CN115291110B - Pile aging prediction method based on feature parameter extraction and aging experience library construction - Google Patents
Pile aging prediction method based on feature parameter extraction and aging experience library construction Download PDFInfo
- Publication number
- CN115291110B CN115291110B CN202210883611.XA CN202210883611A CN115291110B CN 115291110 B CN115291110 B CN 115291110B CN 202210883611 A CN202210883611 A CN 202210883611A CN 115291110 B CN115291110 B CN 115291110B
- Authority
- CN
- China
- Prior art keywords
- pile
- value
- parameters
- vehicle
- aging
- 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
- 230000032683 aging Effects 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000000605 extraction Methods 0.000 title claims abstract description 11
- 238000010276 construction Methods 0.000 title abstract description 6
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 claims description 58
- 229910052739 hydrogen Inorganic materials 0.000 claims description 58
- 239000001257 hydrogen Substances 0.000 claims description 58
- 239000000446 fuel Substances 0.000 claims description 50
- 239000000463 material Substances 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000000630 rising effect Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 230000036541 health Effects 0.000 abstract description 6
- 230000007547 defect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 3
- BVKZGUZCCUSVTD-UHFFFAOYSA-L Carbonate Chemical compound [O-]C([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-L 0.000 description 2
- NBIIXXVUZAFLBC-UHFFFAOYSA-L Phosphate ion(2-) Chemical compound OP([O-])([O-])=O NBIIXXVUZAFLBC-UHFFFAOYSA-L 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 239000012528 membrane Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000032677 cell aging Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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]
- G01R31/367—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]
- G01R31/378—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]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Fuel Cell (AREA)
Abstract
The invention provides a pile aging prediction method based on characteristic parameter extraction and aging experience library construction. The method fully considers various health characteristics affecting the performance of the galvanic pile, thereby ensuring the wide applicability of the aging experience library to different vehicle types and galvanic piles and effectively overcoming a plurality of defects in the prior art.
Description
Technical Field
The invention belongs to the technical field of fuel cell aging and service life detection, and particularly relates to a pile aging prediction method constructed based on characteristic parameter extraction and aging experience library.
Background
For fuel cell stacks, it is necessary to maintain and repair the system in time by effective life prediction means and to ensure the durability of the stack for long-term use. However, in a vehicular environment, the aging process of the fuel cell stack is relatively quick, and the aging causes are complicated, so that the aging trend and the aging period of each vehicle are not very same. In the prior art, an aging prediction model for a fuel cell stack is mainly divided into a mechanism aging model, an empirical aging model and a data driving model, and various models have some defects which are difficult to overcome. For example, the mechanism aging model is extremely difficult in terms of parameter identification; the accuracy of the empirical aging model is not high and the modeling of the aging behavior under the dynamic circulation working condition is not applicable; most of the data driving models are based on single pile modeling, and although the aging state in a shorter time in the future can be predicted, the prediction effect on a longer period is poor. Moreover, these several models are also difficult to use in combination with each other, such as single pile modeling is difficult to record the aging behavior of the pile as an empirical library, and the application of aging modeling knowledge to the aging predictions of other piles is violated. Another difficulty with fuel cell stack aging prediction is that how to construct a characteristic parameter with discrimination is also an important technical problem in the art, since the state of health of the stack cannot be measured directly and needs to be mapped to a health factor by means of the relevant characteristic parameter.
Disclosure of Invention
In view of the above, the present invention provides a method for predicting aging of a galvanic pile based on feature parameter extraction and aging experience library construction, which specifically includes the following steps:
Firstly, a vehicle-mounted terminal of a hydrogen fuel vehicle collects hydrogen fuel pipeline parameters, air pipeline parameters, electric pile power output parameters, electric pile start-stop states and vehicle running state parameters in real time when a fuel cell electric pile 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;
Step three, setting a specific time interval of windows, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval; creating a time sequence of voltage values corresponding to time for each window, and calculating to obtain the aging rate of the galvanic pile in each window according to the initial steady-state voltage value and the terminal steady-state voltage value of the window;
Step four, extracting statistical characteristics including mean value, maximum value, minimum value, variance, peak value number, square sum, information entropy value, over-mean value ratio and the like from the hydrogen fuel pipeline parameters and the air pipeline parameters aiming at each window to serve as operating condition characteristics of the electric pile; extracting the whole vehicle operation condition characteristics related to the loading amplitude, start-stop and idle speed of the vehicle in each window;
Based on the pile operation condition characteristics and the whole vehicle operation condition characteristics extracted from each window, and combining the pile aging rate in each window obtained by the calculation in the step three, constructing an aging experience library of the target pile, and searching the attenuation rate of the corresponding similar pile according to the same pile and the vehicle condition index, thereby realizing the prediction of the residual life of the similar pile.
