A Joint Estimation Method for Multidimensional Battery Health Indicators Applied to Electric Vertical Takeoff and Landing Aircraft

Abstract

The application discloses a battery multi-dimensional health index joint estimation method applied to an electric vertical take-off and landing aircraft, and comprises the following steps: offline acquisition of model input and output parameter matrix; normalization processing of input and output data; training of an estimation model by using a multi-output least square support vector regression algorithm; recording of take-off stage battery current and voltage data; normalization processing of the recorded data and substitution into the estimation model; and inverse normalization processing of the output health index. The method has the following three advantages: (1) the application comprehensively considers battery energy and power recession, and can more comprehensively and accurately reflect the health state of the airborne battery system; (2) the application takes the take-off stage battery current and voltage data set as the input of the estimation model, can comprehensively mine the battery aging information, and improves the accuracy and generalization of the model; and (3) the multi-output least square support vector regression algorithm used in the application improves the estimation accuracy while ensuring the training efficiency of the model.

Description

Technical Field

[0001] This invention belongs to the fields of battery technology and new energy aircraft, and particularly relates to a method for jointly estimating multidimensional health indicators of batteries for electric vertical take-off and landing aircraft. Background Technology

[0002] Electric vertical take-off and landing (eVTOL) aircraft, employing electric propulsion systems, have garnered widespread attention and significant development due to their low noise and zero emissions, gradually becoming a crucial trend in low-altitude aircraft development. Among these technologies, the power battery system, as the primary or sole power source for eVTOLs, directly determines the feasibility and safety of the aircraft. Lithium-ion batteries, with their high specific energy / specific power, low self-discharge rate, and good cycle stability, are increasingly being applied to lightweight and compact aircraft. Compared to electric vehicles and grid energy storage, research on power battery systems for aircraft applications is still in its early stages.

[0003] As the performance of lithium-ion battery systems degrades, their safe operating range gradually changes. To quantitatively describe the energy degradation and power decay characteristics of batteries, capacity and internal resistance parameters are generally selected as health indicators. A decrease in battery capacity directly shortens the aircraft's range; an increase in internal resistance compresses the battery's usable state of charge range, not only reducing the aircraft's range but also decreasing its safety and reliability during landing. Therefore, it is necessary to comprehensively and accurately perceive the capacity and internal resistance parameters of airborne batteries in conjunction with the characteristics of eVTOL operating conditions to ensure efficient, reliable, and safe flight. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a method for jointly estimating multidimensional health indicators of batteries for electric vertical takeoff and landing (EVVT) aircraft. This estimation method is based on battery current and voltage data during the takeoff phase of the EVVT aircraft and employs a multi-output least squares support vector regression algorithm to comprehensively and accurately estimate the three-dimensional health indicators of battery capacity, ohmic internal resistance, and polarization internal resistance.

[0005] The present invention is achieved through the following technical solution.

[0006] This invention provides a method for jointly estimating multidimensional health indicators of batteries for electric vertical takeoff and landing aircraft, the method comprising the following steps:

[0007] S1. Based on the aging test data of the power battery for electric vertical take-off and landing aircraft, obtain the input parameter matrix and output parameter matrix of the battery health index estimation model;

[0008] S2, normalize the input parameter set and output parameter set data obtained in step S1, and use them as the training set for the battery multidimensional health index estimation model.

[0009] S3. Based on the training set obtained in step S2, a multi-dimensional battery health index joint estimation model based on the takeoff phase data of an electric vertical takeoff and landing aircraft is constructed and stored using the multi-output least squares support vector regression algorithm.

[0010] S4 records and stores the battery time-current sequence and time-voltage sequence during the takeoff phase of an electric vertical takeoff and landing aircraft;

[0011] S5. After the takeoff phase ends, the current and voltage data recorded in step S4 are normalized and a normalized input parameter set is formed.

