CN113671386B - Method and device for analyzing durability of hydrogen fuel cell - Google Patents
Method and device for analyzing durability of hydrogen fuel cell Download PDFInfo
- Publication number
- CN113671386B CN113671386B CN202110861037.3A CN202110861037A CN113671386B CN 113671386 B CN113671386 B CN 113671386B CN 202110861037 A CN202110861037 A CN 202110861037A CN 113671386 B CN113671386 B CN 113671386B
- Authority
- CN
- China
- Prior art keywords
- voltage
- data
- target
- fuel cell
- time
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/385—Arrangements for measuring battery or accumulator variables
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Fuel Cell (AREA)
Abstract
The invention relates to the technical field of fuel cells, in particular to a method and a device for analyzing the durability of a hydrogen fuel cell.
Description
Technical Field
The invention relates to the technical field of fuel cells, in particular to a method and a device for analyzing the durability of a hydrogen fuel cell.
Background
The existing whole-vehicle endurance drum test is to perform an endurance test (about 1500 h) for a preset duration according to the urban working condition (ECE working condition) of NEDC, and verify the reliability of the whole hydrogen fuel cell.
Although such data can reflect the battery decay law over a certain period of time, the decay law for the subsequent period of time cannot be readily obtained, and if the duration of the endurance test is prolonged and the test is continued, the time taken is longer and more manpower and material resources are required.
Therefore, how to improve the efficiency and accuracy of the durability analysis of the hydrogen fuel cell is a technical problem to be solved at present.
Disclosure of Invention
The present invention has been made in view of the above problems, and has as its object to provide a method and apparatus for analyzing durability performance of a hydrogen fuel cell that overcomes or at least partially solves the above problems.
In a first aspect, the present invention provides a method for analyzing durability performance of a hydrogen fuel cell, comprising:
performing a durable drum test for a first preset time length on the whole vehicle, and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
based on the related characteristic data and the voltage data, a prediction model for predicting the voltage data in the first preset duration is obtained;
based on the prediction model, sequentially performing extrapolation prediction on each target time in a second preset time period after the first preset time period to obtain a target voltage corresponding to each target time;
obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing with time within the first preset time period;
and analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve.
Further, the relevant characteristic data includes: the current, the temperature of the water inlet of the hydrogen fuel cell stack, the temperature of the water outlet of the hydrogen fuel cell stack and the driving mileage.
Further, the obtaining a prediction model for predicting voltage data within the first preset duration based on the relevant characteristic data and voltage data includes:
acquiring sample data corresponding to each preset time within the first preset time;
extracting relevant characteristic data corresponding to each preset time and relevant characteristic data and voltage data corresponding to the time before the preset time from the sample data as input data, wherein the voltage data corresponding to each preset time in the sample data is used as output data;
and inputting the input data and the output data into an LSTM model to train the LSTM model, so as to obtain a prediction model for predicting voltage data at any moment.
Further, based on the prediction model, performing extrapolation prediction on each target time in a second preset time period after the first preset time period in sequence to obtain a target voltage corresponding to each target time, where the extrapolation prediction includes:
acquiring first relevant characteristic data corresponding to a first target time in a second preset time after the first preset time, and front voltage data and front relevant characteristic data corresponding to a front time of the first target time;
inputting the first related characteristic data, the front related characteristic data and the front voltage data into the preset model to obtain a first target voltage corresponding to the first target moment;
sequentially extrapolation is carried out to obtain an Mth target voltage corresponding to an Mth target time within the second preset time, M is a sorting number corresponding to any target time within the second preset time, and the target voltage corresponding to each target time is obtained.
Further, obtaining first relevant feature data corresponding to a first target time within a second preset time after the first preset time includes:
acquiring a value range of the temperature of a water inlet of the hydrogen fuel cell stack within the first preset time period;
based on the value range, determining the temperature of the water inlet of the hydrogen fuel cell stack corresponding to the first target moment as any value in the value range;
determining that the temperature of the water outlet of the hydrogen fuel cell stack corresponding to the first target moment exceeds the preset temperature of the water inlet of the hydrogen fuel cell stack;
determining a driving mileage corresponding to the first target moment based on the average speed of the whole vehicle and the first target moment;
and determining the current corresponding to the first target moment as a preset constant current.
Further, the obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing with time in the first preset time period includes:
and performing polynomial fitting on the target voltage and the voltage data which change along with time in the first preset time period by adopting a fitting method to obtain a voltage decay curve of the hydrogen fuel cell in the total time period, wherein the total time period is the first preset time period plus the second preset time period.
Further, the durability performance analysis of the hydrogen fuel cell based on the voltage decay curve is characterized by comprising:
obtaining a maximum value and a minimum value of the voltage based on the voltage decay curve;
obtaining a voltage decay rate based on the maximum and minimum values;
and analyzing the durability performance of the hydrogen fuel cell based on the voltage decay rate.
