CN118294815A - Method and device for predicting performance of fuel cell for vehicle and electronic equipment - Google Patents
Method and device for predicting performance of fuel cell for vehicle and electronic equipment Download PDFInfo
- Publication number
- CN118294815A CN118294815A CN202410540812.9A CN202410540812A CN118294815A CN 118294815 A CN118294815 A CN 118294815A CN 202410540812 A CN202410540812 A CN 202410540812A CN 118294815 A CN118294815 A CN 118294815A
- Authority
- CN
- China
- Prior art keywords
- fuel cell
- aging
- vehicle
- performance
- trend
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000000446 fuel Substances 0.000 title claims abstract description 228
- 238000000034 method Methods 0.000 title claims abstract description 93
- 230000032683 aging Effects 0.000 claims abstract description 160
- 230000008569 process Effects 0.000 claims abstract description 36
- 238000005728 strengthening Methods 0.000 claims abstract description 30
- 230000032677 cell aging Effects 0.000 claims abstract description 21
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000015556 catabolic process Effects 0.000 claims abstract description 15
- 238000006731 degradation reaction Methods 0.000 claims abstract description 15
- 230000008859 change Effects 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000000694 effects Effects 0.000 claims description 16
- 230000010287 polarization Effects 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 15
- 239000003054 catalyst Substances 0.000 claims description 14
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 13
- 229910052760 oxygen Inorganic materials 0.000 claims description 13
- 239000001301 oxygen Substances 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 230000000737 periodic effect Effects 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 10
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 8
- 239000007789 gas Substances 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 230000003247 decreasing effect Effects 0.000 claims description 7
- 238000006722 reduction reaction Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- MYMOFIZGZYHOMD-UHFFFAOYSA-N Dioxygen Chemical compound O=O MYMOFIZGZYHOMD-UHFFFAOYSA-N 0.000 claims description 5
- 229910001882 dioxygen Inorganic materials 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 230000002829 reductive effect Effects 0.000 claims description 5
- 230000036961 partial effect Effects 0.000 claims description 4
- 230000036541 health Effects 0.000 abstract description 5
- 210000004027 cell Anatomy 0.000 description 169
- 238000010586 diagram Methods 0.000 description 8
- 230000000670 limiting effect Effects 0.000 description 6
- 230000002787 reinforcement Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 239000012528 membrane Substances 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 229920000642 polymer Polymers 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003090 exacerbative effect Effects 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 239000000376 reactant Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Fuel Cell (AREA)
Abstract
The invention relates to the technical field of fuel cells, and provides a method and a device for predicting performance of a vehicle fuel cell and electronic equipment, wherein the method comprises the following steps: extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition; extracting strengthening features for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence; inputting the aging performance index time sequence and the strengthening characteristic for representing the regularity and/or trend of the aging process of the fuel cell as training data into a preset multi-input convolutional neural network to train a vehicle fuel cell performance prediction model; and predicting the performance degradation condition of the fuel cell based on the trained vehicle fuel cell performance prediction model. The invention can improve the accuracy and reliability of performance prediction under the condition of limited aging data of the fuel cell, and has important significance for health management of the fuel cell for vehicles.
Description
Technical Field
The present invention relates to the field of fuel cell technologies, and in particular, to a method and an apparatus for predicting performance of a fuel cell for a vehicle, and an electronic device.
Background
The fuel cell is used as a main power source of the new energy automobile, has the advantages of high energy density, quick fuel supply and the like, can complement the short plates of the pure electric automobile in the field of long-distance heavy-duty commercial vehicles, and has great development prospect. However, the service life of the vehicle fuel cell is greatly influenced by actual operation conditions, and the operation conditions of the fuel cell are more severe due to complex vehicle-mounted working conditions, so that the performance degradation of the fuel cell is accelerated. The maintenance and replacement cost of the vehicle fuel cell is high, so that the accurate and reliable fuel cell aging prediction is realized, the cost of the fuel cell is reduced, the running efficiency is improved, the basis of constructing a fuel cell automobile fault prediction and health management system is also the key point of improving the durability and reliability of the vehicle fuel cell.
In the operation process of the fuel cell for the vehicle, because the aging phenomenon of the fuel cell involves complex physical and chemical mechanisms, such as corrosion and dissolution of the catalyst, damage to a gas diffusion layer structure, decomposition of a proton exchange membrane polymer and the like, the aging state of the fuel cell at the current moment is difficult to accurately judge, the aging mechanisms of all parts have great difference and the aging rates of all parts are different, for example, the voltage fluctuation under a dynamic circulation working condition can cause corrosion of a catalyst carbon carrier, so that the catalyst is lost; the high potential at idle conditions can cause polymer decomposition, exacerbating proton membrane aging. The equivalent extrapolation method is adopted to predict the aging of the fuel cell by using a single aging index, which is not suitable for the aging prediction under the actual dynamic working condition.
In carrying out the invention, the applicant has found that the prior art has at least the following technical drawbacks:
1) The existing non-invasive aging characteristic expression method of the vehicle fuel cell is inaccurate and has poor generalization capability. For example, at present, fuel cell parameters such as voltage and power which are easy to measure are mostly used for describing the aging characteristics of the fuel cell, but the aging of the fuel cell for the vehicle is greatly influenced by the running condition, and the fuel cell has the characteristics of high nonlinearity, particularly the variable load working condition causes the output voltage of the fuel cell to change along with the load and to present response hysteresis characteristics, and the existing aging estimation method using the voltage and the power as aging indexes has obvious limitations. In addition, the fuel cell aging index constructed based on the strong assumption of the linear variation trend of the limiting current and the ohmic internal resistance lacks the explanation of the ohmic internal resistance and the limiting current density attenuation mechanism, and has limited generalization capability.
