CN116154239B - Multi-level implementation-based hydrogen fuel cell energy conversion method and device - Google Patents

Multi-level implementation-based hydrogen fuel cell energy conversion method and device Download PDF

Info

Publication number
CN116154239B
CN116154239B CN202310399937.XA CN202310399937A CN116154239B CN 116154239 B CN116154239 B CN 116154239B CN 202310399937 A CN202310399937 A CN 202310399937A CN 116154239 B CN116154239 B CN 116154239B
Authority
CN
China
Prior art keywords
catalytic
target
conversion
model
edge
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310399937.XA
Other languages
Chinese (zh)
Other versions
CN116154239A (en
Inventor
向德
李庆先
刘良江
朱宪宇
王晋威
刘青
左从瑞
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.)
Hunan Institute of Metrology and Test
Original Assignee
Hunan Institute of Metrology and Test
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 Hunan Institute of Metrology and Test filed Critical Hunan Institute of Metrology and Test
Priority to CN202310399937.XA priority Critical patent/CN116154239B/en
Publication of CN116154239A publication Critical patent/CN116154239A/en
Application granted granted Critical
Publication of CN116154239B publication Critical patent/CN116154239B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • H01M8/04992—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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • 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
    • H01M8/04694—Processes for controlling fuel cells or fuel cell systems characterised by variables to be controlled
    • H01M8/04858—Electric variables
    • H01M8/04925—Power, energy, capacity or load
    • 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

Abstract

The invention relates to the technical field of hydrogen battery energy conversion, and discloses a hydrogen fuel battery energy conversion method based on multi-level implementation, which comprises the following steps: data cleaning is carried out on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, feature extraction operation is carried out on each target catalytic tube model diagram in the standard hydrogen battery test data set one by one, and catalytic tube features are obtained; updating the standard hydrogen battery test data set by utilizing the characteristics of the catalytic tube to obtain a test characteristic data set; iteratively updating a preset catalytic conversion power model by using the test characteristic data set to obtain a conversion power analysis model; and obtaining primary battery parameters of the hydrogen battery to be converted, generating a target conversion method according to a conversion power analysis model, and converting energy of the hydrogen battery to be converted according to the target conversion method. The invention also provides a hydrogen fuel cell energy conversion device based on multi-level implementation. The invention can improve the energy efficiency ratio of the energy conversion of the hydrogen battery.

Description

Multi-level implementation-based hydrogen fuel cell energy conversion method and device
Technical Field
The invention relates to the technical field of hydrogen battery energy conversion, in particular to a hydrogen fuel battery energy conversion method and device based on multi-level implementation.
Background
The hydrogen fuel cell is a cell manufactured by utilizing the oxidation discharge reaction of hydrogen and oxygen, is environment-friendly, and is one of the development trends of new energy in the future, and the energy conversion efficiency of the hydrogen fuel cell is also changed due to the difference of the reaction rate and the catalytic material of the hydrogen fuel cell.
The existing energy conversion technology of the hydrogen fuel cell is mostly based on a catalyst replacement hydrogen fuel cell energy conversion method, for example, the material type of a catalyst is changed, a low-platinum catalyst is replaced by a platinum-based catalyst or a non-platinum catalyst, in practical application, elements influencing the energy conversion of the hydrogen fuel cell are more, the influence effect is more complex, and the energy conversion method of the hydrogen fuel cell based on the catalyst replacement cannot meet the multi-stage conversion method, which may result in lower energy efficiency when the energy conversion of the hydrogen fuel cell is performed.
Disclosure of Invention
The invention provides a hydrogen fuel cell energy conversion method and device based on multi-level realization, which mainly aim to solve the problem of lower energy efficiency when the energy conversion of a hydrogen fuel cell is carried out.
In order to achieve the above object, the present invention provides a hydrogen fuel cell energy conversion method based on multi-level implementation, including:
acquiring a historical hydrogen battery test data set, performing data cleaning on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, extracting catalytic layer data sets in the standard hydrogen battery test data set one by one to serve as target catalytic layer data sets, and taking a catalytic tube model diagram in the target catalytic layer data sets as a target catalytic tube model diagram;
performing primary filtering operation on the target catalytic tube model diagram to obtain a filtered catalytic tube model diagram, extracting hierarchical features and type features of the filtered catalytic tube model diagram, and converging the hierarchical features and the type features into catalytic tube features of the target catalytic tube model diagram, wherein the extracting of the hierarchical features and the type features of the filtered catalytic tube model diagram comprises the following steps: and carrying out edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
wherein,refers to the edge strength of the edge of the catalytic tube，/>Is the sign of maximum>Refers to the maximum value of the edge intensity, +. >Refers to absolute value symbols, ">Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function; performing edge binarization operation on the reinforced catalytic tube model graph to obtain a catalytic tube edge graph; extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode; extracting hierarchical features from the straight line pipe edge set, extracting pipeline bending features from the curve pipe edge set, and fusing the pipeline primary features and the pipeline bending features into type features;
updating the target catalytic layer data set by utilizing the catalytic tube characteristics to obtain a target catalytic layer characteristic data set, and updating the standard hydrogen battery test data set by utilizing all the target catalytic layer characteristic data sets to obtain a test characteristic data set;
selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, calculating analysis conversion rate corresponding to the target test characteristic data by using a preset catalytic conversion power model, calculating analysis loss values of the catalytic conversion power model according to all the analysis conversion rates and the test characteristic data set, and iteratively updating the catalytic conversion power model by using the analysis loss values to obtain a conversion power analysis model;
Obtaining primary battery parameters of a hydrogen battery to be converted, analyzing a target conversion rate corresponding to the primary battery parameters according to the conversion power analysis model, generating a conversion parameter set corresponding to the primary battery parameters according to the target conversion rate, generating a target conversion method according to the conversion parameter set, and converting energy of the hydrogen battery to be converted according to the target conversion method.