In a second aspect, the present invention also provides a method for analyzing durability of a hydrogen fuel cell, comprising:
the recording module is used for performing a durable drum test for a first preset time length on the whole vehicle and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
the model obtaining module is used for obtaining a prediction model for predicting voltage data in the first preset duration based on the related characteristic data and the voltage data;
the extrapolation prediction module is used for sequentially carrying out extrapolation prediction on each target time within a second preset time period after the first preset time period based on the prediction model so as to obtain a target voltage corresponding to each target time;
the curve obtaining module is used for obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing along with time in the first preset duration;
and the analysis module is used for analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve.
In a third aspect, the present invention also provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the above method steps when executing the program.
In a fourth aspect, the present invention also provides a computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor performs the above method steps.
One or more technical solutions in the embodiments of the present invention at least have the following technical effects or advantages:
the invention provides a method for analyzing the durability performance of a hydrogen fuel cell, which comprises the steps of recording relevant characteristic data and voltage data of the hydrogen fuel cell, which change along with time, by performing a durability drum test on a whole vehicle for a first preset time period, obtaining a prediction model for predicting the voltage data in the first preset time period based on the relevant characteristic data and the voltage data, sequentially performing extrapolation prediction on each target moment in a second preset time period after the first preset time period based on the prediction model so as to obtain a target voltage corresponding to each target moment, obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data, which change along with time in the first preset time period, analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve, further obtaining real data through the whole vehicle drum test, predicting any moment data in the later time period to obtain real prediction data, further predicting the voltage decay trend in the long time period of the hydrogen fuel cell to obtain accurate prediction results, and improving the efficiency of performance decay analysis.
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. Also throughout the drawings, like reference numerals are used to designate like parts. In the drawings:
FIG. 1 is a schematic flow chart showing the steps of a method for analyzing the durability performance of a hydrogen fuel cell according to an embodiment of the present invention;
FIG. 2 illustrates a schematic diagram of ECE cycle conditions in an embodiment of the present invention;
FIG. 3 is a graph showing the voltage of a hydrogen fuel cell over 5000 hours in an embodiment of the present invention;
FIG. 4 is a schematic diagram of a voltage decay curve in an embodiment of the present invention;
fig. 5 is a schematic view showing the structure of a hydrogen fuel cell durability performance analyzing apparatus in an embodiment of the invention;
fig. 6 is a schematic diagram showing the structure of a computer device for realizing the method for analyzing the durability performance of a hydrogen fuel cell in the embodiment of the 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.
Example 1
An embodiment of the present invention provides a method for analyzing durability performance of a hydrogen fuel cell, as shown in fig. 1, including:
s101, performing a durable drum test for a first preset time length on the whole vehicle, and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
s102, obtaining a prediction model for predicting voltage data in a first preset duration based on related characteristic data and voltage data;
s103, based on a prediction model, sequentially performing extrapolation prediction on each target time in a second preset time period after the first preset time period to obtain a target voltage corresponding to each target time;
s104, obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing along with time in the first preset time period;
s105, analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve.
Firstly, in S101, a durable drum test is performed on the whole vehicle for a first preset period of time, and relevant characteristic data and voltage data of the hydrogen fuel cell changing with time are recorded.
Specifically, a whole vehicle emission test endurance drum is used for carrying out a 1500-1800 h endurance test according to ECE working conditions, and the reliability of the whole hydrogen fuel cell is verified according to the running of the first-stage urban working condition (ECE) of the circulating NEDC in BG 18352-2013.
The NEDC is used for placing the vehicle on a test bench and simulating the driving process of the vehicle under different working conditions.
Using the test described above, the ECE cycle loading is shown in fig. 2.
The relevant characteristic data and voltage data of the hydrogen fuel cell with time are obtained by sampling every 20ms period.
After S101, data processing is performed on the recorded characteristic data and voltage data of the hydrogen fuel cell over time.
The data processing specifically comprises the following steps: filling the data, filtering the data, and the like. For some empty data, the last record needs to be used for filling, that is, the empty data is considered to keep the state of the previous data record. And for the case of some abrupt data changes, adopting a sliding algorithm average filtering method to carry out filtering treatment.
The method comprises the steps of filtering an average value of a sliding algorithm, establishing a data buffer area in a memory (RAM), sequentially storing N pieces of sampling data, discarding the earliest collected data every time the N pieces of sampling data are collected, and then calculating an arithmetic average value of the N pieces of data including the new data, so that the new average value can be calculated every time the sampling is carried out, and the data processing speed is increased.
Next, S102 is performed to obtain a prediction model for predicting voltage data for a first preset time period based on the relevant characteristic data and the voltage data.
