CN115084600B - Hydrogen fuel cell stack output performance analysis method based on big data - Google Patents
Hydrogen fuel cell stack output performance analysis method based on big data Download PDFInfo
- Publication number
- CN115084600B CN115084600B CN202210885423.0A CN202210885423A CN115084600B CN 115084600 B CN115084600 B CN 115084600B CN 202210885423 A CN202210885423 A CN 202210885423A CN 115084600 B CN115084600 B CN 115084600B
- Authority
- CN
- China
- Prior art keywords
- pile
- parameters
- stack
- steady
- 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
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 title claims abstract description 64
- 229910052739 hydrogen Inorganic materials 0.000 title claims abstract description 64
- 239000001257 hydrogen Substances 0.000 title claims abstract description 64
- 239000000446 fuel Substances 0.000 title claims abstract description 54
- 238000004458 analytical method Methods 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000012417 linear regression Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 101100508520 Mus musculus Nfkbiz gene Proteins 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000007873 sieving Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 238000012360 testing method Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M8/00—Fuel cells; Manufacture thereof
- H01M8/04—Auxiliary arrangements, e.g. for control of pressure or for circulation of fluids
- H01M8/04298—Processes for controlling fuel cells or fuel cell systems
- H01M8/04305—Modeling, demonstration models of fuel cells, e.g. for training purposes
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M8/00—Fuel cells; Manufacture thereof
- H01M8/04—Auxiliary arrangements, e.g. for control of pressure or for circulation of fluids
- H01M8/04298—Processes for controlling fuel cells or fuel cell systems
- H01M8/04992—Processes for controlling fuel cells or fuel cell systems characterised by the implementation of mathematical or computational algorithms, e.g. feedback control loops, fuzzy logic, neural networks or artificial intelligence
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M2250/00—Fuel cells for particular applications; Specific features of fuel cell system
- H01M2250/20—Fuel cells in motive systems, e.g. vehicle, ship, plane
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/30—Hydrogen technology
- Y02E60/50—Fuel cells
Landscapes
- Engineering & Computer Science (AREA)
- General Chemical & Material Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Fuel Cell (AREA)
- Automation & Control Theory (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
Abstract
The invention provides a method for analyzing output performance of a hydrogen fuel cell pile based on big data, which comprehensively considers influences of real-time operation data of various hydrogen fuel cell piles and whole vehicles on pile performance and attenuation degree, and respectively establishes a unitary linear model reflecting pile attenuation and a multi-element linear prediction model reflecting current performance, thereby realizing prediction on residual life and current performance of the pile adopting similar pile vehicle types according to the operation big data acquired by a real vehicle. The method effectively overcomes the defects that the characteristic index is single and does not accord with the actual working condition in the existing pile performance analysis mode, and the accuracy and the applicability of the method are obviously improved by utilizing a plurality of parameter indexes which are mutually coupled and mutually influenced.
Description
Technical Field
The invention belongs to the technical field of fuel cell operation performance analysis, and particularly relates to a hydrogen fuel cell stack output performance analysis method based on big data.
Background
In the prior art, the performance of a cell stack for a fuel cell mainly utilizes the voltage parameter of the cell stack to continuously reflect the overall performance level of the fuel cell, and realizes the performance test of the cell stack by means of a bench test. However, since the fuel cell stack voltage has frequent recovery and random fluctuation under the dynamic working condition, the fuel cell stack voltage has strong nonlinearity and uncertainty, so that the accuracy of a test result obtained by only considering the voltage parameter cannot meet the requirement. Meanwhile, the working conditions of the test bed and the actual use scene of the galvanic pile on the vehicle are greatly different, and the galvanic pile performance under the actual use condition cannot be comprehensively measured. Under the actual operation condition, the pile performance is obviously affected by parameters such as pile temperature, pressure, hydrogen flow, air flow and the like of the fuel cell. In addition, the fuel cell system is often in a dynamic running condition, and the working states such as frequent start-stop, idling, load change and high load have great influence on the service life of the fuel cell. These control variables are coupled and affect each other, which makes the prior art have great difficulty in evaluating the stack performance and obtaining the relevant parameters of the optimal output performance, so there is a strong need in the art for a more scientific and efficient hydrogen fuel cell stack performance calculation and analysis method that can accurately calculate the current performance level in real time and simultaneously analyze the factors affecting the fuel cell stack performance.