Optionally, the performing data cleaning on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set includes:
screening repeated historical hydrogen battery test data from the historical hydrogen battery test data set to obtain a primary hydrogen battery test data set;
screening class incomplete data from the primary hydrogen battery test data set according to data classes to obtain a secondary hydrogen battery test data set;
performing data unit normalization operation on each secondary hydrogen battery test data in the secondary hydrogen battery test data set to obtain a standard hydrogen battery test data set;
and screening out value field error data from the standard hydrogen battery test data set according to the data value field to obtain a standard hydrogen battery test data set.
Optionally, the performing a primary filtering operation on the target catalytic tube model diagram to obtain a filtered catalytic tube model diagram includes:
Performing picture inclination correction and picture cutting operation on the target catalytic tube model graph to obtain a corrected catalytic tube model graph;
graying the corrected catalytic tube model graph into a primary gray catalytic tube graph, and generating a gray histogram of the primary gray catalytic tube graph;
performing histogram equalization operation on the primary gray catalytic tube graph by using the gray histogram to obtain a standard gray catalytic tube graph;
and carrying out median filtering operation on the standard gray catalytic tube graph to obtain a filtered catalytic tube model graph.
Optionally, the histogram equalization operation is performed on the primary gray catalytic tube graph by using the gray histogram to obtain a standard gray catalytic tube graph, which includes:
calculating the gray level cumulative distribution probability according to the gray level histogram;
calculating an equalized gray value of each pixel in the primary gray catalytic tube diagram according to the gray accumulated distribution probability;
and mapping the balanced gray values back to the primary gray catalytic tube diagram one by one to obtain a standard gray catalytic tube diagram.
Optionally, the extracting the straight tube edge set and the curved tube edge set from the catalytic tube edge map by using an edge detection method includes:
Extracting a primary linear edge set from the catalytic tube edge map by utilizing a linear convolution kernel, and carrying out edge splitting on the catalytic tube edge map by utilizing the primary linear edge set to obtain a primary curve edge set;
performing linear fitting on each primary linear edge in the primary linear edge set to obtain a linear tube edge set;
and performing curve fitting on each primary curve edge in the primary curve edge set to obtain a curve tube edge set.
Optionally, the calculating, using a preset catalytic conversion power model, the analysis conversion rate corresponding to the target test feature data includes:
performing position coding on the target test characteristic data by using a coding layer of a preset catalytic conversion power model to obtain a target test characteristic code;
generating a multi-head vector set of the target test feature codes, and calculating target test attention features corresponding to the multi-head vector set by using an attention mechanism of the catalytic conversion power model;
and vector decoding is carried out on the target test attention characteristic by using a decoding layer of the catalytic conversion power model, and analysis conversion rate corresponding to the target test characteristic data is calculated by using a greedy layer of the catalytic conversion power model.
Optionally, the calculating the analysis loss value of the catalytic conversion power model according to all analysis conversion rates and the test characteristic data set includes:
selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, and calculating the real conversion rate corresponding to the target test characteristic data by using a preset hydrogen energy conversion rate algorithm;
collecting the analysis conversion rate and the real conversion rate corresponding to the target test characteristic data into a target conversion rate group, and collecting all target conversion rate groups into a target conversion rate group set;
calculating an analytical loss value of the catalytic conversion power model according to the target conversion rate group set by using the following analytical loss value algorithm:
；
wherein,refers to the analytical loss value, +.>Refers to->Personal (S)>Refers to the total number of target conversion groups in the target conversion group, +.>Means that the target transformation ratio set is concentrated +.>True conversion of the individual target conversion groups,/->Means that the target transformation ratio set is concentrated +.>Analytical conversion for each of the target conversion groups.
Optionally, the calculating, by using a preset hydrogen energy conversion rate algorithm, the actual conversion rate corresponding to the target test feature data includes:
Utilizing the fan blade number, the fan blade diameter, the fan blade angle, the oxygen fan wind pressure, the catalyst type, the conversion duration, the heat dissipation water pump power, the hot water conversion electric quantity and the hydrogen fuel conversion electric quantity in the target test characteristic data to form a target power data set;
and calculating the real conversion rate corresponding to the target test characteristic data according to the target power data set by using the following hydrogen energy conversion rate algorithm:
；
wherein,means the true conversion corresponding to the target test characteristic data,/for>Is a circumference rate symbol, < >>Means the transformation duration,/->Refers to the diameter of the fan blade, < >>Means the air pressure of the oxygen fan, +.>Is a preset flow coefficient, +.>Means the hydrogen fuel conversion level, +.>Means that the hot water converts electric quantity, +.>Means the power of the heat-dissipating water pump, < >>Means the fan blade angle, +.>Refers to a preset working coefficient, < + >>Means the number of fan blades,/->Is a preset electrical loss coefficient, +.>Is a preset mechanical loss coefficient.
Optionally, the iteratively updating the catalytic conversion power model by using the analysis loss value to obtain a conversion power analysis model includes:
judging whether the analysis loss value is larger than a preset loss value threshold value or not;
If yes, updating the model parameters of the catalytic conversion power model by using the analysis loss value, and returning to the step of calculating the analysis conversion rate corresponding to the target test characteristic data by using the preset catalytic conversion power model;
and if not, taking the updated catalytic conversion power model as a conversion power analysis model.