The relevant data are specifically: current, hydrogen fuel cell stack water inlet temperature, hydrogen fuel cell stack water outlet temperature, and mileage.
The method specifically comprises the step of performing model training and model testing by using data acquired in the first preset time length, namely 1500-1800 hours.
A proper model, such as a cyclic neural network (RNN) model, a variant (LSTM/GRU) model and the like, is selected, and is a model for processing the sequence, wherein the adopted mode is supervised learning, artificial time sequence characteristics are not needed to be constructed, a time sequence curve can be fitted through a deep learning network, a time sequence relation is captured, and characteristic learning and prediction are performed in a long-term dependence mode.
Although conventional Convolutional Neural Networks (CNNs), time Convolutional Networks (TCNs), and conventional time-learned prediction methods can be used, the prediction results are not ideal.
Therefore, in the invention, an LSTM model (long short term record neural network model) in a cyclic neural network model is adopted, wherein a model structure of a many to one is adopted, wherein the many refers to the characteristics of more than one variable characteristic, and the one refers to one variable characteristic.
The data obtained in S101 needs to be processed before training the model, including data normalization processing, so as to accelerate the drop of the training loss and the test loss of the training model. Specifically, the data is normalized by using the minmaxscale () function and the fit_transform () function of the sklearn toolkit in the computer.
Next, the data after normalization processing was divided into two parts in a ratio of 8:2, with data with a ratio of 80% as training data and data with a ratio of 20% as test data.
Wherein the training data comprises input data and output data, and the test data also comprises the input data and the output data.
The specific training process is as follows: acquiring sample data corresponding to each preset time within a first preset time; extracting relevant characteristic data corresponding to each preset time and relevant characteristic data and voltage characteristic data corresponding to the time before the preset time from sample data to serve as input data, and summarizing voltage data corresponding to each preset time of the sample data to serve as output data; and inputting the input data and the output data into the LSTM model to train the LSTM model, so as to obtain a prediction model for predicting the voltage data at any moment.
The input data comprises relevant characteristic data and voltage data corresponding to a first moment and relevant characteristic data corresponding to a second moment, wherein the first moment is a moment before the second moment. The preceding time may be the preceding time or the preceding two times, and of course, the preceding time may be the preceding N times, which is not limited herein.
Taking 1500h as an example, taking one record of every half hour as an example, when the input data is related characteristic data and voltage data corresponding to 0.5h and related characteristic data corresponding to 1h, the output data is voltage data corresponding to 0.5 h. Until the input data is the relevant characteristic data and the voltage data corresponding to 1199.5h and the relevant characteristic data corresponding to 1200h, the output data is the voltage data corresponding to 1200 h. Thereby obtaining the training data.
Similarly, for the data in the last 300h of 1500h, input data and output data are obtained in the manner described above, thereby obtaining a test set.
The input data and the output data in the training set are input into the LSTM model to train the LSTM model, so that a prediction model for predicting voltage data at any moment is obtained.
Then, the predictive model is tested using the input data and the output data in the test set, thereby verifying the accuracy and the reliability of the predictive model.
The data in the training set and the data in the testing set are 3D format data, and the 3D format is [ sample number, time step length and feature number ], wherein the sample number is the number of samples in the training or testing data set; when the time step is time sequence prediction, the current time and the previous time data need to be considered, and if the current time and the previous time data are considered, the time step is 1; if the current time and the last two time data are considered, the time step is 2, and so on. In the invention, only the data of the current moment and the previous moment are considered, and the instant step length is 1. The feature number is the number of input features.
The LSTM model adopted in the invention has 500 initial values of neuron number, and when the neural network model is constructed, the optimal optimizer is selected so as to quickly converge and accurately learn, meanwhile, internal parameters are adjusted, and a loss function is minimized to the greatest extent, wherein Adam has good effect in practical application and exceeds other self-adaptive technologies, so that the LSTM long-term memory neural network algorithm optimizer is Adam.
When the obtained prediction model is tested, the loss evaluation index adopts a mean square error, specifically, the mean square error is made between the predicted value and the true value, and the larger the mean square error is, the worse the effect of prediction is. During training and testing, as the iteration times are increased, the training loss and the testing loss are accelerated to be reduced, and reach a stable value quickly, and the stable value is close to the training loss, so that the training process data convergence condition is better and meets the expectations.
After obtaining the prediction model with better effect, S103 is executed, and based on the prediction model, extrapolation prediction is sequentially performed on each target time within a second preset time after the first preset time, so as to obtain a target voltage corresponding to each target time.
