A method, device and equipment for evaluating the value of a retired battery based on an attenuation trajectory feature, and a storage medium

Abstract

The application discloses a kind of based on attenuation trajectory feature's decommissioned battery value evaluation method, device, equipment and storage medium, it is related to computer technology field, including: first, based on the SOH history sequence of each battery fitting attenuation trajectory, extract four-dimensional feature vector of attenuation rate, acceleration, volatility and current SOH;Second, according to this clustering division attenuation type, and matching corresponding prediction model, predict future SOH trend to construct future attenuation path;Then, based on future path, calculate the expected economic benefit in remaining life, determine individual residual value;Finally, construct asset package planning model, filter battery combination in combination with constraint condition, complete combination value evaluation, to improve the efficiency of the value of decommissioned battery is evaluated.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, device, and storage medium for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics. Background Technology

[0002] Currently, with the rapid development of the energy storage industry, a large number of energy storage power station batteries are gradually withdrawing from main frequency regulation / peak shaving services and entering the secondary utilization market after their capacity decays to 70%~80% of their initial capacity. These retired energy storage batteries still have considerable residual economic value in low-intensity application scenarios (such as user-side peak-valley arbitrage, backup power, and communication base station energy storage). However, the financialization of secondary utilization energy storage batteries faces three serious challenges: (1) Significant individual differences make valuation difficult. Due to different operating conditions in the original energy storage power station (differences in depth of charge and discharge, ambient temperature, and cycle frequency), retired batteries from the same batch have huge differences in SOH (State of Health), internal resistance, and capacity consistency after retirement. The traditional "linear discount based on SOH" method completely ignores the individual differences in the degradation trajectory.

[0003] (2) Lack of asset package optimization methods for securitization. The value of individual batteries used in tiered applications is low and varies greatly, so a large number of batteries need to be packaged into an asset pool to achieve securitization. However, there is currently no systematic method for how to package them to meet the minimum return requirements and maximize the scale of securitizable assets.

[0004] (3) The degradation path of retired batteries is more important than the current state of equilibrium (SOH). Two retired batteries with a current SOH of 75% each, one with stable degradation and the other with accelerated degradation, will have vastly different residual values. Current valuation methods only consider the current SOH snapshot value.

[0005] Furthermore, existing technologies propose a full lifecycle health management and value assessment system for lithium battery assets, which includes a residual value estimation function for secondary use. However, this system only assesses the value of recycled materials and reuse value of individual battery cells.

[0006] As can be seen from the above, improving the efficiency of evaluating the value of retired batteries is an urgent problem to be solved. Summary of the Invention

[0007] In view of this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for evaluating the value of decommissioned batteries based on degradation trajectory characteristics, which can improve the evaluation efficiency in the process of evaluating the value of decommissioned batteries based on degradation trajectory characteristics. The specific solution is as follows: Firstly, this application provides a method for evaluating the value of decommissioned batteries based on degradation trajectory characteristics, including: Based on the historical SOH sequence data of each retired battery, the health status change process of each retired battery is fitted to obtain the decay trajectory, and based on each decay trajectory, a decay trajectory feature vector including decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics and current health status SOH characteristics is determined. Based on the decay trajectory feature vector, cluster analysis and decay type classification are performed on each of the retired batteries to obtain the decay type; Determine the attenuation prediction model corresponding to each of the attenuation types, and use the attenuation prediction model to predict the future SOH change trend of each of the retired batteries, so as to construct the future attenuation path based on the prediction results; Based on each of the future degradation paths, the expected economic benefits generated by each of the retired batteries during their remaining usable lifespan are determined, and the remaining economic value is determined based on the expected economic benefits, so as to construct the individual residual value of each retired battery based on the remaining economic value. Based on the residual value of the individual retired batteries, a planning model for a retired battery asset package is constructed. The retired batteries are then screened using the planning model and based on preset constraints to obtain a battery asset portfolio, which is then used to evaluate the value of the batteries in the battery asset portfolio.

[0008] Optionally, the fitting of the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory includes: Determine the SOH historical sequence data for each time point corresponding to each retired battery; the retired battery refers to the battery used by the energy storage power station within the historical time period. The SOH historical sequence data are arranged according to a preset time order to obtain an SOH time series. Based on the SOH time series, the SOH change process of the retired battery is curve fitted to obtain the corresponding decay trajectory.

[0009] Optionally, determining the attenuation trajectory feature vector based on each of the attenuation trajectories, including attenuation rate characteristics, attenuation acceleration characteristics, attenuation fluctuation characteristics, and current health state (SOH) characteristics, includes: The SOH change trend at each time point is determined based on the aforementioned decay trajectories. Based on each of the decay trajectories, the unit time SOH decrease of each of the retired batteries is determined, and a decay rate feature is constructed based on the unit time SOH decrease. The decay rate change is determined based on the decay rate characteristics corresponding to each consecutive time node, and the decay acceleration characteristics are determined based on the decay rate change. The degree of decay rate change is determined based on the decay rate characteristics corresponding to each of the aforementioned time nodes, and the decay fluctuation characteristics are determined based on the degree of decay rate change. Determine the SOH data of each of the retired batteries at the current time point, and determine the current health status SOH characteristics based on the SOH data of each battery; A decay trajectory feature vector is constructed based on the SOH change trend, decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics, and the current health state SOH characteristics.

[0010] Optionally, the step of performing cluster analysis and degradation type classification on each of the decommissioned batteries based on the degradation trajectory feature vector to obtain the degradation type includes: The attenuation trajectory feature vectors corresponding to each of the retired batteries are standardized to obtain standardized attenuation trajectory feature vectors. Based on the standardized decay trajectory feature vector, the feature distance between each of the retired batteries is determined, and cluster analysis is performed on each of the retired batteries based on the feature distance to obtain the cluster analysis results; Based on the cluster analysis results, the degree of similarity of the degradation behavior of each retired battery is determined. Then, based on the degree of similarity of the degradation behavior, each retired battery is classified into different types to obtain the degradation type. The attenuation types include stable attenuation type that meets preset stable attenuation conditions, accelerated attenuation type that meets preset accelerated attenuation conditions, fluctuating attenuation type that meets preset fluctuating attenuation conditions, and fast attenuation type that meets preset fast attenuation conditions.