In order to solve the above problems, the present invention further provides a hydrogen fuel cell energy conversion device based on multi-level implementation, the device comprising:
the data cleaning module is used for acquiring a historical hydrogen battery test data set, cleaning the data of the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, extracting catalytic layer data sets in the standard hydrogen battery test data set one by one to serve as target catalytic layer data sets, and taking a catalytic tube model diagram in the target catalytic layer data sets as a target catalytic tube model diagram;
the feature extraction module is configured to perform a primary filtering operation on the target catalytic tube model map to obtain a filtered catalytic tube model map, extract a hierarchical feature and a type feature of the filtered catalytic tube model map, and integrate the hierarchical feature and the type feature into a catalytic tube feature of the target catalytic tube model map, where the extracting the hierarchical feature and the type feature of the filtered catalytic tube model map includes: and carrying out edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
Wherein,means the edge strength of the edge of the catalytic tube, < >>Is the sign of maximum>Refers to the maximum value of the edge intensity, +.>Refers to absolute value symbols, ">Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function; performing edge binarization operation on the reinforced catalytic tube model graph to obtain a catalytic tube edge graph; extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode; extracting hierarchical features from the straight line pipe edge set, extracting pipeline bending features from the curve pipe edge set, and fusing the pipeline primary features and the pipeline bending features into type features;
the data updating module is used for updating the target catalytic layer data set by utilizing the catalytic tube characteristics to obtain a target catalytic layer characteristic data set, and updating the standard hydrogen battery test data set by utilizing all the target catalytic layer characteristic data sets to obtain a test characteristic data set;
The model training module is used for selecting test feature data in the test feature data set one by one as target test feature data, calculating analysis conversion rate corresponding to the target test feature data by using a preset catalytic conversion power model, calculating analysis loss values of the catalytic conversion power model according to all the analysis conversion rates and the test feature data set, and performing iterative updating on the catalytic conversion power model by using the analysis loss values to obtain a conversion power analysis model;
the energy conversion module is used for obtaining primary battery parameters of the hydrogen battery to be converted, analyzing target conversion rate corresponding to the primary battery parameters according to the conversion power analysis model, generating a conversion parameter set corresponding to the primary battery parameters according to the target conversion rate, generating a target conversion method according to the conversion parameter set, and carrying out energy conversion on the hydrogen battery to be converted according to the target conversion method.
According to the embodiment of the invention, the historical hydrogen battery test data set is obtained and subjected to data cleaning to obtain the standard hydrogen battery test data set, error and repeated data can be removed from the historical hydrogen battery test data set, so that the accuracy of a subsequent conversion power analysis model is improved, the primary filtering operation is carried out on the target catalytic tube model diagram to obtain the filtered catalytic tube model diagram, more catalytic tube detail characteristics can be reserved, the hierarchical characteristics and the type characteristics of the filtered catalytic tube model diagram are respectively extracted, the hierarchical characteristics and the type characteristics are collected into the catalytic tube characteristics of the target catalytic tube model diagram, the hierarchical characteristics and the type characteristics of the catalytic tube can be subjected to numerical conversion, the subsequent training of the catalytic conversion power model is facilitated, the standard hydrogen battery test data set is updated by utilizing all target catalytic layer characteristic data sets to obtain the test characteristic data set, and the normalized conversion of each data type in the test characteristic data set can be realized, so that the training efficiency is improved.
The analysis conversion rate corresponding to the target test characteristic data is calculated by utilizing a preset catalytic conversion power model, the analysis loss value of the catalytic conversion power model is calculated according to all the analysis conversion rates and the test characteristic data set, the analysis loss value is utilized to update the catalytic conversion power model in an iterative manner to obtain a conversion power analysis model, training of the catalytic conversion power model can be completed, the relation between the test parameters in various test characteristic data and the hydrogen energy conversion rate is obtained, the optimal energy conversion method is convenient to determine subsequently, the primary battery parameters of the hydrogen battery to be converted are obtained, the target conversion rate corresponding to the primary battery parameters is analyzed according to the conversion power analysis model, the conversion parameter set corresponding to the primary battery parameters is generated according to the target conversion rate, the target conversion method is generated according to the conversion parameter set, the energy conversion is carried out on the hydrogen battery to be converted according to the target conversion method, and the corresponding battery parameters when the hydrogen energy conversion rate is maximum can be determined, so that the energy conversion rate of the hydrogen fuel battery is improved. Therefore, the method and the device for converting the energy of the hydrogen fuel cell based on multi-level implementation can solve the problem that the energy efficiency is lower when the energy conversion of the hydrogen fuel cell is carried out.
Drawings
FIG. 1 is a schematic flow chart of a method for converting energy of a hydrogen fuel cell based on a multi-level implementation according to an embodiment of the present application;
FIG. 2 is a flow chart of extracting hierarchical features and type features according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of calculating analytical conversion rate according to an embodiment of the present application;
FIG. 4 is a functional block diagram of a hydrogen fuel cell energy conversion device based on a multi-level implementation according to an embodiment of the present application;
the achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides a hydrogen fuel cell energy conversion method based on multi-level implementation. The implementation main body of the hydrogen fuel cell energy conversion method based on multi-level implementation includes, but is not limited to, at least one of a server, a terminal and the like, which can be configured to implement the method provided by the embodiment of the application. In other words, the hydrogen fuel cell energy conversion method based on multi-level implementation may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a method for converting energy of a hydrogen fuel cell based on multi-level implementation according to an embodiment of the present invention is shown. In this embodiment, the method for converting energy of a hydrogen fuel cell based on multi-level implementation includes:
s1, acquiring a historical hydrogen battery test data set, performing data cleaning on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, extracting catalytic layer data sets in the standard hydrogen battery test data set one by one to serve as target catalytic layer data sets, and taking a catalytic tube model diagram in the target catalytic layer data sets as a target catalytic tube model diagram.
In the embodiment of the present invention, the historical hydrogen cell test data set is a data set composed of a plurality of historical hydrogen cell test data, and each of the historical hydrogen cell test data includes all relevant data of a primary hydrogen cell energy conversion test, such as data of a fan blade number, a fan blade diameter, a fan blade angle, a hydrogen fan wind pressure, an oxygen fan wind pressure, a catalyst type, a conversion duration, a catalytic pipeline model diagram, a heat dissipating water pump power, a hot water conversion electric quantity, a hydrogen fuel conversion electric quantity, and the like.