Further, the hydrogen fuel pipeline parameters collected in the first step specifically include: an in-stack hydrogen pressure (InHP), an out-stack hydrogen pressure (OutHP), and a 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 factor (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 o of a target pile in a range of 50% +/-1A of rated current; the cloud platform judges the current value I t in each frame of data by using a real-time calculation engine: if I t is not equal to I o, screening; if I t is equal to I o, the stack output voltage U t corresponding to the frame data is retrieved, and the voltage value is recorded as a steady-state voltage value at time t.
Further, the third step specifically includes the following steps:
(1) Setting the time interval corresponding to the window as 100 hours, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval;
(2) The first frame of steady-state voltage data V start and the last frame of steady-state voltage data V end are respectively collected for each window, and the aging rate k of the galvanic pile is calculated according to the following formula:
Where Δt is the window time length, and Δt=100deg.H.
Further, in the fourth step, the hydrogen fuel pipeline parameter and the air pipeline parameter are specifically set as follows: an in-stack hydrogen pressure (InHP), an out-stack hydrogen pressure (OutHP), and a 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 following statistical features are extracted respectively:
(1) Average value:
(2) Maximum value:
xmax＝max(xi)i＝1，2，...n
(3) Minimum value:
xmin＝min(xi)i＝1，2，...n
(4) Variance:
wherein x is an original time sequence parameter, n is the frame number of the original time sequence parameter, and a subscript i represents an ith frame;
(5) Number of peaks: for consecutive 3 frames of data of [ x i-1,xi,xi+1 ]; if k i-xi+1 is more than 0 and x i-xi-1 is more than 0, considering x i as a peak value in a small range of three continuous frames, and calculating the number of all peak values in the window;
(6) And (3) square sum:
(7) Information entropy value:
Dividing a probability interval u according to the principle of equidistant segmentation:
a and l are decision parameters of the interval, and default parameters are set as follows: l=10, a e [0,9], the information entropy is calculated as follows:
Wherein p i is the probability of falling within the probability interval i, which is equal to the sum of the number of data samples of the probability interval divided by the number of window samples;
(8) Exceeding the average ratio:
Wherein g i is whether the i frame data exceeds the average value, and the calculation formula is as follows:
further, the whole vehicle operation condition features extracted in the fourth step specifically include:
(1) Average number of loads per hour for a particular magnitude:
Wherein, l i represents whether the rising amplitude or the falling amplitude of the voltage of the hydrogen fuel cell exceeds a threshold value threshold, and if the rising amplitude or the falling amplitude exceeds the threshold value, a specific amplitude loading event is determined, and the calculation formula is as follows:
(2) Average number of start/stop per hour:
In the formula, s i represents whether the working states of the front and rear hydrogen fuel stacks are consistent, if not, a start-stop event is judged, and the calculation formula is as follows:
(3) Average idle time per hour:
wherein m is the number of effective working samples of the hydrogen fuel cell stack, and m is less than or equal to n; t sample represents a data upload time interval every two frames; h i represents whether the pile is in an idle state, and when the pile current is not 0 and the speed is 0, the state of the vehicle is judged to be idle at the moment, and a calculation formula of h i is as follows:
Further, the fourth step further specifically includes:
Firstly, classifying all hydrogen fuel vehicles in a database, wherein classification indexes comprise pile types and pile rated power; wherein the stack category includes a plurality of material classifications based on cell material classification of alkaline hydrogen fuel cells (AFCs), proton exchange membrane hydrogen fuel cells (PEMFCs), hydrogen phosphate fuel cells (PAFCs), molten carbonate hydrogen fuel cells (MCFCs), solid oxide hydrogen fuel cells (SOFCs), and the like; the rated power of the electric pile can be classified into 10 grades according to the rated power of the electric pile of the main stream in the market at equal intervals;
After classification, the vehicle looks up the stack decay rate based on the aging experience library only at the same material classification and power level.
The specific performance index of the galvanic pile aging experience library can be verified based on the mean square error MSE, and the specific calculation formula is as follows:
Where k i is the real voltage value decay rate, The voltage value attenuation rate obtained by looking up the table is obtained, and m is the sample number value.
According to the pile aging prediction method based on characteristic parameter extraction and aging experience library construction, window segmentation, decay rate calculation, health feature extraction and other steps are sequentially carried out on real-time operation data of a hydrogen fuel vehicle pile and a whole vehicle, an aging experience library is constructed, and quick and accurate residual life prediction can be carried out on vehicles of the same type by utilizing vehicle-mounted big data. The method fully considers various health characteristics affecting the performance of the galvanic pile, thereby ensuring the wide applicability of the aging experience library to different vehicle types and galvanic piles and effectively overcoming a plurality of defects in the prior art.
Drawings
FIG. 1 is a block flow diagram of a method provided by the present invention;
FIG. 2 is a schematic view of an operational parameter cut;
FIG. 3 is a schematic diagram of a process for calculating the aging rate of a galvanic pile;
FIG. 4 is a schematic diagram of the feature extraction of the operating conditions of the electric pile and the whole vehicle;
FIG. 5 is a schematic flow chart for predicting the aging of the galvanic pile based on the 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 pile aging prediction method based on characteristic parameter extraction and aging experience library construction provided by the invention, as shown in figure 1, specifically comprises the following steps:
Firstly, a vehicle-mounted terminal of a hydrogen fuel vehicle collects hydrogen fuel pipeline parameters, air pipeline parameters, electric pile power output parameters, electric pile start-stop states and vehicle running state parameters in real time when a fuel cell electric pile 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;
Step three, setting a specific time interval of windows, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval; creating a time sequence of voltage values corresponding to time for each window, and calculating to obtain the aging rate of the galvanic pile in each window according to the initial steady-state voltage value and the terminal steady-state voltage value of the window;
Step four, extracting statistical characteristics including mean value, maximum value, minimum value, variance, peak value number, square sum, information entropy value, over-mean value ratio and the like from the hydrogen fuel pipeline parameters and the air pipeline parameters aiming at each window to serve as operating condition characteristics of the electric pile; extracting the whole vehicle operation condition characteristics related to the loading amplitude, start-stop and idle speed of the vehicle in each window;
Based on the pile operation condition characteristics and the whole vehicle operation condition characteristics extracted from each window, and combining the pile aging rate in each window obtained by the calculation in the step three, constructing an aging experience library of the target pile, and searching the attenuation rate of the corresponding similar pile according to the same pile and the vehicle condition index, thereby realizing the prediction of the residual life of the similar pile.
In a preferred embodiment of the present invention, the hydrogen fuel line parameters collected in step one specifically include: an in-stack hydrogen pressure (InHP), an out-stack hydrogen pressure (OutHP), and a 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 factor (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 o of a target pile in a range of 50% +/-lA of rated current; the cloud platform judges the current value I t in each frame of data by using a real-time calculation engine: if I t is not equal to I o, screening; if I t is equal to I o, the stack output voltage U t corresponding to the frame data is retrieved, and the voltage value is recorded 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 the time interval corresponding to the window to be 100 hours, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval, as shown in fig. 2;
(2) As shown in fig. 3, the first frame of steady-state voltage data V start and the last frame of steady-state voltage data V end are collected separately for each window, and the stack burn-in rate k is calculated according to the following equation:
Where Δt is the window time length, and Δt=100deg.H.
As shown in fig. 4, in a preferred embodiment of the present invention, in the fourth step, the hydrogen fuel line parameter and the air line parameter are specifically: an in-stack hydrogen pressure (InHP), an out-stack hydrogen pressure (OutHP), and a 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 following statistical features are extracted respectively:
(1) Average value:
(2) Maximum value:
xmax＝max(xi)i＝1，2，...n
(3) Minimum value:
xmin＝min(xi)i＝1，2，...n
(4) Variance:
wherein x is an original time sequence parameter, n is the frame number of the original time sequence parameter, and a subscript i represents an ith frame;
(5) Number of peaks: for consecutive 3 frames of data of [ x i-1,xi,xi+1 ]; if x i-xi+1 is more than 0 and x i-xi-1 is more than 0, considering x i as a peak value in a small range of three continuous frames, and calculating the number of all peak values in the window;
(6) And (3) square sum:
(7) Information entropy value:
Dividing a probability interval u according to the principle of equidistant segmentation:
a and l are decision parameters of the interval, and default parameters are set as follows: l=10, a e [0.9], the information entropy is calculated as follows:
Wherein p i is the probability of falling within the probability interval i, which is equal to the sum of the number of data samples of the probability interval divided by the number of window samples;
(8) Exceeding the average ratio:
Wherein g i is whether the i frame data exceeds the average value, and the calculation formula is as follows:
In a preferred embodiment of the present invention, the whole vehicle operation condition features extracted in the fourth step specifically include:
(1) Average number of loads per hour for a particular magnitude:
Wherein, l i represents whether the rising amplitude or the falling amplitude of the voltage of the hydrogen fuel cell exceeds a threshold value threshold, and if the rising amplitude or the falling amplitude exceeds the threshold value, a specific amplitude loading event is determined, and the calculation formula is as follows:
(2) Average number of start/stop per hour:
In the formula, s i represents whether the working states of the front and rear hydrogen fuel stacks are consistent, if not, a start-stop event is judged, and the calculation formula is as follows:
(3) Average idle time per hour:
wherein m is the number of effective working samples of the hydrogen fuel cell stack, and m is less than or equal to n; t sample represents a data upload time interval every two frames; h i represents whether the pile is in an idle state, and when the pile current is not 0 and the speed is 0, the state of the vehicle is judged to be idle at the moment, and a calculation formula of h i is as follows:
in a preferred embodiment of the present invention, the step four further specifically includes:
Firstly, classifying all hydrogen fuel vehicles in a database, wherein classification indexes comprise pile types and pile rated power; wherein the stack category includes a plurality of material classifications based on cell material classification of alkaline hydrogen fuel cells (AFCs), proton exchange membrane hydrogen fuel cells (PEMFCs), hydrogen phosphate fuel cells (PAFCs), molten carbonate hydrogen fuel cells (MCFCs), solid oxide hydrogen fuel cells (SOFCs), and the like; the rated power of the electric pile can be classified into 10 grades according to the rated power of the electric pile of the main stream in the market at equal intervals; the specific performance index of the galvanic pile aging experience library can be verified based on the mean square error MSE, and the specific calculation formula is as follows:
Where k i is the real voltage value decay rate, The voltage value attenuation rate obtained by looking up the table is obtained, and m is the sample number value.
After classification, the vehicle looks up the stack decay rate based on the aging experience library only at the same material classification and power level. As shown in fig. 5, for a certain type of hydrogen fuel cell vehicle, window division is performed on the stack data, and for each stack operation segment, the operating condition characteristic extraction is performed. Firstly, extracting time sequence characteristics of operation conditions of a galvanic pile, extracting 8 types of characteristics from 10 data fields, and obtaining 80 characteristic factors in total. And extracting the time sequence characteristics of the whole vehicle running condition, wherein the total time sequence characteristics are 3 characteristic factors. In total, 83 characteristic factors are obtained to describe characteristic parameters which can be mapped to health factors, and meanwhile, the pile aging rate k of each segment is calculated according to the operation of the step 3. And establishing a stack aging experience library under the conditions of time division, working condition division and stack division of the full life cycle. The experience inventory has equal aging rates (namely aging coefficients) of the galvanic piles under the same model, same using time and same running conditions in a one-to-one correspondence, and the aging state of the new galvanic pile in the future period can be predicted through table lookup.
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 (7)
1. A pile aging prediction method constructed based on characteristic parameter extraction and aging experience library is characterized by comprising the following steps: 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, electric pile power output parameters, electric pile start-stop states and vehicle running state parameters in real time when a fuel cell electric pile 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;
Step three, setting a specific time interval of windows, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval; creating a time sequence of voltage values corresponding to time for each window, and calculating to obtain the aging rate of the galvanic pile in each window according to the initial steady-state voltage value and the terminal steady-state voltage value of the window;
Step four, extracting statistical characteristics including mean value, maximum value, minimum value, variance, peak value number, square sum, information entropy value and exceeding mean value ratio from the hydrogen fuel pipeline parameters and the air pipeline parameters aiming at each window to serve as operating condition characteristics of a galvanic pile; extracting the whole vehicle operation condition characteristics related to the loading amplitude, start-stop and idle speed of the vehicle in each window;
Based on the pile operation condition characteristics and the whole vehicle operation condition characteristics extracted from each window, and combining the pile aging rate in each window obtained by the calculation in the step three, constructing an aging experience library of the target pile, and searching the attenuation rate of the corresponding similar pile according to the same pile and the vehicle condition index, thereby realizing the prediction of the residual life of the similar pile.
2. The method of claim 1, wherein: the hydrogen fuel pipeline parameters collected in the first step specifically comprise: stacking hydrogen pressure, stacking hydrogen pressure and hydrogen temperature; the air pipeline parameters specifically comprise: air humidity, in-stack air pressure, out-stack air pressure, in-stack air temperature, out-stack air temperature, air flow and air excess coefficient; the pile power output parameters specifically include: pile output voltage and pile output current; the vehicle running state parameters specifically include: message time, vehicle speed, accelerator pedal value, and brake pedal value; 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 1, 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 o of a target pile in a range of 50% +/-1A of rated current; the cloud platform judges the current value I t in each frame of data by using a real-time calculation engine: if I t is not equal to I o, screening; if I t is equal to I o, the stack output voltage U t corresponding to the frame data is retrieved, and the voltage value is recorded as a steady-state voltage value at time t.
4. The method of claim 1, wherein: the third step specifically comprises the following steps:
(1) Setting the time interval corresponding to the window as 100 hours, and dividing the operation parameters of the whole life cycle of the target galvanic pile into a plurality of windows according to the time interval;
(2) The first frame of steady-state voltage data V start and the last frame of steady-state voltage data V end are respectively collected for each window, and the aging rate k of the galvanic pile is calculated according to the following formula:
Where Δt is the window time length, and Δt=100deg.H.
5. The method of claim 1, wherein: in the fourth step, specific to the hydrogen fuel pipeline parameter and the air pipeline parameter: stacking hydrogen pressure, stacking hydrogen pressure and hydrogen temperature; the air pipeline parameters specifically comprise: air humidity, in-stack air pressure, out-stack air pressure, in-stack air temperature, out-stack air temperature, air flow and air excess coefficient are respectively extracted to obtain the following statistical characteristics:
(1) Average value:
(2) Maximum value:
xmax＝max(xi)i＝1,2,...n
(3) Minimum value:
xmin＝min(xi)i＝1，2,...n
(4) Variance:
wherein x is an original time sequence parameter, n is the frame number of the original time sequence parameter, and a subscript i represents an ith frame;
(5) Number of peaks: for consecutive 3 frames of data of [ x i-1,xi,xi+1 ]; if x i-xi+1 is more than 0 and x i-xi-1 is more than 0, considering x i as a peak value in a small range of three continuous frames, and calculating the number of all peak values in the window;
(6) And (3) square sum:
(7) Information entropy value:
Dividing a probability interval u according to the principle of equidistant segmentation:
a and l are decision parameters of the interval, and default parameters are set as follows: l=10, a e [0,9], the information entropy is calculated as follows:
Wherein p i is the probability of falling within the probability interval i, which is equal to the sum of the number of data samples of the probability interval divided by the number of window samples;
(8) Exceeding the average ratio:
Wherein g i is whether the i frame data exceeds the average value, and the calculation formula is as follows:
6. The method of claim 5, wherein: the whole vehicle operation condition features extracted in the step four specifically comprise:
(1) Average number of loads per hour for a particular magnitude:
Wherein, l i represents whether the rising amplitude or the falling amplitude of the voltage of the hydrogen fuel cell exceeds a threshold value threshold, and if the rising amplitude or the falling amplitude exceeds the threshold value, a specific amplitude loading event is determined, and the calculation formula is as follows:
(2) Average number of start/stop per hour:
In the formula, s i represents whether the working states of the front and rear hydrogen fuel stacks are consistent, if not, a start-stop event is judged, and the calculation formula is as follows:
(3) Average idle time per hour:
wherein m is the number of effective working samples of the hydrogen fuel cell stack, and m is less than or equal to n; t sample represents a data upload time interval every two frames; h i represents whether the pile is in an idle state, and when the pile current is not 0 and the speed is 0, the state of the vehicle is judged to be idle at the moment, and a calculation formula of h i is as follows:
7. The method of claim 1, wherein: the fourth step further comprises the following steps:
Firstly, classifying all hydrogen fuel vehicles in a database, wherein classification indexes comprise the types of materials of a galvanic pile and rated power level of the galvanic pile;
After classification, the vehicle looks up the stack decay rate based on the aging experience library only at the same material classification and power level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210883611.XA CN115291110B (en) | 2022-07-26 | 2022-07-26 | Pile aging prediction method based on feature parameter extraction and aging experience library construction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210883611.XA CN115291110B (en) | 2022-07-26 | 2022-07-26 | Pile aging prediction method based on feature parameter extraction and aging experience library construction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115291110A CN115291110A (en) | 2022-11-04 |
| CN115291110B true CN115291110B (en) | 2024-07-19 |
Family
ID=83823276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210883611.XA Active CN115291110B (en) | 2022-07-26 | 2022-07-26 | Pile aging prediction method based on feature parameter extraction and aging experience library construction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115291110B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115907237B (en) * | 2023-02-21 | 2023-08-01 | 江苏御传新能源科技有限公司 | Auto-parts production system based on parameter configuration |
| CN118094088B (en) * | 2024-04-17 | 2024-07-12 | 中汽研汽车检验中心（天津）有限公司 | Method, apparatus and storage medium for predicting hydrogen consumption of fuel cell |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105510832A (en) * | 2014-10-14 | 2016-04-20 | 福特全球技术公司 | Electrified vehicle battery state-of-charge monitoring with aging compensation |
| CN106526486A (en) * | 2016-08-30 | 2017-03-22 | 郑州轻工业学院 | Construction method for lithium battery health life model |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101853382B1 (en) * | 2014-11-25 | 2018-04-30 | 자동차부품연구원 | Calculating apparatus of residual life parameter for starting battery |
| CN111458225B (en) * | 2020-04-29 | 2021-09-03 | 同济大学 | Method for predicting service life of proton exchange membrane fuel cell sealing material |
| CN113314738B (en) * | 2021-04-30 | 2023-12-29 | 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 |
-
2022
- 2022-07-26 CN CN202210883611.XA patent/CN115291110B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105510832A (en) * | 2014-10-14 | 2016-04-20 | 福特全球技术公司 | Electrified vehicle battery state-of-charge monitoring with aging compensation |
| CN106526486A (en) * | 2016-08-30 | 2017-03-22 | 郑州轻工业学院 | Construction method for lithium battery health life model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115291110A (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117093879B (en) | A data center intelligent operation management method and system | |
| CN111339712B (en) | Proton exchange membrane fuel cell residual life prediction method | |
| CN115084600B (en) | Hydrogen fuel cell stack output performance analysis method based on big data | |
| CN115291110B (en) | Pile aging prediction method based on feature parameter extraction and aging experience library construction | |
| CN111476435B (en) | Charging pile load prediction method based on density peak value | |
| CN119199366B (en) | Charging pile fault detection system and method based on Internet of things | |
| CN115091963B (en) | Energy consumption estimation method for hydrogen fuel cell vehicles based on energy consumption characteristics and operating condition segmentation | |
| CN113791351A (en) | Lithium battery life prediction method based on transfer learning and difference probability distribution | |
| CN114646888B (en) | Assessment method and system for capacity fading of power battery | |
| CN114384423A (en) | Fuel cell health state identification method based on typical operation parameters | |
| CN117252428A (en) | Transformer operation risk assessment methods, devices, electronic equipment and storage media | |
| CN113379005B (en) | Intelligent energy management system and method for power grid power equipment | |
| CN120044400B (en) | Lithium ion battery thermal runaway early warning method based on multi-parameter monitoring | |
| CN118886333B (en) | Intelligent decision method and system for bridge maintenance | |
| CN116756505B (en) | A photovoltaic equipment intelligent management system and method based on big data | |
| CN115277185B (en) | Operation and maintenance system anomaly detection method based on graph neural network | |
| KR20230145799A (en) | Device and method for predicting defective rate of battery | |
| CN118259179B (en) | Internal resistance characteristic-based power battery cell internal resistance abnormality identification method | |
| CN120669119B (en) | Methods and Systems for Predicting Battery Life of New Energy Vehicles | |
| CN111881419B (en) | Vehicle cold start evaluation method | |
| CN119518728B (en) | Traffic flow and charging station load prediction method based on deep learning | |
| CN121114811B (en) | A method and system for testing the battery performance of intelligent vehicles | |
| CN115308629B (en) | Method and device for testing battery performance, storage medium and electronic equipment | |
| CN119147965B (en) | A method and system for constructing a motor parameter monitoring model | |
| CN118795354B (en) | Battery life state analysis method and system based on charge integration algorithm |
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 |