Disclosure of Invention
In view of the above, the present invention provides a method for analyzing output performance of a hydrogen fuel cell stack based on big data, which specifically includes the following steps:
firstly, a vehicle-mounted terminal of a hydrogen fuel vehicle collects hydrogen fuel pipeline parameters, air pipeline parameters, stack power output parameters, stack start-stop state and vehicle running state parameters in real time when a hydrogen fuel cell stack runs in a full life cycle after leaving a factory, and uploads the running parameters to a cloud platform;
step two, after the cloud platform obtains the operation parameters uploaded by the vehicle-mounted terminal, defining 50% of rated current of a certain target pile as a steady-state current value, and obtaining a steady-state voltage value corresponding to the steady-state current value;
setting a calculation period with a certain length, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence; smoothing and filtering steady-state voltage values in the sequence through a sliding time window; fitting a pile performance attenuation model function by using the voltage data obtained after filtering and applying a unitary linear regression equation, and defining a function slope as a pile performance attenuation rate;
selecting hydrogen fuel pipeline parameters, air pipeline parameters, a stack start-stop state and a vehicle running state parameter as stack performance influence factors and model input, and outputting a stack steady-state voltage value as a model output to construct a hydrogen fuel cell stack performance prediction model based on multiple linear regression; and training the model by using hydrogen fuel pipeline parameters, air pipeline parameters, a starting and stopping state of the electric pile and vehicle running state parameters of the same model of the same electric pile under a running condition in a steady state to obtain the weight of each performance influence factor of the same electric pile.
And inputting hydrogen fuel pipeline parameters, air pipeline parameters, electric pile power output parameters, electric pile start-stop state and vehicle running state parameters acquired by the vehicle-mounted terminal in real time into the electric pile performance attenuation model and the electric pile performance prediction model, so that corresponding electric pile current performance attenuation degree and current prediction performance can be obtained respectively.
Further, the hydrogen fuel pipeline parameters collected in the first step specifically include: the in-stack hydrogen pressure InHP, the out-stack hydrogen pressure OutHP and the hydrogen temperature HT; the air pipeline parameters specifically comprise: air humidity AH, in-stack air pressure InAT, out-stack air pressure OutAT, in-stack air temperature InAT, out-stack air temperature OutAT, air flow AF and air excess coefficient EAC; the pile power output parameters specifically include: pile output voltage U and pile output current I; the vehicle running state parameters specifically include: message time t, vehicle speed v, accelerator pedal value AP and brake pedal value BP; after the cloud platform receives the parameters, the front end processor writes data into Kafka, the offline data is written into HDFS by adopting the Flink, and real-time processing is carried out on the Kafka data by adopting the Storm, so that the real-time dynamic data of the hydrogen fuel vehicle is written into Redis, and the latest dynamic data and static data of the vehicle are written into an elastic search cluster.
Further, the specific process of obtaining the steady-state voltage value in the second step includes:
firstly, setting an output current value I of a target electric pile in a range of 50% +/-1A of rated current o The method comprises the steps of carrying out a first treatment on the surface of the Cloud platform utilizationThe real-time calculation engine judges the current value I in each frame of data t : if I t Not equal to I o Sieving; if I t Equal to I o Retrieving the pile output voltage U corresponding to the frame data t And records the voltage value as a steady-state voltage value at time t.