[0012] S6. Substitute the normalized input parameter set obtained in step S5 into the joint estimation model of battery multidimensional health indicators obtained in step S3 to obtain the model output parameter set.

[0013] S7, the model output parameter set obtained in the denormalization process S6, yields the multidimensional health indicators of the battery.

[0014] In step S1, the battery current and voltage data of the electric vertical takeoff and landing aircraft during the takeoff phase are filtered and organized into an input parameter matrix for the battery health index estimation model, with the input parameter variables having a dimension of 2t. phase / T s , where t phase T is the duration of the takeoff phase. s The sampling period.

[0015] In step S1, the battery capacity, ohmic internal resistance, and the sum of ohmic internal resistance and polarization internal resistance are used as the output parameter matrix of the battery health index estimation model. The battery capacity is obtained by using the ampere-hour integration method on the time-current sequence data during the constant current discharge phase of the reference performance test. The expression for calculating the battery ohmic internal resistance is:

[0016]

[0017] Among them, R o V is the ohmic internal resistance of the battery. t,0 and V t,1 These are the battery terminal voltage values ​​at the initial moment of step current excitation and 1 second later, respectively. L,0 and I L,1 These are the current values ​​at the initial moment of the step current excitation and 1 second later, respectively. The formula for calculating the sum of the battery's ohmic internal resistance and polarization internal resistance is:

[0018]

[0019] Among them, R p V is the internal resistance of the battery polarization. t,30 and I L,30 These are the battery terminal voltage and current values ​​30 seconds after a step current excitation.

[0020] In step S2, the input parameter matrix and the output parameter matrix are combined to form a training set. The number of samples in the training set is equal to the number of aging cycles of the batteries in the dataset. The expression for normalizing the training set parameters is as follows:

[0021]

[0022] Where, x i For the training set parameters, x i,norm For the normalized training set parameters, x i,min and x i,max These are the minimum and maximum values ​​for the same class of data in the training set, respectively.

[0023] In step S3, the training steps for the battery multidimensional health index estimation model based on the multi-output least squares support vector regression algorithm are as follows:

[0024] S3.1 Set the hyperparameter set;

[0025] S3.2 Calculate and obtain the weighting coefficient matrix W and the deviation vector b;

[0026] S3.3 Obtain the battery multidimensional health index estimation model and calculate the training error. The expression for the battery multidimensional health index estimation model is as follows:

[0027]

[0028] Where f(·) is the output regression function of the battery multidimensional health index estimation model, and X is the input parameter matrix. It is a nonlinear mapping function; the training error expression is:

[0029]

[0030] Where RMSE is the root mean square error, y i,ref and y i,est These are the reference and estimated values ​​for battery health indicators, respectively, and l is the number of aging cycles of the batteries in the dataset.

[0031] S3.4 Determine whether the accuracy of the trained model meets the requirements; if it does, proceed to step S3.5; otherwise, proceed to step S3.1.

[0032] S3.5 saves the battery multidimensional health index estimation model.

[0033] In step S7, the expression for the inverse normalization of the model output parameters is:

[0034] y i,anti-norm =y i (y i,max -y i,min )+y i,min

[0035] Among them, y i,anti-norm The set of output parameters for the inverse normalized model, y i The model outputs a parameter set, y i,min and y i,max These are the minimum and maximum values ​​of the model's output parameter set for the same class of datasets, respectively.

[0036] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0037] 1. This invention uses battery capacity, ohmic internal resistance, and the sum of ohmic internal resistance and polarization internal resistance as the output parameter matrix of the battery health index estimation model. It comprehensively considers the characteristics of battery energy degradation and power decay, and can more comprehensively and accurately reflect the health status of the battery system for electric vertical take-off and landing aircraft.

[0038] 2. Unlike battery health status estimation methods based on voltage feature points during flight, this invention uses the battery current and voltage dataset of electric vertical takeoff and landing aircraft during the takeoff phase as the input parameter matrix of the battery health index estimation model, which can more comprehensively and accurately mine battery aging information and improve the accuracy and generalization of the trained model.