The fan blade number refers to the number of fan blades of the hydrogen fan and the oxygen fan, the fan blade diameter refers to the fan blade radius of the fans in the hydrogen gas supply pipeline and the oxygen gas supply pipeline, the fan blade angle refers to the inclination angle of the fan blade, the hydrogen fan wind pressure refers to the wind power generated by the hydrogen fan on the unit area, the oxygen fan wind pressure refers to the wind power generated by the oxygen fan on the unit area, the catalyst type refers to the type of catalyst for the water reverse reaction of the hydrogen fuel cell, the conversion time refers to the total conversion time of the electric power converted by the hydrogen fuel cell, the catalytic pipeline model diagram refers to a model diagram of a catalytic pipeline in a catalytic layer corresponding to the oxidation reaction of the historical hydrogen cell test data, the heat dissipation water pump power refers to the power of a heat dissipation water pump of a heat dissipation system corresponding to the historical hydrogen cell test data, the heat dissipation water conversion power refers to the total conversion fuel converted by a thermoelectric device in the system of the hydrogen cell test data in one-time test, and the conversion time refers to the total heat dissipation power generated by the hydrogen fuel cell test data.
In the embodiment of the present invention, the step of performing data cleaning on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set includes:
screening repeated historical hydrogen battery test data from the historical hydrogen battery test data set to obtain a primary hydrogen battery test data set;
screening class incomplete data from the primary hydrogen battery test data set according to data classes to obtain a secondary hydrogen battery test data set;
performing data unit normalization operation on each secondary hydrogen battery test data in the secondary hydrogen battery test data set to obtain a standard hydrogen battery test data set;
and screening out value field error data from the standard hydrogen battery test data set according to the data value field to obtain a standard hydrogen battery test data set.
In detail, the type incomplete data refers to primary hydrogen battery test data with incomplete data types in the primary hydrogen battery test data set, for example, the primary hydrogen battery test data with missing fan blade numbers, the data unit normalization operation is performed on each secondary hydrogen battery test data in the secondary hydrogen battery test data set, the standard hydrogen battery test data set is obtained by performing unified unit conversion on data units of various data in each secondary hydrogen battery test data, for example, the unit of total conversion duration is unified into hours, and the value range error data refers to standard hydrogen battery test data with at least one data in a data value range of the data in which the data value range is exceeded, namely, the standard hydrogen battery test data in various data of the standard hydrogen battery test data, for example, the standard hydrogen battery test data with the hydrogen fan wind pressure not being twice that of the oxygen fan wind pressure.
In the embodiment of the invention, the historical hydrogen battery test data set is obtained and subjected to data cleaning to obtain the standard hydrogen battery test data set, so that the error and repeated data can be removed from the historical hydrogen battery test data set, and the accuracy of a subsequent conversion power analysis model is improved.
S2, performing primary filtering operation on the target catalytic tube model diagram to obtain a filtered catalytic tube model diagram, respectively extracting hierarchical features and type features of the filtered catalytic tube model diagram, and converging the hierarchical features and the type features into catalytic tube features of the target catalytic tube model diagram.
In the embodiment of the invention, the catalytic tube in the target catalytic tube model diagram is a reaction carrier for releasing hydrogen and oxygen to enable the hydrogen and the oxygen to perform electrochemical reaction, and the length, the layer and the shape of the catalytic tube have great influence on the electrochemical reaction.
In the embodiment of the present invention, the performing a primary filtering operation on the target catalytic tube model diagram to obtain a filtered catalytic tube model diagram includes:
performing picture inclination correction and picture cutting operation on the target catalytic tube model graph to obtain a corrected catalytic tube model graph;
Graying the corrected catalytic tube model graph into a primary gray catalytic tube graph, and generating a gray histogram of the primary gray catalytic tube graph;
performing histogram equalization operation on the primary gray catalytic tube graph by using the gray histogram to obtain a standard gray catalytic tube graph;
and carrying out median filtering operation on the standard gray catalytic tube graph to obtain a filtered catalytic tube model graph.
In detail, the inclination angle of the target catalytic tube model can be identified by using hough transformation, and picture inclination correction is performed according to the inclination angle, and the picture cutting operation refers to cutting a region of interest of the target catalytic tube model picture subjected to inclination correction, and performing picture stretching to obtain a corrected catalytic tube model picture.
In the embodiment of the present invention, the histogram equalization operation is performed on the primary gray scale catalytic tube map by using the gray scale histogram to obtain a standard gray scale catalytic tube map, including:
calculating the gray level cumulative distribution probability according to the gray level histogram;
calculating an equalized gray value of each pixel in the primary gray catalytic tube diagram according to the gray accumulated distribution probability;
and mapping the balanced gray values back to the primary gray catalytic tube diagram one by one to obtain a standard gray catalytic tube diagram.
In detail, the gray histogram of the primary gray catalytic pipe map may be generated by calculating the duty ratio of pixels of respective gray levels in the primary gray catalytic pipe map.
In the embodiment of the present invention, referring to fig. 2, the extracting the hierarchical feature and the type feature of the filtering catalytic tube model map respectively includes:
s21, performing edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
wherein,means the edge strength of the edge of the catalytic tube, < >>Is the sign of maximum>Refers to the maximum value of the edge intensity, +.>Refers to absolute value symbols, ">Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function;
s22, performing edge binarization operation on the reinforced catalytic tube model diagram to obtain a catalytic tube edge diagram;
s23, extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode;
S24, extracting hierarchical features from the edge set of the straight pipe, extracting pipeline bending features from the edge set of the curve pipe, and fusing the pipeline primary features and the pipeline bending features into type features.
In detail, the edge intensity algorithm is utilized to carry out edge enhancement on the filtering catalysis model diagram to obtain an enhanced catalysis tube model diagram, and gray pixel enhancement can be carried out on the edge of the catalysis tube model according to gray standard deviation of each area in the filtering catalysis model diagram, so that more pipeline details are reserved.