Further, the third step specifically includes the following steps:
(1) Setting a calculation period of 1 week, 1 month or 1 year, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence;
(2) Setting the corresponding frame number of the sliding window, sliding the time-steady voltage value sequence according to the time sequence, and setting the steady voltage value U in the sliding window t Smoothing and filtering to obtain a filtered voltage valueWherein time t=1, 2, … …, m, m is the total number of frames of the steady-state voltage value;
(3) With filtered voltage value U' t As a dependent variable time t as an independent variable, the following unitary linear regression equation is constructed:
f(t)＝W·U′ t +B
wherein f (t) is a fitted galvanic pile performance attenuation model function; w is the function slope, namely the pile performance attenuation rate; and B is a bias value of the unitary linear regression equation, and a steady-state voltage value of the galvanic pile when leaving the factory is taken.
Further, the fourth step specifically includes the following steps:
(1) Specifically selecting parameters of a hydrogen fuel pipeline at the time t: in-stack hydrogen pressure InHP t Out-stack hydrogen pressure OutHP t Hydrogen temperature HT t Air line parameters: air humidity AH t In-pile air pressure InAP t Out-pile air-pressure OutAP t In-stack air temperature InAT t OutAT (out-of-stack air temperature) t Air flow rate AF t Air excess coefficient EAC t Stack start-stop state SU t And the time length delta t and the speed v of the galvanic pile t These parameters are entered as a model and set to Xt:
X t ＝(InHP t ，OutHP t ，HT t ，AH t ，InAP t ，OutAP t ，InAT t ，OutAT t ，，EAC t ，SU t ，Δt，v t ) Output Y by taking steady-state voltage value output by pile as model t ，Y t ＝U t The method comprises the steps of carrying out a first treatment on the surface of the The stack use duration deltat is defined as the current time minus the defined initial time;
(2) Grabbing Xt and Yt within a certain time range by an HDFS database to form feature data vectors [ X, Y ], wherein,
constructing a hydrogen fuel cell stack performance prediction model based on multiple linear regression: y=w×x+b, where vector w= [ W InHP W OutHP … W v ]The weight matrix is used for representing the weight of each parameter affecting the pile performance; matrix B is the bias value of the multiple linear regression equation
(3) Firstly, after the vector W and the matrix B are initialized randomly, the final stable weight of each parameter is obtained step by step through the iterative training of the model by the vector [ X, Y ].
If the weight value corresponding to a certain parameter in the trained model is positive, the positive influence of the weight value on the dynamic galvanic pile performance is illustrated; and if the weight value corresponding to the parameter is negative, the negative influence on the performance of the dynamic galvanic pile is indicated. The absolute value of the weight value represents the magnitude of the degree of influence. And establishing a causal chain between the cell stack performance of the bicycle and different parameters through a trained hydrogen fuel cell stack performance prediction model.
According to the analysis method for the output performance of the hydrogen fuel cell stack based on the big data, provided by the invention, the influences of real-time operation data of various hydrogen fuel vehicle stacks and the whole vehicle on the performance and attenuation degree of the stacks are comprehensively considered, and a unitary linear model reflecting the attenuation of the stacks and a multi-element linear prediction model reflecting the current performance are respectively established, so that the prediction of the residual life and the current performance of the stacks adopting similar stack vehicle types can be realized according to the operation big data acquired by a real vehicle. The method effectively overcomes the defects that the characteristic index is single and does not accord with the actual working condition in the existing pile performance analysis mode, and the accuracy and the applicability of the method are obviously improved by utilizing a plurality of parameter indexes which are mutually coupled and mutually influenced.
Drawings
FIG. 1 is a flow chart of the method according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention provides a hydrogen fuel cell stack output performance analysis method based on big data, as shown in figure 1, specifically comprising the following steps:
firstly, a vehicle-mounted terminal of a hydrogen fuel vehicle collects hydrogen fuel pipeline parameters, air pipeline parameters, stack power output parameters, stack start-stop state and vehicle running state parameters in real time when a hydrogen fuel cell stack runs in a full life cycle after leaving a factory, and uploads the running parameters to a cloud platform;
step two, after the cloud platform obtains the operation parameters uploaded by the vehicle-mounted terminal, defining 50% of rated current of a certain target pile as a steady-state current value, and obtaining a steady-state voltage value corresponding to the steady-state current value;
setting a calculation period with a certain length, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence; smoothing and filtering steady-state voltage values in the sequence through a sliding time window; fitting a pile performance attenuation model function by using the voltage data obtained after filtering and applying a unitary linear regression equation, and defining a function slope as a pile performance attenuation rate;
selecting hydrogen fuel pipeline parameters, air pipeline parameters, a stack start-stop state and a vehicle running state parameter as stack performance influence factors and model input, and outputting a stack steady-state voltage value as a model output to construct a hydrogen fuel cell stack performance prediction model based on multiple linear regression; and training the model by using hydrogen fuel pipeline parameters, air pipeline parameters, a starting and stopping state of the electric pile and vehicle running state parameters of the same model of the same electric pile under a running condition in a steady state to obtain the weight of each performance influence factor of the same electric pile.