2) The existing vehicle fuel cell aging characteristic estimation method has large data demand and limited aging early estimation capability. For example, existing fuel cell aging characteristic estimation methods typically employ data-based multi-step advanced predictions to enable online tracking of fuel cell aging conditions. Data-based methods rely on extensive data training to achieve feature understanding and trend tracking of time series. However, in the early stage of fuel cell aging, the amount of aging data which can be used for training a prediction algorithm is limited, aging characteristics are not obvious, so that the prediction algorithm is not trained sufficiently, and meanwhile, aging characteristic information included in the data can interfere with the capability of the algorithm to track the aging state of the fuel cell, so that the overfitting phenomenon is caused.
Disclosure of Invention
The present invention has been made in view of the above problems, and has as its object to provide a method, an apparatus and an electronic device for predicting the performance of a fuel cell for a vehicle, which overcome the above problems.
In one aspect of the present invention, there is provided a fuel cell performance prediction method for a vehicle, the method comprising:
Extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition;
Extracting strengthening features for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence;
inputting the aging performance index time sequence and the strengthening characteristic for representing the regularity and/or trend of the aging process of the fuel cell as training data into a preset multi-input convolutional neural network to train a vehicle fuel cell performance prediction model;
And predicting the performance degradation condition of the fuel cell based on the trained vehicle fuel cell performance prediction model.
Further, extracting an aging performance index time sequence of the vehicle fuel cell under the dynamic working condition comprises the following steps:
Constructing an exchange current density attenuation analytical model according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging empirical model;
acquiring the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell;
And carrying out linearization assumption on the ohmic internal resistance according to the change trend, and establishing a comprehensive aging index model of the vehicle fuel cell by combining the exchange current density attenuation analysis type, wherein the comprehensive aging index model is marked as alpha (t).
Rint＝R0(1+kα(t))
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density,For an intrinsic switching current density,For the reference concentration of oxygen gas,The oxygen concentration is delta b act, the oxygen reduction reaction activation energy is T ref, the reference temperature is gamma 0, the reaction stage number is T, the aging test time is T cyc, the single cycle working condition duration is T, K is a proportionality constant, R is a molar gas constant, T is the thermodynamic temperature, K is a normalization factor, and S min is the normalized minimum electrochemical activity specific surface area.
Further, the method for obtaining the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell comprises the following steps:
Characterizing the output characteristics of the fuel cell by adopting a polarization model of the fuel cell;
And selecting ohmic internal resistance R int and exchange current density i 0 in the polarization model as aging time-varying physical parameters of the fuel cell to be extracted, carrying out a durability experiment on the fuel cell, measuring the voltage of a single cell of the fuel cell, and fitting the polarization model of the fuel cell to obtain the variation trend of each aging physical parameter in the dynamic loading process of the fuel cell.
Further, after establishing the comprehensive aging index of the fuel cell for the vehicle, the method further comprises:
And carrying out fitting solution on the comprehensive aging index of the vehicle fuel cell by adopting a Levenberg-Marquardt optimization algorithm, and taking the obtained aging index fitting result as an aging performance index time sequence of the vehicle fuel cell.
Further, the performing a fitting solution on the comprehensive aging index of the vehicle fuel cell by using a Levenberg-Marquardt optimization algorithm includes:
Defining an initial parameter set p 0 as an initial guess value of a model function f (x, p), wherein f (x, p) is a fitting function of an aging index alpha (t), and setting a search step lambda;
Calculating a residual vector r (p), wherein each element r i (p) is the difference between model prediction and actual data: r i(p)＝yi-f(xi, p), and computing a jacobian matrix J, whose element J ij is the partial derivative of the residual r i with respect to the parameter p j;
Updating the parameter p new by using a preset updating formula, calculating the error square sum S (p new) under the new parameter p new, accepting the new parameter if the S (p new) is smaller than the history S (p), reducing lambda, rejecting the new parameter if the S (p new) is larger than or equal to the S (p), and increasing lambda;
The update formula is as follows:
pnew＝p+(JTJ+λD)-1JTr(p)
Wherein D is a diagonal matrix whose diagonal elements are diagonal elements of J T J;
repeating the steps until the convergence condition is met or the preset iteration times are reached, and outputting the obtained final parameter p as a fitting result.
Further, extracting strengthening features for characterizing regularity and/or trending of the fuel cell aging process based on the aging performance index time series, comprising:
Carrying out characteristic decomposition on the aging performance index time sequence by adopting a three-parameter exponential smoothing method Holt-windows to obtain a grade characteristic component, an increasing and decreasing trend component and a periodic characteristic component corresponding to the aging performance index time sequence;
The grade characteristic component is selected as a first strengthening characteristic for representing the trend of the aging process of the fuel cell, and the periodic characteristic component is selected as a second strengthening characteristic for representing the regularity of the aging process of the fuel cell.
Further, the network structure of the multi-input convolutional neural network comprises a filling layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer which are sequentially arranged;
the filling layer is used for filling and arranging input data of the multi-input convolutional neural network, the convolutional layer is used for identifying characteristic information of the input data after filling and arranging, the pooling layer is used for compressing and dimension-reducing the identified characteristic information, the full-connection layer is used for explaining the dimension-reduced characteristic information, and the output layer is used for realizing time sequence prediction output.
