CN113258104A - Method and device for determining humidity of fuel cell and server - Google Patents

Method and device for determining humidity of fuel cell and server Download PDF

Info

Publication number
CN113258104A
CN113258104A CN202010083879.6A CN202010083879A CN113258104A CN 113258104 A CN113258104 A CN 113258104A CN 202010083879 A CN202010083879 A CN 202010083879A CN 113258104 A CN113258104 A CN 113258104A
Authority
CN
China
Prior art keywords
fuel cell
state data
humidity
current state
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010083879.6A
Other languages
Chinese (zh)
Inventor
臧晓云
柳绪丹
王凯
常亚飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Robert Bosch GmbH
Original Assignee
Robert Bosch GmbH
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Robert Bosch GmbH filed Critical Robert Bosch GmbH
Priority to CN202010083879.6A priority Critical patent/CN113258104A/en
Publication of CN113258104A publication Critical patent/CN113258104A/en
Pending legal-status Critical Current

Links

Images

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
    • H01M8/04313—Processes for controlling fuel cells or fuel cell systems characterised by the detection or assessment of variables; characterised by the detection or assessment of failure or abnormal function
    • 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
    • H01M8/04313—Processes for controlling fuel cells or fuel cell systems characterised by the detection or assessment of variables; characterised by the detection or assessment of failure or abnormal function
    • H01M8/04492—Humidity; Ambient humidity; Water content
    • 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
    • H01M8/04313—Processes for controlling fuel cells or fuel cell systems characterised by the detection or assessment of variables; characterised by the detection or assessment of failure or abnormal function
    • H01M8/04537—Electric variables
    • H01M8/04544—Voltage
    • H01M8/04559—Voltage of fuel cell stacks
    • 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
    • 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 determining the humidity of a fuel cell, which comprises the following steps: receiving current state data for a first fuel cell located on a first vehicle, wherein the current state data comprises Electrochemical Impedance Spectroscopy (EIS) information for the first cell; obtaining historical state data associated with the first fuel cell; generating a humidity estimate for the first fuel cell based on the current state data and historical state data.

Description

Method and device for determining humidity of fuel cell and server
Technical Field
The present invention relates to fuel cell systems, and more particularly to determining humidity in a fuel cell.
Background
A fuel cell, which is a power generating device that directly converts chemical energy into electrical energy by using an electrochemical reaction of oxygen and hydrogen, has been widely spotlighted and developed as a next-generation energy source because it does not consume fossil fuel and has almost zero emission, and among them, a proton exchange membrane type fuel cell (PEMFC), which is one of the most important types, has important advantages of low operating temperature, rapid start-up, etc., and thus is particularly suitable for transportation vehicles.
For proton exchange membrane type fuel cells, maintaining the humidity of a normal proton exchange membrane is very critical, which directly affects the optimum performance of the proton exchange membrane, since the ionic conductivity depends in particular on the hydration level, a greater hydration capacity leading to a higher conductivity and thus a more efficient cell. However, too high a hydration level may also lead to the formation of a liquid water layer, which may cause performance and reliability problems, e.g. the liquid water layer may block porous pathways and thus may lead to voltage losses at high current densities, voltage instability, unreliable start-up at zero temperature, etc. Therefore, on-line monitoring of the humidity in the fuel cell is particularly important.
Humidity sensors are commonly used in the art to detect battery humidity. Sensors are typically provided at the input and output of the fuel cell stack to enable measurement of the humidity inside the stack. However, this brings about two significant drawbacks: firstly, since the sensor can only be installed outside the stack, not inside the stack, it is impossible to effectively measure the actual humidity inside the stack, and the measured humidity does not reflect the actual humidity when the electrochemical reaction occurs in the stack. Secondly, the humidity sensor has high cost, maintenance and repair cost. In addition, the humidity sensor occupies a certain space, and therefore, the application in a specific field such as a fuel cell vehicle is limited to a certain extent.
Disclosure of Invention
The invention provides a detection technology based on Electrochemical Impedance Spectroscopy (EIS), which is a scheme for estimating the humidity of a proton exchange membrane by measuring the impedance of a fuel cell related to frequency. According to the invention, accurate online estimation of humidity is achieved by measuring the state information of the battery in real time using the EIS and the pre-established correlation between other operating states of the battery and humidity.
According to one aspect of the present invention, there is provided a method for determining humidity of a fuel cell, comprising: receiving current state data for a first fuel cell located on a first vehicle, wherein the current state data comprises Electrochemical Impedance Spectroscopy (EIS) information for the first cell; obtaining historical state data associated with the first fuel cell; generating a humidity estimate for the first fuel cell based on the current state data and historical state data.
According to another aspect of the present invention, there is provided a humidity determination method including: acquiring current state data of a fuel cell, wherein the current state data comprises EIS information of the cell; sending the current state data to a remote server; receiving a humidity estimate for the fuel cell from the remote server, wherein the humidity estimate is generated by the remote server executing the method of the present invention.
According to another aspect of the present invention, there is provided an apparatus for estimating humidity of a fuel cell, including: a receiving unit configured to receive current status data of a first fuel cell located on a first vehicle, wherein the current status data includes EIS information of the first fuel cell; an extraction unit configured to acquire historical state data relating to the first fuel cell; an estimation unit configured to generate a humidity estimate for the first fuel cell based on the current state data and historical state data.
According to still another aspect of the present invention, there is provided a server including: a database for storing historical state data of the fuel cell; one or more computing devices configured to implement the humidity estimation method of the present invention by executing machine readable instructions.
According to the scheme of the invention, the installation of the humidity sensor and the related system thereof can be avoided, thereby saving the cost and the communication interface.
Drawings
The accompanying drawings are included to provide a further understanding of the various aspects, and are incorporated in and constitute a part of this specification. The accompanying drawings illustrate various aspects and together with the description serve to explain the principles of the aspects. Other aspects and many of the intended advantages of various aspects will be more readily appreciated as they become better understood by reference to the following detailed description.
FIG. 1 schematically illustrates a graph of Electrochemical Impedance Spectroscopy (EIS) measured under certain conditions versus humidity;
FIG. 2A illustrates an in-vehicle system interconnected to a cloud according to one example;
fig. 2B shows a configuration of a fuel cell system according to an example;
fig. 3 is a configuration diagram of a remote/cloud server according to an embodiment of the present invention;
FIG. 4 is a flow chart of a method of determining fuel cell humidity in accordance with an embodiment of the present invention;
fig. 5 is a flow chart of a method of determining fuel cell humidity according to a further embodiment of the present invention.
Detailed Description
The following describes the apparatus and method provided by the embodiments of the present invention in detail with reference to the accompanying drawings. While the preferred 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.
Electrochemical Impedance Spectroscopy (EIS) is an important performance index of the fuel cell, can comprehensively reflect parameters such as water, oxygen and hydrogen contents in the fuel cell, current passing through a cell stack and the like, and can be used for monitoring the state of the fuel cell in real time. There are various ways in the prior art to determine the EIS, for example, EIS information is typically obtained by applying a small sinusoidal signal to the fuel cell as a fluctuating signal and measuring the voltage response. The current prior art application of EIS information focuses on several aspects: obtaining electrochemical parameters of the fuel cell system, assisting in identifying problems with components within the cell, facilitating cell structural optimization, and selecting appropriate operating conditions, among other things. However, in the present invention, it was also found that EIS information also has a correlation with the humidity inside the battery. For example, fig. 1 schematically shows nyquist plots, i.e., impedance maps EIS at different frequencies, of a fuel cell system at different operating times under different relative humidities, where RH represents relative humidity, re (z) is the real part of EIS, and im (z) is the imaginary part of EIS. It can be seen that EIS has different specific profiles of variation at different humidities.
Therefore, according to an embodiment of the present invention, a scheme of estimating the humidity inside the battery based on the EIS information is proposed. Furthermore, according to an embodiment of the present invention, when estimating the humidity, other physical information in the current state of the fuel cell system to be estimated, such as the operating voltage, current, etc. of the stack, is also taken into account, while further taking into account historical state data about this type of fuel cell system. Therefore, the online accurate estimation of the humidity of the fuel cell system is realized by utilizing the big data based on the collected current state information and the historical state information of the fuel cell system.
Fig. 2A exemplarily illustrates an in-vehicle system interconnected to a cloud. As shown, N fuel cell vehicles V are shown1～VNEach fuel cell vehicle V has network communication capability, such as being interconnected to a cloud server or remote server 100 located in the cloud via a wireless network 300 (e.g., the internet of things IoT or private car networking). These fuel cell vehicles V1～VNEquipped Fuel Cell (FC)1～FCN) Either of the same type, for example with the same type of proton exchange membrane PEM, or possibly equipped with different types of fuel cells FC.
Fig. 2B shows an internal standard architecture of a fuel cell system 200 equipped in an exemplary automotive fuel cell vehicle V. As shown, the fuel cell system 200 generally includes a DC/DC converter 201, a fuel cell control unit FCCU202, and a sensor 203. The DC/DC converter 201 may be used to measure EIS information of the fuel cell FC online and then transmit the EIS data to the fuel cell control unit FCCU202 via the internal CAN bus. In addition, the sensor 203 is used to measure an operating state parameter reflecting the operating state of the cell, such as the operating voltage of the stack, the operating current, the reactant flow rate in the fuel cell (including, for example, the hydrogen flow rate or the air flow rate, etc.), and may also detect an environmental parameter of the cell, such as the current cell operating temperature and pressure. It is to be noted here that in practice the sensors used for measuring the above parameters may be different types of sensors or be determined by means of calculations or the like, and are herein, for simplicity, collectively indicated as sensor 203.
The fuel cell control unit FCCU202 transmits the operating state parameters of the fuel cell FC measured by the sensor 203, the environmental parameters, the EIS information received from the DC/DC201, and other information about the fuel cell FC recorded by the FCCU202, such as the battery life, to the remote server 100 through the communication network 300.
In this way, each fuel cell vehicle V connected to the remote server 1001～VNThe current state information of the fuel cell may be constantly transmitted to the server 100 through the network 300. The server 100 stores the received operation state information with the environment information. It should be noted that, since some vehicles may still be equipped with a humidity sensor, such vehicles may also upload the humidity data detected by the humidity sensor to the server 100. The server 100 records the humidity information of the fuel cell of the corresponding vehicle in association with the state information thereof. Thus, the remote server 100 stores these status information acquired at different times as historical data for analysis.
Fig. 3 shows a configuration diagram of the remote server 100 according to an embodiment of the present invention. As shown, the server 100 includes a database 101, a receiving unit 102, an extracting unit 103, and an estimating unit 104. The database 101 stores historical status data from different fuel cell vehicles V, which may be collected from the respective fuel cell systems 200 by the receiving unit 102 or obtained from other sources, which may be stored in the historical database 1011. As previously mentioned, these status data include not only EIS, but also other battery status data, including but not limited to operating status data and operating environment data, and possibly battery humidity data detected by a humidity sensor in the case where the fuel cell of some vehicles is equipped with a humidity sensor.
According to an example of the present invention, the server 100 may utilize big data processing techniques (e.g., machine learning methods) to train the historical state data to learn the correlation between EIS and other states of the battery and the humidity of the battery, which may be represented by a humidity estimation model HEM, for example, which is a trained big data machine learning model in this example. As a further example, a neural network may be employed to learn historical state big data from a large number of fuel cells FC, thereby training a humidity estimation model HEM for subsequent estimation of the cell humidity. It should be noted here that the humidity estimation model HEM is not limited to a neural network model, but may be any big data machine learning model expressed in other algorithms or mathematical expressions as long as such algorithms or mathematical expression table reflects the correlation of the state of the battery with humidity. In addition, different respective HEMs may be trained for different types or characteristics of fuel cells. The humidity estimation models HEM corresponding to different types or characteristics of batteries may be stored in the model library 1012, for example.
In one example, the historical state data stored in database 101 may be stored in historical database 1011 in the form of data samples, such as (Y, X), where Y represents the humidity of a certain battery at time t, and X ═ X (X)EIS,x1,x2,…xm) Wherein x isEISRepresents the EIS of the battery at time t, and x1,x2,…xmRespectively, corresponding to other state data of the cell at that time t, such as m parameters selected from voltage, operating current, operating life, reactant flow rate in the fuel cell, and cell operating temperature, pressure, etc. The form of data sample is used for storing data for subsequent processing, but the invention is not limited thereto, and the database 101 may also store the historical state data in association with other formats. The database 101 may further classify the data samples according to the type of battery.
The receiving unit 102 is configured to receive, from the FCCU202 in the fuel cell system 200 via the network 300, uploaded current state data (hereinafter, referred to as C _ state) of the fuel cell FC, which may be measured in real time by the sensor 203 in the cell, such as the operating voltage, the current, the operating temperature, and the like of the stack, and a humidity estimation request, which is referred to as HQ; or may be information generated after processing by the DC/DC converter 201, such as EIS. In addition, the current state data C _ state may also contain other attribute information about the battery that changes over time, such as the lifetime. Furthermore, in the received state data C _ state, it is also possible to include type information of the fuel cell, expressed as FC _ ID, and the fuel cells FC on different automobiles may have the same or different FC _ IDs. For the humidity estimation request HQ, the reception unit 102 sends the received current state data C _ state to the extraction unit 103.
The extraction unit 103 parses out the type FC _ ID of the current fuel cell from the current state data C _ state to determine a humidity estimation model HEM that fits the type of fuel cell. For example, the current vehicle V is received at the receiving unit 1021Current state data C _ state1In this case, the extraction unit 103 may extract the current vehicle V based on the current vehicle V obtained by the analysis1Fuel cell system FC1Battery type FC _ ID of1Generating model query requests MQ1And provided to the database 101, the database 101 processing the query request MQ1To determine the battery type FC _ ID1Adapted humidity estimation models, e.g. denoted HEM1. Database 101 is requesting MQ based on queries1Determining a matching humidity estimation model HEM1Thereafter, as an example, the humidity estimation model HEM matched with the current fuel cell may be issued to the estimation unit 1041Is detected.
Furthermore, the extraction unit 103 also utilizes the current state data C _ state1Generating data samples X0＝(x0 EIS,x0 1,x0 2,…x0 m) Here, the superscript 0 represents a current sample generated using the current state data C _ state of the fuel cell. Then, the extraction unit 103 extracts the current sample data sample X0To the estimation unit 104. The estimation unit 104 further comprisesBased on the type FC _ ID of the current fuel cell1The information about the type FC _ ID stored in advance is obtained from the database 10111Data samples of the historical state data of the fuel cell of (1), noted as: xi＝(xi EIS,xi 1,xi 2,…xi m) Where i is 1,2, … P, where P is the total number of historical data samples acquired. The estimation unit 104 then compares (X)0，X1，X2,…XP) As an input, by executing the determined humidity estimation model HEM1Outputting a value H _ EV indicating a humidity estimation of the fuel cell in a current state1For example, the estimation unit 104 may extract the corresponding humidity estimation model HEM based on the indication signal emitted by the database 1011To process (X)0，X1，X2,…XP) To generate H _ EV1。
It is noted here that the data samples XiCan be the current automobile V1Upper fuel cell FC1May be the historical data from the current vehicle V1Upper fuel cell FC1The historical data of fuel cells on other vehicles of the same type may also be a respective part of both. Furthermore, the number of samples P for estimating the humidity can be determined in practice, or a predetermined empirical value, or even adjusted continuously during the implementation of the invention.
It should be noted that the above example is only an example, and the functions of each unit may be redesigned in the practice of the present invention, including the combination of the unit and its functions, etc. For example, the extraction unit 103 may directly extract the current state data C _ state received from the fuel cell1Transmitted to the estimating unit 104, and the estimating unit 104 generates the data sample X representing the current state0＝(x0 EIS,x0 1,x0 2,…x0 m). In addition, the database 101 may also directly store the FC _ ID according to the type of the fuel cell1Determined historical state data samples and humidity estimation model HEM1Transfer ofTo the estimation unit 104.
According to another embodiment of the invention, the trained humidity estimation model HEM is utilized1Estimating the humidity H _ EV of the battery1Thereafter, the server 100 may sample the complete data (Y) in the current state0，X0) Updating into historical data sample base, wherein Y0Representative humidity estimate H _ EV1. When applied, the server 100 may further continuously collect current status data from other batteries, such as the battery FC, via the receiving unit 1022State data C _ state of2. If the cell FC2No humidity estimation is required (e.g. it has installed a humidity sensor), the current state data C _ state is extracted from the current state data by the extraction unit 1032Extracting the fuel cell FC2The current humidity measured by the middle humidity sensor, the EIS of the stack, the operating voltage, the current, the operating temperature, the lifetime, the ambient temperature, etc., and construct a data sample (Y, X) to update the sample library 1011.
According to an example of the present invention, as shown in fig. 3, the server further includes a training unit 105, after a certain amount of updated samples are accumulated in the database 101, the training unit 105 may utilize the updated data samples associated with the same type of fuel cell to train the original humidity estimation model HEM corresponding to the type of fuel cell again to adjust the model, so as to achieve more effective estimation of the actual humidity of the fuel cell, and reflect changes of relevant rules of humidity in real time along with changes of various conditions such as time and environment.
As an example, the server 100 may further include a sending unit 106. Estimated humidity H _ EV is transmitted to the transmission unit 1061As to the vehicle V from1Request for humidity HQ1Is sent back to fuel cell FC in response to1The fuel cell control unit FCCU 202. Upon receiving the humidity H _ EV sent back by the server 100, the FCCU2021Then, corresponding control can be made according to the humidity value. Here, the transmitting unit 106 and the receiving unit 102 may also be combined into one transmitting/receiving interface to implement the data transmitting/receiving function of the server 100.
In the present invention, the remote server 100 is not limited to one physical entity, but may be a plurality of computing devices that are physically separated. For example, the remote server 100 may be any device or cloud server located in the cloud. For example, as a cloud server, one or more computing devices and a database 101 separately provided from or integrated with the computing devices are included, wherein historical State data (here, represented by H _ State) and humidity estimation models HEM matched with each battery type, which are trained in advance, are stored in the database 101. Thus, these computing devices may implement the methods of estimating battery humidity disclosed herein by executing executable programs or instructions to estimate the current humidity of the fuel cell.
In the present invention, the humidity estimation model HEM is trained in advance through machine learning using the historical State data H _ State. As described above, in the data samples (Y, X) included in the historical State data H-State, each sample label Y may be obtained from physical measurement of a humidity sensor in the fuel cell system or estimated based on the data sample X using the initial humidity estimation model HEM. Further, in one example of the invention, the neural network used to learn the historical State data H _ State may be any neural network engine implementation known in the art. For example, the recurrent neural network RNN may be utilized to extract the correlation characteristics of these status data with humidity. The RNN has an advantage in that the previous information is memorized through the connection structure of nodes between each layer and used to affect the output of the following node, so that the timing information in the state data can be sufficiently mined. The structure of the RNN and the units in each RNN can be determined by those skilled in the art according to the actual situation. For example, in a more preferred embodiment, the RNN employs a Bidirectional Recurrent Neural Network (BRNN) and each node may be implemented using different elements, such as a long short term memory neural network (LSTM) or a gated recurrent neural network (GRU).
The humidity estimation method of the present invention is described below with reference to fig. 4, and the method may be implemented by a server 100, such as a cloud or a remote server. In the following description, the vehicle is still usedVehicle V1For example. As shown in fig. 4, the server 100 receives a signal at the vehicle V in step S4011Current state data C _ state transmitted from the upper fuel cell system 2001Wherein the current state data includes cell FC1And also other status data such as operating voltage, operating current, operating life, battery operating temperature, pressure, etc. of the stack. At the same time from cell FC1Receives the determination of the cell FC1Request for internal humidity HQ1. It should be noted here that, in this example, the cell FC1The fuel cell system 200 of (1) is exemplified by explicitly issuing a humidity estimation request HQ1For example, the state C _ state can be associated with1Are sent together. In another example, however, the fuel cell system 200 may not explicitly issue a humidity request, but the server 100 may analyze the current state data C _ state1Whether humidity information is included to determine that a humidity estimation is required. For example, when the state C _ state is determined1When the humidity data is included, the server 100 generates a data sample (Y, X) based on only the state data, and stores the data sample (Y, X) as history data in the sample library 1011. If the humidity data is not contained, a subsequent humidity estimation process is performed to determine the humidity corresponding to the current state.
In step S403, the server 100 extracts and stores fuel cell FC1The relevant historical state data H _ state, which may contain fuel cell FC1May also include historical state data from a plurality of other vehicles ViUpper fuel cell FCiHistory state data of (1), fuel cell FCiAnd fuel cell FC1Of the same type or character (i.e. FC _ ID)i＝FC_ID1)。
Subsequently, in step S405, the server 100 bases on the current state data C _ state1And historical state data H _ state estimation of fuel cell FC1Humidity H _ EV1. According to one example of the invention, a humidity estimation model HEM is used to process the current state data C _ state1And historical state data H _ state, to generate fuelEstimation of humidity H _ EV corresponding to current state of battery1. In this example, the humidity estimation model HEM is a big data machine learning model obtained by training and learning the history data on the type of fuel cell, and corresponds to the type of fuel cell to be estimated. Such a big data machine learning model may be a neural network model. In accordance with this example of the invention, different respective humidity estimation models HEM may be trained for different types of fuel cells and an appropriate HEM selected based on the type of fuel cell at hand when in use. In another example, a universal humidity estimation model may also be trained for different types of fuel cells.
As to FC1Issued estimated humidity request HQ1In response, estimated H _ EV1Sending back FC1Such that the system 200 is based on the estimated H _ EV1Corresponding control is made.
By using current status data as well as historical status data from different vehicles, these collected big data reflect not only the instantaneous operating status of the fuel cell, but also the operating status throughout the life cycle, thus avoiding loss of time information. Moreover, the true humidity level inside the fuel cell can be reflected using the humidity estimation method of the present invention, rather than the humidity at the external outlet or inlet of the fuel cell as is conventional.
Fig. 5 shows a flow chart illustrating a method of estimating battery humidity according to another embodiment, which may be implemented by the server 100 located in the cloud in this example. As shown, in step S501, the cloud server 100 is connected to any vehicle V from the cloudiReceives the current state data C _ state of the vehicleiThe state data C _ stateiThe EIS information of the fuel cell is included, and other operating state parameters, such as operating parameters related to stack operating voltage, operating current, operating life, and environmental information such as temperature, pressure, etc. of the fuel cell operation are also included. If the vehicle ViEquipped with a battery humidity sensor, is in the state data C _ stateiAlso includes a fuel cell FCiThe actual humidity measurement H AV.
In step S503, the cloud server 100 determines whether V-based is requirediCurrent state data C _ state sentiFor which the humidity is determined. For example, the cloud server 100 may detect the status data C _ stateiWhether or not to include fuel cell FCiThe actual humidity measurement H AV. If it is determined that the actual humidity measurement value H _ AV is included, the process proceeds to step S505, where the current state data C _ state is setiStored as historical data, e.g. extracting status data C _ stateiAnd converted to form a sample (Y)i,Xi) And stored in a database. If the state data C _ state is determined in step 503iDoes not include the actual humidity measurement value H _ AV, it is determined that the cell FC is requirediThe humidity value in the current state is estimated, and therefore the step proceeds to S507. As another example, the server 100 may also query the status data C _ stateiWhether or not a humidity estimation request HQ is included to determine whether or not humidity needs to be estimated for the battery. For example, if the status data C _ stateiIf the humidity estimation request Flag is not included, it is determined that it is not necessary to estimate the humidity for the battery of the current vehicle, and otherwise, estimation processing needs to be performed.
In step 507, the cloud server 100 retrieves the current state data C _ stateiIn-situ resolved battery FCiType FC _ ID ofiAnd determining a humidity estimation model HEM adapted to the type of fuel celliAnd with the FCiType FC _ ID ofiA matching set of historical state data samples.
In step 509, the cloud server 100 utilizes the current state data C _ stateiGenerating data samples X0＝(x0 EIS,x0 1,x0 2,…x0 m) Here, the superscript 0 represents the use of the current state data C _ state of the fuel celliThe current sample generated. At the same time, from the FC determined in step 507iType FC _ ID ofiMatching historical state data samplesThis concentrate and draw P data sample, note as: xj＝(xj EIS,xj 1,xj 2,…xj m) Where j is 1,2, … P. Data sample XjMay be the current fuel cell FCiMay be from the fuel cell FCiThe historical data of the fuel cells on other vehicles of the same type may also be a respective part of both.
In step 511, the data sample set (X)0，X1，X2,…XP) As a model input, the humidity estimation model HEM determined in step S507 is executediTo output fuel cell FCiHumidity estimation H _ EV at Current Statei。
In step S513, the cloud server 100 estimates the humidity estimation value H _ EViSending back to fuel cell FCiSo that the fuel cell FCiAnd making a real-time control strategy according to the estimated value. Further, the server 100 also estimates the estimated humidity estimation value H _ EViWith current state data C _ stateiOr generating data samples X based on the current state data0(x0 EIS,x0 1,x0 2,…x0 m) As historical state data samples (Y)0，X0) Stored for later use in fuel cell FCiOr further humidity estimation of other cells, wherein Y0Representative humidity estimate H _ EVi。
According to another embodiment of the invention, the cloud server may further train the humidity estimation model HEM using a continuously updated sample libraryiThereby enabling the humidity estimation model HEMiThe change over time of the correlation between the state of the battery and the humidity can be better reflected. For this purpose, as shown in fig. 5, the method further includes step S515, collecting the FC from the battery FC in the server 100iThe state data samples at a plurality of times and the updated data samples at different times of other batteries can be used for estimating the model HEM of the raw humidityiAnd (5) performing retraining adjustment. Alternatively, to reflect the time series characteristics, the model HEM is evaluated for retraining degreeiMay still retain a portion of the original historical state data samples.
It is to be noted here that, although the present invention has been described in connection with the above preferred embodiments, it is apparent that the present invention is not limited thereto. Further, the modules in fig. 3 may comprise processors, electronics devices, hardware devices, electronics components, logic circuits, memories, software codes, firmware codes, etc., or any combination thereof. Those of skill would further appreciate that the various illustrative logical blocks, units, and method steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both. A software implementation is, for example, a logical device that is formed by a computing device executing computer program instructions. Another embodiment of the invention provides a machine-readable medium having stored thereon machine-readable instructions, which, when executed by a computing device, cause the computing device to perform any of the methods disclosed herein.
It should be noted that not all steps or modules in the structure diagrams of the above-described processes and apparatuses are necessary, and some steps or modules may be omitted or combined according to actual needs. The execution order of the steps is not fixed and can be adjusted as required. The system structure described in the above embodiments may be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by a plurality of physical entities, or some components in a plurality of independent devices may be implemented together.
While the invention has been shown and described in detail in the drawings and in the preferred embodiments, it is not intended to limit the invention to the embodiments disclosed, and it will be apparent to those skilled in the art that various combinations of code auditing means in the various embodiments described above may be used to obtain further embodiments of the invention, which are also within the scope of the invention.

Claims (19)

1. A method for determining humidity of a fuel cell, comprising:
receiving current state data for a first fuel cell located on a first vehicle, wherein the current state data comprises Electrochemical Impedance Spectroscopy (EIS) information for the first cell;
obtaining historical state data associated with the first fuel cell;
generating a humidity estimate for the first fuel cell based on the current state data and historical state data.
2. The method of claim 1, wherein the historical state data associated with the first fuel cell comprises at least one of:
historical state data of the first fuel cell, including historical EIS data of the first fuel cell; and
historical state data of a second fuel cell on the at least one second vehicle, the second fuel cell being of the same type or characteristic as the first fuel cell.
3. The method of claim 2, wherein the current state data and historical state data further comprise at least one of: the working voltage, working current, working life, reactant flow rate, cell working temperature, pressure of the fuel cell stack.
4. The method of one of claims 1 to 3, wherein generating a humidity estimate of the first fuel cell is accomplished by applying a humidity estimation model to the current state data and historical state data, wherein the humidity estimation model is a trained big data machine learning model.
5. The method of claim 4, wherein the big data machine learning model is a neural network model.
6. The method of claim 5, further comprising updating the humidity estimation model with:
current state data of the first fuel cell and a humidity estimate of the first fuel cell;
current state data of the at least one second fuel cell and humidity of the second fuel cell corresponding to the current state data of the second fuel cell measured from a sensor on the second fuel cell or estimated based on the current state data of the second fuel cell; and
the method further comprises the following steps:
updating the historical state data of the first fuel cell with the current state data of the first fuel cell and the humidity estimate.
7. A humidity determination method comprising:
acquiring current state data of a fuel cell, wherein the current state data comprises Electrochemical Impedance Spectroscopy (EIS) information of the fuel cell;
sending the current state data to a remote server;
receiving a humidity estimate for the fuel cell from the remote server, wherein the humidity estimate is generated by the remote server performing the method of one of claims 1-7.
8. The method of claim 7, wherein the remote server is a cloud server.
9. An apparatus for estimating humidity of a fuel cell, comprising:
a receiving unit configured to receive current status data of a first fuel cell located on a first vehicle, wherein the current status data comprises Electrochemical Impedance Spectroscopy (EIS) information of the first fuel cell;
an extraction unit configured to acquire historical state data relating to the first fuel cell;
an estimation unit configured to generate a humidity estimate for the first fuel cell based on the current state data and historical state data.
10. The apparatus of claim 9, wherein
Wherein the historical state data relating to the first fuel cell comprises at least one of:
historical state data of the first fuel cell, including historical EIS data of the first fuel cell; and
historical state data of a second fuel cell on at least one second vehicle, the second fuel cell being of the same type or characteristic as the first fuel cell.
11. The apparatus of claim 10, wherein the current state data and historical state data comprise at least one of: the operating voltage, operating current, operating life, reactant flow rates, and cell operating temperatures and pressures of the fuel cell stack.
12. The apparatus according to any one of claims 9 to 11, wherein the estimation unit estimates the humidity of the first fuel cell by executing a humidity estimation model with the current state data and the historical state data as inputs, wherein the humidity estimation model is a trained big data machine learning model.
13. The apparatus of claim 12, wherein the big data machine learning model is a neural network model.
14. The apparatus of claim 12, further comprising a transmitting unit to transmit the humidity estimate to the first fuel cell.
15. The apparatus of claim 12, further comprising:
a training unit configured to update the humidity estimation model with the following data:
current state data of the first fuel cell and a humidity estimate of the first fuel cell;
the current state data of the at least one second fuel cell and the humidity of the second fuel cell corresponding to the current state data of the second fuel cell measured from a sensor on the second fuel cell or estimated based on the current state data of the second fuel cell.
16. The apparatus according to claim 12, further comprising a storage unit for storing historical state data of the first fuel cell and the second fuel cell and a humidity estimation model,
wherein the storage unit updates the historical state data of the first fuel cell with the current state data of the first fuel cell and the humidity estimate.
17. The device of claim 12, wherein the device is a cloud server networked with the vehicle over a wireless link.
18. The apparatus of claim 12, wherein the humidity estimation model corresponds to a characteristic or type of the fuel cell.
19. A server, comprising:
a database for storing historical state data of the fuel cell;
one or more computing devices configured to implement the method of one of claims 1-6 by executing machine readable instructions.
CN202010083879.6A 2020-02-10 2020-02-10 Method and device for determining humidity of fuel cell and server Pending CN113258104A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010083879.6A CN113258104A (en) 2020-02-10 2020-02-10 Method and device for determining humidity of fuel cell and server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010083879.6A CN113258104A (en) 2020-02-10 2020-02-10 Method and device for determining humidity of fuel cell and server

Publications (1)

Publication Number Publication Date
CN113258104A true CN113258104A (en) 2021-08-13

Family

ID=77219329

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010083879.6A Pending CN113258104A (en) 2020-02-10 2020-02-10 Method and device for determining humidity of fuel cell and server

Country Status (1)

Country Link
CN (1) CN113258104A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113871661A (en) * 2021-09-23 2021-12-31 中国第一汽车股份有限公司 Control method and control device of fuel cell

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113871661A (en) * 2021-09-23 2021-12-31 中国第一汽车股份有限公司 Control method and control device of fuel cell

Similar Documents

Publication Publication Date Title
WO2022151819A1 (en) Clustering analysis-based battery system online fault diagnosis method and system
US10354462B1 (en) Fault diagnosis in power electronics using adaptive PCA
CN112398804B (en) Remote emission management vehicle-mounted terminal data consistency verification platform and verification method
CN115013298B (en) Real-time performance online monitoring system and monitoring method of sewage pump
CN113555591A (en) Fuel cell diagnostic device and corresponding diagnostic method, vehicle and vehicle system
CN112563542B (en) Fuel cell online detection method and detection system
CN115453254B (en) Power quality monitoring method and system based on special transformer acquisition terminal
CN109670549B (en) Data screening method and device for thermal power generating unit and computer equipment
KR20210135281A (en) Battery performance evaluation method and battery performance evaluation device
CN111765593A (en) Air conditioner throttling component fault early warning method and air conditioner
CN117113729A (en) Digital twinning-based power equipment online state monitoring system
CN113258104A (en) Method and device for determining humidity of fuel cell and server
CN113406539A (en) Operation monitoring method and device for intelligent cable connector
CN115320387A (en) Vehicle fuel cell membrane humidity prediction method and system based on multi-source information fusion
CN115248584A (en) Automobile fault remote detection method and system, electronic equipment and storage medium
CN115248382A (en) Control method of proton exchange membrane fuel cell
CN114399336A (en) Natural gas energy metering method and system based on Internet of things
CN114676495A (en) Method, device and equipment for predicting energy consumption of automobile
US11460512B2 (en) Battery module testing
CN108898190A (en) A kind of environmental monitoring method for early warning and system based on environmentally friendly big data
CN114779098B (en) State evaluation method and system for lithium ion battery
CN116031453A (en) On-line estimation method for characteristic frequency impedance of proton exchange membrane fuel cell
CN114119292B (en) Building management system based on Internet of things and big data
CN107710085B (en) Method, device and system for estimating damage level of electric device
CN117295960A (en) Battery performance evaluation device and battery performance evaluation method

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