Specifically, first relevant characteristic data corresponding to a first target time in a second preset time after a first preset time is obtained, and front voltage data and front relevant characteristic data corresponding to a front time of the first target time are obtained;
inputting the first related characteristic data, the front related characteristic data and the front voltage data into a preset model to obtain a first target voltage corresponding to a first target moment;
sequentially extrapolation is carried out to obtain an Mth target voltage corresponding to an Mth target time within a second preset time, M is a sequencing number corresponding to any target time within the second preset time, and the target voltage corresponding to each target time is obtained.
In a specific embodiment, the first preset duration is 1500h, and the relevant characteristic parameters and the voltage data corresponding to the last moment, i.e. 1500h, are known.
First, first relevant feature data corresponding to a first target time within a second preset time after a first preset time is obtained, wherein the second preset time is 3500h, and the total time length of the first preset time plus the second preset time is 5000h. The first target time within the second preset time is 1500.5h, and the first relevant feature data corresponding to 1500.5h can be obtained through estimation.
Specifically, the first relevant characteristic data includes: current, hydrogen fuel cell stack water inlet temperature, hydrogen fuel cell stack water outlet temperature, and mileage.
Firstly, acquiring a value range of the temperature of a water inlet of a hydrogen fuel cell stack within a first preset time period; then, based on the value range, determining the temperature of the water inlet of the hydrogen fuel cell stack corresponding to the first target moment as any value in the value range. Here, the value range may be segmented, and values may be randomly taken from the segmented result.
And for the temperature of the water outlet of the hydrogen fuel cell stack, the temperature of the water inlet of the hydrogen fuel cell stack corresponding to the first target moment is higher than a preset temperature, and the temperature of the water outlet of the hydrogen fuel cell stack is obtained, wherein the preset temperature is 1-2 ℃.
For the driving mileage, the driving mileage corresponding to the first target time is determined based on the average driving speed of the whole vehicle and the first target time, for example, the average driving speed is 22km/h, and if the first target time is 1500.5h, the corresponding driving mileage is 33011km.
For the current, the current is a preset constant value, for example, a constant value randomly selected from 26A to 28A.
Then, the front voltage data and the front related characteristic data corresponding to the front moment of the first target moment can be accurately predicted by the prediction model.
After the data are known, the first relevant characteristic data, the pre-relevant characteristic data and the pre-voltage data are input into the prediction model, so that a first target voltage corresponding to the first target time is obtained.
Taking 1500h as the previous time and 1500.5h as the first target time, the first relevant characteristic data corresponding to 1500.5h, the previous relevant characteristic data corresponding to 1500h and the previous voltage data are input into a prediction model to obtain the target voltage corresponding to 1500.5 h.
Then extrapolate again to obtain the first relevant feature data and the first target voltage (obtained by extrapolation prediction in the front) corresponding to 1500.5h and the second relevant feature data (obtained by prediction according to the method described above) corresponding to 1501h, so as to input the data into a prediction model to obtain the second target voltage corresponding to 1501h, and similarly, the M target voltage corresponding to the M target time within a second preset time period can be obtained, wherein M is the sequence number corresponding to any target time within the second preset time period, so as to obtain the target voltage corresponding to each target time. Thus, a target voltage corresponding to any one of the time points within 3500h after 1500h was obtained.
Thus, the target voltage corresponding to each target time within 3500h after 1500h, specifically as shown in fig. 3, can be extrapolated, thereby obtaining a curve of the voltage change with time of the hydrogen fuel cell within 5000h.
Then, S104 is performed to obtain a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data varying with time within the first preset time period. I.e. a voltage decay curve of the voltage data over time in 5000h.
Specifically, as the data have larger floating according to the voltage decay diagram obtained by the actual data, burrs exist at the edge of the curve, and the performance analysis of the hydrogen fuel cell cannot be accurately performed, a fitting method is adopted to perform polynomial fitting on the target voltage and the voltage data which change with time in a first preset time period, so as to obtain a voltage decay curve of the hydrogen fuel cell in a total time period, wherein the total time period is the time period of the first preset time period plus a second preset time period, namely 5000 hours.
Specifically, a third-order polynomial fitting may be adopted, or a fifth-order polynomial fitting may be adopted, which is not limited herein, and the more the degree of the polynomial is, the higher the accuracy of the fitting is.
When a fifth order polynomial is used, the formula is: y=1.332 e -16 x 5 -1.932e -12 x 4 +9.512e -9 x 3 -1.82e -5 x 2 +0.009053x+274.5
Wherein x is any time in the range of 0 to 5000 hours, and y is a voltage value corresponding to any time.
A smooth curve is thus obtained, namely a voltage decay curve, on which inflection point coordinate positions (324, 276), (2131, 269), (3461, 271), (5691, 258) are obtained, as shown in fig. 4.
Finally, S105 is performed to analyze the durability performance of the hydrogen fuel cell based on the voltage decay curve. The method specifically comprises the following steps:
obtaining a maximum value and a minimum value of the voltage based on the voltage decay curve; next, a voltage decay rate is obtained based on the maximum value and the minimum value, and the durability performance of the hydrogen fuel cell is analyzed based on the voltage decay rate.