[0011] Optionally, determining the degradation prediction model corresponding to each of the degradation types, and using the degradation prediction model to predict the future SOH change trend of each of the retired batteries, so as to construct a future degradation path based on the prediction results, includes: The first decay behavior characteristic corresponding to the stable decay type is determined. Then, a preset exponential extrapolation model is used to predict the future SOH change trend of the retired battery corresponding to the stable decay type based on the first decay behavior characteristic, and the corresponding first prediction result is obtained. Determine the second degradation behavior characteristic corresponding to the accelerated degradation type, and then use a preset exponential accelerated degradation model and based on the second degradation behavior characteristic to predict the future SOH change trend of the retired battery corresponding to the accelerated degradation type, and obtain the corresponding second prediction result. The third decay behavior feature corresponding to the fluctuating decay type is determined. Then, a preset Gaussian process regression model is used to predict the future SOH change trend of the retired battery corresponding to the fluctuating decay type based on the third decay behavior feature, and the corresponding third prediction result is obtained. The fourth decay behavior feature corresponding to the rapid decay type is determined. Then, the future SOH change trend of the retired battery corresponding to the rapid decay type is predicted by using a preset lower bound extrapolation model and based on the fourth decay behavior feature, and the corresponding fourth prediction result is obtained. Based on the first prediction result, the second prediction result, the third prediction result, and the fourth prediction result, a corresponding future decay path is constructed.

[0012] Optionally, the step of determining the expected economic benefits generated by each retired battery during its remaining usable life based on each of the future degradation paths, and determining the remaining economic value based on the expected economic benefits, to construct the individual residual value of the retired battery based on each of the remaining economic values, includes: Based on the future degradation path, the time when the SOH of each retired battery drops to the preset end of its tiered utilization life is determined, and the remaining usable life corresponding to each retired battery is determined based on the time. The expected operating cycle is determined based on the remaining usable life of each of the retired batteries, and the expected economic benefits generated by each of the retired batteries during the remaining usable life are determined based on the expected operating cycle. Based on a preset discounting calculation method, the expected economic benefits corresponding to each time point in the remaining usable life are discounted to obtain the economic value corresponding to each time point. The economic values ​​of each battery are summed to obtain the residual economic value of each retired battery over its lifetime, and the individual residual value of each retired battery is constructed based on the residual economic value of each battery.

[0013] Optionally, the step of constructing a retired battery asset package planning model based on the individual residual value of the retired batteries, and using the retired battery asset package planning model and screening each retired battery based on preset constraints to obtain a battery asset portfolio, for value assessment of the batteries in the battery asset portfolio, includes: A retirement battery asset package planning model is constructed based on the residual value of each individual retired battery, and the optimization objective is to maximize the total residual value of each retired battery. Then, a corresponding battery selection scheme is generated based on the optimization objective and the retirement battery asset package planning model. Based on the preset cash flow coverage constraints and the battery screening scheme, the first retired battery to be processed that meets the preset future revenue coverage requirements is selected. Based on the preset attenuation type diversification constraint, the proportion of retired batteries corresponding to each attenuation type in the asset portfolio is determined, and the second retired battery to be processed is determined based on the proportion and the first retired battery to be processed. Based on the preset SOH threshold constraint, the retired batteries in the second batch of batteries to be processed that have a current health status SOH lower than the preset health status threshold are removed to obtain the third batch of batteries to be processed, and a battery asset portfolio is constructed based on the third batch of batteries to be processed. Based on the battery asset portfolio, risk indicators for the asset package are determined, including the weighted average SOH of retired batteries, the composition of degradation types, the cash flow coverage multiple, the probability of stress scenario coverage, and the degradation correlation matrix. Based on the asset package risk indicators, the batteries in the battery asset portfolio are assessed for asset securitization, and the assessment results are obtained.

[0014] Secondly, this application provides a device for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics, comprising: The decay trajectory feature vector generation module is used to fit the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory, and to determine the decay trajectory feature vector based on each decay trajectory, including decay rate features, decay acceleration features, decay fluctuation features and current health status SOH features. The attenuation type determination module is used to perform cluster analysis and attenuation type classification on each of the retired batteries based on the attenuation trajectory feature vector to obtain the attenuation type; The future degradation path construction module is used to determine the degradation prediction model corresponding to each degradation type, and use the degradation prediction model to predict the future SOH change trend of each retired battery, so as to construct the future degradation path based on the prediction results. The expected economic benefit determination module is used to determine the expected economic benefits generated by each of the retired batteries within their remaining usable life based on each of the future degradation paths, and to determine the remaining economic value based on the expected economic benefits, so as to construct the individual residual value of the retired battery based on each of the remaining economic values. The battery asset portfolio generation module is used to construct a retired battery asset package planning model based on the residual value of the individual retired batteries, and to use the retired battery asset package planning model and preset constraints to screen each retired battery to obtain a battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio.

[0015] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics.

[0016] Fourthly, this application provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics.

[0017] As can be seen from the above, before conducting the valuation of retired batteries based on degradation trajectory characteristics, this application needs to fit the health status change process of each retired battery based on the historical SOH sequence data of each retired battery to obtain a degradation trajectory, and determine a degradation trajectory feature vector based on each degradation trajectory, including degradation rate characteristics, degradation acceleration characteristics, degradation volatility characteristics, and current health status SOH characteristics; perform cluster analysis and degradation type classification on each retired battery based on the degradation trajectory feature vector to obtain a degradation type; determine the degradation prediction model corresponding to each degradation type, and use the degradation prediction model to predict the future SOH change trend of each retired battery, so as to construct a future degradation path based on the prediction results; determine the expected economic benefits generated by each retired battery in the remaining usable life based on each future degradation path, and determine the remaining economic value based on the expected economic benefits, so as to construct the individual residual value of each retired battery; construct a retired battery asset package planning model based on the individual residual value of the retired battery, and use the retired battery asset package planning model and the preset constraints to screen each retired battery to obtain a battery asset portfolio, so as to conduct a valuation of the batteries in the battery asset portfolio.