And inputting hydrogen fuel pipeline parameters, air pipeline parameters, electric pile power output parameters, electric pile start-stop state and vehicle running state parameters acquired by the vehicle-mounted terminal in real time into the electric pile performance attenuation model and the electric pile performance prediction model, so that corresponding electric pile current performance attenuation degree and current prediction performance can be obtained respectively.
In a preferred embodiment of the present invention, the hydrogen fuel line parameters collected in step one specifically include: the in-stack hydrogen pressure InHP, the out-stack hydrogen pressure OutHP and the hydrogen temperature HT; the air pipeline parameters specifically comprise: air humidity AH, in-stack air pressure InAP, out-stack air pressure OutAP, in-stack air temperature InAAT, out-stack air temperature OutAT, air flow AF and air excess coefficient EAC; the pile power output parameters specifically include: pile output voltage U and pile output current I; the vehicle running state parameters specifically include: message time t, vehicle speed v, accelerator pedal value AP and brake pedal value BP; after the cloud platform receives the parameters, the front end processor writes data into Kafka, the offline data is written into HDFS by adopting the Flink, and real-time processing is carried out on the Kafka data by adopting the Storm, so that the real-time dynamic data of the hydrogen fuel vehicle is written into Redis, and the latest dynamic data and static data of the vehicle are written into an elastic search cluster.
In a preferred embodiment of the present invention, the specific process of obtaining the steady-state voltage value in the second step includes:
firstly, setting an output current value I of a target electric pile in a range of 50% +/-1A of rated current o The method comprises the steps of carrying out a first treatment on the surface of the The cloud platform judges a current value I in each frame of data by using a real-time calculation engine t : if I t Not equal to I o Sieving; if I t Equal to I o Retrieving the pile output voltage U corresponding to the frame data t And records the voltage value as a steady-state voltage value at time t.
In a preferred embodiment of the present invention, the third step specifically comprises the steps of:
(1) Setting a calculation period of 1 week, 1 month or 1 year, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence;
(2) Setting the corresponding frame number of the sliding window, sliding the time-steady voltage value sequence according to the time sequence, and setting the steady voltage value U in the sliding window t Smoothing and filtering to obtain a filtered voltage valueWherein time t=1, 2, … …, m, m is the total number of frames of the steady-state voltage value;
(3) With filtered voltage value U' t As a dependent variable time t as an independent variable, the following unitary linear regression equation is constructed:
f(t)＝W·U′ t +B
wherein f (t) is a fitted galvanic pile performance attenuation model function; w is the function slope, namely the pile performance attenuation rate; and B is a bias value of the unitary linear regression equation, and a steady-state voltage value of the galvanic pile when leaving the factory is taken.