In another aspect of the present invention, there is provided a fuel cell performance prediction apparatus for a vehicle, the apparatus comprising:
The performance index extraction module is used for extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition;
The characteristic strengthening module is used for extracting strengthening characteristics for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence;
The model training module is used for inputting the aging performance index time sequence and the strengthening characteristics for representing the regularity and/or trend of the aging process of the fuel cell into a preset multi-input convolutional neural network as training data to train a vehicle fuel cell performance prediction model;
and the prediction module is used for predicting the performance degradation condition of the fuel cell based on the trained vehicle fuel cell performance prediction model.
Further, the performance index extraction module includes:
The first construction unit is used for constructing an exchange current density attenuation analysis type according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging experience model;
the acquisition unit is used for acquiring the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell;
the second construction unit is used for carrying out linearization assumption on the ohmic internal resistance according to the change trend, and constructing a comprehensive aging index model of the vehicle fuel cell by combining the exchange current density attenuation analysis type, and is marked as alpha (t):
Rint＝R0(1+kα(t))
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density, For an intrinsic switching current density,For the reference concentration of oxygen gas,The oxygen concentration is delta b act, the oxygen reduction reaction activation energy is T ref, the reference temperature is gamma 0, the reaction stage number is T, the aging test time is T cyc, the single cycle working condition duration is T, K is a proportionality constant, R is a molar gas constant, T is the thermodynamic temperature, K is a normalization factor, and S min is the normalized minimum electrochemical activity specific surface area.
In another aspect of the invention, an electronic device is provided that includes a memory, a processor, and a computer program stored on the memory and executable on the processor; the computer program, when executed by the processor, implements the steps of the fuel cell performance prediction method for a vehicle as described above.
According to the method, the device and the electronic equipment for predicting the performance of the vehicle fuel cell, provided by the embodiment of the invention, aiming at the conditions that the data of the vehicle fuel cell in the early stage of performance degradation is incomplete and the performance degradation trend is not obvious, the aging characteristic strengthening method is provided for extracting the regularity phenomenon in the aging process of the fuel cell, and is used as the input of a training fuel cell performance predicting algorithm to strengthen the learning capability of the predicting algorithm, so that the accuracy and the reliability of the fuel cell performance prediction can be improved under the condition that the fuel cell aging data are limited. Meanwhile, the calculation efficiency is improved, the resource waste is avoided, and the method has important significance for building the vehicle fuel cell health management system.
The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. In the drawings:
FIG. 1 is a flowchart of a method for predicting performance of a fuel cell for a vehicle according to an embodiment of the present invention;
Fig. 2 (a) is a schematic diagram of a variation trend of ohmic internal resistance in a dynamic loading process of a fuel cell according to an embodiment of the present invention;
fig. 2 (b) is a schematic diagram showing a change trend of the exchange current density in the dynamic loading process of the fuel cell according to the embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a fitting effect of the exchange current density decay analysis according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a fitting result of the comprehensive aging index according to the embodiment of the present invention;
FIG. 5 is an exploded view of an aging performance index of a fuel cell for a vehicle according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a fuel cell performance prediction model based on reinforcement features according to an embodiment of the present invention;
FIG. 7 is a flowchart of a specific implementation of a prediction of performance of a vehicle fuel cell based on enhanced feature extraction according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a fuel cell performance prediction system for a vehicle according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
It will be understood by those skilled in the art that all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs unless defined otherwise. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The embodiment of the invention provides a method for predicting the performance of a vehicle fuel cell, which can accurately monitor and predict the working state of the fuel cell and improve the accuracy of performance prediction under the condition of incomplete data in the early stage of performance degradation of the fuel cell by extracting and correcting the aging characteristics of the vehicle fuel cell and adopting a characteristic reinforcement technology to carry out reinforcement characteristic extraction. As shown in fig. 1, the method for predicting the performance of the fuel cell for the vehicle provided by the invention comprises the following steps:
s1, extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition.
The aging performance index time sequence is obtained by fitting an aging characteristic curve of the vehicle fuel cell under a dynamic working condition, and can embody multi-component aging parameters.
S2, extracting strengthening characteristics for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence.
S3, inputting the aging performance index time sequence and the strengthening characteristic for representing the regularity and/or trend of the aging process of the fuel cell into a preset multi-input convolutional neural network as training data to train a vehicle fuel cell performance prediction model.
S4, predicting the performance degradation condition of the fuel cell based on a trained vehicle fuel cell performance prediction model.
According to the method for predicting the performance of the vehicle fuel cell, provided by the embodiment of the invention, aiming at the conditions that the vehicle fuel cell is incomplete in early data of performance degradation and the performance degradation trend is not obvious, the aging characteristic strengthening method is provided for extracting the regularity phenomenon in the aging process of the fuel cell, and is used as the input of a training fuel cell performance prediction algorithm to strengthen the learning capacity of the prediction algorithm, so that the accuracy and the reliability of the performance prediction of the fuel cell can be improved under the condition that the aging data of the fuel cell is limited. Meanwhile, the calculation efficiency is improved, the resource waste is avoided, and the method has important significance for building the vehicle fuel cell health management system.
In the embodiment of the present invention, the extracting the aging performance index time sequence of the vehicle fuel cell under the dynamic working condition in the step S1 specifically includes the steps not shown in the following drawings:
s11, constructing an exchange current density attenuation analysis type according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging experience model.