[0018] Therefore, this application first needs to fit the health status change process of each retired battery based on the historical SOH sequence data of each retired battery to obtain the decay trajectory, and then determine the decay trajectory feature vector based on each decay trajectory, including decay rate characteristics, decay acceleration characteristics, decay volatility characteristics, and current health status SOH characteristics; secondly, cluster analysis and decay type classification are performed on each retired battery based on the decay trajectory feature vector to obtain the decay type; a decay prediction model corresponding to each decay type is determined, and the decay prediction model is used to predict the future SOH change trend of each retired battery, so as to construct the future decay path based on the prediction results; then, the expected economic benefits generated by each retired battery in the remaining usable life are determined based on each future decay path, and the remaining economic value is determined based on the expected economic benefits, so as to construct the individual residual value of each retired battery; finally, a retired battery asset package planning model is constructed based on the individual residual value of the retired battery, and the retired battery is screened based on the retired battery asset package planning model and preset constraints to obtain a battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio. This improves the efficiency of assessing the value of retired batteries during the valuation process, thereby enhancing the user experience. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart of a method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics disclosed in this application; Figure 2 This is a schematic diagram of a retired battery value assessment device based on attenuation trajectory characteristics disclosed in this application. Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Currently, with the rapid development of the energy storage industry, a large number of energy storage power station batteries have gradually withdrawn from main frequency regulation / peak shaving services and entered the secondary utilization market after their capacity decays to 70%~80% of their initial capacity. These retired energy storage batteries still have considerable residual economic value in low-intensity application scenarios. However, the financialization of secondary utilization energy storage batteries faces three serious challenges: (1) significant individual differences make valuation difficult; (2) there is a lack of asset package optimization methods for securitization; and (3) the degradation path of retired batteries is more important than the current state of energy consumption (SOH). To address these challenges, this application provides a method for evaluating the value of retired batteries based on degradation trajectory characteristics, which can improve the efficiency of evaluating the value of retired batteries during the valuation process.

[0023] See Figure 1 As shown in the figure, this invention discloses a method for evaluating the value of decommissioned batteries based on decay trajectory characteristics, including: Step S11: Fit the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory, and determine the decay trajectory feature vector based on each decay trajectory, including decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics and current health status SOH characteristics.

[0024] In this embodiment, the first step is to obtain the SOH (State of Health) historical sequence data for each retired battery. SOH is a key indicator characterizing the battery's current performance state relative to its brand-new state, typically expressed as a percentage (100% indicates the battery condition conforms to specifications). For each retired battery, this embodiment needs to fit a degradation trajectory and extract a four-dimensional degradation feature vector based on its SOH historical sequence during its operation at the original energy storage power station.

[0025] Specifically, the SOH historical sequence data for each time point corresponding to each retired battery is determined; the retired battery refers to the battery used by the energy storage power station within a historical time period. Then, the SOH historical sequence data are arranged according to a preset time order to obtain an SOH time series, and the SOH change process of the retired battery is curve-fitted based on the SOH time series to obtain the corresponding decay trajectory.

[0026] After obtaining the decay trajectory, a decay trajectory feature vector is determined based on each decay trajectory, including decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics, and current health state (SOH) characteristics. The specific steps are as follows: The SOH change trend at each time point is determined based on the decay trajectory described above.

[0027] Based on the aforementioned degradation trajectories, the unit time SOH decrease for each retired battery is determined, and a degradation rate characteristic is constructed based on this unit time SOH decrease. The degradation rate characteristic reflects the magnitude of SOH decrease within a unit time and is a fundamental indicator characterizing the rate of battery degradation.

[0028] The decay rate change is determined based on the decay rate characteristics corresponding to each consecutive time point, and the decay acceleration characteristics are determined based on the decay rate change. The decay acceleration characteristics reflect the changing trend of the decay rate itself: if the acceleration is positive, it indicates that the decay is accelerating; if the acceleration is negative or close to zero, it indicates that the decay is stable or decelerating.

[0029] The degree of decay rate change is determined based on the decay rate characteristics corresponding to each of the aforementioned time nodes, and the decay volatility characteristics are determined based on the degree of decay rate change. The decay volatility characteristics reflect the smoothness or the severity of fluctuations in the decay rate sequence, and can be quantified using indicators such as the coefficient of variation (CV).

[0030] Determine the SOH data of each retired battery at the current time point, and determine the current health status SOH characteristics based on the SOH data.

[0031] A decay trajectory feature vector is constructed based on the SOH change trend, decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics, and the current health state SOH characteristics.

[0032] The final four-dimensional decay eigenvector can be expressed as: ; in, The SOH characteristic represents the current health status. Indicates the decay rate characteristic, Indicates the characteristics of decaying acceleration. This indicates the characteristics of decaying volatility.

[0033] Step S12: Based on the attenuation trajectory feature vector, perform cluster analysis and attenuation type classification on each of the retired batteries to obtain the attenuation type.

[0034] In this embodiment, the present application requires cluster analysis and attenuation type classification of each retired battery based on the attenuation trajectory feature vector to obtain the attenuation type. The specific steps are as follows: The attenuation trajectory feature vectors corresponding to each of the aforementioned retired batteries are standardized to obtain standardized attenuation trajectory feature vectors. The standardization process uses the Z-score standardization method, which involves subtracting the mean and dividing by the standard deviation to eliminate the influence of dimensional differences on distance calculation.

[0035] Based on the standardized decay trajectory feature vector, the feature distance between each retired battery is determined, and cluster analysis is performed on each retired battery based on the feature distance to obtain the cluster analysis results. Preferably, the cluster analysis uses the K-means++ clustering algorithm, with Euclidean distance as the similarity measure.

[0036] The specific implementation of the K-means++ clustering algorithm is as follows: Four initial cluster centers are selected using the K-means++ initialization method. This method selects initial centers through a probability distribution (points farther from existing centers have a higher probability of being selected), avoiding the sensitivity of standard K-means to initial values. The assignment and update steps are iteratively executed until the cluster center displacement is less than a preset displacement threshold or the maximum number of iterations (100) is reached.

[0037] In one specific implementation, the choice of the number of clusters k=4 is based on a joint criterion of the elbow method (selecting the inflection point of the SSE descent curve) and the silhouette coefficient (silhouette score > 0.5): the silhouette coefficient corresponding to each k value is calculated within the range of k=2 to k=8, and the k value with the largest silhouette coefficient is selected as the optimal number of clusters. In actual retired battery datasets, k=4 is usually optimal. Other clustering algorithms (DBSCAN (Density-Based Spatial Clustering of Applications with Noise), hierarchical clustering, Gaussian Mixture Model (GMM)) can all be used as alternatives under the guidance of k=4.

[0038] Based on the clustering analysis results, the similarity of the degradation behavior of each retired battery is determined. Then, based on the similarity of the degradation behavior, each retired battery is classified into a degradation type. The degradation types include stable degradation type (Type A) that meets preset stable degradation conditions, accelerated degradation type (Type B) that meets preset accelerated degradation conditions, fluctuating degradation type (Type C) that meets preset fluctuating degradation conditions, and rapid degradation type (Type D) that meets preset rapid degradation conditions.