In a preferred embodiment of the present invention, the fourth step specifically comprises the steps of:
(1) Specifically selecting parameters of a hydrogen fuel pipeline at the time t: in-stack hydrogen pressure InHP t Out-stack hydrogen pressure OutHP t Hydrogen temperature HT t Air line parameters: air humidity AH t In-pile air pressure InAP t Go out of the stackPress OutAP t In-stack air temperature InAT t OutAT (out-of-stack air temperature) t Air flow rate AF t Air excess coefficient EAC t Stack start-stop state SU t And the time length delta t and the speed v of the galvanic pile t These parameters are entered as a model and set to Xt:
X t ＝(InHP t ，OutHP t ，HT t ，AH t ，InAP t ，OutAP t ，InAT t ，OutAT t ，，EAC t ，SU t ，Δt，v t ) Taking a steady-state voltage value output by a galvanic pile as a model output Yt, tt=ut; the stack use duration deltat is defined as the current time minus the defined initial time;
(2) Grabbing Xt and Yt within a certain time range by an HDFS database to form feature data vectors [ X, Y ], wherein,
constructing a hydrogen fuel cell stack performance prediction model based on multiple linear regression: y=w×x+b, where vector w= [ W InHP W OutHP … W v ]The weight matrix is used for representing the weight of each parameter affecting the pile performance; matrix B is the bias value of the multiple linear regression equation
(3) Firstly, after the vector W and the matrix B are initialized randomly, the final stable weight of each parameter is obtained step by step through the iterative training of the model by the vector [ X, Y ].
In an implementation of the invention, the performance index of the model may be verified by the mean square error MSE, which is calculated as follows:wherein Y is i For a true steady state voltage value, +.>A voltage value predicted for the multiple linear model; n is the sample number value.
It should be understood that, the sequence number of each step in the embodiment of the present invention does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present invention.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. The hydrogen fuel cell stack output performance analysis method based on big data is characterized in that: the method specifically comprises the following steps:
firstly, a vehicle-mounted terminal of a hydrogen fuel vehicle collects hydrogen fuel pipeline parameters, air pipeline parameters, stack power output parameters, stack start-stop state and vehicle running state parameters in real time when a hydrogen fuel cell stack runs in a full life cycle after leaving a factory, and uploads the running parameters to a cloud platform;
step two, after the cloud platform obtains the operation parameters uploaded by the vehicle-mounted terminal, defining 50% of rated current of a certain target pile as a steady-state current value, and obtaining a steady-state voltage value corresponding to the steady-state current value;
setting a calculation period with a certain length, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence; smoothing and filtering steady-state voltage values in the sequence through a sliding time window; fitting a pile performance attenuation model function by using the voltage data obtained after filtering and applying a unitary linear regression equation, and defining a function slope as a pile performance attenuation rate;
selecting hydrogen fuel pipeline parameters, air pipeline parameters, a stack start-stop state and a vehicle running state parameter as stack performance influence factors and model input, and outputting a stack steady-state voltage value as a model output to construct a hydrogen fuel cell stack performance prediction model based on multiple linear regression; training the model by using hydrogen fuel pipeline parameters, air pipeline parameters, a starting and stopping state of the electric pile and vehicle running state parameters of the same model of the same electric pile under a running condition in a steady state to obtain the weight of each performance influence factor of the same electric pile;
and inputting hydrogen fuel pipeline parameters, air pipeline parameters, electric pile power output parameters, electric pile start-stop state and vehicle running state parameters acquired by the vehicle-mounted terminal in real time into the electric pile performance attenuation model and the electric pile performance prediction model, so that corresponding electric pile current performance attenuation degree and current prediction performance can be obtained respectively.
2. The method of claim 1, wherein: the hydrogen fuel pipeline parameters collected in the first step specifically comprise: the in-stack hydrogen pressure InHP, the out-stack hydrogen pressure OutHP and the hydrogen temperature HT; the air pipeline parameters specifically comprise: air humidity AH, in-stack air pressure InAP, out-stack air pressure OutAP, in-stack air temperature InAT, out-stack air temperature OutAT, air flow AF and air excess coefficient EAC; the pile power output parameters specifically include: pile output voltage U and pile output current I; the vehicle running state parameters specifically include: message time t, vehicle speed v, accelerator pedal value AP and brake pedal value BP; after the cloud platform receives the parameters, the front end processor writes data into Kafka, the offline data is written into HDFS by adopting the Flink, and real-time processing is carried out on the Kafka data by adopting the Storm, so that the real-time dynamic data of the hydrogen fuel vehicle is written into Redis, and the latest dynamic data and static data of the vehicle are written into an elastic search cluster.