Specifically, the maximum value and the minimum value are subjected to difference, and the difference value and the maximum value are subjected to quotient to obtain the voltage attenuation rate.
For example, if the voltage decay rate obtained by this is large, it is determined that the hydrogen fuel cell has poor durability; if the obtained voltage decay rate is small, it is determined that the hydrogen fuel cell has excellent durability.
One or more technical solutions in the embodiments of the present invention at least have the following technical effects or advantages:
the invention provides a method for analyzing the durability performance of a hydrogen fuel cell, which comprises the steps of recording relevant characteristic data and voltage data of the hydrogen fuel cell, which change along with time, by performing a durability drum test on a whole vehicle for a first preset time period, obtaining a prediction model for predicting the voltage data in the first preset time period based on the relevant characteristic data and the voltage data, sequentially performing extrapolation prediction on each target moment in a second preset time period after the first preset time period based on the prediction model so as to obtain a target voltage corresponding to each target moment, obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data, which change along with time in the first preset time period, analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve, further obtaining real data through the whole vehicle drum test, predicting any moment data in the later time period to obtain real prediction data, further predicting the voltage decay trend in the long time period of the hydrogen fuel cell to obtain accurate prediction results, and improving the efficiency of performance decay analysis.
Example two
Based on the same inventive concept, the present invention also provides a hydrogen fuel cell durability analysis apparatus, as shown in fig. 5, comprising:
the recording module 501 is used for performing a durable drum test for a first preset time length on the whole vehicle and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
a model obtaining module 502, configured to obtain a prediction model for predicting voltage data within the first preset duration based on the relevant feature data and voltage data;
an extrapolation prediction module 503, configured to sequentially extrapolate, based on the prediction model, each target time within a second preset duration after the first preset duration, so as to obtain a target voltage corresponding to each target time;
a curve obtaining module 504, configured to obtain a voltage degradation curve of the hydrogen fuel cell based on the target voltage and the voltage data that changes with time in the first preset duration;
an analysis module 505 for analyzing the hydrogen fuel cell durability performance based on the voltage decay curve.
In an alternative embodiment, the relevant characteristic data includes: the current, the temperature of the water inlet of the hydrogen fuel cell stack, the temperature of the water outlet of the hydrogen fuel cell stack and the driving mileage.
In an alternative embodiment, the model obtaining module 502 includes:
the first acquisition unit is used for acquiring sample data corresponding to each preset time within the first preset time;
the extraction unit is used for extracting relevant characteristic data corresponding to each preset time and relevant characteristic data and voltage data corresponding to the time before the preset time from the sample data to be used as input data, and voltage data corresponding to each preset time in the sample data to be used as output data;
the training unit is used for inputting the input data and the output data into an LSTM model so as to train the LSTM model and obtain a prediction model for predicting voltage data at any moment.
In an alternative embodiment, extrapolation prediction module 503 includes:
the second acquisition unit is used for acquiring first relevant characteristic data corresponding to a first target time in a second preset time after the first preset time, and front voltage data and front relevant characteristic data corresponding to a front time of the first target time;
the first extrapolation prediction unit is used for inputting the first related characteristic data, the front related characteristic data and the front voltage data into the preset model to obtain a first target voltage corresponding to the first target moment;
the sequential extrapolation prediction unit is used for sequentially extrapolating to obtain an Mth target voltage corresponding to an Mth target time within the second preset time, wherein M is a sorting number corresponding to any target time within the second preset time so as to obtain a target voltage corresponding to each target time.
In an alternative embodiment, the first acquisition unit comprises:
the first acquisition subunit is used for acquiring the value range of the temperature of the water inlet of the hydrogen fuel cell stack within the first preset time period;
the first determining subunit is used for determining that the temperature of the water inlet of the hydrogen fuel cell stack corresponding to the first target moment is any value in the value range based on the value range;
the second determining subunit is used for determining that the temperature of the water outlet of the hydrogen fuel cell stack corresponding to the first target moment exceeds the preset temperature of the water inlet of the hydrogen fuel cell stack;
the third determining subunit is used for determining the driving mileage corresponding to the first target moment based on the average speed of the whole vehicle and the first target moment;
and the fourth determination subunit is used for determining that the current corresponding to the first target moment is a preset constant current.
In an alternative embodiment, the curve obtaining module 504 is configured to:
and performing polynomial fitting on the target voltage and the voltage data which change along with time in the first preset time period by adopting a fitting method to obtain a voltage decay curve of the hydrogen fuel cell in the total time period, wherein the total time period is the first preset time period plus the second preset time period.