[0039] It is worth mentioning that, based on statistical data of actual degradation of a large number of retired energy storage batteries, the criteria for the four-type classification are determined by the typical value ranges of the following four-dimensional feature vectors. Each feature is first standardized using Z-scores, and after clustering, the classification boundary is represented by the original physical quantities after de-standardization: Type A (robust decay type): Four-dimensional characteristic criteria: (1) Monthly average decay rate SOH loss / month < 0.15% (absolute value), and the standard deviation of the decay rate in the past 3 months < 0.05% / month, indicating slow and stable decay; (2) decay acceleration a < 0.01% / month² (acceleration close to zero or negative value), indicating that the decay is not accelerating, and may even be slightly decelerating (typical of the stable growth stage of SEI (Solid Electrolyte Interphase) membrane); (3) decay volatility CV < 0.3 (coefficient of variation), indicating that the time series of decay rate is smooth and there are no abnormal jumps; (4) current SOH > 72% (relatively high among retired batteries), indicating that the absolute health status at the time of retirement is good. Typical proportion: about 30% to 40% of the total retired batteries.

[0040] Type B (Accelerated Degradation): Four-dimensional characteristic criteria: (1) Monthly average degradation rate of 0.10-0.30% / month, with a monotonically increasing trend in the degradation rate over the past 3 months; (2) Degradation acceleration a > 0.03% / month², indicating that degradation is accelerating (typical SEI film rupture or electrolyte consumption acceleration stage), and the acceleration has been positive for the past 3 months; (3) Degradation volatility CV is 0.3-0.6, and the recent increase in volatility reflects a change in the degradation mode; (4) The current SOH is usually 65%-75%, which is close to Type A but the degradation trend is completely different. Typical proportion: about 20%-30% of the total retired batteries. This type of battery is a high-risk type that is of great concern.

[0041] Type C (Fluctuating Attenuation Type): Four-dimensional characteristic criteria: (1) The monthly average attenuation rate has no fixed range, but the key difference from Type A / B is that the time series of attenuation rate shows high volatility; (2) The sign of attenuation acceleration frequently switches (alternating between positive and negative) in the recent period, indicating that the attenuation mode is unstable; (3) Attenuation volatility CV > 0.6 (high coefficient of variation), and obvious abnormal jumps appear in the attenuation rate series (which may be related to BMS (Battery Management System) recording interruption, maintenance events, or intermittent thermal management failures); (4) The current SOH distribution is relatively wide, usually 55% to 80%. Typical proportion: about 15% to 25% of the total retired batteries.

[0042] Type D (Rapid Degradation): Four-dimensional characteristic criteria: (1) Average monthly degradation rate > 0.30% / month (absolute value), the fastest degradation among all retired batteries; (2) Degradation acceleration can be positive or close to zero (high degradation rate has reached steady state or is still accelerating), but the key is the high absolute value; (3) Degradation volatility CV < 0.4 (although the degradation is fast, it is relatively stable), which is different from the high volatility of Type C; (4) Current SOH is usually < 65%, and some batteries are even below 55%. Typical proportion: about 10% to 15% of the total retired batteries. These batteries are usually close to the economic boundary of cascade utilization.

[0043] The above classification boundary values ​​are empirical statistical values ​​based on retired lithium iron phosphate (LFP) energy storage batteries. For retired batteries of ternary lithium batteries or other chemical systems, the thresholds can be adjusted proportionally according to the typical degradation characteristics of that chemical system.

[0044] Step S13: Determine the attenuation prediction model corresponding to each of the attenuation types, and use the attenuation prediction model to predict the future SOH change trend of each of the retired batteries, so as to construct the future attenuation path based on the prediction results.

[0045] In this embodiment, the present application requires determining a degradation prediction model corresponding to each degradation type, and using the degradation prediction model to predict the future SOH (State of Health) trend of each retired battery, so as to construct a future degradation path based on the prediction results. Each degradation type is matched with a different future degradation prediction model to reflect the heterogeneous degradation behavior of the retired battery. The specific steps are as follows: The first degradation behavior characteristic corresponding to the stable degradation type is determined. Then, a preset exponential extrapolation model is used to predict the future SOH change trend of the retired battery corresponding to the stable degradation type based on the first degradation behavior characteristic, and the corresponding first prediction result is obtained.

[0046] Type A prediction model is an adaptive exponential extrapolation model, and its form is as follows: SOH(t) = SOH0 - k A × t α ; Where α ∈ [0.85, 1.0], k A The model was obtained by fitting the SOH sequence from the past 6 months using the least squares method. The value of α reflects the slightly decelerating trend of type A cells (a physical characteristic of slower degradation after SEI film stabilization). Prediction confidence interval: The 95% confidence interval width is ±8% of the predicted value (based on the statistical distribution of historical prediction errors for type A cells). Model update frequency: Parameters are refitted quarterly or when the cumulative decrease in SOH exceeds 2%.

[0047] The second degradation behavior characteristic corresponding to the accelerated degradation type is determined. Then, a preset exponential accelerated degradation model is used to predict the future SOH change trend of the retired battery corresponding to the accelerated degradation type based on the second degradation behavior characteristic, and the corresponding second prediction result is obtained.

[0048] Type B prediction model is an exponentially accelerating recession model (with early warning mechanism), and the model form is as follows: SOH(t) = SOH0 - k B × exp(β × t); Where β > 0 is the acceleration coefficient, k B The baseline decay rate is used. Parameter fitting: Nonlinear least squares fitting (Levenberg-Marquardt algorithm) is performed using SOH sequences from the past 6 months, with the fitting objective being to determine k. BAnd β. It's worth noting a key feature: the fitted value of β directly determines the pessimism of the prediction: the larger the β, the shorter the predicted remaining lifetime. Early warning mechanism: If the fitted value of β exceeds 0.05 / month, an acceleration warning is triggered, and the remaining lifetime prediction uses the upper limit of the 95% confidence interval of β (i.e., the most pessimistic acceleration scenario). Prediction confidence interval: The 95% confidence interval width is ±15% of the predicted value (wider than Type A, due to the additional uncertainty introduced by the acceleration trend).

[0049] The third decay behavior characteristic corresponding to the fluctuating decay type is determined. Then, a preset Gaussian process regression model is used to predict the future SOH change trend of the retired battery corresponding to the fluctuating decay type based on the third decay behavior characteristic, and the corresponding third prediction result is obtained.