3. The method of claim 2, wherein: the specific process for obtaining the steady-state voltage value in the second step comprises the following steps:
firstly, setting an output current value I of a target electric pile in a range of 50% +/-lA of rated current o The method comprises the steps of carrying out a first treatment on the surface of the Cloud platform utilizing real-time computing engineJudging the current value I in each frame of data t : if I t Not equal to I o Sieving; if I t Equal to I o Retrieving the pile output voltage U corresponding to the frame data t And records the voltage value as a steady-state voltage value at time t.
4. A method as claimed in claim 3, wherein: the third step specifically comprises the following steps:
(1) Setting a calculation period of 1 week, 1 month or 1 year, collecting all steady-state voltage values and corresponding time in the target pile operation parameters in the calculation period, and establishing a time-steady-state voltage value sequence;
(2) Setting the corresponding frame number of the sliding window, sliding the time-steady voltage value sequence according to the time sequence, and setting the steady voltage value U in the sliding window t Smoothing and filtering to obtain a filtered voltage valueWhere time t=1, 2, &.. m is the total frame number of the steady-state voltage value;
(3) With filtered voltage value U' t As a dependent variable time t as an independent variable, the following unitary linear regression equation is constructed:
f(t)＝W·U′ t +B
wherein f (t) is a fitted galvanic pile performance attenuation model function; w is the function slope, namely the pile performance attenuation rate; and B is a bias value of the unitary linear regression equation, and a steady-state voltage value of the galvanic pile when leaving the factory is taken.
5. The method of claim 4, wherein: the fourth step comprises the following steps:
(1) Specifically selecting parameters of a hydrogen fuel pipeline at the time t: in-stack hydrogen pressure InHP t Out-stack hydrogen pressure OutHP t Hydrogen temperature HT t Air line parameters: air humidity AH t In-pile air pressure InAP t Out-pile air-pressure OutAP t In-stack air temperature InAT t Temperature of the empty reactorOutAT t Air flow rate AF t Air excess coefficient EAC t Stack start-stop state SU t And the time length delta t and the speed v of the galvanic pile t These parameters are entered as a model and set to Xt:
X t ＝(InHP t ，OutHP t ，HT t ，AH t ，InAP t ，OutAP t ，InAT t ，OutAT t ，，EAC t ，SU t ，Δt，v t ) Output Y by taking steady-state voltage value output by pile as model t ，Y t ＝U t The method comprises the steps of carrying out a first treatment on the surface of the The stack use duration deltat is defined as the current time minus the defined initial time;
(2) Grabbing Xt and Yt within a certain time range by an HDFS database to form feature data vectors [ X, Y ], wherein,
constructing a hydrogen fuel cell stack performance prediction model based on multiple linear regression: y=w×x+b, where vector w= [ W InHP W OutHP ...W v ]The weight matrix is used for representing the weight of each parameter affecting the pile performance; matrix B is the bias value of the multiple linear regression equation
(3) Firstly, after the vector W and the matrix B are initialized randomly, the final stable weight of each parameter is obtained step by step through the iterative training of the model by the vector [ X, Y ].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885423.0A CN115084600B (en) | 2022-07-26 | 2022-07-26 | Hydrogen fuel cell stack output performance analysis method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885423.0A CN115084600B (en) | 2022-07-26 | 2022-07-26 | Hydrogen fuel cell stack output performance analysis method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115084600A CN115084600A (en) | 2022-09-20 |
| CN115084600B true CN115084600B (en) | 2023-11-24 |
Family
ID=83242818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210885423.0A Active CN115084600B (en) | 2022-07-26 | 2022-07-26 | Hydrogen fuel cell stack output performance analysis method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115084600B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115312895B (en) * | 2022-10-12 | 2022-12-20 | 启东市航新实用技术研究所 | Method for monitoring steady state of battery pack of new energy vehicle |