In detail, performing edge binarization operation on the enhanced catalytic tube model graph to obtain a catalytic tube edge graph, namely changing the gray value of a pixel point with a gray value smaller than a preset gray threshold value into 0 and changing the gray value of a pixel point with a gray value larger than or equal to the gray threshold value into 255, wherein extracting the primary characteristic of the pipeline from the catalytic tube edge graph is to extract the dimension reduction characteristic of the catalytic tube edge graph by using a preset convolution layer and a pooling layer.
In detail, the extracting the primary pipeline characteristics from the catalytic tube edge map refers to rolling and normalizing the catalytic tube edge map to obtain the primary pipeline characteristics.
In detail, the method for extracting the straight tube edge set and the curve tube edge set from the catalytic tube edge map by using the edge detection method comprises the following steps:
extracting a primary linear edge set from the catalytic tube edge map by utilizing a linear convolution kernel, and carrying out edge splitting on the catalytic tube edge map by utilizing the primary linear edge set to obtain a primary curve edge set;
performing linear fitting on each primary linear edge in the primary linear edge set to obtain a linear tube edge set;
and performing curve fitting on each primary curve edge in the primary curve edge set to obtain a curve tube edge set.
In detail, a primary linear edge set may be extracted from the catalytic tube edge map by using a hough or linear operator and a linear convolution kernel, a linear fitting may be performed on each primary linear edge in the primary linear edge set by using a column Ma Suanfa to obtain a linear tube edge set, and a curve fitting may be performed on each primary curve edge in the primary curve edge set by using a least square method to obtain a curve tube edge set.
In the embodiment of the invention, the primary filtering operation is carried out on the target catalytic tube model diagram to obtain the filtered catalytic tube model diagram, more catalytic tube detail characteristics can be reserved, the hierarchical characteristics and the type characteristics of the filtered catalytic tube model diagram are respectively extracted, and are collected into the catalytic tube characteristics of the target catalytic tube model diagram, so that the hierarchical characteristics and the type characteristics of the catalytic tube can be converted in numerical value, and the subsequent training of a catalytic conversion power model is facilitated.
And S3, updating the target catalytic layer data set by utilizing the catalytic tube characteristics to obtain a target catalytic layer characteristic data set, and updating the standard hydrogen battery test data set by utilizing all the target catalytic layer characteristic data sets to obtain a test characteristic data set.
In the embodiment of the invention, the step of updating the target catalytic layer data set by using the catalytic tube characteristics to obtain the target catalytic layer characteristic data set refers to the step of replacing the hierarchical characteristics and the type characteristics in the catalytic tube characteristics with the target catalytic tube model diagrams in the target catalytic layer data set to obtain the target catalytic layer characteristic data set, the step of updating the standard hydrogen cell test data set by using all the target catalytic layer characteristic data sets to obtain the test characteristic data set refers to the step of updating the corresponding standard hydrogen cell test data one by using the target catalytic layer characteristic data set to obtain the test characteristic data set.
In the embodiment of the invention, the standard hydrogen battery test data set is updated by utilizing all the target catalytic layer characteristic data sets to obtain the test characteristic data set, and the normalized conversion of each data type in the test characteristic data set can be realized, so that the training efficiency is improved.
S4, selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, calculating analysis conversion rate corresponding to the target test characteristic data by using a preset catalytic conversion power model, calculating analysis loss values of the catalytic conversion power model according to all the analysis conversion rates and the test characteristic data set, and performing iterative updating on the catalytic conversion power model by using the analysis loss values to obtain a conversion power analysis model.
In the embodiment of the invention, the catalytic conversion power model may be a transducer network model with a self-attention mechanism introduced, the analysis conversion rate refers to a ratio of hydrogen and oxygen volumes to net-growth electric energy corresponding to target test feature data calculated by the catalytic conversion power model, wherein the net-growth electric energy refers to net-income electric energy obtained by subtracting heat dissipation water pump energy consumption, hydrogen fan energy consumption and oxygen fan energy consumption after adding the hydrogen fuel conversion electric quantity to the hot water conversion electric quantity.
In the embodiment of the present invention, referring to fig. 3, the calculating, by using a preset catalytic conversion power model, the analysis conversion rate corresponding to the target test feature data includes:
S31, performing position coding on the target test characteristic data by using a coding layer of a preset catalytic conversion power model to obtain a target test characteristic code;
s32, generating a multi-head vector set of the target test feature codes, and calculating target test attention features corresponding to the multi-head vector set by using an attention mechanism of the catalytic conversion power model;
and S33, vector decoding is carried out on the target test attention characteristic by using a decoding layer of the catalytic conversion power model, and analysis conversion rate corresponding to the target test characteristic data is calculated by using a greedy layer of the catalytic conversion power model.
In detail, the coding layer may be an Embedding function of a Tensorflow, and the position coding of the target test feature data is performed by using a coding layer of a preset catalytic conversion power model, and the target test feature coding is performed by coding oxygen fan power, fan blade number, fan blade diameter, fan blade angle, oxygen fan wind pressure, catalyst type, conversion duration, hierarchical feature and type feature and cooling water pump power in the target test feature data to obtain a target test feature coding; the generating the multi-head vector set of the target test feature code refers to generating a query vector, a key vector and a value vector of the target test feature code, and collecting the query vector, the key vector and the value vector into the multi-head vector set, and the decoding layer may be a Decoder of a transformer.
In the embodiment of the present invention, the calculating the analysis loss value of the catalytic conversion power model according to all the analysis conversion rates and the test feature data set includes:
selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, and calculating the real conversion rate corresponding to the target test characteristic data by using a preset hydrogen energy conversion rate algorithm;
collecting the analysis conversion rate and the real conversion rate corresponding to the target test characteristic data into a target conversion rate group, and collecting all target conversion rate groups into a target conversion rate group set;
calculating an analytical loss value of the catalytic conversion power model according to the target conversion rate group set by using the following analytical loss value algorithm:
；
wherein,refers to the analytical loss value, +.>Refers to->Personal (S)>Refers to the total number of target conversion groups in the target conversion group, +.>Means that the target transformation ratio set is concentrated +.>True conversion of the individual target conversion groups,/->Means that the target transformation ratio set is concentrated +.>Analytical conversion for each of the target conversion groups.
