied to the battery sequentially according to each frequency in a preset frequency set, while the infrared temperature field on the battery surface is collected simultaneously; wherein, the infrared temperature field is a sequence containing multiple infrared frames;

[0018] Under a fixed observation angle and a fixed observation distance, semantic segmentation is performed on each infrared frame in the infrared temperature field to generate a battery binary matrix; where element 1 represents the battery region in the infrared frame and element 0 represents the non-battery region in the infrared frame.

[0019] For each infrared frame, calculate the average temperature of the region:

[0020] Multiply the temperature value of each pixel in the infrared frame by the corresponding element in the battery binary matrix, and sum all the product results to obtain the total sum of pixel temperature products; divide the total sum of pixel temperature products by the number of 1s in the battery binary matrix to obtain the regional average temperature of the infrared frame.

[0021] The regional average temperature is calculated for all infrared frames at each frequency, resulting in a regional average temperature sequence for each frequency.

[0022] Furthermore, for the regional average temperature corresponding to each frequency in the preset frequency set, the amplitude and phase of the first harmonic and the amplitude and phase of the second harmonic are extracted, including:

[0023] Determine each frequency Corresponding reference phase And calculate:

[0024] First harmonic cosine reference ;

[0025] First harmonic sine reference ;

[0026] Second harmonic cosine reference ;

[0027] Second harmonic sine reference ;

[0028] Where n represents frequency The nth infrared frame;

[0029] The first-harmonic quadrature components are calculated based on the first-harmonic cosine reference and the first-harmonic sine reference, as follows:

[0030]

[0031]

[0032] in, Represents frequency The next harmonic cosine quadrature component, Represents frequency Next harmonic sinusoidal quadrature components, Represents frequency Total number of infrared frames below Represents frequency The regional average temperature series below, Represents frequency The nth infrared frame with the Hanning window applied below;

[0033] The second harmonic quadrature components are calculated based on the second harmonic cosine reference and the second harmonic sine reference, as follows:

[0034]

[0035]

[0036] in, Represents frequency Lower second harmonic cosine quadrature components, Represents frequency Lower second harmonic sinusoidal quadrature components;

[0037] Calculate frequency The corresponding first harmonic amplitude First harmonic phase ;

[0038] Calculate frequency The corresponding second harmonic amplitude Second harmonic phase .

[0039] Furthermore, a phase-equal penetration mapping is established from the second harmonic phase on the first harmonic phase dimension, including:

[0040] For each first harmonic phase and second harmonic phase, perform phase expansion processing within the range of zero to two times the circumferential phase to obtain the corresponding first harmonic expanded phase and second harmonic expanded phase;

[0041] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the first harmonic to obtain the first monotonic phase function.

[0042] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the second harmonic to obtain the second monotonic phase function.

[0043] The inverse function of the first monotonic phase function is calculated to obtain the inverse function, and the domain of the inverse function is limited to the range of the second monotonic phase function to ensure that each second target harmonic phase corresponds to a unique first target harmonic frequency;

[0044] Each frequency in the preset frequency set is used as the second target harmonic frequency;

[0045] The second target harmonic phase corresponding to the second target harmonic frequency is calculated using the second monotonic phase function.

[0046] The harmonic frequency of the first target corresponding to the harmonic phase of the second target is calculated using an inverse function.

[0047] Each second target harmonic frequency in the preset frequency set is paired with its corresponding first target harmonic frequency to form a mapping pair.

[0048] Furthermore, based on the phase equal penetration mapping, the corresponding equal-phase first harmonic frequency is determined for each second harmonic frequency; the second harmonic amplitude and the corresponding first harmonic amplitude of the equal-phase first harmonic frequency constitute a thermal state index sequence, including:

[0049] Within the frequency range corresponding to the preset frequency set, the amplitude of the first harmonic is interpolated by monotonic spline to obtain the first monotonic amplitude function;

[0050] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the second harmonic amplitude to obtain the second monotonic amplitude function;

[0051] For each mapping pair, perform the following operations:

[0052] Input the second target harmonic frequency in the mapping pair into the second monotonic amplitude function to obtain the second target harmonic amplitude corresponding to the second target harmonic frequency;

[0053] The first target harmonic frequency in the mapping pair is input into the first monotonic amplitude function along with the equal phase first target harmonic frequency of the first monotonic phase function to obtain the corresponding first target harmonic amplitude.

[0054] The thermal state index is obtained by dividing the second target harmonic amplitude by the first target harmonic amplitude.

[0055] Furthermore, the double-harmonic phase crossover frequency is further determined, including:

[0056] Determine the second target harmonic frequency corresponding to each thermal state index;

[0057] Each second target harmonic frequency and thermal state index is combined to form a binary pair;

[0058] Arrange the binary pairs in ascending order according to the second target harmonic frequency to form a binary pair sequence;

[0059] Traverse each binary pair until you find the first binary pair in which the second target harmonic frequency is ≤1 and the target harmonic frequency in the next binary pair is ≥1. Then take the corresponding binary pair sequence as the target pair.

[0060] Calculate the logarithmic frequency of the second target harmonic frequency in the target pair, and take the natural exponent of the logarithmic frequency to obtain the double harmonic phase crossover frequency.

[0061] Furthermore, the battery's operating data is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle, including:

[0062] Measure the actual lifespan of each battery in the battery swapping station and use the actual lifespan as the corresponding operating quantity of the battery;

[0063] The operating quantity of each battery and the dual-harmonic phase crossover frequency are combined to form a matching pair;

[0064] Using the double-harmonic phase crossover frequency as the independent variable and the running quantity in the matched pair as the dependent variable, the lifetime function corresponding to the entire life cycle is obtained by fitting all matched pairs.

[0065] Furthermore, the predicted value of the dual harmonic phase crossover frequency of the target battery is obtained to determine the predicted lifetime of the target battery, including:

[0066] Obtain the dual harmonic phase crossover frequency of the target battery;

[0067] The biharmonic phase crossover frequency of the target battery is input into the lifetime function to obtain the corresponding predicted lifetime.