[0039] 3. The battery multidimensional health index joint estimation method based on multi-output least squares support vector regression algorithm constructed in this invention takes into account the coupling relationship between the output variables, and improves the overall accuracy of the model output while ensuring the model training efficiency. Attached Figure Description

[0040] Figure 1 This is a flowchart of the battery health index joint estimation method of the present invention.

[0041] Figure 2 This is a flowchart illustrating the training process of the battery multidimensional health index estimation model based on the multi-output least squares support vector regression algorithm of this invention.

[0042] Figure 3 To compare the estimated battery capacity obtained using this invention with the actual measured value.

[0043] Figure 4 To compare the estimated value of the battery internal resistance obtained by using this invention with the actual measured value. Detailed Implementation

[0044] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0045] This invention provides a method for jointly estimating multi-dimensional health indicators of batteries for electric vertical takeoff and landing aircraft, enabling efficient and safe management of airborne power batteries.

[0046] A method for jointly estimating multidimensional health indicators of batteries for electric vertical takeoff and landing aircraft is as follows: Figure 1 As shown, this method is mainly divided into two parts: the first part is the offline training part, and the second part is the online state estimation part; the two parts will be further explained below.

[0047] The offline training component includes the following steps:

[0048] 1) Based on the aging test data of the power battery for electric vertical take-off and landing aircraft, obtain the input parameter matrix and output parameter matrix of the battery health index estimation model;

[0049] The battery capacity, ohmic internal resistance, and the sum of ohmic internal resistance and polarization internal resistance are used as the output parameter matrix of the battery health index estimation model. The battery capacity is obtained by using the ampere-hour integration method on the time-current sequence data during the constant current discharge phase of the reference performance test. The expression for calculating the battery ohmic internal resistance is as follows:

[0050]

[0051] Among them, R o V is the ohmic internal resistance of the battery. t ,0 and V t I and I represent the battery terminal voltage values ​​at the initial moment of step current excitation and 1 second later, respectively. L ,0 and I L , 1 represent the current values ​​at the initial moment of the step current excitation and 1 second later, respectively. The formula for calculating the sum of the battery's ohmic internal resistance and polarization internal resistance is:

[0052]

[0053] Among them, R p V is the internal resistance of the battery polarization. t,30 and I L,30 These are the battery terminal voltage and current values ​​30 seconds after a step current excitation.

[0054] 2) Normalize the input parameter set and output parameter set data obtained in step 1) and use them as the training set for the battery multidimensional health index estimation model.

[0055] The input parameter matrix and output parameter matrix are combined to form the training set. The number of samples in the training set is equal to the number of aging cycles of the batteries in the dataset. The expression for normalizing the parameters of the training set is as follows:

[0056]

[0057] Where, x i For the training set parameters, x i,norm For the normalized training set parameters, x i,min and x i,max These are the minimum and maximum values ​​for the same class of data in the training set, respectively.

[0058] 3) Based on the training set obtained in step 2), a multi-dimensional battery health index joint estimation model based on the takeoff phase data of electric vertical takeoff and landing aircraft is constructed and stored using the multi-output least squares support vector regression algorithm.

[0059] The training steps for the battery multidimensional health index estimation model based on the multi-output least squares support vector regression algorithm are as follows: Figure 2 As shown, specifically:

[0060] 3.1) Set the hyperparameter set;

[0061] 3.2) Calculate and obtain the weighting coefficient matrix W and the deviation vector b;

[0062] 3.3) Obtain the battery multidimensional health index estimation model and calculate the training error.

[0063] The expression of type is:

[0064]

[0065] Where f(·) is the output regression function of the battery multidimensional health index estimation model, and X is the input parameter matrix. It is a nonlinear mapping function; the training error expression is:

[0066]

[0067] Where RMSE is the root mean square error, y i,ref and y i,est These are the reference and estimated values ​​for battery health indicators, respectively, and l is the number of aging cycles of the batteries in the dataset.