In the embodiment of the invention, the analysis loss value of the catalytic conversion power model is calculated according to the target conversion rate group set by utilizing the analysis loss value algorithm, so that the loss value of each test characteristic data in the test characteristic data set can be counted, and the characterization of the analysis loss value is improved.
Specifically, the calculating, by using a preset hydrogen energy conversion rate algorithm, the actual conversion rate corresponding to the target test feature data includes:
utilizing the fan blade number, the fan blade diameter, the fan blade angle, the oxygen fan wind pressure, the catalyst type, the conversion duration, the heat dissipation water pump power, the hot water conversion electric quantity and the hydrogen fuel conversion electric quantity in the target test characteristic data to form a target power data set;
and calculating the real conversion rate corresponding to the target test characteristic data according to the target power data set by using the following hydrogen energy conversion rate algorithm:
；
wherein,means the true conversion corresponding to the target test characteristic data,/for>Is a circumference rate symbol, < >>Means the transformation duration,/->Refers to the diameter of the fan blade, < >>Means the air pressure of the oxygen fan, +.>Is a preset flow coefficient, +.>Means the hydrogen fuel conversion level, +.>Means that the hot water converts electric quantity, +.>Means the power of the heat-dissipating water pump, < >>Means the fan blade angle, +.>Refers to a preset working coefficient, < + >>Means the number of fan blades,/->Is a preset electrical loss coefficient, +.>Is a preset mechanical loss coefficient.
In the embodiment of the invention, the real conversion rate corresponding to the target test characteristic data is calculated according to the target power data set by utilizing the hydrogen energy conversion rate algorithm, so that the energy consumption of the heat dissipation water pump and the heat energy conversion influence of hot water can be removed, the ratio of hydrogen, oxygen volume and conversion energy can be analyzed more accurately, and the more accurate hydrogen energy conversion rate can be obtained.
In detail, the step of iteratively updating the catalytic conversion power model by using the analysis loss value to obtain a conversion power analysis model includes:
judging whether the analysis loss value is larger than a preset loss value threshold value or not;
if yes, updating the model parameters of the catalytic conversion power model by using the analysis loss value, and returning to the step of calculating the analysis conversion rate corresponding to the target test characteristic data by using the preset catalytic conversion power model;
and if not, taking the updated catalytic conversion power model as a conversion power analysis model.
In detail, model parameters of the catalytic conversion power model may be updated according to the analysis loss value using a gradient descent algorithm.
In the embodiment of the invention, the analysis conversion rate corresponding to the target test characteristic data is calculated by using the preset catalytic conversion power model, the analysis loss value of the catalytic conversion power model is calculated according to all the analysis conversion rates and the test characteristic data set, the catalytic conversion power model is iteratively updated by using the analysis loss value to obtain the conversion power analysis model, and the training of the catalytic conversion power model can be completed, so that the relation between the test parameters in various test characteristic data and the hydrogen energy conversion rate is obtained, and the subsequent determination of the optimal energy conversion method is facilitated.
S5, acquiring primary battery parameters of the hydrogen battery to be converted, analyzing target conversion rate corresponding to the primary battery parameters according to the conversion power analysis model, generating a conversion parameter set corresponding to the primary battery parameters according to the target conversion rate, generating a target conversion method according to the conversion parameter set, and converting energy of the hydrogen battery to be converted according to the target conversion method.
In the embodiment of the invention, the primary battery parameters refer to a plurality of data in the data such as the number of fan blades, the diameter of the fan blades, the angle of the fan blades, the wind pressure of the hydrogen fan, the wind pressure of the oxygen fan, the type of the catalyst, the conversion duration, the catalytic pipeline model diagram, the power of the heat dissipating water pump, the conversion electric quantity of hot water, the conversion electric quantity of hydrogen fuel and the like of the hydrogen battery to be converted.
In detail, the analysis of the target conversion rate corresponding to the primary battery parameter according to the conversion power analysis model refers to the combination and replenishment of the primary battery parameter by using a simulated annealing method, and the obtained maximum analysis conversion rate is taken as the target conversion rate, the generation of the conversion parameter set corresponding to the primary battery parameter according to the target conversion rate refers to the generation of the test characteristic data corresponding to the target conversion rate as the conversion parameter set corresponding to the primary battery parameter, and the generation of the target conversion method according to the conversion parameter set refers to the setting of the battery parameters such as the number of fan blades, the diameter of fan blades, the angle of fan blades, the wind pressure of a hydrogen fan, the wind pressure of an oxygen fan, the type of catalyst, the conversion duration, a catalytic pipeline model diagram, the power of a heat dissipation water pump, the conversion power of hot water, the conversion power of hydrogen fuel and the like of the hydrogen battery to be converted according to the conversion parameter set.
In the embodiment of the invention, the primary battery parameters of the hydrogen battery to be converted are obtained, the target conversion rate corresponding to the primary battery parameters is analyzed according to the conversion power analysis model, the conversion parameter group corresponding to the primary battery parameters is generated according to the target conversion rate, the target conversion method is generated according to the conversion parameter group, and the energy conversion is carried out on the hydrogen battery to be converted according to the target conversion method, so that the corresponding battery parameters when the hydrogen energy conversion rate is maximum can be determined, and the energy conversion rate of the hydrogen fuel battery is improved.