S12, acquiring the change trend of the ohmic internal resistance in the dynamic loading process of the fuel cell. Further, the method for obtaining the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell comprises the following steps: characterizing the output characteristics of the fuel cell by adopting a polarization model of the fuel cell; and selecting ohmic internal resistance and exchange current density in the polarization model as aging time-varying physical parameters of the fuel cell to be extracted, carrying out a durability experiment on the fuel cell, measuring the voltage of a single cell of the fuel cell, and fitting the polarization model of the fuel cell to obtain the variation trend of each aging physical parameter in the dynamic loading process of the fuel cell.
And S13, carrying out linearization assumption on the ohmic internal resistance according to the change trend, and establishing a comprehensive aging index model of the vehicle fuel cell by combining the exchange current density attenuation analysis type, wherein the comprehensive aging index model is marked as alpha (t).
In this embodiment, the half-mechanism polarization model of a single fuel cell is as follows:
Vcell＝ENerst-Eohm-Eact-Econc (1)
Wherein V cell is the fuel cell monolithic cell voltage. E Nerst is the reversible Nerst potential, derived from the Nerst equation; ohmic overpotential E ohm represents the ohmic loss during operation of the stack, and is related to proton exchange membrane thickness, conductivity, etc.; the activation overpotential E act represents the activation energy for overcoming the electrochemical reaction in the low potential area, is influenced by the activity of the catalyst, is obtained by a Butler-Volmar equation, is related to the exchange current density, and is influenced by the temperature of the fuel cell, the activity surface area of the catalyst and the like; the concentration difference overpotential E conc is caused by the rapid decrease of the concentration of the reactant on the electrode surface in the high potential region, reflects the mass transfer capacity of the fuel cell, and can be obtained according to Fick's law and is related to the limiting current density. The output characteristics of the fuel cell are represented by adopting a polarization model of the fuel cell, and the specific expression is as follows:
Wherein i is the total current density (A/cm 2),Rint is ohmic internal resistance (Ω cm 2), R is molar gas constant (8.3145J/mol/K), T is thermodynamic temperature (K), α is charge transfer coefficient, F is Faraday constant (96485C/mol), i 0 is exchange current density (A/cm 2), B is concentration difference coefficient (V), and i L is limiting current density (A/cm 2).
According to the invention, R int、i0 in a polarization model is selected as an aging time-varying physical parameter of the fuel cell to be extracted, the voltage of the fuel cell is measured through a durability experiment, the polarization model is fitted, and the change trend of the aging parameter of the parameter in the dynamic loading process of the fuel cell is obtained, as shown in fig. 2, (a) is the change trend of the ohmic internal resistance in the dynamic loading process of the fuel cell, and (b) is the change trend of the exchange current density in the dynamic loading process of the fuel cell. The liquid crystal display device shows a certain stable rising trend, and the total rising trend is about 23%. The exchange current density shows a typical nonlinear change trend, the overall change amplitude reaches 87%, the aging test shows a rapid trend in the first 200 hours, the subsequent trend of the decrease is smooth, and the overall exponential function change trend is similar.
According to the exchange current density calculation model, see formula 3, and the catalyst layer electrochemical surface area aging empirical model, see formula 4:
K＝KSC×KT×KRH×KV (5)
obtaining an exchange current density attenuation analysis type for representing the performance aging of the fuel cell:
And (3) carrying out linearization assumption on the ohmic internal resistance aging trend by combining the ohmic internal resistance aging trend analysis result, and establishing a fuel cell comprehensive aging index model which is marked as alpha (t) by combining with exchange current density attenuation analysis.
Rint＝R0(1+kα(t)) (7)
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density,For an intrinsic exchange current density (a/cm 2),For the reference concentration of oxygen gas,In order to achieve the oxygen concentration,The catalyst Pt effective area (cm 2),Δbact is oxygen reduction reaction activation energy (J/mol), T ref is reference temperature (K), gamma 0 is reaction series, N is dynamic circulation working condition loading times, T is aging test time, T cyc is single circulation working condition duration (h),ECSA is electrochemical activity specific surface area (A/cm 2), ECSA is electrochemical activity specific surface area (cm 2/cm2),SN is normalized electrochemical activity specific surface area, K is proportionality constant, K SC is proportionality constant under standard working condition, K T is temperature correction coefficient, K RH is relative humidity correction coefficient, K V is circulation working condition correction coefficient, R is molar gas constant, T is thermodynamic temperature, K is normalization factor, and S min is normalized minimum electrochemical activity specific surface area.
The fitting effect of the exchange current density decay analytical model is shown in figure 3.
In an embodiment of the present invention, after the comprehensive aging index of the fuel cell for the vehicle is established, the method further includes: and carrying out fitting solution on the comprehensive aging index of the vehicle fuel cell by adopting a Levenberg-Marquardt optimization algorithm, and taking the obtained aging index fitting result as an aging performance index time sequence of the vehicle fuel cell.
In the actual operation process of the vehicle fuel cell, the vehicle fuel cell can be aged gradually along with the increase of the service time, and the output performance and the service life of the battery are affected. In extracting the performance index of the fuel cell for the vehicle, it is assumed that the aging state of the fuel cell is constant for a period of time, that is, the effect of aging on the performance is uniform and stable for this period of time. On the basis, a numerical solution method of a nonlinear least square problem is adopted, namely a Levenberg-Marquardt optimization Levenberg-Marquardt algorithm is adopted, and the fuel cell comprehensive aging index model is subjected to fitting solution.