[0068] Furthermore, through collaborative interaction between the cloud and the battery swapping station, visualized monitoring of the battery is achieved based on the predicted lifespan of the target battery, including:

[0069] Send the predicted lifespan of the target battery to the cloud;

[0070] When the predicted lifespan of the target battery is less than or equal to the preset lifespan threshold, a monitoring alarm is sent to the battery swapping station. The alarm includes the battery number of the target battery and the predicted lifespan of the target battery.

[0071] Secondly, a battery lifecycle health monitoring system for a battery swapping station, implementing the battery lifecycle health monitoring method for a battery swapping station as described in any one of the claims, includes:

[0072] Within a historical period:

[0073] The data acquisition module applies small sinusoidal currents to the battery sequentially according to a preset frequency set within a preset charge window to acquire the infrared temperature sequence corresponding to the battery, and further obtains the regional average temperature.

[0074] The harmonic extraction module extracts the first harmonic amplitude and the first harmonic phase, and the second harmonic amplitude and the second harmonic phase, respectively, for the average temperature of the region corresponding to each frequency in the preset frequency set.

[0075] The phase mapping module establishes a phase-equal penetration mapping on the dimension of the first harmonic phase using the second harmonic phase;

[0076] The phase crossover module determines the equal-phase first harmonic frequency corresponding to each second harmonic frequency based on the phase equal penetration mapping; the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency constitute a thermal state index sequence to further determine the dual-harmonic phase crossover frequency.

[0077] The lifetime function construction module obtains the battery's operating data, combines it with the corresponding dual-harmonic phase crossover frequency for pairing and fitting, and forms the lifetime function corresponding to the entire life cycle.

[0078] During the monitoring period:

[0079] The battery life prediction module obtains the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted life of the target battery.

[0080] The battery life monitoring module enables visualized monitoring of batteries based on the predicted lifespan of the target battery through collaborative interaction between the cloud and the battery swapping station.

[0081] The beneficial effects of this invention are as follows: By applying a small sinusoidal current within the controlled charging window of the battery and combining it with infrared temperature field data, the amplitude and phase of the double harmonics are extracted to establish a phase-equal penetration mapping, thereby obtaining the thermal state index and the double harmonic phase crossover frequency, and constructing a full life cycle life function based on this. This application breaks through the limitations of traditional methods that rely on voltage, current, capacity, or single-point temperature detection, and can reveal health degradation signals caused by increased battery internal resistance and changes in heating patterns earlier and more accurately, and achieve cross-site collaborative visual monitoring in the cloud. Therefore, this application significantly improves the accuracy and real-time performance of battery full life cycle health prediction in battery swapping station scenarios, ensuring the safety and controllability of battery operation. Attached Figure Description

[0082] Figure 1 This is a flowchart of a battery lifecycle health monitoring method for battery swapping stations according to the present invention;

[0083] Figure 2 This is a block diagram of a battery lifecycle health monitoring system for a battery swapping station according to the present invention. Detailed Implementation

[0084] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0085] like Figure 1 As shown, a method for monitoring the health of batteries throughout their entire lifecycle at a battery swapping station includes:

[0086] Within a historical period:

[0087] Within a preset charging window, small sinusoidal currents are applied to the battery sequentially according to a preset frequency set to collect the corresponding infrared temperature sequence of the battery, and further obtain the regional average temperature.

[0088] For the average temperature of the region corresponding to each frequency in the preset frequency set, the amplitude and phase of the first harmonic and the amplitude and phase of the second harmonic are extracted respectively.

[0089] A phase-equal penetration mapping is established on the dimension of the first harmonic phase using the second harmonic phase;

[0090] The equal-phase first harmonic frequency corresponding to each second harmonic frequency is determined based on the phase equal-penetration mapping; the thermal state index sequence is constructed by the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency to further determine the double harmonic phase crossover frequency.

[0091] The battery's operating volume is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle.

[0092] During the monitoring period:

[0093] Obtain the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted lifetime of the target battery;

[0094] Through collaborative interaction between the cloud and the battery swapping station, the battery can be visualized and monitored based on the predicted lifespan of the target battery.

[0095] In one embodiment of the present invention, within a preset charging window, small sinusoidal currents are sequentially applied to the battery according to a preset frequency set to collect the infrared temperature sequence corresponding to the battery, and further obtain the regional average temperature, including:

[0096] Within a preset charging window, a small sinusoidal current of the corresponding frequency is applied to the battery sequentially according to each frequency in a preset frequency set, while the infrared temperature field on the battery surface is collected simultaneously; wherein, the infrared temperature field is a sequence containing multiple infrared frames;

[0097] Under a fixed observation angle and a fixed observation distance, semantic segmentation is performed on each infrared frame in the infrared temperature field to generate a battery binary matrix; where element 1 represents the battery region in the infrared frame and element 0 represents the non-battery region in the infrared frame.

[0098] For each infrared frame, calculate the average temperature of the region:

[0099] Multiply the temperature value of each pixel in the infrared frame by the corresponding element in the battery binary matrix, and sum all the product results to obtain the total sum of pixel temperature products; divide the total sum of pixel temperature products by the number of 1s in the battery binary matrix to obtain the regional average temperature of the infrared frame.

[0100] The regional average temperature is calculated for all infrared frames at each frequency, resulting in a regional average temperature sequence for each frequency.

[0101] In detail, the state of charge (SOC) is a key factor affecting the battery's internal resistance and thermal characteristics. Presetting a SOC window ensures that the interference of SOC on the thermal response is consistent across different data collection times and different batteries, avoiding incomparable temperature data due to SOC fluctuations. For example, if the window is set to 0.5 to 0.6 (i.e., SOC 50% to 60%), all data collection will be performed within this range, eliminating differences in SOC.

[0102] In detail, it refers to a small-amplitude sinusoidal current. "Small-amplitude" means that the current amplitude is controlled within a range that does not cause a significant temperature rise in the battery (avoiding the risk of thermal runaway) while still producing a detectable temperature response, thus ensuring battery safety and ensuring a linear correlation between the temperature response and the current excitation. "Sinusoidal current" is chosen because a sinusoidal signal has a single frequency component, avoiding frequency aliasing problems caused by complex waveforms.