According to the embodiment of the invention, the historical hydrogen battery test data set is obtained and subjected to data cleaning to obtain the standard hydrogen battery test data set, error and repeated data can be removed from the historical hydrogen battery test data set, so that the accuracy of a subsequent conversion power analysis model is improved, the primary filtering operation is carried out on the target catalytic tube model diagram to obtain the filtered catalytic tube model diagram, more catalytic tube detail characteristics can be reserved, the hierarchical characteristics and the type characteristics of the filtered catalytic tube model diagram are respectively extracted, the hierarchical characteristics and the type characteristics are collected into the catalytic tube characteristics of the target catalytic tube model diagram, the hierarchical characteristics and the type characteristics of the catalytic tube can be subjected to numerical conversion, the subsequent training of the catalytic conversion power model is facilitated, the standard hydrogen battery test data set is updated by utilizing all target catalytic layer characteristic data sets to obtain the test characteristic data set, and the normalized conversion of each data type in the test characteristic data set can be realized, so that the training efficiency is improved.
The analysis conversion rate corresponding to the target test characteristic data is calculated by utilizing a preset catalytic conversion power model, the analysis loss value of the catalytic conversion power model is calculated according to all the analysis conversion rates and the test characteristic data set, the analysis loss value is utilized to update the catalytic conversion power model in an iterative manner to obtain a conversion power analysis model, training of the catalytic conversion power model can be completed, the relation between the test parameters in various test characteristic data and the hydrogen energy conversion rate is obtained, the optimal energy conversion method is convenient to determine subsequently, the primary battery parameters of the hydrogen battery to be converted are obtained, the target conversion rate corresponding to the primary battery parameters is analyzed according to the conversion power analysis model, the conversion parameter set corresponding to the primary battery parameters is generated according to the target conversion rate, the target conversion method is generated according to the conversion parameter set, the energy conversion is carried out on the hydrogen battery to be converted according to the target conversion method, and the corresponding battery parameters when the hydrogen energy conversion rate is maximum can be determined, so that the energy conversion rate of the hydrogen fuel battery is improved. Therefore, the hydrogen fuel cell energy conversion method based on multi-level implementation can solve the problem of lower energy efficiency when the hydrogen fuel cell energy conversion is carried out.
Fig. 4 is a functional block diagram of a hydrogen fuel cell energy conversion device according to an embodiment of the present invention.
The hydrogen fuel cell energy conversion device 100 based on multi-level implementation of the present invention can be installed in an electronic device. Depending on the functions implemented, the hydrogen fuel cell energy conversion device 100 based on multi-level implementation may include a data cleansing module 101, a feature extraction module 102, a data update module 103, a model training module 104, and an energy conversion module 105. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the data cleaning module 101 is configured to obtain a historical hydrogen cell test data set, perform data cleaning on the historical hydrogen cell test data set to obtain a standard hydrogen cell test data set, extract catalytic layer data sets in the standard hydrogen cell test data set one by one as a target catalytic layer data set, and use a catalytic tube model diagram in the target catalytic layer data set as a target catalytic tube model diagram;
The feature extraction module 102 is configured to perform a primary filtering operation on the target catalytic tube model map to obtain a filtered catalytic tube model map, extract a hierarchical feature and a type feature of the filtered catalytic tube model map, and aggregate the hierarchical feature and the type feature into a catalytic tube feature of the target catalytic tube model map, where the extracting the hierarchical feature and the type feature of the filtered catalytic tube model map includes: and carrying out edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
wherein,means the edge strength of the edge of the catalytic tube, < >>Is the sign of maximum>Refers to the maximum value of the edge intensity, +.>Refers to absolute value symbols, ">Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function; performing edge binarization operation on the reinforced catalytic tube model graph to obtain a catalytic tube edge graph; extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode; extracting hierarchical features from the straight tube edge set, extracting pipeline bending features from the curved tube edge set, and bending the pipeline primary features and the pipeline Fusing the features into type features;
the data updating module 103 is configured to update the target catalytic layer data set by using the catalytic tube feature to obtain a target catalytic layer feature data set, and update the standard hydrogen cell test data set by using all the target catalytic layer feature data sets to obtain a test feature data set;
the model training module 104 is configured to select test feature data in the test feature data set one by one as target test feature data, calculate an analysis conversion rate corresponding to the target test feature data by using a preset catalytic conversion power model, calculate an analysis loss value of the catalytic conversion power model according to all the analysis conversion rates and the test feature data set, and iteratively update the catalytic conversion power model by using the analysis loss value to obtain a conversion power analysis model;
the energy conversion module 105 is configured to obtain a primary battery parameter of a hydrogen battery to be converted, analyze a target conversion rate corresponding to the primary battery parameter according to the conversion power analysis model, generate a conversion parameter set corresponding to the primary battery parameter according to the target conversion rate, generate a target conversion method according to the conversion parameter set, and perform energy conversion on the hydrogen battery to be converted according to the target conversion method.