In an alternative embodiment, the analysis module 505 includes:
a first obtaining unit configured to obtain a maximum value and a minimum value of a voltage based on the voltage decay curve;
a second obtaining unit configured to obtain a voltage attenuation rate based on the maximum value and the minimum value;
and an analysis unit configured to analyze durability performance of the hydrogen fuel cell based on the voltage decay rate.
Example III
Based on the same inventive concept, a third embodiment of the present invention provides a computer device, as shown in fig. 6, including a memory 604, a processor 602, and a computer program stored in the memory 604 and capable of running on the processor 602, where the processor 602 implements the steps of the above-mentioned hydrogen fuel cell durability performance analysis method when executing the program.
Where in FIG. 6, a bus architecture (represented by bus 600), bus 600 may include any number of interconnected buses and bridges, with bus 600 linking together various circuits, including one or more processors, represented by processor 602, and memory, represented by memory 604. Bus 600 may also link together various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., as are well known in the art and, therefore, will not be described further herein. The bus interface 606 provides an interface between the bus 600 and the receiver 601 and transmitter 603. The receiver 601 and the transmitter 603 may be the same element, i.e. a transceiver, providing a means for communicating with various other apparatus over a transmission medium. The processor 602 is responsible for managing the bus 600 and general processing, while the memory 604 may be used to store data used by the processor 602 in performing operations.
Example IV
Based on the same inventive concept, an embodiment of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above-described hydrogen fuel cell durability performance analysis method.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general-purpose systems may also be used with the teachings herein. The required structure for a construction of such a system is apparent from the description above. In addition, the present invention is not directed to any particular programming language. It will be appreciated that the teachings of the present invention described herein may be implemented in a variety of programming languages, and the above description of specific languages is provided for disclosure of enablement and best mode of the present invention.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be construed as reflecting the intention that: i.e., the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the apparatus of the embodiments may be adaptively changed and disposed in one or more apparatuses different from the embodiments. The modules or units or components of the embodiments may be combined into one module or unit or component and, furthermore, they may be divided into a plurality of sub-modules or sub-units or sub-components. Any combination of all features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or units of any method or apparatus so disclosed, may be used in combination, except insofar as at least some of such features and/or processes or units are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
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, in the following claims, any of the claimed embodiments can be used in any combination.
Various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that some or all of the functions of some or all of the components in a hydrogen fuel cell durability performance analyzing apparatus, a computer device according to an embodiment of the invention may be implemented in practice using a microprocessor or Digital Signal Processor (DSP). The present invention can also be implemented as an apparatus or device program (e.g., a computer program and a computer program product) for performing a portion or all of the methods described herein. Such a program embodying the present invention may be stored on a computer readable medium, or may have the form of one or more signals. Such signals may be downloaded from an internet website, provided on a carrier signal, or provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third, etc. do not denote any order. These words may be interpreted as names.
Claims (9)
1. A method for analyzing durability performance of a hydrogen fuel cell, comprising:
performing a durable drum test for a first preset time length on the whole vehicle, and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
based on the related characteristic data and the voltage data, a prediction model for predicting the voltage data in the first preset duration is obtained;
based on the prediction model, sequentially performing extrapolation prediction on each target time in a second preset time period after the first preset time period to obtain a target voltage corresponding to each target time; sequentially performing extrapolation prediction on each target time within a second preset time period after the first preset time period based on the prediction model to obtain a target voltage corresponding to each target time, including:
acquiring first relevant characteristic data corresponding to a first target time in a second preset time after the first preset time, and front voltage data and front relevant characteristic data corresponding to a front time of the first target time;
inputting the first related characteristic data, the front related characteristic data and the front voltage data into the prediction model to obtain a first target voltage corresponding to the first target moment;
sequentially extrapolation is carried out to obtain an Mth target voltage corresponding to an Mth target time within the second preset time, M is a sorting number corresponding to any target time within the second preset time, and the target voltage corresponding to each target time is obtained;
obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing with time within the first preset time period;
and analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve.
2. The method of claim 1, wherein the relevant characteristic data comprises: the current, the temperature of the water inlet of the hydrogen fuel cell stack, the temperature of the water outlet of the hydrogen fuel cell stack and the driving mileage.
3. The method of claim 1, wherein the obtaining a prediction model for predicting voltage data for the first preset duration based on the relevant characteristic data and voltage data comprises:
acquiring sample data corresponding to each preset time within the first preset time;
extracting relevant characteristic data corresponding to each preset time and relevant characteristic data and voltage data corresponding to the time before the preset time from the sample data as input data, wherein the voltage data corresponding to each preset time in the sample data is used as output data;
and inputting the input data and the output data into an LSTM model to train the LSTM model, so as to obtain a prediction model for predicting voltage data at any moment.