[0103] In detail, the preset frequency set consists of multiple positive frequencies arranged in ascending order. Applying these frequencies sequentially avoids frequency interference caused by the simultaneous action of currents of different frequencies, ensuring that the temperature field collected at each frequency corresponds only to the excitation response of the current at that frequency, thus guaranteeing the purity of the single-frequency thermal signal.

[0104] In detail, a single infrared frame is easily affected by environmental noise, while a continuous infrared frame sequence can reduce noise interference through subsequent processing (such as averaging and filtering), and at the same time can reflect the dynamic changes of temperature over time, which is consistent with the characteristic of temperature changing periodically over time under sinusoidal current excitation.

[0105] In detail, the pixel range of the battery area in the infrared image is directly related to the observation angle and distance. Changes in the angle or distance will cause the same battery to appear in different pixel positions and sizes in the infrared frame. Fixing these two parameters can ensure that the battery area is positioned consistently at the pixel level in different infrared frames and different acquisition scenarios.

[0106] In detail, infrared imaging includes non-battery components (such as wires and supports) and the environmental background around the battery. The temperature of these non-battery areas is unrelated to the battery's own heating; if they are not removed, the temperature data will contain a lot of interference. Semantic segmentation uses image recognition algorithms (deep learning segmentation) to distinguish between battery and non-battery areas.

[0107] In detail, the pixel temperatures outside the battery area are automatically eliminated through multiplication, retaining only the pixel temperatures of the battery body area to ensure that all temperature data used in the calculation comes from the battery's own heat generation. A summation operation yields the cumulative temperature value of all pixels within the battery body; dividing by the number of valid pixels normalizes the cumulative value to obtain the average temperature of the battery body. This average temperature avoids misjudging the overall thermal response due to abnormal temperatures of individual pixels and better reflects the overall thermal state of the battery.

[0108] In detail, the current excitation at each frequency will cause the battery to produce periodic temperature changes at the corresponding frequency, which will be reflected in continuous infrared frames. By generating a regional average temperature sequence for each frequency, the average temperature in the spatial dimension can be combined with the changes in the temporal dimension to form a data form that reflects the dynamic response of battery temperature over time under a specific frequency excitation.

[0109] In one embodiment of the present invention, for the average regional temperature corresponding to each frequency in a preset frequency set, the first harmonic amplitude and the first harmonic phase, and the second harmonic amplitude and the second harmonic phase are extracted respectively, including:

[0110] Determine each frequency Corresponding reference phase And calculate:

[0111] First harmonic cosine reference ;

[0112] First harmonic sine reference ;

[0113] Second harmonic cosine reference ;

[0114] Second harmonic sine reference ;

[0115] Where n represents frequency The nth infrared frame;

[0116] Detailed reference phase This is to compensate for the phase deviation caused by system delay. In the entire link of current excitation application and infrared temperature acquisition, there are signal transmission delays (such as current control module response delay and infrared sensor imaging delay). These delays will cause the phase of the harmonic components in the temperature sequence to shift from the theoretical phase of current excitation and temperature response.

[0117] The reference phase needs to be based on the previous calibration experiment, that is, on a standard battery with a known healthy state, a specific frequency current is applied and the temperature sequence is collected. By comparing the difference between the theoretical excitation phase and the actual temperature harmonic phase, the reference phase at that frequency is determined. The reference phase remains fixed in subsequent acquisitions at the same frequency to ensure that the data collected at different times and from different batteries are comparable in phase.

[0118] In detail, the reference signal is a periodic signal that perfectly matches the target harmonic frequency, and its design is based on the harmonic characteristics of the battery's temperature response: that is, when a frequency of [missing information] is applied to the battery... When the current is sinusoidal, the battery's heating power (proportional to the square of the current) will include 2. The frequency components (due to sin²x=(1-cos2x) / 2) thus lead to the simultaneous presence of these components in the temperature response. (First harmonic, generated by the dynamic balance between heat generation and heat dissipation) and 2 (Second harmonic, directly generated by the periodicity of the square of the current) harmonic components.

[0119] The specific design logic for the four reference signals is as follows:

[0120] First harmonic sine / cosine reference: frequency is This frequency coincides with the first harmonic frequency in the temperature response and is used to extract the fundamental component at that frequency; in the phase term... The corresponding signal changes periodically over time ( (timestamp of the nth frame), superimposed Implement system delay compensation.

[0121] Second harmonic sine and cosine reference: frequency is 2 (i.e., 4πf) k The corresponding angular frequency is consistent with the second harmonic frequency in the temperature response, and is used to extract the second harmonic component; the phase term 2 This is because the phase shift of the second harmonic is twice that of the first harmonic (the system delay has a linear relationship with the phase of different frequency harmonics), and synchronous compensation is required to ensure phase alignment.

[0122] The first-harmonic quadrature components are calculated based on the first-harmonic cosine reference and the first-harmonic sine reference, as follows:

[0123]

[0124]

[0125] in, Represents frequency The next harmonic cosine quadrature component, Represents frequency Next harmonic sinusoidal quadrature components, Represents frequency Total number of infrared frames below Represents frequency The regional average temperature series below, Represents frequency The nth infrared frame with the Hanning window applied below;

[0126] In detail, the Hanning window is used to eliminate spectral leakage interference. Since infrared frame acquisition is a discrete process, if the period of the temperature sequence does not strictly match the number of acquisition frames, the energy of harmonic frequencies will diffuse to adjacent frequencies (i.e., spectral leakage), affecting the accuracy of amplitude and phase calculations. The Hanning window is a smooth window function whose coefficient gradually increases from 0 to 1 and then decreases back to 0. By suppressing abrupt changes at both ends of the sequence, it significantly reduces spectral leakage, ensuring that the extracted harmonic components are energy-concentrated.

[0127] In detail, the first harmonic reference is strongly correlated only with the first harmonic component in the temperature series, and weakly correlated or uncorrelated with other frequency components (such as DC component, noise, and second harmonic component). By multiplying and summing, the energy of the first harmonic component can be accumulated, while other irrelevant components are suppressed, thus achieving frequency screening.