Further, the specific steps of performing fitting solution on the comprehensive aging index of the vehicle fuel cell by adopting the Levenberg-Marquardt optimization algorithm are as follows:
step 1, initializing parameters
Defining an initial parameter set p 0 as an initial guess value of a model function f (x, p), wherein f (x, p) is a fitting function of an aging index alpha (t), and setting a search step lambda;
step 2, calculating error and jacobian matrix
Calculating a residual vector r (p), wherein each element r i (p) is the difference between model prediction and actual data: r i(p)＝yi-f(xi, p), and computing a jacobian matrix J, whose element J ij is the partial derivative of the residual r i with respect to the parameter p j;
step 3, updating parameters
Updating the parameter p new by using a preset updating formula, wherein the updating formula is as follows:
pnew＝p+(JTJ+λD)-1JTr(p)
Wherein D is a diagonal matrix whose diagonal elements are diagonal elements of J T J;
Step 4, evaluating new parameters
Calculating the sum of squares of errors S (p new) under the new parameter p new, if S (p new) is smaller than the history S (p), determining that the new parameter is better, accepting the new parameter, reducing lambda, and if S (p new) is greater than or equal to S (p), rejecting the new parameter and increasing lambda;
step 5, iterating
Repeating the steps until the convergence condition is met or the preset iteration times are reached, and outputting the obtained final parameter p as a fitting result.
The fitting method is adopted to fit the comprehensive aging index of the durability data set of the vehicle fuel cell, which is obtained in advance, and the fitting result is shown in fig. 4.
In the embodiment of the present invention, the step S2 of extracting the strengthening features for characterizing the regularity and/or the trend of the aging process of the fuel cell based on the aging performance index time sequence specifically includes the following steps:
S21, carrying out characteristic decomposition on the aging performance index time sequence by adopting a three-parameter exponential smoothing method Holt-winter to obtain a grade characteristic component, an increasing and decreasing trend component and a periodic characteristic component corresponding to the aging performance index time sequence.
In order to reduce nonlinear factor interference in the aging process of the vehicle fuel cell, the extracted aging performance index time sequence of the fuel cell is decomposed by adopting a Holt-winter algorithm, so that the performance index enhancement of the fuel cell is realized, and a decomposition model of the Holt-winter method is as follows:
bt＝β(lt-lt-1)+(1-β)bt-1 (11)
The performance index after decomposition is expressed as:
formulas (10) - (12) are gradation components, increasing/decreasing trend components, and periodic components update formulas, and formula 13 is an update formula for recursively time-series data.
Wherein l t is a rank component, b t is an increasing and decreasing trend component, s t is a periodic component, alpha, beta, gamma is a smoothing coefficient, h is a time period, t is an aging test time, y t is a time series predicted value at time t+h, and m is a period length in a time series.
And calculating the minimum value of the Matlab fminsearch multiple functions to obtain the optimizing function for calibrating the alpha, beta and gamma smoothing coefficients. The degradation of the aging performance index results in a grade trend, an increasing and decreasing trend, and a cycle characteristic of the index, which are shown in fig. 5. The invention solves the problems of insignificant aging characteristics and estimation of aging characteristics under the condition of insufficient aging data of the fuel cell for vehicles based on the characteristic strengthening technology, lays a foundation for establishing a fuel cell vehicle fault prediction and health management system, and is a key point for improving the durability and reliability of the fuel cell for vehicles.
S22, selecting the grade characteristic component as a first strengthening characteristic for representing the trend of the fuel cell aging process, and selecting the periodic characteristic component as a second strengthening characteristic for representing the regularity of the fuel cell aging process.
Further, the network structure of the multi-input convolutional neural network in the step S3 comprises a filling layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer which are sequentially arranged; the filling layer is used for filling and arranging input data of the multi-input convolutional neural network, the convolutional layer is used for identifying characteristic information of the input data after filling and arranging, the pooling layer is used for compressing and dimension-reducing the identified characteristic information, the full-connection layer is used for explaining the dimension-reduced characteristic information, and the output layer is used for realizing time sequence prediction output.
The invention adopts the multi-input convolutional neural network to train the processed aging data and predicts and tracks the performance decline condition of the fuel cell. The input of the multi-input convolutional neural network module is a multi-element time sequence with a certain time step, and the multi-element time sequence comprises decomposed periodic characteristics, grade characteristics and aging index time sequences. The input data is subjected to filling arrangement, multi-volume lamination characteristic identification, pooling layer characteristic information compression dimension reduction and full-connection layer characteristic information interpretation, and finally the time sequence prediction output is realized, and the prediction process is shown. The convolution layer calculation rule is:
Wherein f is an activation function, w s is a weight value of a convolution filter, b s is a bias parameter, x is a convolution layer input, s is a convolution layer number, q and u are convolution kernel matrix sizes, and a ReLU function is adopted as the activation function between neuron nodes. Fig. 6 shows a schematic diagram of a fuel cell performance prediction model based on the reinforcement characteristics, from which the evaluation results of the fuel cell performance prediction are shown in table 1:
TABLE 1 rating results for multiple input performance prediction models
In Table 1, MAPE is the mean absolute percentage error and RMSE is the root mean square error.
The method for predicting the performance of the vehicle fuel cell solves the problem of how to accurately judge the current aging state of the fuel cell in the vehicle fuel cell power system and predicts the aging trend of the fuel cell, thereby being responsible for making corresponding control and maintenance strategies subsequently and prolonging the service life of the vehicle fuel cell. The specific implementation process is as shown in fig. 7:
1. And solving the aging parameters of the vehicle fuel cell, combining the ohm internal resistance change rule and an exchange current density aging empirical formula suitable for the dynamic working condition, and adopting the input signal segmentation fitting to obtain the comprehensive aging index of the vehicle fuel cell under the dynamic working condition.
2. And the short-term prediction accuracy of the multivariable convolutional neural network on the aging trend of the fuel cell is improved by strengthening the early aging data trend and the reversible aging characteristic of the fuel cell.
The invention provides a non-invasive vehicle fuel cell aging characteristic representation method, which is characterized in that an aging characteristic curve of a vehicle fuel cell under a dynamic working condition is fitted, an electrochemical active surface area attenuation empirical formula of a catalyst layer is introduced into an exchange current density analysis formula, and the method is different from an original single aging index.
Further, the invention provides a vehicle fuel cell performance prediction method based on reinforcement feature learning, which fully considers nonlinear features of aging indexes, decomposes the nonlinear features to obtain a regular trend, and is used for training a prediction algorithm.
For the purposes of simplicity of explanation, the methodologies are shown and described as a series of acts, it is to be understood and appreciated by one of ordinary skill in the art that the methodologies are not limited by the order of acts, as some acts may, in accordance with the methodologies, take place in other order or concurrently. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.
Another embodiment of the present invention also provides an apparatus for predicting performance of a fuel cell for a vehicle, the apparatus including a functional module for implementing the method for predicting performance of a fuel cell for a vehicle according to any one of the above. Fig. 8 schematically illustrates a structural schematic diagram of a fuel cell performance prediction apparatus for a vehicle according to an embodiment of the present invention, and referring to fig. 8, the fuel cell performance prediction apparatus for a vehicle according to an embodiment of the present invention specifically includes a performance index extraction module 201, a feature enhancement module 202, a model training module 203, and a prediction module 204, where:
The performance index extraction module 201 is configured to extract an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition;
a feature enhancement module 202 for extracting enhancement features for characterizing regularity and/or trend of the fuel cell aging process based on the aging performance index time series;
the model training module 203 is configured to input the aging performance index time sequence and the strengthening feature for characterizing regularity and/or trending of the aging process of the fuel cell as training data to a preset multi-input convolutional neural network to train the vehicle fuel cell performance prediction model;
The prediction module 204 is configured to predict a fuel cell performance degradation condition based on a trained vehicle fuel cell performance prediction model.
In the embodiment of the present invention, the performance index extraction module 201 includes a first construction unit, an acquisition unit, and a second construction unit, where:
The first construction unit is used for constructing an exchange current density attenuation analysis type according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging experience model;
the acquisition unit is used for acquiring the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell;
the second construction unit is used for carrying out linearization assumption on the ohmic internal resistance according to the change trend, and constructing the comprehensive aging index of the vehicle fuel cell by combining the exchange current density attenuation analysis type, and is marked as alpha (t):
Rint＝R0(1+kα(t))
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density, For an intrinsic switching current density,For reference oxygen concentration, c O2 is oxygen concentration, Δb act is oxygen reduction reaction activation energy, T ref is reference temperature, γ 0 is reaction progression, T is aging test time, T cyc is single cycle operating condition duration, K is proportionality constant, R is molar gas constant, T is thermodynamic temperature, K is normalization factor, and S min is normalized minimum electrochemical activity specific surface area.
In this embodiment, the obtaining unit is specifically configured to characterize an output characteristic of the fuel cell by using a polarization model of the fuel cell; and selecting ohmic internal resistance and exchange current density in the polarization model as aging time-varying physical parameters of the fuel cell to be extracted, carrying out a durability experiment on the fuel cell, measuring the voltage of a single cell of the fuel cell, and fitting the polarization model of the fuel cell to obtain the variation trend of each aging physical parameter in the dynamic loading process of the fuel cell.
In this embodiment, the performance index extraction module 201 further includes a calculation unit, where the calculation unit is configured to perform a fitting solution on the comprehensive aging index of the vehicle fuel cell by using a levenberg-marquardt optimization algorithm after the comprehensive aging index of the vehicle fuel cell is established, and use the obtained fitting result of the aging index as the aging performance index time sequence of the vehicle fuel cell.
Further, the computing unit is specifically configured to execute the following fitting solution method:
Defining an initial parameter set p 0 as an initial guess value of a model function f (x, p), wherein f (x, p) is a fitting function of an aging index alpha (t), and setting a search step lambda;
Calculating a residual vector r (p), wherein each element r i (p) is the difference between model prediction and actual data: r i(p)＝yi-f(xi, p), and computing a jacobian matrix J, whose element J ij is the partial derivative of the residual r i with respect to the parameter p j;
Updating the parameter p new by using a preset updating formula, calculating the error square sum S (p new) under the new parameter p new, accepting the new parameter if the S (p new) is smaller than the history S (p), reducing lambda, rejecting the new parameter if the S (p new) is larger than or equal to the S (p), and increasing lambda;
The update formula is as follows:
pnew＝p+(JTJ+λD)-1JTr(p)
Wherein D is a diagonal matrix whose diagonal elements are diagonal elements of J T J;
repeating the steps until the convergence condition is met or the preset iteration times are reached, and outputting the obtained final parameter p as a fitting result.
In the embodiment of the present invention, the feature enhancement module 202 specifically includes a feature decomposition unit and a feature selection unit, where:
the characteristic decomposition unit is used for carrying out characteristic decomposition on the aging performance index time sequence by adopting a three-parameter exponential smoothing method Holt-windows to obtain a grade characteristic component, an increasing and decreasing trend component and a periodic characteristic component corresponding to the aging performance index time sequence;
And the characteristic selecting unit is used for selecting the grade characteristic component as a first strengthening characteristic for representing the trend of the fuel cell aging process and selecting the periodic characteristic component as a second strengthening characteristic for representing the regularity of the fuel cell aging process.
In the embodiment of the present invention, the network structure of the multi-input convolutional neural network trained by the model training module 203 includes a filling layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer which are sequentially arranged;
the filling layer is used for filling and arranging input data of the multi-input convolutional neural network, the convolutional layer is used for identifying characteristic information of the input data after filling and arranging, the pooling layer is used for compressing and dimension-reducing the identified characteristic information, the full-connection layer is used for explaining the dimension-reduced characteristic information, and the output layer is used for realizing time sequence prediction output.
In a specific implementation process, the embodiment of the device can refer to the embodiment of the method, and has corresponding technical effects.
In another aspect of the invention, an electronic device is provided that includes a memory, a processor, and a computer program stored on the memory and executable on the processor; the computer program, when executed by the processor, implements the steps of the fuel cell performance prediction method for a vehicle as described above. Such as steps S1-S4 shown in fig. 1. Or the processor, when executing the computer program, implements the functions of the modules in the embodiments of the fuel cell performance prediction apparatus for vehicles described above, such as the performance index extraction module 201, the feature enhancement module 202, the model training module 203, and the prediction module 204 shown in fig. 8.
In the specific implementation process, the embodiment of the equipment can refer to the embodiment of the method, and has the corresponding technical effects
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, any of the claimed embodiments can be used in any combination.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by others; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1.A method for predicting performance of a fuel cell for a vehicle, the method comprising:
Extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition;
Extracting strengthening features for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence;
inputting the aging performance index time sequence and the strengthening characteristic for representing the regularity and/or trend of the aging process of the fuel cell as training data into a preset multi-input convolutional neural network to train a vehicle fuel cell performance prediction model;
And predicting the performance degradation condition of the fuel cell based on the trained vehicle fuel cell performance prediction model.
2. The method of claim 1, wherein extracting the time series of aging performance indicators for the vehicular fuel cell under dynamic conditions comprises:
Constructing an exchange current density attenuation analytical model according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging empirical model;
acquiring the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell;
And carrying out linearization assumption on ohmic internal resistance according to the change trend, and establishing a comprehensive aging index model of the vehicle fuel cell by combining the exchange current density attenuation analysis type, wherein the comprehensive aging index model is marked as alpha (t):
Rint＝R0(1+kα(t))
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density, For an intrinsic switching current density,For the reference concentration of oxygen gas,The oxygen concentration is delta b act, the oxygen reduction reaction activation energy is T ref, the reference temperature is gamma 0, the reaction stage number is T, the aging test time is T cyc, the single cycle working condition duration is T, K is a proportionality constant, R is a molar gas constant, T is the thermodynamic temperature, K is a normalization factor, and S min is the normalized minimum electrochemical activity specific surface area.
3. The method of claim 2, wherein obtaining a trend of ohmic internal resistance change during dynamic loading of the fuel cell comprises:
Characterizing the output characteristics of the fuel cell by adopting a polarization model of the fuel cell;
And selecting ohmic internal resistance R int and exchange current density i 0 in the polarization model as aging time-varying physical parameters of the fuel cell to be extracted, carrying out a durability experiment on the fuel cell, measuring the voltage of a single cell of the fuel cell, and fitting the polarization model of the fuel cell to obtain the variation trend of each aging physical parameter in the dynamic loading process of the fuel cell.
4. The method of claim 2, wherein after establishing the integrated degradation indicator for the fuel cell for the vehicle, the method further comprises:
And carrying out fitting solution on the comprehensive aging index of the vehicle fuel cell by adopting a Levenberg-Marquardt optimization algorithm, and taking the obtained aging index fitting result as an aging performance index time sequence of the vehicle fuel cell.
5. The method of claim 4, wherein the performing a fitting solution on the integrated aging index of the fuel cell for the vehicle using the levenberg-marquardt optimization algorithm comprises:
Defining an initial parameter set p 0 as an initial guess value of a model function f (x, p), wherein f (x, p) is a fitting function of an aging index alpha (t), and setting a search step lambda;
Calculating a residual vector r (p), wherein each element r i (p) is the difference between model prediction and actual data: r i(p)＝yi-f(xi, p), and computing a jacobian matrix J, whose element J ij is the partial derivative of the residual r i with respect to the parameter p j;
Updating the parameter p new by using a preset updating formula, calculating the error square sum S (p new) under the new parameter p new, accepting the new parameter if the S (p new) is smaller than the history S (p), reducing lambda, rejecting the new parameter if the S (p new) is larger than or equal to the S (p), and increasing lambda;
The update formula is as follows:
pnew＝p+(JTJ+λD)-1JTr(p)
Wherein D is a diagonal matrix whose diagonal elements are diagonal elements of J T J;
repeating the steps until the convergence condition is met or the preset iteration times are reached, and outputting the obtained final parameter p as a fitting result.
6. The method according to claim 1, wherein extracting strengthening features for characterizing regularity and/or trending of a fuel cell aging process based on the aging performance index time series comprises:
Carrying out characteristic decomposition on the aging performance index time sequence by adopting a three-parameter exponential smoothing method Holt-windows to obtain a grade characteristic component, an increasing and decreasing trend component and a periodic characteristic component corresponding to the aging performance index time sequence;
The grade characteristic component is selected as a first strengthening characteristic for representing the trend of the aging process of the fuel cell, and the periodic characteristic component is selected as a second strengthening characteristic for representing the regularity of the aging process of the fuel cell.