| CN117590242B (en) * | 2024-01-18 | 2024-04-26 | 未势能源科技有限公司 | Method and device for detecting consistency of galvanic pile, storage medium and electronic device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069741A (en) * | 2020-09-25 | 2020-12-11 | 国网四川省电力公司经济技术研究院 | Fuel cell stack transient aging prediction method, device, equipment and storage medium |
| CN112563542A (en) * | 2020-12-08 | 2021-03-26 | 上海重塑能源科技有限公司 | Fuel cell online detection method and detection system |
| CN113314738A (en) * | 2021-04-30 | 2021-08-27 | 金龙联合汽车工业（苏州）有限公司 | Method for evaluating running health state of hydrogen fuel cell engine system |
| CN114566686A (en) * | 2020-11-27 | 2022-05-31 | 中国科学院大连化学物理研究所 | Method for estimating state and predicting service life of fuel cell |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9806363B2 (en) * | 2013-02-21 | 2017-10-31 | Korea Insitute Of Energy Research | Apparatus and method for softsensing fuel cell system |
-
2022
- 2022-07-26 CN CN202210885423.0A patent/CN115084600B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069741A (en) * | 2020-09-25 | 2020-12-11 | 国网四川省电力公司经济技术研究院 | Fuel cell stack transient aging prediction method, device, equipment and storage medium |
| CN114566686A (en) * | 2020-11-27 | 2022-05-31 | 中国科学院大连化学物理研究所 | Method for estimating state and predicting service life of fuel cell |
| CN112563542A (en) * | 2020-12-08 | 2021-03-26 | 上海重塑能源科技有限公司 | Fuel cell online detection method and detection system |
| CN113314738A (en) * | 2021-04-30 | 2021-08-27 | 金龙联合汽车工业（苏州）有限公司 | Method for evaluating running health state of hydrogen fuel cell engine system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115084600A (en) | 2022-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115084600B (en) | Hydrogen fuel cell stack output performance analysis method based on big data | |
| WO2021185308A1 (en) | Online determination method and system for state of health of power battery pack of electric vehicle | |
| CN111707951B (en) | Battery pack consistency evaluation method and system | |
| CN111999657B (en) | Method for evaluating driving mileage of lithium ion battery of electric vehicle in residual life | |
| CN111007399B (en) | Lithium battery state of charge prediction method based on improved generation countermeasure network | |
| CN109001640B (en) | Data processing method and device for power battery | |
| CN111381170A (en) | Electric vehicle battery pack health state prediction method and system based on big data | |
| CN112858919A (en) | Battery system online fault diagnosis method and system based on cluster analysis | |
| CN111301426A (en) | Method for predicting energy consumption in future driving process based on GRU network model | |
| CN114290960B (en) | Method and device for acquiring battery health of power battery and vehicle | |
| CN107783057B (en) | Power battery SOC estimation method and estimation system | |
| CN111191824B (en) | Power battery capacity attenuation prediction method and system | |
| CN110850298A (en) | Lithium battery SOH estimation method and system based on data driving | |
| CN113406524B (en) | Inconsistent fault diagnosis method and system for power battery system | |
| CN114818831B (en) | Bidirectional lithium ion battery fault detection method and system based on multi-source perception | |
| CN114662954A (en) | Vehicle performance evaluation system | |
| CN113657651A (en) | Diesel vehicle emission prediction method, medium and equipment based on deep migration learning | |
| Park et al. | Capacity estimation of Li-ion batteries using constant current charging voltage with multilayer perceptron | |
| CN112782594A (en) | Method for estimating SOC (state of charge) of lithium battery by data-driven algorithm considering internal resistance | |
| CN115091963A (en) | Hydrogen fuel cell automobile energy consumption estimation method based on energy consumption characteristics and working condition segmentation | |
| CN109782182B (en) | Online estimation method and device for energy state of series battery pack | |
| CN113837473A (en) | Charging equipment fault rate analysis system and method based on BP neural network | |
| CN113505932A (en) | Power battery capacity algorithm based on big data technology evaluation | |
| CN115481796A (en) | Method for predicting remaining service life of battery based on Bayesian hybrid neural network | |
| CN115097344A (en) | Battery health state terminal cloud collaborative estimation method based on constant voltage charging segments |
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 |