[0050] The C-type prediction model is a Gaussian Process Regression (GPR) model (wide confidence interval). The model is a non-parametric Gaussian process regression, and the kernel function is a composite kernel consisting of a radial basis function (RBF) kernel and a white noise kernel. K(t i , t j ) = σ² × exp(-|t i -t j |² / 2L²) + σ n ² × δ ij ; Where σ² is the signal variance (reflecting the overall amplitude of attenuation), L is the length scale (reflecting the time dependence of attenuation), and σ n ² represents the noise variance (reflecting volatility). Hyperparameters are learned from the SOH sequence over the past 12 months by maximizing the log-marginal likelihood (L-BFGS optimizer (Limited-memory Broyden-Fletcher-Goldfarb-Shanno algorithm)). Predicted output: The predicted mean is used as a point estimate, and the predicted variance is directly used as a quantification of uncertainty: the predicted variance for C-type batteries is naturally greater than that for A-type and B-type batteries. Predicted confidence interval: The 95% confidence interval is ±22% of the predicted value (the widest of the three, reflecting the high uncertainty caused by high volatility).

[0051] Subsequently, the fourth decay behavior characteristic corresponding to the rapid decay type is determined. Then, a preset lower bound extrapolation model is used to predict the future SOH change trend of the retired battery corresponding to the rapid decay type based on the fourth decay behavior characteristic, and the corresponding fourth prediction result is obtained.

[0052] The Type D prediction model is a conservative lower bound extrapolation model, and its form is as follows: SOH(t) = SOH0 - k Dmin × t; Where, k Dmin This represents the maximum (not the average) monthly degradation rate over the past 6 months, i.e., the most conservative degradation rate estimate. Physical basis: D-type batteries degrade rapidly and relatively steadily (CV < 0.4). Extrapolating the maximum value provides a conservative but reasonable lower bound for residual value—avoiding overestimation of residual value leading to unexpected losses in the asset package, while also avoiding the complete abandonment of batteries with residual value. When the predicted SOH drops to 50% (the absolute lower limit of tiered utilization), it is marked as the end of the economic life. No confidence interval: The output is a conservative point estimate, and the selection weight of D-type batteries can be further reduced in asset package optimization.

[0053] The mathematical forms and parameters of the four differentiated prediction models mentioned above are all derived from statistical analysis of the actual degradation behavior of retired energy storage batteries. The model selection is automatically determined by the aforementioned K-means++ clustering classification, requiring no manual intervention. For batteries with ambiguous boundaries (cluster membership degree < 0.6), a weighted average of the prediction results of the two closest types can be used, with the weights being the normalized values ​​of the membership degrees.

[0054] Finally, based on the first prediction result, the second prediction result, the third prediction result, and the fourth prediction result, a corresponding future decay path is constructed.

[0055] Step S14: Determine the expected economic benefits generated by each of the retired batteries within their remaining usable lifespan based on each of the future degradation paths, and determine the remaining economic value based on the expected economic benefits, so as to construct the individual residual value of each retired battery based on the remaining economic value.

[0056] In this embodiment, the present application requires determining the expected economic benefits of each retired battery within its remaining usable lifespan based on each of the future degradation paths, and determining the remaining economic value based on the expected economic benefits, so as to construct the individual residual value of each retired battery based on the remaining economic value. The specific steps are as follows: Based on the future degradation path, the time when the SOH of each retired battery drops to the preset end point of its secondary use life is determined, and the remaining usable life of each retired battery is determined based on the time. The end point of the secondary use life can be preset as SOH dropping to 50% (the absolute lower limit of secondary use) or other thresholds set according to specific application scenarios.

[0057] The expected operating cycle is determined based on the remaining usable life of each of the retired batteries, and the expected economic benefits generated by each retired battery during the remaining usable life are determined based on the expected operating cycle. The expected economic benefits include, but are not limited to, the revenue generated by the batteries in secondary utilization scenarios through peak-valley arbitrage, backup power services, and other methods.

[0058] Based on a preset discounting method, the expected economic returns corresponding to each time point within the remaining usable lifespan are discounted to obtain the economic value corresponding to each time point. The discounting calculation takes into account the time value of money, discounting the expected returns of each future period to the current point in time.

[0059] The economic values ​​of each battery are summed to obtain the residual economic value of each retired battery over its lifetime, and the individual residual value of each retired battery is constructed based on the residual economic value of each battery.

[0060] Specifically, for each retired battery cell i, a corresponding prediction model is selected based on its degradation type to determine the remaining usable lifespan T. i The residual value is calculated using the discounted cash flow method over the remaining useful life: ; in, V i This represents the individual residual value of retired battery i. C tgt Indicates the target capacity. This represents the predicted SOH of battery i at time t. P ( t () represents the unit capacity revenue price at time t. η Indicates energy conversion efficiency. M ( t () represents the maintenance cost at time t. r This represents the discount rate.

[0061] Step S15: Construct a retired battery asset package planning model based on the residual value of the individual retired batteries, and use the retired battery asset package planning model and preset constraints to screen each retired battery to obtain a battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio.

[0062] In this embodiment, a retirement battery asset package planning model is constructed based on the residual value of each individual retired battery. This model is then used to screen each retired battery based on preset constraints to obtain a battery asset portfolio. The batteries within this portfolio are then valued. The specific steps are as follows: A retirement battery asset package planning model is constructed based on the residual value of each individual retired battery, and the optimization objective is set as maximizing the total residual value of each retired battery. Then, a corresponding battery selection scheme is generated based on the optimization objective and the retirement battery asset package planning model.

[0063] The objective function to be optimized is: ; in, ∈{0,1} is the decision variable, representing whether retired battery i is selected into the asset package.

[0064] Based on preset cash flow coverage constraints and the battery screening scheme, the first batch of decommissioned batteries that meet the preset future revenue coverage requirements are selected. The cash flow coverage constraint requires that the expected cash flows generated by the asset package in future periods can cover the corresponding operating costs and repayment requirements.

[0065] The proportion of retired batteries corresponding to each degradation type in the asset portfolio is determined based on preset degradation type diversification constraints. Then, based on these proportions and the first set of retired batteries to be processed, a second set of retired batteries to be processed is determined. The degradation type diversification constraint requires that the proportion of each degradation type in the asset portfolio not exceed 40%, in order to reduce the concentration risk of the asset portfolio for a specific degradation type.

[0066] Based on a preset State of Health (SOH) threshold constraint, retired batteries in the second batch of batteries to be processed whose current State of Health (SOH) is lower than the preset health state threshold are removed, resulting in a third batch of retired batteries to be processed. A battery asset portfolio is then constructed based on these third batch of retired batteries. The SOH threshold constraint ensures that the current health state of the selected batteries is not lower than the minimum requirement for cascade utilization.

[0067] For the aforementioned 0-1 integer programming problem, this application provides two solution strategies for different sizes of retired battery cells: Strategy 1 (Exact Solution, Applicable to Sections N≤500): Use Branch and Bound (B&B or BnB). Branch and bound is the most commonly used algorithm for solving integer programming problems. It can be regarded as an algorithm that exhaustively searches for feasible solutions. However, unlike exhaustive search, the branch and bound algorithm calculates the upper or lower bound of a branch before searching for it. If the bound is not better than the current best solution, then the branch is discarded. Commercial solvers such as CPLEX or Gurobi (MIP, Mixed Integer Programming) or the open-source solver SCIP (Solving Constraint Integer Programs) can be used. In the preprocessing stage, variable reduction is performed to eliminate obviously infeasible batteries (SOH < SOH_threshold or negative residuals), compressing the problem size to 60%–80% of the original. Typical solution time: approximately 5 seconds for N=200, approximately 120 seconds for N=500 (Intel i7 / 32GB RAM).

[0068] Strategy Two (Heuristic Solution, Applicable to Sections N>500): Use a Genetic Algorithm (GA) to find an approximate optimal solution. Encoding Scheme: Binary chromosomes, length N. ∈{0,1} indicates whether battery i is selected. Parameter settings: Population size 200, roulette wheel selection, two-point crossover (probability 0.8), bit flip mutation (probability 0.05), elite retention strategy (the 5 best individuals in each generation directly enter the next generation). Constraint handling: Cash flow coverage constraint and decentralization constraint are integrated into the fitness function through a penalty function method—the fitness of individuals that violate the constraints is multiplied by a penalty coefficient (0.5 / number of constraints violated), making them disadvantaged in selection. Termination condition: After 500 iterations or the fitness of the best individual fluctuates by <1% in the last 50 generations.

[0069] It is worth mentioning that the choice of the above two strategies does not constitute a limitation on the embodiments of this application: any algorithm capable of solving the above 0-1 Integer Programming problem (such as simulated annealing, particle swarm optimization, ant colony optimization) falls within the protection scope of this application.

[0070] Finally, based on the battery asset portfolio, risk indicators for the asset package are determined, including the weighted average state of equilibrium (SOH) of retired batteries, the composition of degradation types, the cash flow coverage multiple, the probability of stress scenario coverage, and the degradation correlation matrix. Based on these risk indicators, an asset securitization assessment is conducted on the batteries in the battery asset portfolio, yielding the assessment results. These risk indicators can be directly used as underlying asset data inputs for ABS (Asset-Backed Securities) pricing.

[0071] As can be seen from the above, the embodiments of this application first need to fit the health status change process of each retired battery based on the historical SOH sequence data of each retired battery to obtain the decay trajectory, and determine the decay trajectory feature vector based on each decay trajectory, including decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics, and current health status SOH characteristics; secondly, cluster analysis and decay type classification are performed on each retired battery based on the decay trajectory feature vector to obtain the decay type; a decay prediction model corresponding to each decay type is determined, and the decay prediction model is used to predict the future SOH change trend of each retired battery, so as to construct the future decay path based on the prediction results; then, the expected economic benefits generated by each retired battery in the remaining usable life are determined based on each future decay path, and the remaining economic value is determined based on the expected economic benefits, so as to construct the individual residual value of each retired battery; finally, a retired battery asset package planning model is constructed based on the individual residual value of the retired battery, and the retired battery is screened based on the retired battery asset package planning model and preset constraints to obtain the battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio. This improves the efficiency of assessing the value of retired batteries during the valuation process, thereby enhancing the user experience.

[0072] Accordingly, see Figure 2 As shown, this application also provides a device for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics, comprising: The decay trajectory feature vector generation module 11 is used to fit the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory, and to determine the decay trajectory feature vector based on each decay trajectory, including decay rate features, decay acceleration features, decay fluctuation features and current health status SOH features. The attenuation type determination module 12 is used to perform cluster analysis and attenuation type classification on each of the retired batteries based on the attenuation trajectory feature vector to obtain the attenuation type. The future degradation path construction module 13 is used to determine the degradation prediction model corresponding to each degradation type, and use the degradation prediction model to predict the future SOH change trend of each retired battery, so as to construct the future degradation path based on the prediction result. The expected economic benefit determination module 14 is used to determine the expected economic benefit generated by each of the retired batteries in the remaining usable life based on each of the future degradation paths, and to determine the remaining economic value based on the expected economic benefit, so as to construct the individual residual value of the retired battery based on each of the remaining economic values. The battery asset portfolio generation module 15 is used to construct a retired battery asset package planning model based on the residual value of the individual retired batteries, and to use the retired battery asset package planning model and preset constraints to screen each retired battery to obtain a battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio.

[0073] In some specific embodiments, the attenuation trajectory feature vector generation module 11 may specifically include: The SOH historical sequence acquisition unit is used to determine the SOH historical sequence data of each time node corresponding to each retired battery; the retired battery is the battery of the energy storage power station after it has been used within a historical time period.

[0074] The decay trajectory fitting unit is used to arrange the historical SOH sequence data of each battery according to a preset time order to obtain the SOH time series, and to perform curve fitting on the SOH change process of the retired battery based on the SOH time series to obtain the corresponding decay trajectory.

[0075] In some specific embodiments, the attenuation trajectory feature vector generation module 11 may specifically include: The SOH change trend determination unit is used to determine the SOH change trend at each time node based on each of the decay trajectories. The decay rate feature construction unit is used to determine the unit time SOH decrease of each retired battery based on each decay trajectory, so as to construct the decay rate feature based on the unit time SOH decrease. The decay acceleration feature determination unit is used to determine the decay rate change based on the decay rate feature corresponding to each consecutive time node, so as to determine the decay acceleration feature based on the decay rate change. The decay volatility characteristic determination unit is used to determine the degree of decay rate change based on the decay rate characteristics corresponding to each of the time nodes, so as to determine the decay volatility characteristic based on the degree of decay rate change. The current health status (SOH) feature determination unit is used to determine the SOH data of each of the retired batteries at the current time node, so as to determine the current health status (SOH) feature based on the SOH data of each battery. The decay trajectory feature vector generation subunit is used to construct a decay trajectory feature vector based on the SOH change trend, the decay rate feature, the decay acceleration feature, the decay fluctuation feature, and the current health state SOH feature.

[0076] In some specific embodiments, the attenuation type determination module 12 may specifically include: The standardization processing unit is used to standardize the attenuation trajectory feature vectors corresponding to each of the retired batteries to obtain standardized attenuation trajectory feature vectors. The clustering analysis unit is used to determine the feature distance between each of the retired batteries based on the standardized decay trajectory feature vector, and to perform clustering analysis on each of the retired batteries based on the feature distance to obtain the clustering analysis results; The type classification unit is used to determine the degree of similarity of the degradation behavior of each retired battery based on the clustering analysis results, and then classify each retired battery into a type based on the degree of similarity of degradation behavior to obtain the degradation type; wherein, the degradation type includes a stable degradation type that meets a preset stable degradation condition, an accelerated degradation type that meets a preset accelerated degradation condition, a fluctuating degradation type that meets a preset fluctuating degradation condition, and a rapid degradation type that meets a preset rapid degradation condition.

[0077] In some specific embodiments, the future attenuation path construction module 13 may specifically include: The first prediction unit is used to determine the first decay behavior characteristics corresponding to the stable decay type, and then use a preset exponential extrapolation model and based on the first decay behavior characteristics to predict the future SOH change trend of the retired battery corresponding to the stable decay type, and obtain the corresponding first prediction result. The second prediction unit is used to determine the second degradation behavior characteristics corresponding to the accelerated degradation type, and then use a preset exponential accelerated degradation model and based on the second degradation behavior characteristics to predict the future SOH change trend of the retired battery corresponding to the accelerated degradation type, and obtain the corresponding second prediction result. The third prediction unit is used to determine the third decay behavior characteristics corresponding to the fluctuating decay type, and then use a preset Gaussian process regression model and based on the third decay behavior characteristics to predict the future SOH change trend of the retired battery corresponding to the fluctuating decay type, and obtain the corresponding third prediction result. The fourth prediction unit is used to determine the fourth decay behavior characteristics corresponding to the rapid decay type, and then use a preset lower bound extrapolation model and the fourth decay behavior characteristics to predict the future SOH change trend of the retired battery corresponding to the rapid decay type, so as to obtain the corresponding fourth prediction result. The future decay path construction unit is used to construct a corresponding future decay path based on the first prediction result, the second prediction result, the third prediction result and the fourth prediction result.

[0078] In some specific embodiments, the expected economic benefit determination module 14 may specifically include: The remaining usable life determination unit is used to determine the time when the SOH of each retired battery drops to the preset end of the tiered utilization life based on the future degradation path, so as to determine the remaining usable life corresponding to each retired battery based on the time. An expected economic benefit calculation unit is used to determine the expected operating cycle based on the remaining usable life of each of the retired batteries, and to determine the expected economic benefit generated by each of the retired batteries during the remaining usable life based on the expected operating cycle. The discount calculation unit is used to perform discount calculation on the expected economic benefits corresponding to each time point in the remaining usable life period based on a preset discount calculation method, so as to obtain the economic value corresponding to each time point. The individual residual value construction unit is used to accumulate the economic values ​​of each battery to obtain the remaining economic value of each retired battery over its lifetime, and to construct the individual residual value of the retired battery based on the remaining economic value of each battery.

[0079] In some specific embodiments, the battery asset portfolio generation module 15 may specifically include: The optimization target setting unit is used to construct a retired battery asset package planning model based on the residual value of each retired battery, and set the maximization of the total residual value of each retired battery as the optimization target. Then, based on the optimization target and the retired battery asset package planning model, a corresponding battery screening scheme is generated. A cash flow constraint screening unit is used to screen the first decommissioned battery that meets the preset future revenue coverage requirements based on preset cash flow coverage constraints and the battery screening scheme. A type-diversification constraint screening unit is used to determine the proportion of retired batteries corresponding to each of the attenuation types in the asset portfolio based on preset attenuation type diversification constraint conditions, so as to determine the second retired battery to be processed based on the proportion and the first retired battery to be processed. The SOH threshold constraint screening unit is used to remove retired batteries in the second batch of batteries to be processed whose current health status SOH is lower than the preset health status threshold based on the preset SOH threshold constraint conditions, so as to obtain the third batch of retired batteries to be processed, and to construct a battery asset portfolio based on the third batch of retired batteries to be processed. The risk indicator assessment unit is used to determine the asset package risk indicators based on the battery asset portfolio, including the weighted average SOH of retired batteries, the composition of degradation types, the cash flow coverage multiple, the probability of stress scenario coverage, and the degradation correlation matrix. Based on the asset package risk indicators, the unit performs asset securitization assessment on the batteries in the battery asset portfolio to obtain the assessment results.

[0080] Furthermore, embodiments of this application also disclose an electronic device, Figure 3This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the decommissioned battery value assessment method based on attenuation trajectory characteristics disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0081] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0082] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0083] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the retired battery value assessment method based on attenuation trajectory characteristics executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0084] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned disclosed method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0085] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0086] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0087] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0088] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0089] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for evaluating the value of decommissioned batteries based on degradation trajectory characteristics, characterized in that, include: Based on the historical SOH sequence data of each retired battery, the health status change process of each retired battery is fitted to obtain the decay trajectory, and based on each decay trajectory, a decay trajectory feature vector including decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics and current health status SOH characteristics is determined. Based on the decay trajectory feature vector, cluster analysis and decay type classification are performed on each of the retired batteries to obtain the decay type; Determine the attenuation prediction model corresponding to each of the attenuation types, and use the attenuation prediction model to predict the future SOH change trend of each of the retired batteries, so as to construct the future attenuation path based on the prediction results; Based on each of the future degradation paths, the expected economic benefits generated by each of the retired batteries during their remaining usable lifespan are determined, and the remaining economic value is determined based on the expected economic benefits, so as to construct the individual residual value of each retired battery based on the remaining economic value. Based on the residual value of the individual retired batteries, a planning model for a retired battery asset package is constructed. The retired batteries are then screened using the planning model and based on preset constraints to obtain a battery asset portfolio, which is then used to evaluate the value of the batteries in the battery asset portfolio.

2. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to claim 1, characterized in that, The process of fitting the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory includes: Determine the SOH historical sequence data for each time point corresponding to each retired battery; the retired battery refers to the battery used by the energy storage power station within the historical time period. The SOH historical sequence data are arranged according to a preset time order to obtain an SOH time series. Based on the SOH time series, the SOH change process of the retired battery is curve fitted to obtain the corresponding decay trajectory.

3. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to claim 1, characterized in that, The determination of the attenuation trajectory feature vector based on each of the attenuation trajectories, including attenuation rate characteristics, attenuation acceleration characteristics, attenuation fluctuation characteristics, and current health state (SOH) characteristics, includes: The SOH change trend at each time point is determined based on the aforementioned decay trajectories. Based on each of the decay trajectories, the unit time SOH decrease of each of the retired batteries is determined, and a decay rate feature is constructed based on the unit time SOH decrease. The decay rate change is determined based on the decay rate characteristics corresponding to each consecutive time node, and the decay acceleration characteristics are determined based on the decay rate change. The degree of decay rate change is determined based on the decay rate characteristics corresponding to each of the aforementioned time nodes, and the decay fluctuation characteristics are determined based on the degree of decay rate change. Determine the SOH data of each of the retired batteries at the current time point, and determine the current health status SOH characteristics based on the SOH data of each battery; A decay trajectory feature vector is constructed based on the SOH change trend, decay rate characteristics, decay acceleration characteristics, decay fluctuation characteristics, and the current health state SOH characteristics.

4. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to claim 1, characterized in that, The method involves clustering and classifying the attenuation types of each retired battery based on the attenuation trajectory feature vector to obtain attenuation types, including: The attenuation trajectory feature vectors corresponding to each of the retired batteries are standardized to obtain standardized attenuation trajectory feature vectors. Based on the standardized decay trajectory feature vector, the feature distance between each of the retired batteries is determined, and cluster analysis is performed on each of the retired batteries based on the feature distance to obtain the cluster analysis results; Based on the cluster analysis results, the degree of similarity of the degradation behavior of each retired battery is determined. Then, based on the degree of similarity of the degradation behavior, each retired battery is classified into different types to obtain the degradation type. The attenuation types include stable attenuation type that meets preset stable attenuation conditions, accelerated attenuation type that meets preset accelerated attenuation conditions, fluctuating attenuation type that meets preset fluctuating attenuation conditions, and fast attenuation type that meets preset fast attenuation conditions.

5. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to claim 4, characterized in that, The step of determining the attenuation prediction model corresponding to each of the attenuation types, and using the attenuation prediction model to predict the future SOH change trend of each of the retired batteries, so as to construct the future attenuation path based on the prediction results, includes: The first decay behavior characteristic corresponding to the stable decay type is determined. Then, a preset exponential extrapolation model is used to predict the future SOH change trend of the retired battery corresponding to the stable decay type based on the first decay behavior characteristic, and the corresponding first prediction result is obtained. Determine the second degradation behavior characteristic corresponding to the accelerated degradation type, and then use a preset exponential accelerated degradation model and based on the second degradation behavior characteristic to predict the future SOH change trend of the retired battery corresponding to the accelerated degradation type, and obtain the corresponding second prediction result. The third decay behavior feature corresponding to the fluctuating decay type is determined. Then, a preset Gaussian process regression model is used to predict the future SOH change trend of the retired battery corresponding to the fluctuating decay type based on the third decay behavior feature, and the corresponding third prediction result is obtained. The fourth decay behavior feature corresponding to the rapid decay type is determined. Then, the future SOH change trend of the retired battery corresponding to the rapid decay type is predicted by using a preset lower bound extrapolation model and based on the fourth decay behavior feature, and the corresponding fourth prediction result is obtained. Based on the first prediction result, the second prediction result, the third prediction result, and the fourth prediction result, a corresponding future decay path is constructed.

6. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to claim 1, characterized in that, The process of determining the expected economic benefits of each retired battery over its remaining usable life based on each of the future degradation paths, and determining the remaining economic value based on the expected economic benefits, to construct the individual residual value of each retired battery based on the remaining economic value, includes: Based on the future degradation path, the time when the SOH of each retired battery drops to the preset end of its tiered utilization life is determined, and the remaining usable life corresponding to each retired battery is determined based on the time. The expected operating cycle is determined based on the remaining usable life of each of the retired batteries, and the expected economic benefits generated by each of the retired batteries during the remaining usable life are determined based on the expected operating cycle. Based on a preset discounting calculation method, the expected economic benefits corresponding to each time point in the remaining usable life are discounted to obtain the economic value corresponding to each time point. The economic values ​​of each battery are summed to obtain the residual economic value of each retired battery over its lifetime, and the individual residual value of each retired battery is constructed based on the residual economic value of each battery.

7. The method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics according to any one of claims 1 to 6, characterized in that, The process involves constructing a retirement battery asset package planning model based on the individual residual values ​​of the retired batteries, and then using this model to screen each retired battery based on preset constraints to obtain a battery asset portfolio. The process also includes evaluating the value of the batteries within the battery asset portfolio. A retirement battery asset package planning model is constructed based on the residual value of each individual retired battery, and the optimization objective is to maximize the total residual value of each retired battery. Then, a corresponding battery selection scheme is generated based on the optimization objective and the retirement battery asset package planning model. Based on the preset cash flow coverage constraints and the battery screening scheme, the first retired battery to be processed that meets the preset future revenue coverage requirements is selected. Based on the preset attenuation type diversification constraint, the proportion of retired batteries corresponding to each attenuation type in the asset portfolio is determined, and the second retired battery to be processed is determined based on the proportion and the first retired battery to be processed. Based on the preset SOH threshold constraint, the retired batteries in the second batch of batteries to be processed that have a current health status SOH lower than the preset health status threshold are removed to obtain the third batch of batteries to be processed, and a battery asset portfolio is constructed based on the third batch of batteries to be processed. Based on the battery asset portfolio, risk indicators for the asset package are determined, including the weighted average SOH of retired batteries, the composition of degradation types, the cash flow coverage multiple, the probability of stress scenario coverage, and the degradation correlation matrix. Based on the asset package risk indicators, the batteries in the battery asset portfolio are assessed for asset securitization, and the assessment results are obtained.

8. A device for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics, characterized in that, include: The decay trajectory feature vector generation module is used to fit the health status change process of each retired battery based on the SOH historical sequence data of each retired battery to obtain the decay trajectory, and to determine the decay trajectory feature vector based on each decay trajectory, including decay rate features, decay acceleration features, decay fluctuation features and current health status SOH features. The attenuation type determination module is used to perform cluster analysis and attenuation type classification on each of the retired batteries based on the attenuation trajectory feature vector to obtain the attenuation type; The future degradation path construction module is used to determine the degradation prediction model corresponding to each degradation type, and use the degradation prediction model to predict the future SOH change trend of each retired battery, so as to construct the future degradation path based on the prediction results. The expected economic benefit determination module is used to determine the expected economic benefits generated by each of the retired batteries within their remaining usable life based on each of the future degradation paths, and to determine the remaining economic value based on the expected economic benefits, so as to construct the individual residual value of the retired battery based on each of the remaining economic values. The battery asset portfolio generation module is used to construct a retired battery asset package planning model based on the residual value of the individual retired batteries, and to use the retired battery asset package planning model and preset constraints to screen each retired battery to obtain a battery asset portfolio, so as to evaluate the value of the batteries in the battery asset portfolio.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the method for evaluating the value of decommissioned batteries based on attenuation trajectory characteristics as described in any one of claims 1 to 7.

Images