4. The method of claim 1, wherein obtaining first relevant feature data corresponding to a first target time within a second preset duration after the first preset duration comprises:
acquiring a value range of the temperature of a water inlet of the hydrogen fuel cell stack within the first preset time period;
based on the value range, determining the temperature of the water inlet of the hydrogen fuel cell stack corresponding to the first target moment as any value in the value range;
determining that the temperature of the water outlet of the hydrogen fuel cell stack corresponding to the first target moment exceeds the preset temperature of the water inlet of the hydrogen fuel cell stack;
determining a driving mileage corresponding to the first target moment based on the average speed of the whole vehicle and the first target moment;
and determining the current corresponding to the first target moment as a preset constant current.
5. The method of claim 1, wherein the deriving a voltage decay curve of the hydrogen fuel cell based on the target voltage and the time-varying voltage data within the first preset time period comprises:
and performing polynomial fitting on the target voltage and the voltage data which change along with time in the first preset time period by adopting a fitting method to obtain a voltage decay curve of the hydrogen fuel cell in the total time period, wherein the total time period is the first preset time period plus the second preset time period.
6. The method of claim 1, wherein said analyzing said hydrogen fuel cell durability performance based on said voltage decay curve comprises:
obtaining a maximum value and a minimum value of the voltage based on the voltage decay curve;
obtaining a voltage decay rate based on the maximum and minimum values;
and analyzing the durability performance of the hydrogen fuel cell based on the voltage decay rate.
7. A method for analyzing durability performance of a hydrogen fuel cell, comprising:
the recording module is used for performing a durable drum test for a first preset time length on the whole vehicle and recording relevant characteristic data and voltage data of the hydrogen fuel cell along with time variation;
the model obtaining module is used for obtaining a prediction model for predicting voltage data in the first preset duration based on the related characteristic data and the voltage data;
the extrapolation prediction module is used for sequentially carrying out extrapolation prediction on each target time within a second preset time period after the first preset time period based on the prediction model so as to obtain a target voltage corresponding to each target time;
the extrapolation prediction module further includes:
the second acquisition unit is used for acquiring first relevant characteristic data corresponding to a first target time in a second preset time after the first preset time, and front voltage data and front relevant characteristic data corresponding to a front time of the first target time;
the first extrapolation prediction unit is used for inputting the first relevant characteristic data, the front relevant characteristic data and the front voltage data into the prediction model to obtain a first target voltage corresponding to the first target moment;
the sequential extrapolation prediction unit is used for sequentially extrapolating to obtain an Mth target voltage corresponding to an Mth target time within the second preset time, wherein M is a sorting number corresponding to any target time within the second preset time so as to obtain a target voltage corresponding to each target time;
the curve obtaining module is used for obtaining a voltage decay curve of the hydrogen fuel cell based on the target voltage and the voltage data changing along with time in the first preset duration;
and the analysis module is used for analyzing the durability performance of the hydrogen fuel cell based on the voltage decay curve.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1-6 when the program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110861037.3A CN113671386B (en) | 2021-07-29 | 2021-07-29 | Method and device for analyzing durability of hydrogen fuel cell |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110861037.3A CN113671386B (en) | 2021-07-29 | 2021-07-29 | Method and device for analyzing durability of hydrogen fuel cell |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113671386A CN113671386A (en) | 2021-11-19 |
| CN113671386B true CN113671386B (en) | 2023-09-26 |
Family
ID=78540760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110861037.3A Active CN113671386B (en) | 2021-07-29 | 2021-07-29 | Method and device for analyzing durability of hydrogen fuel cell |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113671386B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116644870B (en) * | 2023-07-27 | 2024-01-16 | 佛山市顺德区裕安燃气具实业有限公司 | Method and device for realizing gas information prediction |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005235420A (en) * | 2004-02-17 | 2005-09-02 | Nippon Telegr & Teleph Corp <Ntt> | Life prediction method of nickel-hydrogen storage battery |
| CN104868143A (en) * | 2014-02-24 | 2015-08-26 | 现代自动车株式会社 | Method and apparatus for diagnosing state of fuel cell system |
| DE102017205038A1 (en) * | 2017-03-24 | 2018-09-27 | Bayerische Motoren Werke Aktiengesellschaft | Control unit and method for checking a fuel cell |