[0068] 3.4) Determine if the accuracy of the trained model meets the requirements; if it does, proceed to step 3.5); otherwise, proceed to step 3.1).

[0069] 3.5) Save the battery multidimensional health index estimation model.

[0070] The online status estimation section includes the following steps:

[0071] 1) During the takeoff phase of an electric vertical takeoff and landing aircraft, record and store the battery time-current sequence and time-voltage sequence;

[0072] 2) After the takeoff phase is completed, the current and voltage data recorded in step 1) are normalized and a normalized input parameter set is formed.

[0073] 3) Substitute the normalized input parameter set obtained in step 2) into the joint estimation model of battery multidimensional health indicators obtained in step 3) of the offline training part to obtain the model output parameter set;

[0074] 4) The model output parameter set obtained in step 3) is processed by inverse normalization to obtain the multidimensional health index of the battery.

[0075] The expression for the inverse normalization of the model output parameters is as follows:

[0076] y i,anti-norm =y i (y i,max -y i,min )+y i,min

[0077] Among them, y i,anti-norm The set of output parameters for the inverse normalized model, y i The model outputs a parameter set, y i,min and y i,max These are the minimum and maximum values ​​of the model's output parameter set for the same class of datasets, respectively.

[0078] In this embodiment, the applied object is a ternary lithium-ion battery with a nominal capacity of 3.0 Ah. However, it is not limited to this in practical applications. The estimated values ​​of battery capacity and internal resistance obtained using this invention are compared with the actual measured values ​​as follows: Figure 3 and Figure 4 As shown, the estimated battery capacity and internal resistance values ​​can track the reference values ​​well and capture the corresponding nonlinear evolution trends. Therefore, this method can effectively achieve a comprehensive and accurate characterization of battery aging status.

[0079] In summary, the present invention provides a method for jointly estimating multidimensional health indicators of batteries for electric vertical takeoff and landing (EVVT) aircraft, comprising: obtaining input parameter matrices and output parameter matrices of a battery health indicator estimation model based on aging test data of the power batteries used in EVVT; normalizing the input parameter set and output parameter set data to serve as the training set for the battery multidimensional health indicator estimation model; constructing and storing a battery multidimensional health indicator joint estimation model based on EVVT takeoff phase data using a multi-output least squares support vector regression algorithm; recording and storing battery time-current and time-voltage sequences during the EVVT takeoff phase; normalizing the recorded current and voltage data after the takeoff phase to form a normalized input parameter set; substituting the normalized input parameter set into the battery multidimensional health indicator joint estimation model to obtain the model output parameter set; and performing inverse normalization on the model output parameter set to obtain the battery multidimensional health indicators. The estimation method proposed in this invention has the following three advantages: (1) This invention comprehensively considers the characteristics of battery energy decay and power attenuation, and can more comprehensively and accurately reflect the health status of the battery system used in electric vertical take-off and landing aircraft; (2) Compared with the battery health status estimation method based on voltage feature points during flight, the battery health index estimation method based on the battery current and voltage dataset during take-off provided by this invention can more comprehensively and accurately mine battery aging information, and improve the accuracy and generalization of the trained model; (3) The battery multidimensional health index joint estimation method constructed in this invention considers the coupling relationship between the output variables, and improves the overall accuracy of the model output while ensuring the model training efficiency.

[0080] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0081] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0082] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for multi-dimensional battery health index joint estimation applied to an electric vertical take-off and landing aircraft, characterized in that, The method comprises the following steps: S1, according to the power battery aging test data of the electric vertical take-off and landing aircraft, obtaining the input parameter matrix and the output parameter matrix of the battery health index estimation model; S2, normalizing the input parameter set and the output parameter set data obtained in step S1 as the training set of the battery multi-dimensional health index estimation model; S3, according to the training set obtained in step S2, adopting a multi-output least square support vector regression algorithm to construct and store a battery multi-dimensional health index joint estimation model based on the take-off stage data of the electric vertical take-off and landing aircraft; S4, recording and storing the time-current sequence and time-voltage sequence of the battery during the take-off stage of the electric vertical take-off and landing aircraft; S5, when the take-off stage ends, normalizing the current and voltage data recorded in step S4 and forming a normalized input parameter set; S6, substituting the normalized input parameter set obtained in step S5 into the battery multi-dimensional health index joint estimation model obtained in step S3 to obtain a model output parameter set; S7, inversely normalizing the model output parameter set obtained in step S6 to obtain the battery multi-dimensional health index; In step S1, the battery capacity, ohmic resistance and the sum of the ohmic resistance and the polarization resistance are taken as the output parameter matrix of the battery health index estimation model, wherein the battery capacity is obtained by adopting the ampere-hour integration method on the time-current sequence data in the constant current discharge stage of the reference performance test; the calculation expression of the battery ohmic resistance is: wherein R o is the battery ohmic internal resistance, V t,0 and V t,1 are the battery terminal voltage values at the initial moment and after 1 second, respectively, of the step current excitation, I L,0 and I L,1 are the current values at the initial moment and after 1 second, respectively, of the step current excitation, and the calculation expression of the sum of the battery ohmic internal resistance and the polarization internal resistance is: wherein R p is the battery polarization resistance, V t,30 and I L,30 are the battery terminal voltage and current values, respectively, after 30 seconds of step current excitation.

2. The battery multi-dimensional state of health joint estimation method for electric vertical take-off and landing aircraft according to claim 1, characterized in that, The battery current and voltage data of the electric vertical take-off and landing aircraft in the take-off stage in the step S1 are screened and arranged into a battery health index estimation model input parameter matrix, and the input parameter variable dimension is 2t phase / T s , wherein t phase is the duration of the take-off stage, and T s is a sampling period.

3. The battery multi-dimensional state of health joint estimation method for electric vertical take-off and landing aircraft according to claim 1, characterized in that, In step S2, the input parameter matrix and the output parameter matrix are combined to form a training set, the number of training set samples is the number of aging cycles of the battery in the data set, and the expression for normalizing the training set parameters is: wherein x i is the training set parameter, x i,norm is the normalized training set parameter, x i,min and x i,max are the minimum and maximum values, respectively, of the same class of data set in the training set.

4. The battery multi-dimensional state of health joint estimation method for electric vertical take-off and landing aircraft according to claim 1, characterized in that, In step S3, the training steps of the battery multi-dimensional health index estimation model based on the multi-output least square support vector regression algorithm are: S3.1 setting a hyperparameter set; S3.2 calculating the weight coefficient matrix W and the bias vector b; S3.3 obtaining the battery multi-dimensional health index estimation model and calculating the training error, and the expression of the battery multi-dimensional health index estimation model is: Wherein, f(·) is the battery multi-dimensional health index estimation model output regression function, X is the input parameter matrix, is a nonlinear mapping function; the training error expression is: wherein RMSE is the root mean square error, y i,ref and y i,est are the reference and estimated values of the battery health indicator, respectively, and / is the number of aging cycles of the battery in the dataset. S3.4 judging whether the training model precision meets the requirements; if it meets the requirements, executing step S3.5; otherwise, executing step S3.1; S3.5 saving the battery multi-dimensional health index estimation model.

5. The battery multi-dimensional state of health joint estimation method for electric vertical take-off and landing aircraft according to claim 1, characterized in that, In step S7, the inverse normalization expression of the model output parameter is: y i,anti-norm = y i (y i,max - y i,min ) + y i,min where y i,anti-norm is the de-normalized model output parameter set, y i is the model output parameter set, y i,min and y i,max are the minimum and maximum values of the model output parameter set corresponding to the same class of data set, respectively.

Images