[0128] coefficient The result is used for normalization. The summation result is proportional to the number of acquisition frames N. k Proportional (the more frames, the larger the sum), multiplied by It can eliminate the influence of frame number on component size, making the amplitude of quadrature components at different frequencies and frame numbers comparable; the coefficient 2 is because the effective value of the cosine and sine reference signals is √2, and multiplying by 2 can make the amplitude of the final quadrature component match the actual amplitude of the harmonic.

[0129] The second harmonic quadrature components are calculated based on the second harmonic cosine reference and the second harmonic sine reference, as follows:

[0130]

[0131]

[0132] in, Represents frequency Lower second harmonic cosine quadrature components, Represents frequency Lower second harmonic sinusoidal quadrature components;

[0133] In detail, the frequency of the second harmonic sine / cosine reference is 2. It is strongly correlated only with the second harmonic component in the temperature sequence. The energy of the second harmonic can be separated by orthogonal correlation operation, and other frequency components can be suppressed.

[0134] Calculate frequency The corresponding first harmonic amplitude First harmonic phase ;

[0135] Calculate frequency The corresponding second harmonic amplitude Second harmonic phase .

[0136] In detail, the first harmonic sine / cosine reference (or second harmonic sine / cosine reference) represents the components of the harmonic in two orthogonal directions (cosine and sine), much like the two legs of a right triangle; the amplitude is like the hypotenuse of the right triangle, representing the intensity of the harmonic response. A larger amplitude indicates a more significant temperature harmonic response at that frequency, directly reflecting the battery's thermal characteristics at that frequency. For example, a higher internal resistance may result in a larger first harmonic amplitude.

[0137] Phase reflects the degree of lag in the temperature harmonic response relative to the current excitation. The atan² function can determine the specific value of the phase in the range of 0 to 2π based on the sign of the orthogonal components. Phase lag is mainly caused by the impedance characteristics of the battery (the combined effect of resistance, capacitance, and inductance). The phase lag will change systematically for batteries in different health states (such as the increased internal resistance of aging batteries).

[0138] In one embodiment of the present invention, establishing a phase-equal penetration mapping based on the second harmonic phase on the first harmonic phase dimension includes:

[0139] For each first harmonic phase and second harmonic phase, perform phase expansion processing within the range of zero to two times the circumferential phase to obtain the corresponding first harmonic expanded phase and second harmonic expanded phase;

[0140] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the first harmonic to obtain the first monotonic phase function.

[0141] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the second harmonic to obtain the second monotonic phase function.

[0142] In detail, the original extracted first and second harmonic phases exhibit a periodic jump problem. Because the phase is periodic, it restarts counting from 0 after exceeding 2π. For example, when the frequency increases, causing the phase lag to increase from around 2π to slightly greater than 2π, the original phase data will suddenly jump to 0, forming a phase discontinuity. This phase discontinuity disrupts the continuity of phase variation with frequency.

[0143] Phase unrolling eliminates periodic jumps. The algorithm identifies jump points in the original phase that exceed 2π or fall below 0. The phase value after the jump is superimposed or subtracted by 2π, so that the phase data shows a continuous monotonic change trend in the range of 0 to 2π.

[0144] In detail, the frequencies in the preset frequency set are discrete, and the phase of the first harmonic expansion only corresponds to these discrete frequencies. It cannot reflect the phase value at any frequency between two discrete frequencies. Therefore, it is necessary to obtain the phase relationship in the continuous frequency range through interpolation.

[0145] In detail, since the phase change with frequency has physical monotonicity, monotonic spline interpolation is chosen. In a battery, the higher the frequency, the shallower the current penetrates the battery, and the smaller the phase lag (due to the impedance characteristics of the shallow surface layer, the phase response is faster). Therefore, the phase of the first harmonic unfolded shows a monotonically decreasing trend with increasing frequency (or a monotonically increasing trend in specific scenarios, depending on the battery structure, but overall it must be monotonic). Monotonic spline interpolation can ensure the continuous smoothness of the interpolation function while strictly adhering to monotonicity, avoiding fluctuations where the phase increases with increasing frequency, and ensuring that the first monotonic phase function can truly reflect the intrinsic relationship between the phase and frequency of the first harmonic.

[0146] Similarly, calculate the second monotonic phase function.

[0147] The inverse function of the first monotonic phase function is calculated to obtain the inverse function, and the domain of the inverse function is limited to the range of the second monotonic phase function to ensure that each second target harmonic phase corresponds to a unique first target harmonic frequency;

[0148] The first monotonic phase function is a function of frequency to phase (input frequency, output phase). What is needed is to know the second harmonic phase and find the corresponding first harmonic frequency, that is, the phase to frequency mapping. Therefore, the input-output relationship must be reversed through an inverse function.

[0149] The domain of the second monotonic phase function is the range of all possible second harmonic phases (since the second target harmonic phases all originate from this function). If the domain of the inverse function is not limited, the input second target harmonic phase may exceed the effective range of the inverse function, resulting in the inability to find the corresponding first target harmonic frequency, or finding multiple frequencies (violating uniqueness). Limiting the domain of the inverse function to the domain of the second monotonic phase function ensures that each second target harmonic phase is within the effective input range of the inverse function. Furthermore, since the first monotonic phase function is monotonic (and the inverse function of a monotonic function is also monotonic), each phase must correspond to a unique frequency, thus guaranteeing the uniqueness and validity of the mapping.

[0150] Each frequency in the preset frequency set is used as the second target harmonic frequency;

[0151] In detail, the preset frequency set serves as the frequency reference. Previous current excitation, temperature acquisition, and harmonic extraction were all based on frequencies in this set. These frequencies were selected as the second target harmonic frequencies to ensure data consistency and physical correlation throughout the entire process.

[0152] The second target harmonic phase corresponding to the second target harmonic frequency is calculated using the second monotonic phase function.

[0153] In detail, the second target harmonic phase is the input parameter for establishing the mapping. The corresponding first harmonic frequency needs to be found based on the second harmonic phase; therefore, it is necessary to first obtain the accurate phase value corresponding to each second target harmonic frequency. Since the second monotonic phase function is continuous, even if the second target harmonic frequency is a discrete preset frequency, an accurate phase value can be obtained through function calculation, avoiding errors from discrete phase data.

[0154] The harmonic frequency of the first target corresponding to the harmonic phase of the second target is calculated using an inverse function.

[0155] In detail, the phase of the second target harmonic frequency is equal to the phase corresponding to the first target harmonic frequency (because the first target harmonic frequency comes from the inverse function, the input of the inverse function is the phase of the second target harmonic, and the phase obtained by substituting the output frequency into the first monotonic phase function must be equal to the input phase).

[0156] Equal phase means equal current penetration depth. In a battery, harmonic phase lag reflects the impedance response when current propagates inside the battery, and the same phase lag corresponds to the same propagation depth (i.e., penetration depth). Therefore, the current penetration depth corresponding to the first target harmonic frequency found through the inverse function is exactly equal to the penetration depth corresponding to the second target harmonic frequency, achieving alignment of equal penetration depth.

[0157] Each second target harmonic frequency in the preset frequency set is paired with its corresponding first target harmonic frequency to form a mapping pair.

[0158] In one embodiment of the present invention, the equal-phase first harmonic frequency corresponding to each second harmonic frequency is determined according to the phase equal-penetration mapping; a thermal state index sequence is constructed by the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency, including:

[0159] Within the frequency range corresponding to the preset frequency set, the amplitude of the first harmonic is interpolated by monotonic spline to obtain the first monotonic amplitude function;

[0160] In detail, the first harmonic amplitude corresponds only to discrete frequency points in a preset frequency set, while the first target harmonic frequency obtained through phase equal penetration mapping may not be within this discrete set (for example, the preset frequencies are 1Hz and 3Hz, while the first target harmonic frequency may be 2.2Hz). If discrete amplitude data is used directly, it is impossible to obtain the first harmonic amplitude at these non-preset frequencies, resulting in the inability to complete the amplitude comparison at equal penetration depths.

[0161] Since the amplitude of the first harmonic exhibits physical monotonicity with frequency, monotonic spline interpolation is chosen. In a battery, the higher the current frequency, the shallower the penetration depth (acting only on the battery surface). The thermal resistance and thermal capacity characteristics of the surface are relatively stable, and the thermal response generated by the current decays faster at high frequencies. Therefore, the amplitude of the first harmonic decreases monotonically with increasing frequency (or increases monotonically in specific scenarios, depending on the battery structure, but overall it is monotonic). Monotonic spline interpolation ensures the interpolation function is continuous and smooth while strictly adhering to monotonicity, avoiding fluctuations where the amplitude increases with increasing frequency. This ensures that the first monotonic amplitude function accurately outputs the amplitude of the first harmonic at any frequency, and that this amplitude conforms to the actual thermal response of the battery.

[0162] Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the second harmonic amplitude to obtain the second monotonic amplitude function;

[0163] In detail, the second harmonic is generated by the periodic heating of the square of the current, and its amplitude also has physical monotonicity as the frequency changes (frequency increases → penetration depth decreases → thermal response intensity changes monotonically). Therefore, a continuous second monotonic amplitude function needs to be obtained by monotonic spline interpolation.

[0164] For each mapping pair, perform the following operations:

[0165] Input the second target harmonic frequency in the mapping pair into the second monotonic amplitude function to obtain the second target harmonic amplitude corresponding to the second target harmonic frequency;

[0166] The first target harmonic frequency in the mapping pair is input into the first monotonic amplitude function along with the equal phase first target harmonic frequency of the first monotonic phase function to obtain the corresponding first target harmonic amplitude.

[0167] The thermal state index is obtained by dividing the second target harmonic amplitude by the first target harmonic amplitude.

[0168] In detail, the second target harmonic amplitude is the benchmark for the second harmonic response intensity in the comparison of equal penetration depths:

[0169] The second target harmonic frequency comes from a preset frequency set and serves as the reference frequency for current excitation and temperature acquisition. Its corresponding second harmonic amplitude directly reflects the battery's thermal response at that frequency (related to battery internal resistance and polarization characteristics). Calculation using a second monotonic amplitude function avoids deviations in discrete amplitudes caused by noise and acquisition errors. The continuous function smooths the fluctuations in discrete data through interpolation, resulting in the output second target harmonic amplitude that better represents the battery's true thermal response intensity at that frequency.

[0170] The first target harmonic frequency is in phase with the second target harmonic frequency (with the same penetration depth). Therefore, the corresponding first target harmonic amplitude is the thermal response intensity of the first harmonic at the same penetration depth.

[0171] If the same first harmonic frequency as the second target harmonic frequency is directly used (e.g., if the second target harmonic frequency is 2Hz, directly using the 2Hz first harmonic amplitude), the penetration depth will differ due to the frequency difference. In this case, the amplitude ratio includes frequency interference and cannot reflect the true internal state of the battery. However, the first target harmonic frequency is obtained through phase-equal penetration mapping, ensuring consistent penetration depth and eliminating interference from frequency differences. Because the first target harmonic frequency may not be within the preset frequency set, its amplitude must be calculated using the first monotonic amplitude function.

[0172] In detail, thermal state indicators are related to battery health status:

[0173] The amplitude ratio is a dimensionless parameter that can eliminate interference from external factors such as excitation power (e.g., amplitude fluctuations of small sinusoidal currents), battery surface emissivity (interference from surface characteristics obtained by infrared temperature acquisition), and ambient temperature (which has a proportional effect on the first / second harmonic amplitude). These factors have a synchronous effect on the amplitudes of the second and first target harmonics, and the ratio calculation can cancel out their effects, retaining only information related to the internal characteristics of the battery.

[0174] Battery health degradation manifests as increased internal resistance, which leads to increased irreversible heat generation. The second harmonic amplitude is more strongly correlated with polarization heating (related to internal resistance), while the first harmonic amplitude is related to the overall thermal response. As the battery ages, the growth rate of the second harmonic amplitude is faster than that of the first harmonic amplitude, causing the thermal state indicators to change systematically with the degree of aging (usually showing an increasing trend). Therefore, the thermal state indicator sequence can directly reflect changes in battery health.

[0175] In one embodiment of the present invention, further determining the dual-harmonic phase crossover frequency includes:

[0176] Determine the second target harmonic frequency corresponding to each thermal state index;

[0177] Each second target harmonic frequency and thermal state index is combined to form a binary pair;

[0178] Arrange the binary pairs in ascending order according to the second target harmonic frequency to form a binary pair sequence;

[0179] Traverse each binary pair until you find the first binary pair in which the second target harmonic frequency is ≤1 and the target harmonic frequency in the next binary pair is ≥1. Then take the corresponding binary pair sequence as the target pair.

[0180] Calculate the logarithmic frequency of the second target harmonic frequency in the target pair, and take the natural exponent of the logarithmic frequency to obtain the double harmonic phase crossover frequency.

[0181] In detail, the thermal state index is the ratio of the amplitude of the second target harmonic to the amplitude of the first target harmonic of equal phase, and the calculation of this ratio depends on a specific second target harmonic frequency. Therefore, it is necessary to first establish the correspondence between the thermal state index and the second target harmonic frequency:

[0182] In detail, the thermal state indices of a battery exhibit a clear trend with frequency. Generally, the lower the frequency, the deeper the current penetration (acting within the battery), the larger the first harmonic amplitude (related to the overall thermal response), and the less than 1 the thermal state index (the ratio of the second to the first amplitude). As the frequency increases, the penetration depth becomes shallower (acting on the surface, where surface polarization is more pronounced), the second harmonic amplitude (related to polarization heat) increases faster, and the index gradually increases to greater than 1. Therefore, arranging the binary pairs in ascending order of frequency allows the index to exhibit a monotonically increasing trend with increasing frequency.

[0183] In detail, when the thermal state index sequence is traversed point by point with frequency, the thermal state index sequence transitions from less than or equal to 1 to greater than or equal to 1, which means that there must exist a frequency point in this interval such that the temperature response of the second harmonic is completely equal to the temperature response of the first harmonic under the condition of equal phase.

[0184] Therefore, traversing each pair of binary components until finding the first interval where the ratio of the previous pair is ≤ 1 and the ratio of the next pair is ≥ 1, essentially locates the equilibrium position where the thermal response shifts from second harmonic dominance to first harmonic dominance. This position corresponds to the critical point of the battery's internal thermal and chemical processes under its current operating state. Before this point, the second harmonic component is weak, mainly exhibiting low-order thermal diffusion behavior; after this point, the second harmonic component significantly strengthens, reflecting the increase in internal impedance and the cumulative effect of thermal nonlinearity.

[0185] In detail, in thermodynamics and diffusion processes, frequency often reflects the strength of material or interface processes by orders of magnitude. The logarithmic domain can compress nonlinear distributions into approximately linear relationships, thus more accurately capturing crossover points. The frequency solution obtained in the logarithmic domain actually corresponds to the critical position of equal energy penetration depth. Mapping this frequency solution back to the original frequency space through the natural exponent yields the biharmonic phase crossover frequency. At the biharmonic phase crossover frequency, the second harmonic nonlinear thermal effect and the first harmonic linear thermal effect of the battery reach equilibrium, reflecting the boundary between the battery's internal resistance evolution and thermal response mechanism. The drift process of the biharmonic phase crossover frequency is the most direct and sensitive quantitative indicator of the battery's degradation state throughout its entire life cycle.

[0186] In one embodiment of the present invention, the battery's operating data is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle, including:

[0187] Measure the actual lifespan of each battery in the battery swapping station and use the actual lifespan as the corresponding operating quantity of the battery;

[0188] The operating quantity of each battery and the dual-harmonic phase crossover frequency are combined to form a matching pair;

[0189] Using the double-harmonic phase crossover frequency as the independent variable and the running quantity in the matched pair as the dependent variable, the lifetime function corresponding to the entire life cycle is obtained by fitting all matched pairs.

[0190] In detail, a full lifecycle mapping of the lifetime function was established by introducing a pairing relationship between battery operating quantity and dual harmonic phase crossover frequency. Specifically, firstly, the actual lifespan of each battery was measured under the actual operating environment of the battery swapping station. This actual lifespan directly quantifies the real working cycles that the battery can complete, and thus serves as an accurate characterization of operating quantity. Subsequently, the operating quantity obtained by the battery at a specific testing stage was combined with the corresponding dual harmonic phase crossover frequency to form a one-to-one matching pair. Each matching pair reflects the battery's response under a specific degradation state. Furthermore, within the sample range of all batteries, these matching pairs were used to construct a global fit, with the dual harmonic phase crossover frequency as the independent variable and the operating quantity as the dependent variable, to obtain the complete lifetime function. The lifetime function can present the regular relationship between the operating quantity and the dual harmonic phase crossover frequency throughout the entire lifecycle, thus providing a mathematical and continuous mapping model to describe and predict the degradation trajectory of the battery from its initial state to the end of its lifespan.

[0191] In detail, the actual lifespan of each battery in the battery swapping station can be determined through cycle life testing.

[0192] In one embodiment of the present invention, obtaining the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted lifetime of the target battery includes:

[0193] Obtain the dual harmonic phase crossover frequency of the target battery;

[0194] The biharmonic phase crossover frequency of the target battery is input into the lifetime function to obtain the corresponding predicted lifetime.

[0195] In detail, the target battery is monitored and its biharmonic phase crossover frequency is calculated. The biharmonic phase crossover frequency reflects the critical equilibrium point between the thermal process and electrochemical impedance of the battery in its current state, and is a core indicator characterizing the battery's health evolution. Subsequently, the biharmonic phase crossover frequency is used as input to a pre-established lifetime function. This lifetime function is a full-lifecycle mapping model obtained by fitting the operating data of a large number of batteries with the biharmonic phase crossover frequency, and it can describe the continuous relationship between crossover frequency and lifetime under different states. Through this method, the lifetime function outputs the lifetime value corresponding to the biharmonic phase crossover frequency, directly giving the predicted lifetime of the target battery, realizing a quantitative process of deriving the future usable lifetime of the battery from physically measurable quantities.

[0196] In one embodiment of the present invention, through collaborative interaction between the cloud and the battery swapping station, visual monitoring of the battery is achieved based on the predicted lifespan of the target battery, including:

[0197] Send the predicted lifespan of the target battery to the cloud;

[0198] When the predicted lifespan of the target battery is less than or equal to the preset lifespan threshold, a monitoring alarm is sent to the battery swapping station. The alarm includes the battery number of the target battery and the predicted lifespan of the target battery.

[0199] In detail, the predicted lifespan of the target battery is transmitted to the cloud via a communication link. Upon receiving the data, the cloud stores and compares it in real time to ensure centralized management and scheduling of the predicted lifespan information. When the cloud determines that the predicted lifespan of the target battery is less than or equal to a set lifespan threshold, it means that the battery may not meet safety and performance requirements in the short term. The cloud then proactively issues a monitoring alarm command to the battery swapping station. This alarm command clearly includes the target battery's serial number and its corresponding predicted lifespan value, enabling the swapping station to immediately identify the specific battery and take targeted measures, such as arranging replacement in advance or strengthening operational monitoring. This collaborative interaction between the cloud and the swapping station not only ensures that lifespan prediction information is fully utilized at the system level but also forms a closed-loop mechanism from data collection and cloud-based judgment to station-level execution, thereby achieving visualized safety monitoring and proactive early warning throughout the battery's entire lifecycle.

[0200] Example 2

[0201] like Figure 2 As shown, a battery lifecycle health monitoring system for a battery swapping station, implementing any of the battery lifecycle health monitoring methods for battery swapping stations as described in any one of the claims, includes:

[0202] Within a historical period:

[0203] The data acquisition module applies small sinusoidal currents to the battery sequentially according to a preset frequency set within a preset charge window to acquire the infrared temperature sequence corresponding to the battery, and further obtains the regional average temperature.

[0204] The harmonic extraction module extracts the first harmonic amplitude and the first harmonic phase, and the second harmonic amplitude and the second harmonic phase, respectively, for the average temperature of the region corresponding to each frequency in the preset frequency set.

[0205] The phase mapping module establishes a phase-equal penetration mapping on the dimension of the first harmonic phase using the second harmonic phase;

[0206] The phase crossover module determines the equal-phase first harmonic frequency corresponding to each second harmonic frequency based on the phase equal penetration mapping; the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency constitute a thermal state index sequence to further determine the dual-harmonic phase crossover frequency.

[0207] The lifetime function construction module obtains the battery's operating data, combines it with the corresponding dual-harmonic phase crossover frequency for pairing and fitting, and forms the lifetime function corresponding to the entire life cycle.

[0208] During the monitoring period:

[0209] The battery life prediction module obtains the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted life of the target battery.

[0210] The battery life monitoring module enables visualized monitoring of batteries based on the predicted lifespan of the target battery through collaborative interaction between the cloud and the battery swapping station.

[0211] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A method for monitoring the health of batteries throughout their entire lifecycle at a battery swapping station, characterized in that, include: Within a historical timeframe: Within a preset charging window, small sinusoidal currents are applied to the battery sequentially according to a preset frequency set to collect the corresponding infrared temperature sequence of the battery, and further obtain the regional average temperature, including: Within a preset charging window, a small sinusoidal current of the corresponding frequency is applied to the battery sequentially according to each frequency in a preset frequency set, while the infrared temperature field on the battery surface is collected simultaneously; wherein, the infrared temperature field is a sequence containing multiple infrared frames; Under a fixed observation angle and a fixed observation distance, semantic segmentation is performed on each infrared frame in the infrared temperature field to generate a battery binary matrix; where element 1 represents the battery region in the infrared frame and element 0 represents the non-battery region in the infrared frame. For each infrared frame, calculate the average temperature of the region: Multiply the temperature value of each pixel in the infrared frame by the corresponding element in the battery binary matrix, and sum all the product results to obtain the total sum of pixel temperature products; divide the total sum of pixel temperature products by the number of 1s in the battery binary matrix to obtain the regional average temperature of the infrared frame. For all infrared frames at each frequency, the regional average temperature is calculated to obtain the regional average temperature sequence corresponding to each frequency. For the average regional temperature corresponding to each frequency in the preset frequency set, the amplitude and phase of the first harmonic and the amplitude and phase of the second harmonic are extracted, including: Determine each frequency Corresponding reference phase And calculate: First harmonic cosine reference ; First harmonic sine reference ; Second harmonic cosine reference ; Second harmonic sine reference ; Where n represents frequency The nth infrared frame; The first-harmonic quadrature components are calculated based on the first-harmonic cosine reference and the first-harmonic sine reference, as follows: in, Represents frequency The next harmonic cosine quadrature component, Represents frequency Next harmonic sinusoidal quadrature components, Represents frequency Total number of infrared frames below Represents frequency The regional average temperature series below, Represents frequency The nth infrared frame with the Hanning window applied below; The second harmonic quadrature components are calculated based on the second harmonic cosine reference and the second harmonic sine reference, as follows: in, Represents frequency Lower second harmonic cosine quadrature components, Represents frequency Lower second harmonic sinusoidal quadrature components; Calculate frequency The corresponding first harmonic amplitude First harmonic phase ; Calculate frequency The corresponding second harmonic amplitude Second harmonic phase ; A phase-equal penetration mapping is established on the dimension of the first harmonic phase using the second harmonic phase; The equal-phase first harmonic frequency corresponding to each second harmonic frequency is determined based on the phase equal-penetration mapping; the thermal state index sequence is constructed by the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency to further determine the double harmonic phase crossover frequency. The battery's operating volume is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle. During the monitoring period: Obtain the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted lifetime of the target battery; Through collaborative interaction between the cloud and the battery swapping station, the battery can be visualized and monitored based on the predicted lifespan of the target battery.

2. The method for monitoring the health of batteries throughout their entire lifecycle at a battery swapping station according to claim 1, characterized in that, A phase-equal penetration mapping is established from the second harmonic phase along the first harmonic phase dimension, including: For each first harmonic phase and second harmonic phase, perform phase expansion processing within the range of zero to two times the circumferential phase to obtain the corresponding first harmonic expanded phase and second harmonic expanded phase; Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the first harmonic to obtain the first monotonic phase function. Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the expanded phase of the second harmonic to obtain the second monotonic phase function. The inverse function of the first monotonic phase function is calculated to obtain the inverse function, and the domain of the inverse function is limited to the range of the second monotonic phase function to ensure that each second target harmonic phase corresponds to a unique first target harmonic frequency; Each frequency in the preset frequency set is used as the second target harmonic frequency; The second target harmonic phase corresponding to the second target harmonic frequency is calculated using the second monotonic phase function. The harmonic frequency of the first target corresponding to the harmonic phase of the second target is calculated using an inverse function. Each second target harmonic frequency in the preset frequency set is paired with its corresponding first target harmonic frequency to form a mapping pair.

3. The method for monitoring the health of a battery throughout its entire lifecycle at a battery swapping station according to claim 2, characterized in that, The equal-phase first harmonic frequency corresponding to each second harmonic frequency is determined based on the phase equal penetration mapping. A sequence of thermal state indicators is constructed using the second harmonic amplitude and the corresponding equal-phase first harmonic frequency's first harmonic amplitude, including: Within the frequency range corresponding to the preset frequency set, the amplitude of the first harmonic is interpolated by monotonic spline to obtain the first monotonic amplitude function; Within the frequency range corresponding to the preset frequency set, monotonic spline interpolation is performed on the second harmonic amplitude to obtain the second monotonic amplitude function. For each mapping pair, perform the following operations: Input the second target harmonic frequency in the mapping pair into the second monotonic amplitude function to obtain the second target harmonic amplitude corresponding to the second target harmonic frequency; The first target harmonic frequency in the mapping pair is input into the first monotonic amplitude function along with the equal phase first target harmonic frequency of the first monotonic phase function to obtain the corresponding first target harmonic amplitude. The thermal state index is obtained by dividing the second target harmonic amplitude by the first target harmonic amplitude.

4. The method for monitoring the health of a battery throughout its entire lifecycle at a battery swapping station according to claim 3, characterized in that, Further determination of the double-harmonic phase crossover frequency, including: Determine the second target harmonic frequency corresponding to each thermal state index; Each second target harmonic frequency and thermal state index is combined to form a binary pair; Arrange the binary pairs in ascending order according to the second target harmonic frequency to form a binary pair sequence; Traverse each binary pair until you find the first binary pair in which the second target harmonic frequency is ≤1 and the target harmonic frequency in the next binary pair is ≥1. Then take the corresponding binary pair sequence as the target pair. Calculate the logarithmic frequency of the second target harmonic frequency in the target pair, and take the natural exponent of the logarithmic frequency to obtain the double harmonic phase crossover frequency.

5. The method for monitoring the health of batteries throughout their entire lifecycle at a battery swapping station according to claim 4, characterized in that, The battery's operating data is obtained, and then matched and fitted with the corresponding dual-harmonic phase crossover frequency to form a lifetime function corresponding to the entire life cycle, including: Measure the actual lifespan of each battery in the battery swapping station and use the actual lifespan as the corresponding operating quantity of the battery; The operating quantity of each battery and the dual-harmonic phase crossover frequency are combined to form a matching pair; Using the double-harmonic phase crossover frequency as the independent variable and the running quantity in the matched pair as the dependent variable, the lifetime function corresponding to the entire life cycle is obtained by fitting all matched pairs.

6. The method for monitoring the health of a battery throughout its entire lifecycle at a battery swapping station according to claim 5, characterized in that, Obtain the predicted dual-harmonic phase crossover frequency of the target battery to determine the predicted lifetime of the target battery, including: Obtain the dual harmonic phase crossover frequency of the target battery; The biharmonic phase crossover frequency of the target battery is input into the lifetime function to obtain the corresponding predicted lifetime.

7. The method for monitoring the health of a battery throughout its entire lifecycle at a battery swapping station according to claim 6, characterized in that, Through collaborative interaction between the cloud and battery swapping stations, visualized monitoring of batteries is achieved based on the predicted lifespan of the target battery, including: Send the predicted lifespan of the target battery to the cloud; When the predicted lifespan of the target battery is less than or equal to the preset lifespan threshold, a monitoring alarm is sent to the battery swapping station. The alarm includes the battery number of the target battery and the predicted lifespan of the target battery.

8. A battery lifecycle health monitoring system for a battery swapping station, implementing the battery lifecycle health monitoring method for a battery swapping station as described in any one of claims 1-7, characterized in that, include: Within a historical timeframe: The data acquisition module applies small sinusoidal currents to the battery sequentially according to a preset frequency set within a preset charge window to acquire the infrared temperature sequence corresponding to the battery, and further obtains the regional average temperature. The harmonic extraction module extracts the first harmonic amplitude and the first harmonic phase, and the second harmonic amplitude and the second harmonic phase, respectively, for the average temperature of the region corresponding to each frequency in the preset frequency set. The phase mapping module establishes a phase-equal penetration mapping on the dimension of the first harmonic phase using the second harmonic phase; The phase crossover module determines the equal-phase first harmonic frequency corresponding to each second harmonic frequency based on the phase equal penetration mapping; the second harmonic amplitude and the first harmonic amplitude of the corresponding equal-phase first harmonic frequency constitute a thermal state index sequence to further determine the dual-harmonic phase crossover frequency. The lifetime function construction module obtains the battery's operating data, combines it with the corresponding dual-harmonic phase crossover frequency for pairing and fitting, and forms the lifetime function corresponding to the entire life cycle. During the monitoring period: The battery life prediction module obtains the predicted value of the dual harmonic phase crossover frequency of the target battery to determine the predicted life of the target battery. The battery life monitoring module enables visualized monitoring of batteries based on the predicted lifespan of the target battery through collaborative interaction between the cloud and the battery swapping station.

Images