7. The method of claim 1, wherein the network structure of the multi-input convolutional neural network comprises a filling layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer which are sequentially arranged;
the filling layer is used for filling and arranging input data of the multi-input convolutional neural network, the convolutional layer is used for identifying characteristic information of the input data after filling and arranging, the pooling layer is used for compressing and dimension-reducing the identified characteristic information, the full-connection layer is used for explaining the dimension-reduced characteristic information, and the output layer is used for realizing time sequence prediction output.
8. A fuel cell performance prediction apparatus for a vehicle, characterized by comprising:
The performance index extraction module is used for extracting an aging performance index time sequence of the vehicle fuel cell under a dynamic working condition;
The characteristic strengthening module is used for extracting strengthening characteristics for representing regularity and/or trend of the fuel cell aging process based on the aging performance index time sequence;
The model training module is used for inputting the aging performance index time sequence and the strengthening characteristics for representing the regularity and/or trend of the aging process of the fuel cell into a preset multi-input convolutional neural network as training data to train a vehicle fuel cell performance prediction model;
and the prediction module is used for predicting the performance degradation condition of the fuel cell based on the trained vehicle fuel cell performance prediction model.
9. The apparatus of claim 8, wherein the performance index extraction module comprises:
The first construction unit is used for constructing an exchange current density attenuation analysis type according to a preset exchange current density calculation model and a catalyst layer electrochemical surface area aging experience model;
the acquisition unit is used for acquiring the variation trend of the ohmic internal resistance in the dynamic loading process of the fuel cell;
and the second construction unit is used for carrying out linearization assumption on the ohmic internal resistance according to the change trend and constructing the comprehensive aging index of the vehicle fuel cell by combining the exchange current density attenuation analysis type, and is marked as alpha (t).
Rint＝R0(1+kα(t))
Wherein: r int is ohm internal resistance, R 0 is ohm internal resistance initial value, i 0 is exchange current density,For an intrinsic switching current density,For the reference concentration of oxygen gas,The oxygen concentration is delta b act, the oxygen reduction reaction activation energy is T ref, the reference temperature is gamma 0, the reaction stage number is T, the aging test time is T cyc, the single cycle working condition duration is T, K is a proportionality constant, R is a molar gas constant, T is the thermodynamic temperature, K is a normalization factor, and S min is the normalized minimum electrochemical activity specific surface area.
10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor; the computer program, when executed by the processor, implements the steps of the fuel cell performance prediction method for a vehicle as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410540812.9A CN118294815A (en) | 2024-04-30 | 2024-04-30 | Method and device for predicting performance of fuel cell for vehicle and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410540812.9A CN118294815A (en) | 2024-04-30 | 2024-04-30 | Method and device for predicting performance of fuel cell for vehicle and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118294815A true CN118294815A (en) | 2024-07-05 |
Family
ID=91677819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410540812.9A Pending CN118294815A (en) | 2024-04-30 | 2024-04-30 | Method and device for predicting performance of fuel cell for vehicle and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118294815A (en) |
-
2024
- 2024-04-30 CN CN202410540812.9A patent/CN118294815A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lyu et al. | Li-ion battery state of health estimation and remaining useful life prediction through a model-data-fusion method | |
| Tan et al. | Online state-of-health estimation of lithium-ion battery based on dynamic parameter identification at multi timescale and support vector regression | |
| Pan et al. | Performance degradation prediction of proton exchange membrane fuel cell using a hybrid prognostic approach | |
| Shi et al. | Battery health management using physics-informed machine learning: Online degradation modeling and remaining useful life prediction | |
| Tian et al. | Lithium-ion battery health estimation with real-world data for electric vehicles | |
| US11965935B2 (en) | Method and apparatus for operating a system for providing predicted states of health of electrical energy stores for a device using machine learning methods | |
| CN113009349A (en) | Lithium ion battery health state diagnosis method based on deep learning model | |
| CN115201686B (en) | Lithium ion battery health state assessment method under incomplete charge and discharge data | |
| CN115015789B (en) | Lithium ion battery health state estimation method and device based on incremental capacitance curve | |
| CN115308608A (en) | All-vanadium redox flow battery voltage prediction method, device and medium | |
| CN115935659A (en) | Fuel cell stack service life prediction method and system and electronic equipment | |
| CN114814589A (en) | Method and device for predicting remaining service life of PEMFC | |
| CN115629314B (en) | Battery parameter and state joint estimation method and system based on improved Jaya | |
| CN114839538A (en) | Method for extracting degradation characteristics of lithium ion battery for estimating residual life | |
| Li et al. | Degradation prediction of proton exchange membrane fuel cell based on the multi-inputs Bi-directional long short-term memory | |
| López-Villanueva et al. | A fractional-order model for calendar aging with dynamic storage conditions | |
| CN113722898A (en) | Durability evaluation method and device for fuel cell stack | |
| CN117825970A (en) | Battery degradation analysis method, device, equipment and storage medium | |
| Hong et al. | State of health prediction for proton exchange membrane fuel cells combining semi-empirical model and machine learning | |
| CN117761558A (en) | Lithium battery health state estimation method and device based on multi-learning machine fusion | |
| CN117630673A (en) | Method and device for initially providing an aging state model for an energy store | |
| CN114594397B (en) | Method and system for estimating health state of lithium ion battery based on generalization model | |
| CN118294815A (en) | Method and device for predicting performance of fuel cell for vehicle and electronic equipment | |
| CN116819364A (en) | Battery health determination method, application method, electronic device and storage medium | |
| CN115701545A (en) | Method and device for providing a calculated and predicted state of aging of an electrical energy store |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |