A method and device for operation and maintenance control of a fuel cell engine

Abstract

The application provides a kind of operation and maintenance control method and device for fuel cell engine, belong to fuel cell technical field, solve the problem that only real-time optimization for specific control target in prior art leads to strong limitation.The method comprises: obtaining the historical trace data of all control parameters and design parameters of fuel cell engine;Obtain the operation scene parameter and running performance characterization data of fuel cell engine at the current time, and extract the time distribution feature and spatial distribution feature of each performance characterization data respectively;The time distribution feature, spatial distribution feature, operation scene parameter, and historical trace data of control parameter and design parameter are respectively input into the information fusion and parameter optimization model trained in advance, to obtain the operation and maintenance state of fuel cell engine at the current time and the optimal control parameter and optimal design parameter under the operation scene parameter;Control fuel cell engine to adjust the working state to the optimal control parameter.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fuel cell, in particular to a method and device for operation and maintenance control of fuel cell engine. BACKGROUND

[0002] The fuel cell engine generally includes a stack and peripheral hydrogen, air, cooling and other components. The stack includes a proton exchange membrane, a catalyst layer, a gas diffusion layer, a bipolar plate, etc. Since the theoretical voltage of a single cell is only 1.23 V, several hundred cells are usually connected in parallel to achieve high-power output.

[0003] At present, the fuel cell engine is still in the initial stage of commercial development, the product iteration speed is fast, the test time is short, and the core parameters of the key components, such as the I / C ratio of the ionomer and the carbon carrier of the membrane electrode, the equivalent mass of the resin, the type of the catalyst, and the pore size distribution of the catalyst layer, have not reached the optimal design combination, so the actual effect cannot match the customer's demand.

[0004] The prior art considers the optimization of the fuel cell engine, which is usually for specific control targets, such as real-time optimization of economy and power, or optimization of key design parameters by means of experiments or neural networks, which has strong limitations. SUMMARY

[0005] In view of the above analysis, the embodiments of the present application aim to provide a method and device for operation and maintenance control of fuel cell engine to solve the problem that the prior art only optimizes real-time for specific control targets, resulting in strong limitations.

[0006] In one aspect, the embodiments of the present application provide a method for operation and maintenance control of fuel cell engine, comprising the following steps:

[0007] Obtain historical trace data of all control parameters and design parameters of the fuel cell engine;

[0008] Obtain operation scene parameters and running performance characterization data of the fuel cell engine at the current time, and extract the time distribution characteristics and spatial distribution characteristics of each performance characterization data respectively;

[0009] Input the time distribution characteristics, spatial distribution characteristics, operation scene parameters, and historical trace data of control parameters and design parameters into the information fusion and parameter optimization model trained in advance respectively, to obtain the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters and optimal design parameters under the operation scene parameters;

[0010] Control the fuel cell engine to adjust the working state to the optimal control parameters.

[0011] The beneficial effects of the above technical solutions are as follows: a fuel cell operation and maintenance state tracing method is provided, which characterizes the operation scene, and through accumulation and mining of the time and space dimensions of the operation performance representation data, optimization iteration and evaluation of the combination matching of the engine control parameters and design parameters can be realized. The method can accelerate the design iteration speed of the fuel cell engine, fully utilize the on-board measured data, and realize global integrated data processing of multi-dimensional and multi-level information of design parameters, control parameters and operation scene parameters.

[0012] Based on the further improvement of the above method, the control parameters include the hydrogen pressure into the stack, the air flow into the stack and the coolant temperature; and,

[0013] The design parameters include at least one of the I / C ratio of the membrane electrode, the EW value, the flow channel design parameter of the bipolar plate, the selection parameter of the carbon paper, the VIN code of the vehicle, and the engine type number.

[0014] Further, the operation scene parameters include at least one of the vehicle type, the region, the weather state, the working condition information, and the vehicle information of the engine; and,

[0015] The operation performance representation data includes at least one of the average single piece voltage drop rate of the fuel cell engine, the impedance value rise rate, and the number of single low phenomena of the impedance value.

[0016] Further, the step of obtaining the historical tracing data of all control parameters and design parameters of the fuel cell engine further includes:

[0017] Determine all core components that affect the operation and maintenance state of the fuel cell engine, and sequentially number them;

[0018] Obtain the historical monitoring data of the control parameters and design parameters of each numbered core component;

[0019] The historical monitoring data of the control parameters and design parameters are sequentially filtered and feature extracted, and the maximum change amplitude, change rate and importance score of the control parameters and design parameters under various operation scene parameters are obtained respectively as the historical tracing data.

[0020] Further, the step of respectively extracting the time distribution feature and the space distribution feature of each performance representation data further includes:

[0021] Respectively obtain the maximum value, minimum value, average change amount and real-time change rate of each performance representation data, perform data fusion, and take the fusion result as the time distribution feature of each performance representation data;

[0022] The concentration, dispersion, skewness and kurtosis of each performance characterization data at the preset position are acquired respectively, data fusion is performed, and the fusion result is taken as the spatial distribution feature of each performance characterization data.

[0023] Further, the step of inputting the time distribution feature, the spatial distribution feature, the operation scene parameter, and the historical trace data of the control parameter and the design parameter into the information fusion and parameter optimization model trained in advance respectively to obtain the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameter and the optimal design parameter under the operation scene parameter further comprises:

[0024] determining the importance ranking of all performance characterization data;

[0025] establishing a time distribution feature matrix and a spatial distribution feature matrix of all performance characterization data according to the importance ranking of all performance characterization data;

[0026] inputting the time distribution feature matrix, the spatial distribution feature matrix, and the historical trace data of the operation scene parameter, the control parameter and the design parameter into the information fusion and parameter optimization model trained in advance to obtain the working state of each core component of the fuel cell engine at the current time, the prediction result of whether the core component is damaged, and the optimal control parameter and the optimal design parameter under the operation scene parameter.

[0027] Further, the method further comprises the following steps:

[0028] determining the occurrence probability of each operation scene parameter;

[0029] obtaining the optimal design parameter of the fuel cell engine under the comprehensive working condition of the vehicle according to the optimal design parameter under each operation scene parameter and the occurrence probability.

[0030] Compared with the prior art, the present application can at least achieve one of the following beneficial effects:

[0031] 1. The failure of each key component of the fuel cell engine can be traced, and the application scene can be characterized and modeled.

[0032] 2. In addition to obtaining the optimal control parameter under the current working condition, the optimal design parameter of the fuel cell engine under the comprehensive working condition of the vehicle can also be obtained, which reduces the test cost and shortens the product development cycle.

[0033] 3. Using the method, the fuel cell engine can be put into the market without reaching the full verification and optimal combination design, and can be continuously improved and matched with the customer's use demand in the use process.

[0034] In one aspect, the embodiment of the present application provides an operation and maintenance control device for a fuel cell engine, comprising:

[0035] a scene parameter acquisition unit configured to acquire an operation scene parameter of the fuel cell engine at the current time and send the operation scene parameter to the cloud platform processing unit;

[0036] an engine performance data acquisition unit configured to acquire running performance representation data of the fuel cell engine at the current time and send the running performance representation data to the cloud platform processing unit;

[0037] the cloud platform processing unit is configured to acquire historical trace data of all control parameters and design parameters of the fuel cell engine, extract time distribution features and space distribution features of each performance representation data, respectively, and input the time distribution features, the space distribution features, the operation scene parameter, and the historical trace data of the control parameters and the design parameters into a pre-trained information fusion and parameter optimization model to obtain an operation and maintenance state of the fuel cell engine at the current time and optimal control parameters and optimal design parameters, send the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters to the engine control unit, and send the operation scene parameter and the optimal design parameters to the computer terminal;

[0038] the engine control unit is configured to display whether a key component of the engine is faulty according to the received operation and maintenance state and adjust a working state of the key component of the engine according to the optimal control parameters;

[0039] the computer terminal is configured to receive and display the operation scene parameter and the optimal design parameters.

[0040] The technical scheme has the following beneficial effects: the fuel cell operation and maintenance state trace device features an operation scene, and can realize optimization iteration and evaluation of combination matching of engine control parameters and design parameters by means of accumulation and mining of time and space dimension data of running performance representation data. The device can accelerate design iteration speed of the fuel cell engine, fully utilize on-board measured data, and realize global integrated data processing of multi-dimension and multi-level information of design parameters, control parameters, and operation scene parameters.

[0041] The scene parameter acquisition unit further comprises:

[0042] a GPS navigation unit built-in high-precision map and mounted on a vehicle on which the fuel cell engine is located, configured to acquire regional, weather state, and working condition information of the fuel cell engine at the current time;

[0043] a vehicle information acquisition unit configured to acquire a vehicle type and on-board information of the fuel cell engine at the current time, wherein the on-board information includes a number of passengers, weights of the passengers, whether the vehicle contains fragile goods, and a weight of the goods; and

[0044] The engine performance data acquisition unit further comprises:

[0045] A fuel cell stack single piece voltage monitoring device is used to acquire stack single piece voltage data and send to the data processing subunit;

[0046] A fuel cell impedance measurement device is used to acquire stack impedance data and send to the data processing subunit;

[0047] The data processing subunit is used to derive an average single piece voltage drop rate from the stack single piece voltage data and an impedance value rise rate and impedance value single low phenomenon occurrence times from the stack impedance data.

[0048] Further, the cloud platform processing unit executes the following procedures:

[0049] Acquire historical trace data of all control parameters and design parameters of the fuel cell engine;

[0050] Respectively extract the time distribution characteristics and spatial distribution characteristics of each performance characterization data;

[0051] According to the input of the above-mentioned time distribution characteristics, spatial distribution characteristics, operation scene parameters, and historical trace data of control parameters and design parameters into the pre-trained information fusion and parameter optimization model, the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters and optimal design parameters are derived;

[0052] The operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters are sent to the engine control unit; and according to the optimal design parameters under each operation scene parameter, the optimal design parameters of the fuel cell engine under the comprehensive working condition of the vehicle are derived and sent to the computer terminal.

[0053] Compared with the prior art, the present application can at least achieve one of the following beneficial effects:

[0054] 1. The failure of each key component of the fuel cell engine can be traced, and the application scene can be characterized and modeled.

[0055] 2. In addition to obtaining the optimal control parameters under the current working condition, the optimal design parameters of the fuel cell engine under the comprehensive working condition of the vehicle are also obtained, which reduces the test cost.

[0056] 3. After the vehicle is loaded with the operation and maintenance control device, the fuel cell can be put into the market without reaching the full verification and optimal combination design, and can be continuously improved during use to match the customer's use demand.

[0057] 4. With the accumulation and mining of data in time and space dimensions of the cloud platform, the optimization iteration and evaluation of the combination matching of the key parameters of the core components and the engine control parameters are realized.

[0058] The summary is provided to introduce a selection of concepts in a simplified form, which are further described below in the detailed description. The summary is not intended to identify key or essential features of the disclosure, nor is it intended to limit the scope of the disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0059] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the figures, in which exemplary embodiments of the present disclosure are shown.

[0060] Figure 1 An operation and maintenance control method step schematic diagram for a fuel cell engine according to embodiment 1 is shown;

[0061] Figure 2 An operation and maintenance control device composition schematic diagram for a fuel cell engine according to embodiment 3 is shown. DETAILED DESCRIPTION

[0062] Embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the present disclosure are shown. While several embodiments of the disclosure are described, it should be understood that the disclosure can be embodied in various forms and should not be limited by 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.

[0063] The term "comprising" and variations thereof as used herein are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to those elements, but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Unless otherwise specified, the term "or" as used herein is intended to mean "and / or". The term "based on" means "based, at least in part, on". The terms "one example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "a first", "a second", etc. can refer to different or the same objects. Other explicit or implicit definitions can also be included below.

[0064] Embodiment 1

[0065] In one embodiment of the present disclosure, an operation and maintenance control method for a fuel cell engine is disclosed, as shown in the accompanying drawings, comprising the following steps: Figure 1

[0066] S1. Obtain the historical tracking data of all control parameters and design parameters of the fuel cell engine;

[0067] ​S2. Obtain the operating scene parameters and the running performance characterization data of the fuel cell engine at the current time, and extract the time distribution characteristics and the spatial distribution characteristics of each performance characterization data, respectively;

[0068] S3. Input the time distribution characteristics, the spatial distribution characteristics, the operating scene parameters, and the historical tracing data of the control parameters and the design parameters into the information fusion and parameter optimization model trained in advance, respectively, to obtain the operation and maintenance state of the fuel cell engine at the current time, and the optimal control parameters and the optimal design parameters under the operating scene parameters;

[0069] S4. Control the fuel cell engine to adjust the working state to the optimal control parameters. The optimal design parameters can be provided to the R&D personnel to speed up the product design cycle.

[0070] Specifically, the selection or combination of the control parameters and the design parameters of the fuel cell engine is not unique, and can be set according to actual needs. The control parameters can refer to patent CN202110570980.9, the design parameters can refer to patent CN202010981030.0, or embodiment 2.

[0071] The selection or combination of the operating scene parameters is also not unique. In addition to the operating scene parameters described in embodiment 2, exemplary, refer to patent CN202111390601.4. In the scene where the working condition changes little, the signal light can be selected as the operating scene parameter and data is obtained.

[0072] The selection or combination of the time distribution characteristics and the spatial distribution characteristics is also not unique. In addition to the time distribution characteristics and the spatial distribution characteristics described in embodiment 2, exemplary, it can be that the engine A (membrane electrode model E1, software version number H1) is loaded on a heavy truck and operates in Chongli as the spatial distribution, and experiences winter and urban road conditions as the time distribution, which can be understood by those skilled in the art.

[0073] The information fusion and parameter optimization model can use any of the existing neural networks HN, CNN, DCNN, DN, GAN, RNN, etc. The input data in the training data is the time distribution characteristics and the spatial distribution characteristics of each performance characterization data under different operating scene parameters, and the historical tracing data of the control parameters and the design parameters. The output data is the optimal control parameters and the optimal design parameters of the set target under the corresponding operating scene parameters calibrated in the laboratory, and the set target includes the most power saving, the fastest starting speed, the highest energy utilization efficiency, etc., which can be set according to actual needs.

[0074] Compared with the prior art, the embodiment provides a fuel cell operation and maintenance state tracing method, which characterizes an operation scene, and through accumulation and mining of time and space dimensions of operation performance representation data, can realize optimization iteration and evaluation of combination matching of engine control parameters and design parameters. The method can accelerate the design iteration speed of the fuel cell engine, fully utilize the on-vehicle measured data, and realize global integrated data processing of multi-dimensional and multi-level information of design parameters, control parameters and operation scene parameters.

[0075] Embodiment 2

[0076] On the basis of the method of embodiment 1, the control parameters include hydrogen pressure into the stack, air flow into the stack and cooling liquid temperature, which can be uniquely marked by the software version number of the fuel cell control system. For example, different start-up sequences are divided into corresponding program versions H01 and H02.

[0077] Preferably, the design parameters include I / C ratio (mass ratio of resin and carbon) of the membrane electrode, EW value (acidity of the membrane), flow channel design parameters of the bipolar plate, selection parameters of the carbon paper, VIN code of the vehicle, engine type number, etc.

[0078] The I / C ratio and the EW value can be uniquely marked by the model of the membrane electrode. Specifically, for the membrane electrode with I / C of 0.5 and 0.8, the model of the membrane electrode is E01 and E02, respectively.

[0079] For the selection of the design parameters, a person skilled in the art can make a simple migration according to professional knowledge. The historical tracing data of the control parameters and the design parameters of the fuel cell engine are obtained through 4G, 5G and the like, and the operation performance representation data uploaded by the fuel cell engine control system is received.

[0080] Preferably, the operation scene parameters include at least one of the vehicle type, region, weather state, working condition information and vehicle information. Exemplarily, the weather and working condition information of the vehicle operation are obtained through a GPS navigation unit with high-precision map and a weather forecast module, and the information is fused and processed in a cloud platform, and the cloud platform is provided with an information library of the operation scene parameters.

[0081] Specifically, the vehicle type can be divided into passenger cars, heavy trucks, logistics vehicles and the like, which are marked as 1, 2, 3 and the like, respectively. The region can be divided into Zhangjiakou urban area, Zhangjiakou Chongli and the like, which are marked as a, b and the like, respectively. The weather state includes weather information (sunny, rainy, snowy) and season information (spring, summer, autumn, winter), which are marked respectively. The working condition information can be divided into high-speed, rural area and provincial road, which are marked as aa, bb, cc and the like, respectively. The vehicle information includes passenger capacity and load capacity, which are marked as AI, BI and the like, respectively.

[0082] Preferably, the operation performance characterization data includes at least one of the average single-cell voltage drop rate, the impedance value rise rate, and the number of single low phenomena of the impedance value of the fuel cell engine.

[0083] Preferably, the operation performance characterization data further includes the temperature, pressure, humidity of the hydrogen entering the stack and the air entering the stack, the output current and voltage of the stack, etc.

[0084] Specifically, the operation performance characterization data is measured by corresponding sensors. For example, the average single-cell voltage drop rate can be obtained by data processing of the voltage signal collected by the fuel cell stack single-cell voltage monitoring device (see patent CN200510086690.8, etc.), and the impedance value rise rate and the number of single low phenomena of the impedance value can be obtained by data processing of the impedance signal collected by the existing fuel cell impedance measurement system (see patents CN201911348485.2 and CN202010068829.0, etc.).

[0085] Preferably, step S1 further comprises:

[0086] S11. Determine all core components that affect the operation and maintenance state of the fuel cell engine, and sequentially number them (involved in labeling and digitizing the core components); the core components include the air inlet control device (air compressor and electric control valve) arranged at the air inlet of the stack, the hydrogen inlet control device (electric control valve) arranged at the hydrogen inlet of the stack, and the cooling liquid temperature regulating device (including a radiator and a heater in parallel) arranged at the cooling liquid end of the stack.

[0087] S12. Obtain the historical monitoring data of the control parameters and design parameters of each numbered core component.

[0088] S13. Filter and feature extract the historical monitoring data of the control parameters and design parameters in sequence (involved in labeling and digitizing the control parameters), respectively obtain the maximum change amplitude, change rate, and importance score of the control parameters and design parameters under various operation scenario parameters as historical tracing data.

[0089] Preferably, step S2 can be further refined as:

[0090] S21. Obtain the operation scenario parameters and operation performance characterization data of the fuel cell engine at the current time;

[0091] S22. Respectively obtain the maximum value, minimum value, average change amount, and real-time change rate of each performance characterization data, perform data fusion (there are many existing data fusion methods, for example, see the method described in patent CN202010976642.0 or data superposition), and take the fusion result as the time distribution characteristics of each performance characterization data.

[0092] S23. Respectively acquire the concentration, dispersion, skewness and kurtosis of each performance characterization data at the preset position, perform data fusion, and take the fusion result as the spatial distribution feature of each performance characterization data.

[0093] Specifically, the concentration can be represented by one of the data mode, median, quantile, mean, geometric mean, and trimmed mean. Concentration is also known as "data center position", "concentration number" and the like. It is a representative value of a set of data, which can be representative of certain characteristics of the population, indicating the common nature and general level under certain time and space conditions.

[0094] The dispersion can be represented by one of the range, internal distance, variance, standard deviation, and dispersion coefficient of the data. Dispersion refers to the degree of difference between the values of the observed variables, and is an index for measuring the size of the risk.

[0095] Skewness refers to the skewness of the asymmetric distribution.

[0096] Kurtosis is also known as peak state, which represents the characteristic number of the peak value of the probability density distribution curve at the mean value. In a direct view, kurtosis reflects the sharpness of the peak. The kurtosis of the sample is compared with the normal distribution. If the kurtosis is greater than three, the peak shape is relatively sharp, and the normal distribution peak is steep. Conversely.

[0097] Preferably, step S3 further comprises:

[0098] S31. Determine the importance ranking of all performance characterization data;

[0099] S32. According to the importance ranking of all performance characterization data, respectively establish the time distribution feature matrix and the space distribution feature matrix of all performance characterization data;

[0100] S33. Input the above time distribution feature matrix, space distribution feature matrix, and historical trace data of operation scene parameters, and control parameters, design parameters into the pre-trained information fusion and parameter optimization model to obtain the working state of each core component of the fuel cell engine at the current time, the prediction result of whether it is possible to be damaged, and the optimal control parameter and optimal design parameter under the operation scene parameter.

[0101] Specifically, for example, the optimal control parameter for running in Chongli, Zhangjiakou in winter is software version H1, and the membrane electrode model E1 has the lowest voltage drop rate in all seasons in Zhangjiakou urban area.

[0102] Preferably, the method further comprises the following steps:

[0103] S5. Determine the occurrence probability of each operation scene parameter;

[0104] S6. According to the optimal design parameters under each operating scenario parameter and the above-mentioned occurrence probability, the comprehensive operating condition optimal design parameters of the fuel cell engine in the vehicle are obtained.

[0105] For example, the optimal design parameters under each operating scenario parameter are multiplied by the above-mentioned occurrence probability, and then summed up to obtain the comprehensive operating condition optimal design parameters of the fuel cell engine in the vehicle.

[0106] Compared with embodiment 1, the method of the present embodiment has the following beneficial effects:

[0107] 1. The failure of each key component of the fuel cell engine can be traced back, and the application scene can be characterized and modeled.

[0108] 2. In addition to obtaining the optimal control parameters of the current operating condition, the comprehensive operating condition optimal design parameters of the fuel cell engine in the vehicle are also obtained, which reduces the test cost and shortens the product development cycle.

[0109] 3. Using the method, the fuel cell engine can be put into the market without reaching the full verification and optimal combination design, and can be continuously improved in the use process to match the customer's use demand.

[0110] Embodiment 3

[0111] The application also discloses a running and maintenance control device for a fuel cell engine, which comprises a scene parameter acquisition unit, an engine performance data acquisition unit, a cloud platform processing unit, an engine control unit (a vehicle client) and a computer terminal (a research and development personnel client), as shown in Figure 2 The scene parameter acquisition unit, the engine performance data acquisition unit and the engine control unit are all mounted in the vehicle and connected with the cloud platform processing unit located at a remote end through a wireless network, and the computer terminal can be arranged in a laboratory of a research and development personnel.

[0112] The scene parameter acquisition unit is used for acquiring the operating scenario parameters of the fuel cell engine at the current time and sending the operating scenario parameters to the cloud platform processing unit.

[0113] The engine performance data acquisition unit is used for acquiring the running performance characteristic data of the fuel cell engine at the current time and sending the running performance characteristic data to the cloud platform processing unit.

[0114] The cloud platform processing unit is configured to acquire historical tracking data of all control parameters and design parameters of the fuel cell engine, extract time distribution features and space distribution features of each performance characterization data, and input the time distribution features, the space distribution features, operation scene parameters, and the historical tracking data of the control parameters and the design parameters into a pre-trained information fusion and parameter optimization model to obtain an operation and maintenance state of the fuel cell engine at the current time and optimal control parameters and optimal design parameters of the fuel cell engine at the current time, and send the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters to the engine control unit, and send the operation scene parameters and the optimal design parameters to the computer terminal.

[0115] The engine control unit is configured to display whether a key component of the engine fails according to the received operation and maintenance state, and adjust an operating state of the key component of the engine according to the optimal control parameters.

[0116] The computer terminal is configured to receive and display the operation scene parameters and the optimal design parameters.

[0117] Compared with the prior art, the embodiment provides a tracking device for an operation and maintenance state of a fuel cell, and features operation scene parameters, and through accumulation and mining of time and space dimensions of performance characterization data, optimization iteration and evaluation of combination matching of control parameters and design parameters of the engine can be realized. The method can accelerate the design iteration speed of the fuel cell engine, fully utilize on-board measured data, and realize global integrated data processing of multi-dimensional and multi-level information of the design parameters, the control parameters, and the operation scene parameters.

[0118] Embodiment 4

[0119] The device of embodiment 3 is improved, the key components of the engine include at least one of an air intake control device (an air compressor and an electric control valve), a hydrogen intake control device (an electric control valve), and a cooling liquid temperature regulation device (a radiator and a heater connected in parallel). The output end of the engine control unit is connected to the control end of each of the key components of the engine. The control parameters include hydrogen pressure into a stack, air flow into a stack, and cooling liquid temperature.

[0120] Preferably, the design parameters include an I / C ratio (mass ratio of resin and carbon) of a membrane electrode, an EW value (acidity of a membrane), a flow channel design parameter of a bipolar plate, a selection parameter of carbon paper, a VIN code of a vehicle, an engine type number, and the like.

[0121] Preferably, the performance characterization data include an average single-piece voltage drop rate of the fuel cell engine, an impedance value rise rate, and a number of single low phenomena of the impedance value.

[0122] Preferably, the scene parameter acquisition unit further comprises a GPS navigation unit and a vehicle information acquisition unit.

[0123] The GPS navigation unit is built-in with high-precision maps and is mounted on a vehicle where the fuel cell engine is located, and is used to acquire the region, weather state, and working condition information where the fuel cell engine is located at the current time.

[0124] The vehicle information acquisition unit is used to acquire the vehicle information (obtained by input or equipped with sensors) and the system built-in data equipped by the fuel cell engine at the current time. The vehicle information includes the number of passengers (input), the weight of passengers (measured by a whole vehicle weight sensor or input), whether it contains fragile goods (input), the weight of goods (measured by a whole vehicle weight sensor or input), etc.

[0125] The above information is fused and processed in the cloud platform processing unit, and the cloud platform is built-in with an information base of operating scene parameters.

[0126] Preferably, the engine performance data acquisition unit further comprises a fuel cell stack single piece voltage monitoring device, a fuel cell impedance measuring device, and a data processing subunit. The output ends of the fuel cell stack single piece voltage monitoring device and the fuel cell impedance measuring device are respectively connected to the input end of the data processing subunit.

[0127] The fuel cell stack single piece voltage monitoring device is used to acquire the stack single piece voltage data (continuous data in a preset period) and send it to the data processing subunit. See patent CN200510086690.8, etc.

[0128] The fuel cell impedance measuring device is used to acquire the stack impedance data (continuous data in a preset period) and send it to the data processing subunit. See patents CN201911348485.2 and CN202010068829.0, etc.

[0129] The data processing subunit is used to derive the average single piece voltage drop rate according to the stack single piece voltage data, and derive the impedance value rise rate and the number of times of single low phenomenon of impedance value according to the stack impedance data.

[0130] Preferably, the engine performance data acquisition unit further comprises a temperature-pressure-humidity integrated sensor, a current sensor, and a voltage sensor.

[0131] The temperature-pressure-humidity integrated sensor is respectively arranged on the inner wall of the hydrogen inlet and air inlet pipeline of the stack, and is used to acquire the temperature, pressure, and humidity of the hydrogen entering the stack and the air entering the stack.

[0132] The current sensor and the voltage sensor are respectively arranged at the power supply end of the stack, and are used to acquire the output current and voltage of the stack.

[0133] Preferably, the engine control unit has a sound alarm module for displaying different engine key component faults. Different sound alarms can be issued for different engine key component faults to prompt the driver to timely repair.

[0134] Preferably, the cloud platform processing unit executes the following procedures:

[0135] SS1. Obtain historical trace data of all control parameters and design parameters of the fuel cell engine;

[0136] SS2. Extract the time distribution characteristics and spatial distribution characteristics of each performance characterization data, respectively;

[0137] SS3. According to the input of the time distribution characteristics, spatial distribution characteristics, operation scene parameters, and historical trace data of control parameters and design parameters into the pre-trained information fusion and parameter optimization model, the operation and maintenance state of the fuel cell engine at the current time, and the optimal control parameters and optimal design parameters are obtained;

[0138] SS4. Send the operation and maintenance state of the fuel cell engine at the current time and the optimal control parameters to the engine control unit; and according to the optimal design parameters under each operation scene parameter, obtain the optimal design parameters of the fuel cell engine under the comprehensive working condition of the vehicle, and send them to the computer terminal and push them to the R&D personnel.

[0139] Compared with Embodiment 3, the device provided in the present embodiment has the following beneficial effects:

[0140] 1. The faults of each key component of the fuel cell engine can be traced, and the application scene can be characterized and modeled.

[0141] 2. In addition to obtaining the optimal control parameters under the current working condition, the optimal design parameters of the fuel cell engine under the comprehensive working condition of the vehicle can also be obtained, which reduces the test cost and shortens the product development cycle.

[0142] 3. After the vehicle is loaded with the operation and maintenance control device, the fuel cell can be put into the market without reaching the full verification and optimal combination design, and can be continuously improved during use to meet the customer's use requirements.

[0143] 4. With the accumulation and mining of time and space dimension data of the cloud platform, the optimization iteration and evaluation of the combination matching of the key component key parameters and the engine control parameters can be realized.

[0144] Having described above several embodiments of the disclosure, any modifications and variations that fall within the scope of the described embodiments are also intended to be within the scope of the disclosure. As will be apparent to those skilled in the art, some modifications and variations to the embodiments described above can be practiced while staying within the scope and spirit of the described embodiments. The foregoing description of the described embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the described embodiments to the precise form disclosed. Many modifications and variations are possible in light of the above teachings. It is intended that the disclosed embodiments be limited only by the claims.

Claims

1. A method for operation and maintenance control of a fuel cell engine, characterized in that, Includes the following steps: Obtain historical traceability data of all control parameters and design parameters of the fuel cell engine; Obtain the current operating scenario parameters and performance characterization data of the fuel cell engine, and extract the temporal and spatial distribution characteristics of each performance characterization data. The historical data of the above-mentioned time distribution characteristics, spatial distribution characteristics, operation scenario parameters, control parameters, and design parameters are respectively input into the pre-trained information fusion and parameter optimization model to obtain the current operation and maintenance status of the fuel cell engine and the optimal control parameters and optimal design parameters under the operation scenario parameters. Control the fuel cell engine to adjust its operating state to the above-mentioned optimal control parameters; The control parameters include the reactor hydrogen pressure, reactor air flow rate, and coolant temperature; and... The design parameters include at least one of the following: I / C ratio of the membrane electrode, EW value, flow channel design parameters of the bipolar plate, selection parameters of carbon paper, vehicle VIN code, and engine type number. The operational scenario parameters include at least one of the following: engine vehicle model, region, weather conditions, operating conditions, and vehicle information; and... The operational performance characterization data includes at least one of the following: average single-cell voltage drop rate of fuel cell engine, impedance rise rate, and number of times impedance values ​​are low. The step of obtaining historical traceability data of all control parameters and design parameters of the fuel cell engine further includes: Identify all core components that affect the operational status of the fuel cell engine and number them sequentially. Obtain historical monitoring data of the control parameters and design parameters of each core component with a specific number; The historical monitoring data of the above control parameters and design parameters are filtered and feature extracted in sequence to obtain the maximum change range, change rate and importance score of the above control parameters and design parameters under various operational scenario parameters, which are used as historical traceability data.

2. The operation and maintenance control method for a fuel cell engine according to claim 1, characterized in that, The step of extracting the temporal and spatial distribution features of each performance characterization data further includes: The maximum value, minimum value, average change and real-time change rate of each performance characterization data are obtained separately, and the data are fused. The fused result is used as the time distribution feature of each performance characterization data. The concentration, dispersion, skewness and kurtosis of each performance characterization data at a preset location are obtained separately, and the data are fused. The fusion result is used as the spatial distribution feature of each performance characterization data.

3. The operation and maintenance control method for a fuel cell engine according to claim 2, characterized in that, The step of inputting the aforementioned historical data on time distribution characteristics, spatial distribution characteristics, operational scenario parameters, control parameters, and design parameters into a pre-trained information fusion and parameter optimization model to derive the current operational status of the fuel cell engine and the optimal control parameters and optimal design parameters under the operational scenario parameters further includes: Determine the importance ranking of all performance characterization data; Based on the importance ranking of all performance characterization data, time distribution feature matrices and spatial distribution feature matrices of all performance characterization data are constructed respectively. The aforementioned time distribution feature matrix, spatial distribution feature matrix, and historical traceability data of operation scenario parameters, control parameters, and design parameters are input into a pre-trained information fusion and parameter optimization model to obtain the current working status of each core component of the fuel cell engine, the prediction results of whether it may be damaged, and the optimal control parameters and optimal design parameters under the operation scenario parameters.

4. The operation and maintenance control method for a fuel cell engine according to claim 3, characterized in that, It also includes the following steps: Determine the probability of occurrence of parameters for each operational scenario; Based on the optimal design parameters under each operating scenario and the above-mentioned occurrence probabilities, the optimal design parameters for the fuel cell engine under the comprehensive operating conditions of this vehicle are obtained.

5. A maintenance control device for a fuel cell engine, used to implement the maintenance control method for a fuel cell engine according to any one of claims 1-4, characterized in that, include: The scenario parameter acquisition unit is used to acquire the current operating scenario parameters of the fuel cell engine and send them to the cloud platform processing unit. The engine performance data acquisition unit is used to acquire the current operating performance characterization data of the fuel cell engine and send it to the cloud platform processing unit. The cloud platform processing unit is used to acquire historical traceability data of all control parameters and design parameters of the fuel cell engine; and to extract the temporal and spatial distribution characteristics of each performance characterization data. Furthermore, based on the historical traceability data of the aforementioned time distribution characteristics, spatial distribution characteristics, operation scenario parameters, control parameters, and design parameters, respectively input into the pre-trained information fusion and parameter optimization model, the current operation and maintenance status of the fuel cell engine and the optimal control parameters and optimal design parameters are obtained. The current operation and maintenance status and optimal control parameters of the fuel cell engine are sent to the engine control unit, and the operation scenario parameters and optimal design parameters are sent to the computer terminal. An engine control unit is used to display whether key engine components are faulty based on the received maintenance status, and to adjust the operating status of key engine components according to the optimal control parameters. A computer terminal is used to receive and display the operational scenario parameters and optimal design parameters.

6. The operation and maintenance control device for a fuel cell engine according to claim 5, characterized in that, The scene parameter acquisition unit further includes: The GPS navigation unit, with built-in high-precision maps, is installed on the vehicle where the fuel cell engine is located to obtain the current location, weather conditions, and operating status of the fuel cell engine. The vehicle information collection unit is used to acquire the vehicle model and onboard information of the fuel cell engine at the current moment; the onboard information includes the number of passengers, passenger weight, whether fragile items are included, and cargo weight; and... The engine performance data acquisition unit further includes: A fuel cell stack single-cell voltage monitoring device is used to acquire single-cell voltage data and send it to the data processing subunit. A fuel cell impedance measurement device is used to acquire stack impedance data and send it to the data processing subunit. The data processing subunit is used to derive the average voltage drop rate of a single cell based on the voltage data of the individual cells in the fuel cell stack, and to derive the impedance rise rate and the number of times the impedance value is low based on the impedance data of the fuel cell stack.

7. The operation and maintenance control device for a fuel cell engine according to claim 5 or 6, characterized in that, The cloud platform processing unit executes the following program: Obtain historical traceability data of all control parameters and design parameters of the fuel cell engine; Extract the temporal and spatial distribution features of each performance characterization data point; By inputting the historical data of the above-mentioned time distribution characteristics, spatial distribution characteristics, operation scenario parameters, control parameters, and design parameters into the pre-trained information fusion and parameter optimization model, the current operation and maintenance status of the fuel cell engine and the optimal control parameters and optimal design parameters are obtained. The current operating status and optimal control parameters of the fuel cell engine are sent to the engine control unit; and, based on the optimal design parameters under each operating scenario, the optimal design parameters of the fuel cell engine under the comprehensive operating conditions of this vehicle are obtained and sent to the computer terminal.

Images