| CN109683093A (en) * | 2019-02-01 | 2019-04-26 | 清华大学 | Fuel battery service life prediction technique, prediction meanss and computer readable storage medium |
| CN111046550A (en) * | 2019-12-10 | 2020-04-21 | 电子科技大学 | Fuel cell voltage prediction system based on space-time characteristics |
| CN111146478A (en) * | 2019-12-22 | 2020-05-12 | 同济大学 | Method for predicting residual service life of proton exchange membrane fuel cell stack |
| CN112659927A (en) * | 2020-10-14 | 2021-04-16 | 中国汽车技术研究中心有限公司 | Method, apparatus, device and medium for determining performance characteristics of fuel cell stack for vehicle |
| CN112687926A (en) * | 2019-10-18 | 2021-04-20 | 中国科学院大连化学物理研究所 | Method and system for predicting attenuation trend and predicting remaining service life of fuel cell |
| CN112763916A (en) * | 2021-01-04 | 2021-05-07 | 北京空间飞行器总体设计部 | Method for predicting future working conditions of lithium ion battery pack for space |
-
2021
- 2021-07-29 CN CN202110861037.3A patent/CN113671386B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005235420A (en) * | 2004-02-17 | 2005-09-02 | Nippon Telegr & Teleph Corp <Ntt> | Life prediction method of nickel-hydrogen storage battery |
| CN104868143A (en) * | 2014-02-24 | 2015-08-26 | 现代自动车株式会社 | Method and apparatus for diagnosing state of fuel cell system |
| DE102017205038A1 (en) * | 2017-03-24 | 2018-09-27 | Bayerische Motoren Werke Aktiengesellschaft | Control unit and method for checking a fuel cell |
| CN109683093A (en) * | 2019-02-01 | 2019-04-26 | 清华大学 | Fuel battery service life prediction technique, prediction meanss and computer readable storage medium |
| CN112687926A (en) * | 2019-10-18 | 2021-04-20 | 中国科学院大连化学物理研究所 | Method and system for predicting attenuation trend and predicting remaining service life of fuel cell |
| CN111046550A (en) * | 2019-12-10 | 2020-04-21 | 电子科技大学 | Fuel cell voltage prediction system based on space-time characteristics |
| CN111146478A (en) * | 2019-12-22 | 2020-05-12 | 同济大学 | Method for predicting residual service life of proton exchange membrane fuel cell stack |
| CN112659927A (en) * | 2020-10-14 | 2021-04-16 | 中国汽车技术研究中心有限公司 | Method, apparatus, device and medium for determining performance characteristics of fuel cell stack for vehicle |
| CN112763916A (en) * | 2021-01-04 | 2021-05-07 | 北京空间飞行器总体设计部 | Method for predicting future working conditions of lithium ion battery pack for space |
Non-Patent Citations (3)
| Title |
|---|
| 模拟车况下的质子交换膜燃料电池耐久性研究;陆鑫 等;电源技术;第33卷(第3期);178-181 * |
| 道路环境下质子交换膜燃料电池系统性能衰退统计研究;张新丰;沈勇;杨瑞;章桐;;电工技术学报(06) * |
| 采用数学模型估计氢镍蓄电池的劣化程度;高爽 等;电源技术;第42卷(第6期);828-831 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113671386A (en) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109001640B (en) | Data processing method and device for power battery | |
| CN109976998B (en) | Software defect prediction method and device and electronic equipment | |
| KR102206324B1 (en) | Methods for ascertaining a model of a starting variable of a technical system | |
| CN111914936B (en) | Data characteristic enhancement method and device for corpus data and computer equipment | |
| CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
| CN111191824B (en) | Power battery capacity attenuation prediction method and system | |
| CN113671386B (en) | Method and device for analyzing durability of hydrogen fuel cell | |
| CN114911788B (en) | Data interpolation method and device and storage medium | |
| CN113743661A (en) | Method, system, equipment and storage medium for predicting online capacity of lithium ion battery | |
| CN115656840A (en) | Method, device, system and storage medium for predicting battery charging remaining time | |
| CN115407211A (en) | Online prediction method and system for health state of lithium battery of electric vehicle | |
| CN113406496B (en) | Battery capacity prediction method, system, device and medium based on model migration | |
| CN112906883A (en) | Hybrid precision quantization strategy determination method and system for deep neural network | |
| CN116224091B (en) | Battery cell fault detection method and device, electronic equipment and storage medium | |
| CN114778141B (en) | Method, device and equipment for formulating durability test of automobile electric drive assembly rack | |
| CN111401551A (en) | Weak supervision self-learning method based on reinforcement learning | |
| CN116466253A (en) | Method and device for analyzing attenuation of fuel cell | |
| CN115600728A (en) | Annual carbon emission estimation method and device for power battery | |
| CN114841553A (en) | Fault diagnosis method, system, equipment and storage medium for generator rotating rectifier | |
| CN117129895A (en) | Battery state of health calculation method, device, storage medium and vehicle | |
| CN113821452A (en) | Intelligent test method for dynamically generating test case according to test performance of tested system | |
| CN114819449A (en) | Method, device and equipment for predicting defrosting performance of vehicle and storage medium | |
| CN117555812B (en) | Cloud platform automatic testing method and system | |
| CN112861951B (en) | Image neural network parameter determining method and electronic equipment | |
| CN113609777B (en) | Vehicle behavior analysis method and device based on vehicle-mounted sensing equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |