A lithium battery state of health decay rate-based lease pricing dynamic adjustment system

Abstract

This invention relates to the field of intelligent processing technology for lithium battery management and leasing business, specifically a dynamic adjustment system for leasing pricing based on the decay rate of lithium battery health status; it includes: a value calibration module: using sparse Bayesian learning to analyze data flow impedance characteristics and establish a value benchmark; a loss model construction module: establishing a damage potential energy surface model and converting physical parameters into depreciation coordinates; a behavior quantification module: calculating the value dissipation gradient vector and the consumption behavior velocity vector; a pricing strategy generation module: calculating the pricing adjustment coefficient based on vector coupling and generating dynamic rate instructions; and a transaction and control module: issuing instructions and adjusting the power limit threshold according to risk. This invention utilizes the potential energy-damping principle to transform physical loss into a dual constraint of economic leverage and physical current limiting, forcing users to return to a safe potential energy zone and extending asset life.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lithium battery management and leasing business intelligent processing, in particular to a leasing pricing dynamic adjustment system based on lithium battery health state decay rate. BACKGROUND

[0002] In the new energy leasing application scenario, the leasing operation platform relies on an accurate asset value evaluation system to ensure the profitability and asset safety of the leasing business. The management system usually needs to combine battery operation data and user behavior characteristics to monitor the health state of the leased lithium battery in real time.

[0003] For leasing pricing adjustment, existing schemes generally use fixed rates or simple threshold-based step pricing architecture. That is, through the battery management system, basic physical quantities such as voltage and current are collected. Once the value exceeds the pre-set static threshold, a fixed penalty rate is charged, or only the remaining capacity SOH of the battery is used for periodic depreciation calculation. Although this scheme has certain feasibility under low frequency or standard working conditions, it relies too much on static rules and lacks dynamic mapping of the microscopic physical loss process. When encountering complex user behaviors such as high-frequency large-rate discharge, illegal overcharging and over-discharging, simple threshold judgment cannot quantify the irreversible damage to the internal electrochemical structure of the battery caused by these transient behaviors, resulting in a serious disconnection between the charging results and the actual asset depreciation. In addition, the traditional one-price or lagging depreciation evaluation mode cannot provide real-time feedback on the intensity of user operations, and cannot form an effective economic lever to constrain user behavior, resulting in hidden costs that cannot be covered in time, and thus causing premature asset scrap or operational loss risks. Therefore, how to establish a dynamic pricing mechanism based on the physical loss mechanism to accurately quantify the value dissipation caused by user behavior while achieving closed-loop control of economic regulation and physical limitation has become a technical problem that needs to be solved. SUMMARY

[0004] To solve the above technical problems, the present application provides a leasing pricing dynamic adjustment system based on lithium battery health state decay rate. Specifically, the technical solution of the present application includes:

[0005] Leasing asset data acquisition and value calibration module: used to obtain the operation data stream of the leased lithium battery, which is used as the input basis for asset valuation, and based on the sparse Bayesian learning algorithm to analyze the electrochemical impedance spectrum characteristics implied in the data stream to establish the asset micro-value benchmark for the current leasing period;

[0006] Asset loss model construction module: used to establish a damage potential surface model for quantifying leasing costs. This model maps the physical operation parameters of the lithium battery into asset depreciation dimension coordinates and defines the health state decay rate as the asset value dissipation potential.

[0007] a user consumption behavior quantification module for locating a current asset state point in the damage potential surface model, calculating a value dissipation gradient vector along a user usage behavior direction, and generating a consumption behavior velocity vector in combination with a change rate of the user behavior;

[0008] a dynamic pricing strategy generation module for calculating a pricing adjustment factor for adjusting a rental fee based on a coupling relationship between the value dissipation gradient vector and the consumption behavior velocity vector, and converting the pricing adjustment factor into a real-time dynamic fee rate instruction;

[0009] a transaction execution and risk control module for issuing the dynamic fee rate instruction to a settlement terminal to execute price adjustment, and synchronously adjusting a power limitation threshold of a battery management system according to a risk determination result, so as to form a closed-loop control for inhibiting high devaluation behavior of the asset by means of economic adjustment and physical limitation.

[0010] Preferably, the modules are realized through the following methods:

[0011] S1, collecting voltage, current and temperature data of the lithium battery in a billing period through a rental management system interface, and constructing a low-frequency operation data set for asset evaluation;

[0012] S2, inputting the low-frequency operation data set into a rental asset data collection and value calibration module, performing feature reconstruction operation, and generating a high-dimensional state matrix reflecting current physical properties of the asset;

[0013] S3, mapping the high-dimensional state matrix to a preset asset damage model, determining a current asset depreciation state coordinate, and analyzing a value dissipation potential gradient corresponding to the coordinate point;

[0014] S4, quantifying stress intensity applied by the user to the rental asset based on time dimension change of the current data, generating a consumption behavior velocity vector, and calculating a dot product of the consumption behavior velocity vector and the value dissipation potential gradient as a billing basis;

[0015] S5, calculating a pricing adjustment factor, i.e. a pricing adjustment factor, based on the calculation result of the dot product, using a nonlinear mapping function, and generating a dynamic fee rate instruction containing price adjustment information;

[0016] S6, sending the dynamic fee rate instruction to a user transaction terminal for publicity and settlement, and triggering a power dynamic clamping mechanism of the battery management system according to a risk level in the instruction.

[0017] Preferably, S1 specifically comprises:

[0018] S11, configuring a data capture unit at an output end of the rental asset, and recording voltage instantaneous value, current instantaneous value and temperature value in real time according to a preset billing sampling frequency;

[0019] S12, time stamp alignment is performed on the collected transaction basis data, and invalid data frames are removed to ensure the integrity of the billing data;

[0020] S13, normalization processing is performed on the time stamp aligned data to eliminate dimensional differences, and a standardized low-frequency operation data set is constructed for subsequent valuation algorithm calling.

[0021] Preferably, S2 specifically comprises:

[0022] S21, an inverse problem solving model based on sparse Bayesian learning is constructed, the low-frequency operation data set is taken as an asset observation vector, and the electrochemical impedance spectrum features are taken as sparse hidden variables to be calculated;

[0023] S22, a relevant vector machine is introduced as a kernel function, and the voltage response and current excitation in the asset observation vector are subjected to deconvolution operation to deduce the ohmic impedance, charge transfer impedance and diffusion impedance parameters affecting the asset value;

[0024] S23, the above impedance parameters are fused to reconstruct a high-dimensional state matrix representing the micro-electrochemical reaction rate of the asset.

[0025] Preferably, S3 specifically comprises:

[0026] S31, a three-dimensional coordinate system for asset value assessment is established, and the X-axis is defined as the depth of discharge, the Y-axis is defined as the discharge rate, and the Z-axis is defined as the health state decay rate;

[0027] S32, based on historical full life cycle experimental data, an acceleration surface under different use condition combinations is fitted as an asset loss model;

[0028] S33, the high-dimensional state matrix is analyzed, the current depth of discharge value and discharge rate value are extracted, and they are projected to the X-Y plane of the asset loss model to determine the current asset value state point;

[0029] S34, the partial derivatives of the current state point on the asset loss model along the X-axis and Y-axis directions are calculated, and the value dissipation potential gradient vector of the current position is synthesized.

[0030] Preferably, S4 specifically comprises:

[0031] S41, first-order difference operation is performed on the current data to obtain the current change rate, and the consumption behavior speed vector describing the intensity of user use is constructed in combination with the depth of discharge change rate;

[0032] S42, the scalar product of the value dissipation potential gradient vector and the consumption behavior speed vector is calculated as a dissipation action index;

[0033] S43, judge the value range of the dissipation action index: if the index is greater than zero, it is determined that the user behavior is increasing the asset depreciation risk; if the index is less than or equal to zero, it is determined that the user behavior is in the asset value safety area.

[0034] Preferably, S5 specifically comprises:

[0035] S51, defining a benchmark rental rate coefficient, dynamically calculating a rate gain value based on the size of the dissipation action index;

[0036] S52, if the dissipation action index is greater than the preset asset safety threshold, the difference between the dissipation action index and the asset safety threshold is taken as the independent variable of the exponential function, the rate gain value showing an exponential growth trend is calculated, and the gain value is superimposed on the benchmark rental rate coefficient to generate a pricing adjustment coefficient;

[0037] S53, if the dissipation action index is less than or equal to the preset asset safety threshold, the benchmark rental rate coefficient is maintained unchanged or a preferential attenuation value is calculated according to a preset linear proportion;

[0038] S54, mapping the pricing adjustment coefficient to a monetary unit to generate a dynamic rate instruction containing the current settlement price and the future price trend prediction.

[0039] Preferably, S6 specifically comprises: S61, issuing the dynamic rate instruction to the display interaction interface of the user end through the Internet of Things communication module to refresh the current rental price; S62, synchronously analyzing the pricing adjustment coefficient, if the coefficient exceeds the preset asset protection threshold, a risk blocking signal is generated; if the coefficient does not exceed the preset asset protection threshold, the current running state is maintained; S63, in response to the generated risk blocking signal, a software locking instruction is sent to the underlying control system through the risk blocking mechanism to dynamically reduce the maximum allowed discharge current threshold of the lithium battery, limit the user's high-risk use permission, and force the rental asset running track to return to the low-loss area.

[0040] Compared with the prior art, the present application has the following beneficial effects:

[0041] 1. The present application introduces the sparse Bayesian learning algorithm and the inverse problem solving mechanism, effectively solving the problem of high cost of battery microstate monitoring in the traditional leasing mode; Unlike relying on expensive high-frequency electrochemical workstations, the present system uses ordinary low-frequency operating data streams such as voltage and current to reconstruct and analyze the implicit electrochemical impedance spectrum characteristics, achieving accurate calibration of the internal microstate of the battery without increasing the cost of hardware deployment, reducing the operation threshold, and ensuring the high accuracy of asset valuation.

[0042] 2、The application constructs an asset loss model and visual economic cost gradient based on potential energy-damping principle, overcoming the defect that the static threshold judgment in the prior art cannot quantify the transient behavior damage; the system defines the health state decay rate as asset value dissipation potential energy, and converts the abstract battery aging physical process into specific algebraic operation through mathematical mapping, so that each high-rate discharge or illegal operation can be accurately quantified as specific monetary cost, eliminating the asset abuse risk caused by the traditional flat rate mode.

[0043] 3、The application innovatively adopts a dynamic pricing strategy of converting physical work into economic billing, solving the problem that the billing result is disconnected with the actual asset depreciation; by calculating the dot product of the value dissipation gradient vector and the consumption behavior speed vector, a dynamic rate instruction containing price adjustment information is generated, and a nonlinear mapping function is used to convert physical loss into economic leverage in real time, and this price damper mechanism can make the user feel direct economic pain, thereby forcing the user to actively reduce the high-loss operation intensity, achieving dynamic balance between business interests and asset preservation.

[0044] 4、The application establishes a closed-loop control mechanism of economic regulation and physical limitation, significantly improving the full life cycle life and safety of the leased asset; in the transaction execution and risk control module, the system can not only issue dynamic rate instructions for price adjustment, but also adjust the power limit threshold of the battery management system according to the risk judgment result, directly triggering the physical fuse mechanism when the economic means fails to force the discharge current to be reduced, completely solving the problem of asset safety out of control caused by extreme user behavior. BRIEF DESCRIPTION OF DRAWINGS

[0045] The application will be further explained in conjunction with the accompanying drawings and embodiments:

[0046] Figure 1 is a system block diagram of the system of the application. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical scheme and advantages of the application clearer and more apparent, the application will be further described in detail below in conjunction with specific embodiments.

[0048] Embodiment 1:

[0049] Please refer to Figure 1 A dynamic adjustment system for leasing pricing based on lithium battery health state decay rate, comprising:

[0050] The lease asset data acquisition and value calibration module is configured to obtain the operation data stream of the lithium battery as the target of the lease, and the operation data stream is used as the input basis for asset valuation, and the electrochemical impedance spectrum characteristics hidden in the data stream are analyzed based on the sparse Bayesian learning algorithm to determine the micro asset value benchmark in the current lease period.

[0051] The asset loss model construction module is configured to establish a damage potential surface model for quantifying the lease cost, which maps the physical operation parameters of the lithium battery to the asset depreciation dimension coordinates, and defines the health state decay rate as the asset value dissipation potential.

[0052] The user consumption behavior quantification module is configured to locate the current asset state point in the damage potential surface model, calculate the value dissipation gradient vector in the direction of the user's use behavior, and generate the consumption behavior speed vector combined with the change rate of the user's behavior.

[0053] The dynamic pricing strategy generation module is configured to calculate the pricing adjustment coefficient for adjusting the lease cost based on the coupling relationship between the value dissipation gradient vector and the consumption behavior speed vector, and convert the pricing adjustment coefficient into real-time dynamic rate instructions.

[0054] The transaction execution and risk control module is configured to issue the dynamic rate instructions to the settlement terminal to execute the price adjustment, and to adjust the power limit threshold of the battery management system according to the risk determination result, thereby forming a closed-loop control to suppress the high devaluation behavior of the asset by means of economic regulation and physical limitation.

[0055] The embodiment provides a lease pricing dynamic adjustment system based on the health state decay rate of a lithium battery, which aims to solve the contradiction between battery life protection and high-frequency monitoring cost in the existing lease mode. The system is configured with a lease asset data acquisition and value calibration module, which obtains the operation data stream of the lithium battery as the target of the lease through a hardware interface, and uses the operation data stream as the input basis for asset valuation. A sparse Bayesian learning algorithm is built in, and the inverse problem solving capability of the algorithm is used to analyze the electrochemical impedance spectrum characteristics hidden in the data stream, thereby determining the micro asset value benchmark in the current lease period.

[0056] The system is configured with an asset loss model construction module for establishing a damage potential surface model for quantifying lease cost, which converts the physical operation parameters of the lithium battery into asset depreciation dimension coordinates through mathematical mapping, and defines the health state decay rate as asset value dissipation potential, thereby constructing a value loss topographic map; on this basis, a user consumption behavior quantification module locates the current asset state point in the damage potential surface model, calculates the value dissipation gradient vector in the direction of user use behavior, and generates a consumption behavior speed vector in combination with the change rate of user behavior, to represent the intensity of user operation in vector form; a dynamic pricing strategy generation module calculates a pricing adjustment coefficient for adjusting the lease cost based on the coupling relationship between the value dissipation gradient vector and the consumption behavior speed vector, converts the physical loss into an economic lever in real time, and converts the pricing adjustment coefficient into real-time dynamic rate instructions; a transaction execution and risk control module is responsible for issuing the dynamic rate instructions to the settlement terminal to execute the price adjustment, and synchronously adjusting the power limitation threshold of the battery management system according to the risk judgment result, to form a closed-loop control for inhibiting high-value-depreciation behavior of the asset through economic regulation and physical limitation.

[0057] The system constructed in the embodiment establishes an asset protection mechanism based on the principle of potential energy-damping in the leasing scenario by introducing the dissipation structure theory; the mechanism ingeniously reconstructs the microscopic state by using low-frequency data, avoids the deployment cost of expensive high-frequency electrochemical workstations, and converts the abstract physical process of battery aging into a visual economic cost gradient; when the user operation is in a high-loss potential energy area, the economic resistance and physical current-limiting measures generated by the system automatically form a synergistic effect, forcing the user behavior to return to a low-loss safe potential energy area, thereby significantly prolonging the full life cycle of the leased asset without relying on manual intervention.

[0058] Embodiment 2:

[0059] The modules are realized through the following methods:

[0060] S1, collect the voltage, current and temperature data of the lithium battery in the billing period through the interface of the lease management system, and construct a low-frequency operation data set for asset evaluation;

[0061] S2, input the low-frequency operation data set into the lease asset data acquisition and value calibration module, perform feature reconstruction operation, and generate a high-dimensional state matrix reflecting the current physical properties of the asset;

[0062] S3, map the high-dimensional state matrix to the preset asset loss model, determine the current asset depreciation state coordinates, and analyze the value dissipation potential gradient corresponding to the coordinate point;

[0063] S4, based on the time dimension change of the current data, quantifying the stress intensity exerted by the user on the rental asset, generating a consumption behavior speed vector, and calculating the dot product of the value dissipation potential gradient as the basis for charging;

[0064] S5, based on the calculation result of the dot product, using a nonlinear mapping function to calculate the pricing adjustment factor, i.e. the pricing adjustment coefficient, to generate a dynamic rate instruction containing price adjustment information;

[0065] S6, send the dynamic rate instruction to the user transaction terminal for publicizing and settlement, and trigger the power dynamic clamping mechanism of the battery management system according to the risk level in the instruction.

[0066] The embodiment details the specific execution logic of the rental pricing dynamic adjustment system based on the lithium battery state of health decay rate; S1 step collects the voltage, current and temperature data of the lithium battery in the billing period through the rental management system interface, and constructs a low-frequency operation data set for asset evaluation, which serves as the basis for the subsequent calculation of the physical base; S2 step inputs the low-frequency operation data set into the rental asset data acquisition and value calibration module, performs feature reconstruction operation, compensates for the lack of sampling frequency using algorithmic power, and generates a high-dimensional state matrix reflecting the current physical properties of the asset; S3 step maps the high-dimensional state matrix to the preset asset wear model, determines the current asset depreciation state coordinates, and analyzes the value dissipation potential gradient corresponding to the coordinate point using differential geometry method, indicating the direction of the fastest value loss under the current state; Based on this, S4 step quantifies the stress intensity exerted by the user on the rental asset based on the time dimension change of the current data, generates a consumption behavior speed vector, and calculates the dot product of the value dissipation potential gradient as the basis for charging, realizing the conversion of physical work to economic charging; S5 step, based on the calculation result of the dot product, uses a nonlinear mapping function to calculate the pricing adjustment factor, i.e. the pricing adjustment coefficient, to generate a dynamic rate instruction containing price adjustment information; S6 step sends the dynamic rate instruction to the user transaction terminal for publicizing and settlement, and triggers the power dynamic clamping mechanism of the battery management system according to the risk level in the instruction;

[0067] The embodiment converts complex electrochemical protection requirements into a standardized data processing pipeline, especially the dot product calculation introduced in S4, which ingeniously simplifies multi-dimensional user behavior into a single scalar indicator, greatly improving the real-time response speed of the system; The process ensures strict synchronization between the charging logic and the physical wear, and this charging method based on physical work principle can accurately quantify the specific monetary cost of each high-rate discharge or illegal operation, eliminating the asset abuse risk caused by the one-off price in the traditional rental mode.

[0068] Embodiment 3:

[0069] S1 specifically includes:

[0070] S11. Configure a data capture unit at the output end of the leased asset to record the instantaneous voltage, instantaneous current and temperature values ​​in real time according to the preset billing sampling frequency.

[0071] S12. Timestamp alignment is performed on the collected basic transaction data, and invalid data frames are removed to ensure the integrity of the billing data.

[0072] S13. Normalize the timestamp-aligned data to eliminate dimensional differences and construct a standardized low-frequency operation dataset for subsequent valuation algorithms.

[0073] This embodiment further specifies the low-frequency operation dataset construction process in step S1. Step S11 involves configuring a data capture unit at the output end of the leased asset to record instantaneous voltage, current, and temperature values ​​in real time according to a preset billing sampling frequency, ensuring real-time acquisition of the original physical quantities. Step S12 timestamps the collected transaction data and uses interpolation algorithms to remove invalid data frames to ensure the integrity of the billing data and eliminate timing misalignments caused by sensor response delays. Step S13 normalizes the aligned data to eliminate dimensional differences and constructs a standardized low-frequency operation dataset for subsequent valuation algorithms. The normalization formula is as follows:

[0074]

[0075] Among them, to prevent the occurrence of problems under constant temperature or steady-state operating conditions. To address the case where the denominator is zero, a minimal correction operator is introduced into the denominator. This is to ensure the numerical stability of the algorithm; The physical meaning is the dimensionless data value after normalization; The data source is sensor data, and its physical meaning is the original voltage, current, or temperature value. The physical meaning refers to the physical limit boundary value of this type of data;

[0076] This embodiment eliminates noisy data caused by sensor jitter or communication packet loss through rigorous data cleaning and standardization, providing a high-quality input benchmark for subsequent sparse Bayesian learning algorithms. This preprocessing mechanism ensures that high signal-to-noise ratio feature data can still be obtained under the limited computing power of the vehicle BMS, preventing algorithm weight imbalance caused by differences in magnitude, thereby ensuring the accuracy of asset micro-state reconstruction.

[0077] Example 4:

[0078] S2 specifically includes:

[0079] S21. Construct an inverse problem-solving model based on sparse Bayesian learning, using the low-frequency running dataset as the asset observation vector and the electrochemical impedance spectroscopy features as the sparse latent variables to be calculated.

[0080] S22. Introduce a correlation vector machine as the kernel function to perform deconvolution operation on the voltage response and current excitation in the asset observation vector, and infer the ohmic impedance, charge transfer impedance and diffusion impedance parameters that affect the asset value.

[0081] S23. By integrating the above impedance parameters, a high-dimensional state matrix characterizing the microscopic electrochemical reaction rate of the asset is reconstructed.

[0082] This embodiment further specifies the feature reconstruction operation in step S2, aiming to solve the mathematical loop in solving the inverse problem; this operation executes step S21, constructing a discretized observation equation based on temporal convolution, which serves as the input model for sparse Bayesian learning:

[0083]

[0084] in, This is the normalized voltage observation vector obtained in step S1; The Toeplitz convolution matrix, constructed using normalized current data, is used to characterize the system's input excitation; where This represents the number of sampling points for voltage observation data. The preset number of impedance spectrum feature discretization points; Toeplitz convolution matrix. The specific construction logic is as follows: its first column elements are composed of normalized current vectors. Composition, the first row of elements consists of Composition, in which The number of The remaining elements in the matrix satisfy This structure ensures that matrix operations are equivalent to time-domain discrete convolution, enabling precise alignment of current excitation and voltage response in time phase. The eigenvector of the electrochemical impedance spectrum to be reconstructed is the discrete weight of the relaxation time distribution function DRT, which represents the microstate inside the battery. It is a Gaussian white noise vector;

[0085] S22. Introducing a correlation vector machine as the kernel function, deconvolution is performed on the voltage response and current excitation in the asset observation vector to infer the ohmic impedance, charge transfer impedance, and diffusion impedance parameters that affect asset value; weight vector. Each element in Assign independent zero-mean Gaussian prior distributions ,in The hyperparameters are defined as follows: The hyperparameters are iteratively updated by maximizing the marginal likelihood function. hyperparameters The iterative update employs the fast marginal likelihood maximization operator, and its specific update formula is as follows: ;in, Let be the posterior mean of the weight vector, satisfying If the algorithm fails to achieve convergence accuracy within the preset 500 iterations, the system will forcibly call the stable feature vector from the previous billing cycle. Make minor compensation adjustments to ensure that real-time settlement instructions are not interrupted due to computing power fluctuations;

[0086] S23. By integrating the above impedance parameters, a high-dimensional state matrix characterizing the microscopic electrochemical reaction rate of the asset is reconstructed; before generating the final impedance parameters, the system performs a dimensionless weighting operation to restore the dimensionless weights. Converted into impedance values ​​with physical units. The recovery formula is: ,in All of these are the normalized boundary values ​​used in step S13 of Example 3, to ensure that the reconstructed microscopic parameters truly reflect the ohmic resistance of the battery. ;

[0087] This embodiment uses an explicit linear convolution equation. By using sparse prior settings, the abstract inverse problem is transformed into a computable convex optimization problem, ensuring that the microscopic impedance characteristics inside the battery can be calculated using conventional VI data even with the limited computing power of the BMS.

[0088] Example 5:

[0089] S3 specifically includes:

[0090] S31. Establish a three-dimensional coordinate system for asset valuation, defining the X-axis as the depth of discharge, the Y-axis as the discharge rate, and the Z-axis as the rate of decay of the health status.

[0091] S32. Based on historical full life cycle experimental data, fit the decay acceleration surface under different combinations of usage conditions as an asset depreciation model.

[0092] S33. Analyze the high-dimensional state matrix, extract the current discharge depth value and discharge rate value, project them onto the XY plane of the asset loss model, and determine the current asset value state point.

[0093] S34. Calculate the partial derivatives of the current state point along the X and Y axes in the asset loss model, and synthesize the value dissipation potential energy gradient vector at the current position.

[0094] This embodiment is a further specification of the asset loss model mapping in step S3, with the focus on clarifying the mathematical expression of the potential energy surface function to support the analytical calculation of the gradient; the mapping process executes step S31 to establish a three-dimensional coordinate system for asset value assessment, defining the X-axis as the depth of discharge (DOD), the Y-axis as the discharge rate (C-rate), and the Z-axis as the rate of decay of the healthy state.

[0095] Step S32, based on historical experimental data, uses bivariate polynomial regression to fit the decay rate surface function. As an asset depreciation model:

[0096]

[0097] The source is a fixed regression coefficient determined by the least squares method, which characterizes the inherent physical degradation characteristics of this type of battery; The physical meanings are the discharge depth variable and the discharge rate variable, respectively.

[0098] Step S33 analyzes the high-dimensional state matrix and extracts the current discharge depth value. With discharge rate value ;

[0099] Step S34 is based on the fitting function. The parsing expression directly calculates the current position. Partial derivatives at the point, composite value dissipated potential energy gradient vector :

[0100]

[0101] The source is extracted from step S33, and its physical meaning is the current discharge depth value and discharge rate value.

[0102] This embodiment concretizes the fuzzy loss potential energy surface into a bivariate polynomial function, so that the gradient calculation no longer depends on the abstract description, but is transformed into specific algebraic operations, ensuring that the system can accurately quantify the direction of the fastest loss of asset value.

[0103] Example 6:

[0104] S4 specifically includes:

[0105] S41. Perform first-order difference operation on the current data to obtain the current change rate, and combine it with the discharge depth change rate to construct a consumption behavior speed vector describing the intensity of user usage.

[0106] S42. Calculate the scalar product of the value dissipation potential energy gradient vector and the consumption behavior velocity vector as an indicator of dissipation action.

[0107] S43. Determine the numerical range of the dissipation effect indicator: If the indicator is greater than zero, it is determined that the user's behavior is increasing the risk of asset depreciation; if the indicator is less than or equal to zero, it is determined that the user's behavior is within the safe range of asset value.

[0108] This embodiment further specifies the quantization of consumer behavior in step S4, aiming to ensure alignment between the physical dimensions and the dimensions of the gradient vector; this quantization process executes step S41, based on the sampling time interval. Calculate the rate of change of discharge depth and rate of change of discharge rate Construct a two-dimensional consumer behavior velocity vector :

[0109]

[0110] The source is the analysis result of step S3, and the physical meaning is the current moment. Compared to the previous moment The depth of discharge; The source is the analysis result of step S3, and the physical meaning is the current moment. Compared to the previous moment The discharge rate; The source is the system clock, and its physical meaning is the sampling time interval;

[0111] This vector clearly characterizes the speed and direction of the user's movement in the depth-rate plane;

[0112] Step S42 calculates the gradient vector of the value dissipation potential energy. From S34 Steps and Consumer Behavior Velocity Vector The dot product is used as an indicator of dissipative action. :

[0113]

[0114] : These are the components of the gradient vector on the discharge depth axis and the rate axis, respectively; : These are the components of the velocity vector on the depth of discharge axis and the rate axis, respectively;

[0115] S43. Indicators for determining dissipative action The value: If This indicates that the user is engaging in high-intensity operations that accelerate battery aging; if This indicates that the user's operation is within the value maintenance or recovery range; to eliminate the billing failure blind spot that may occur when the dot product operation is performed drastically but orthogonal to the gradient direction, the system adds velocity vector magnitude compensation logic: calculating the Euclidean norm of the velocity vector. ,like Exceeding the preset physical impact threshold ,in The value is taken as 80% of the maximum pulse discharge rate change rate specified in the battery model's datasheet, even if the dot product... The system still forcibly generates impact correction coefficients. The calculation formula is: ,in This is a preset impact penalty factor to cover the mechanical stress loss caused by high-frequency fluctuations to the battery separator; this coefficient will be passed to step S5 to participate in the final pricing decision.

[0116] This embodiment uses a strict definition of the velocity vector. The physical meaning of the components, making them related to the gradient vector. The coordinate system is fully orthogonal and aligned, avoiding the invalidation of the physical meaning of the dot product due to unclear dimensional definitions, and ensuring the mathematical rigor of the billing logic.

[0117] Example 7:

[0118] S5 specifically includes:

[0119] S51. Define the benchmark rental rate coefficient and dynamically calculate the rate gain value based on the magnitude of the dissipation effect index.

[0120] S52. If the dissipation effect index is greater than the preset asset safety threshold, the difference between the dissipation effect index and the asset safety threshold is used as the independent variable of the exponential function to calculate the rate gain value with an exponential growth trend, and the gain value is superimposed on the benchmark rental rate coefficient to generate the pricing adjustment coefficient.

[0121] S53. If the dissipation effect index is less than or equal to the preset asset safety threshold, the benchmark rental rate coefficient shall remain unchanged or the discount attenuation value shall be calculated according to the preset linear ratio.

[0122] S54. Map the pricing adjustment coefficient to a currency unit and generate a dynamic rate instruction that includes the current settlement price and future price trend prediction.

[0123] This embodiment further specifies the dynamic pricing strategy generation in step S5. This strategy executes step S51, defining a benchmark rental rate coefficient and dynamically calculating the rate gain value based on the magnitude of the dissipation effect index. In step S52, if the dissipation effect index exceeds a preset asset safety threshold, the difference between the dissipation effect index and the asset safety threshold is used as the independent variable of an exponential function to calculate the rate gain value, which exhibits an exponential growth trend, and an adjustment coefficient based on wear and tear is calculated. This gain value is then added to the benchmark lease rate coefficient to generate a wear-and-tear adjustment coefficient. The system will use this coefficient Impact correction factor generated in step S4 The two values ​​are compared, and the larger value is taken as the final pricing adjustment coefficient. ,Right now ;

[0124] The formula for calculating the adjustment coefficient is as follows:

[0125]

[0126] in, The source is calculated, and the physical meaning is an adjustment coefficient calculated based on cumulative wear. The source is calculated, and the physical meaning is the final pricing adjustment coefficient; The source is a preset constant, and its physical meaning is the benchmark rate coefficient; The source is the operational strategy setting, and the physical meanings are the adjustment sensitivity factor and the price steepness factor, respectively. Among them, the price steepness factor... The physical dimension is set to time (second) to neutralize the dissipative action index. The unit, that is This ensures that the independent variable of the exponential function is a dimensionless real number; it is used to control the magnitude and speed of price increases. The source is step S42, and its physical meaning is the current dissipation effect index; The source is statistical data from battery life tests, and its physical meaning is the asset safety threshold. Mathematical constants, the base of the natural logarithm; asset safety threshold. The quantitative acquisition method is as follows: based on the cycle life test curve of this battery model, its performance in... Discharge rate and Calculated under standard operating conditions for depth of discharge Value distribution, taking its cumulative distribution function. The quantile value is used as the threshold; this method of value selection ensures that normal usage behavior will not trigger punitive price increases, and only abnormal operations that exceed the standard wear slope will be subject to economic damping.

[0127] Step S53 handles the safety zone situation, responding to the dissipation effect index being less than or equal to the preset asset safety threshold, maintaining the benchmark rental rate coefficient unchanged or calculating the discount reduction value according to the preset linear ratio; Step S54, responding to the aforementioned comparison results, adjusts the final pricing coefficient. Mapped to currency units, generating dynamic rate instructions that include the current settlement price and future price trend predictions;

[0128] This embodiment introduces an exponential function to construct a nonlinear price damper. When a user attempts to perform high-damage operations, the price will spike instantly. This price pain will be fed back to the user through a psychological mechanism, prompting them to actively reduce the intensity of use. This mechanism achieves proactive protection of battery life at the economic level, transforming the user's profit-seeking instinct into the motivation to protect assets, and realizing a dynamic balance between commercial interests and asset preservation.

[0129] Example 8:

[0130] S6 specifically includes:

[0131] S61. The dynamic rate instruction is sent to the user's display interface via the Internet of Things communication module to refresh the current rental price;

[0132] S62. Synchronously analyze the pricing adjustment coefficient. If the coefficient exceeds the preset asset protection threshold, a risk blocking signal is generated; if the coefficient does not exceed the preset asset protection threshold, the current operating status is maintained.

[0133] S63. In response to the generated risk blocking signal, a software lock command is sent to the underlying control system through the risk blocking mechanism to dynamically reduce the maximum allowable discharge current threshold of the lithium battery, restrict the user's high-risk usage rights, and force the leased asset's operating trajectory to return to the low-loss area.

[0134] This embodiment further specifies the transaction execution and risk management in step S6. Step S61 involves sending dynamic rate instructions to the user's display interface via the IoT communication module, updating the current rental price in real time to ensure the user's right to know. Step S62 simultaneously analyzes the pricing adjustment coefficient. If this coefficient exceeds a preset asset protection threshold, the threshold is determined by the ratio of the maximum permissible instantaneous power to the rated power in the battery specifications, combined with a safety factor setting; its value is fixed as the base rate coefficient. Three to five times the value of the coefficient is used to identify extreme violations that cannot be effectively constrained by economic means; a risk blocking signal is generated, and if the coefficient does not exceed the preset asset protection threshold, the current operating state is maintained; step S63 executes physical closed-loop control, and in response to the generated risk blocking signal, a software lock command is sent to the underlying control system through the risk blocking mechanism to dynamically reduce the maximum allowable discharge current threshold of the lithium battery, restrict the user's high-risk usage rights, and force the leased asset's operating trajectory to return to the low-loss area;

[0135] This embodiment constitutes a line of defense for the system, namely the circuit breaker mechanism. When economic means fail, that is, when users abuse assets regardless of cost, the system directly takes over physical permissions and prevents irreversible thermal runaway or lithium plating damage to the battery by forcibly reducing power. This forced intervention at the physical level completely solves the problem of asset security loss of control caused by extreme user behavior in the leasing scenario, and ensures that the battery always operates within the physical safety boundary.

[0136] It should be noted that 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A dynamic adjustment system for leasing pricing based on the degradation rate of lithium battery health status, characterized in that, include: The leased asset data acquisition and valuation module is used to acquire the operating data stream of the lithium battery of the leased asset. The operating data stream serves as the input basis for asset valuation. Based on the sparse Bayesian learning algorithm, the module analyzes the electrochemical impedance spectral characteristics hidden in the data stream to establish the asset micro-value benchmark for the current lease period. Asset depreciation model building module: used to build a damage potential energy surface model for quantifying leasing costs. This model maps the physical operating parameters of lithium batteries to asset depreciation dimension coordinates and defines the rate of health decay as the asset value dissipation potential energy. User consumption behavior quantification module: used to locate the current asset state point in the damage potential energy surface model, calculate the value dissipation gradient vector along the user's usage behavior direction, and generate a consumption behavior velocity vector by combining the rate of change of user behavior. Dynamic pricing strategy generation module: Based on the coupling relationship between the value dissipation gradient vector and the consumption behavior velocity vector, it calculates the pricing adjustment coefficient for adjusting rental fees and converts the pricing adjustment coefficient into real-time dynamic rate instructions; The transaction execution and risk control module is used to send dynamic rate instructions to the settlement terminal to execute price adjustments, and to adjust the power limit threshold of the battery management system in sync with the risk assessment results. It forms a closed-loop control to suppress high asset depreciation through a combination of economic regulation and physical restrictions.

2. The dynamic adjustment system for rental pricing based on the lithium battery health status degradation rate according to claim 1, characterized in that, The modules are connected in the following way: S1. Collect voltage, current and temperature data of lithium batteries during the billing cycle through the leasing management system interface to construct a low-frequency operation dataset for asset evaluation; S2. Input the low-frequency operation dataset into the leased asset data collection and value assessment module, perform feature reconstruction calculation, and generate a high-dimensional state matrix that reflects the current physical attributes of the asset. S3. Map the high-dimensional state matrix to the preset asset depreciation model, determine the current asset depreciation state coordinates, and analyze the value dissipation potential energy gradient corresponding to the coordinate point. S4. Based on the time dimension change of current data, quantify the stress intensity exerted by users on leased assets, generate a consumption behavior speed vector, and calculate its dot product with the value dissipation potential energy gradient as the basis for billing. S5. Based on the calculation results of the dot product, the pricing adjustment factor, i.e. the pricing adjustment coefficient, is calculated using a nonlinear mapping function to generate a dynamic rate instruction containing price adjustment information. S6. Send the dynamic rate instruction to the user's transaction terminal for public display and settlement, and trigger the power dynamic clamping mechanism of the battery management system according to the risk level in the instruction.

3. The dynamic adjustment system for rental pricing based on the lithium battery health state degradation rate according to claim 2, characterized in that, S1 specifically includes: S11. Configure a data capture unit at the output end of the leased asset to record the instantaneous voltage, instantaneous current and temperature values ​​in real time according to the preset billing sampling frequency. S12. Timestamp alignment is performed on the collected basic transaction data, and invalid data frames are removed to ensure the integrity of the billing data. S13. Normalize the timestamp-aligned data to eliminate dimensional differences and construct a standardized low-frequency operation dataset for subsequent valuation algorithms.

4. The dynamic adjustment system for rental pricing based on the degradation rate of lithium battery health status according to claim 3, characterized in that, S2 specifically includes: S21. Construct an inverse problem-solving model based on sparse Bayesian learning, using the low-frequency running dataset as the asset observation vector and the electrochemical impedance spectroscopy features as the sparse latent variables to be calculated. S22. Introduce a correlation vector machine as the kernel function to perform deconvolution operation on the voltage response and current excitation in the asset observation vector, and infer the ohmic impedance, charge transfer impedance and diffusion impedance parameters that affect the asset value. S23. By integrating the above impedance parameters, a high-dimensional state matrix characterizing the microscopic electrochemical reaction rate of the asset is reconstructed.

5. The dynamic adjustment system for rental pricing based on the lithium battery health state degradation rate according to claim 4, characterized in that, S3 specifically includes: S31. Establish a three-dimensional coordinate system for asset valuation, defining the X-axis as the depth of discharge, the Y-axis as the discharge rate, and the Z-axis as the rate of decay of the health status. S32. Based on historical full life cycle experimental data, fit the decay acceleration surface under different combinations of usage conditions as an asset depreciation model. S33. Analyze the high-dimensional state matrix, extract the current discharge depth value and discharge rate value, project them onto the XY plane of the asset loss model, and determine the current asset value state point. S34. Calculate the partial derivatives of the current state point along the X and Y axes in the asset loss model, and synthesize the value dissipation potential energy gradient vector at the current position.

6. The dynamic adjustment system for rental pricing based on the degradation rate of lithium battery health status according to claim 5, characterized in that, S4 specifically includes: S41. Perform first-order difference operation on the current data to obtain the current change rate, and combine it with the discharge depth change rate to construct a consumption behavior speed vector describing the intensity of user usage. S42. Calculate the scalar product of the value dissipation potential energy gradient vector and the consumption behavior velocity vector as an indicator of dissipation action. S43. Determine the numerical range of the dissipation effect indicator: If the indicator is greater than zero, it is determined that the user's behavior is increasing the risk of asset depreciation; if the indicator is less than or equal to zero, it is determined that the user's behavior is within the safe range of asset value.

7. The dynamic adjustment system for rental pricing based on the degradation rate of lithium battery health status according to claim 6, characterized in that, S5 specifically includes: S51. Define the benchmark rental rate coefficient and dynamically calculate the rate gain value based on the magnitude of the dissipation effect index. S52. If the dissipation effect index is greater than the preset asset safety threshold, the difference between the dissipation effect index and the asset safety threshold is used as the independent variable of the exponential function to calculate the rate gain value with an exponential growth trend, and the gain value is superimposed on the benchmark rental rate coefficient to generate the pricing adjustment coefficient. S53. If the dissipation effect index is less than or equal to the preset asset safety threshold, the benchmark rental rate coefficient shall remain unchanged or the discount attenuation value shall be calculated according to the preset linear ratio. S54. Map the pricing adjustment coefficient to a currency unit and generate a dynamic rate instruction that includes the current settlement price and future price trend prediction.

8. The leasing pricing dynamic adjustment system based on lithium battery health state degradation rate according to claim 7, characterized in that, S6 specifically includes: S61. The dynamic rate instruction is sent to the user's display interface via the Internet of Things communication module to refresh the current rental price; S62. Synchronously analyze the pricing adjustment coefficient. If the coefficient exceeds the preset asset protection threshold, a risk blocking signal is generated; if the coefficient does not exceed the preset asset protection threshold, the current operating status is maintained. S63. In response to the generated risk blocking signal, a software lock command is sent to the underlying control system through the risk blocking mechanism to dynamically reduce the maximum allowable discharge current threshold of the lithium battery, restrict the user's high-risk usage rights, and force the leased asset's operating trajectory to return to the low-loss area.

Images