In detail, each module in the hydrogen fuel cell energy conversion device 100 based on multi-level implementation in the embodiment of the present invention adopts the same technical means as the above-mentioned hydrogen fuel cell energy conversion method based on multi-level implementation in fig. 1 to 3, and can produce the same technical effects, which are not described herein.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Multiple units or means as set forth in the system embodiments may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims (7)

1. A hydrogen fuel cell energy conversion method based on multi-level implementation, the method comprising:
s1: acquiring a historical hydrogen battery test data set, performing data cleaning on the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, extracting catalytic layer data sets in the standard hydrogen battery test data set one by one to serve as target catalytic layer data sets, and taking a catalytic tube model diagram in the target catalytic layer data sets as a target catalytic tube model diagram;
s2: performing primary filtering operation on the target catalytic tube model diagram to obtain a filtered catalytic tube model diagram, extracting hierarchical features and type features of the filtered catalytic tube model diagram, and converging the hierarchical features and the type features into catalytic tube features of the target catalytic tube model diagram, wherein the extracting of the hierarchical features and the type features of the filtered catalytic tube model diagram comprises the following steps:
S21: and carrying out edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
wherein,means the edge strength of the edge of the catalytic tube, < >>Is the sign of the maximum value，/>Refers to the maximum value of the edge intensity, +.>Refers to absolute value symbols, ">Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function;
s22: performing edge binarization operation on the reinforced catalytic tube model graph to obtain a catalytic tube edge graph;
s23: extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode;
s24: extracting hierarchical features from the straight line pipe edge set, extracting pipeline bending features from the curve pipe edge set, and fusing the pipeline primary features and the pipeline bending features into type features;
s3: updating the target catalytic layer data set by utilizing the catalytic tube characteristics to obtain a target catalytic layer characteristic data set, and updating the standard hydrogen battery test data set by utilizing all the target catalytic layer characteristic data sets to obtain a test characteristic data set;
S4: selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, calculating analysis conversion rate corresponding to the target test characteristic data by using a preset catalytic conversion power model, calculating analysis loss values of the catalytic conversion power model according to all the analysis conversion rates and the test characteristic data set, and iteratively updating the catalytic conversion power model by using the analysis loss values to obtain a conversion power analysis model;
s5: obtaining primary battery parameters of a hydrogen battery to be converted, analyzing a target conversion rate corresponding to the primary battery parameters according to the conversion power analysis model, generating a conversion parameter set corresponding to the primary battery parameters according to the target conversion rate, generating a target conversion method according to the conversion parameter set, and converting energy of the hydrogen battery to be converted according to the target conversion method;
wherein said calculating an analytical loss value for said catalytic conversion power model based on all analytical conversions and said test signature data set comprises:
selecting test characteristic data in the test characteristic data set one by one as target test characteristic data, and calculating the real conversion rate corresponding to the target test characteristic data by using a preset hydrogen energy conversion rate algorithm;
Collecting the analysis conversion rate and the real conversion rate corresponding to the target test characteristic data into a target conversion rate group, and collecting all target conversion rate groups into a target conversion rate group set;
calculating an analytical loss value of the catalytic conversion power model according to the target conversion rate group set by using the following analytical loss value algorithm:
；
wherein,refers to the analytical loss value, +.>Refers to->Personal (S)>Refers to the total number of target conversion groups in the target conversion group, +.>Means that the target transformation ratio set is concentrated +.>True conversion of the individual target conversion groups,/->Means that the target transformation ratio set is concentrated +.>Analytical conversions of the target conversion groups;
the calculating the real conversion rate corresponding to the target test characteristic data by using a preset hydrogen energy conversion rate algorithm comprises the following steps:
utilizing the fan blade number, the fan blade diameter, the fan blade angle, the oxygen fan wind pressure, the catalyst type, the conversion duration, the heat dissipation water pump power, the hot water conversion electric quantity and the hydrogen fuel conversion electric quantity in the target test characteristic data to form a target power data set;
and calculating the real conversion rate corresponding to the target test characteristic data according to the target power data set by using the following hydrogen energy conversion rate algorithm:
；
Wherein,means the true conversion corresponding to the target test characteristic data,/for>Is a circumference rate symbol, < >>Means the transformation duration,/->Refers to the diameter of the fan blade, < >>Means the air pressure of the oxygen fan, +.>Is a preset flow coefficient, +.>Means the hydrogen fuel conversion level, +.>Means that the hot water converts electric quantity, +.>Means the power of the heat-dissipating water pump, < >>Means the fan blade angle, +.>Refers to a preset working coefficient, < + >>Means the number of fan blades,/->Is a preset electrical loss coefficient, +.>Is a preset mechanical loss coefficient;
and iteratively updating the catalytic conversion power model by using the analysis loss value to obtain a conversion power analysis model, wherein the method comprises the following steps of:
judging whether the analysis loss value is larger than a preset loss value threshold value or not;
if yes, updating the model parameters of the catalytic conversion power model by using the analysis loss value, and returning to the step of calculating the analysis conversion rate corresponding to the target test characteristic data by using the preset catalytic conversion power model;
and if not, taking the updated catalytic conversion power model as a conversion power analysis model.
2. The multi-level implementation-based hydrogen fuel cell energy conversion method according to claim 1, wherein the performing data cleaning on the historical hydrogen cell test data set to obtain a standard hydrogen cell test data set comprises:
Screening repeated historical hydrogen battery test data from the historical hydrogen battery test data set to obtain a primary hydrogen battery test data set;
screening class incomplete data from the primary hydrogen battery test data set according to data classes to obtain a secondary hydrogen battery test data set;
performing data unit normalization operation on each secondary hydrogen battery test data in the secondary hydrogen battery test data set to obtain a standard hydrogen battery test data set;
and screening out value field error data from the standard hydrogen battery test data set according to the data value field to obtain a standard hydrogen battery test data set.
3. The method for converting energy of hydrogen fuel cell based on multi-level implementation of claim 1, wherein said performing primary filtering operation on said target catalytic tube model map to obtain a filtered catalytic tube model map comprises:
performing picture inclination correction and picture cutting operation on the target catalytic tube model graph to obtain a corrected catalytic tube model graph;
graying the corrected catalytic tube model graph into a primary gray catalytic tube graph, and generating a gray histogram of the primary gray catalytic tube graph;
performing histogram equalization operation on the primary gray catalytic tube graph by using the gray histogram to obtain a standard gray catalytic tube graph;
And carrying out median filtering operation on the standard gray catalytic tube graph to obtain a filtered catalytic tube model graph.
4. The method for converting energy of a hydrogen fuel cell based on multi-level implementation of claim 3, wherein performing histogram equalization operation on the primary gray scale catalytic tube map by using the gray scale histogram to obtain a standard gray scale catalytic tube map comprises:
calculating the gray level cumulative distribution probability according to the gray level histogram;
calculating an equalized gray value of each pixel in the primary gray catalytic tube diagram according to the gray accumulated distribution probability;
and mapping the balanced gray values back to the primary gray catalytic tube diagram one by one to obtain a standard gray catalytic tube diagram.
5. The multi-level implementation-based hydrogen fuel cell energy conversion method according to claim 1, wherein the extracting the straight line tube edge set and the curved line tube edge set from the catalytic tube edge map by using an edge detection method comprises:
extracting a primary linear edge set from the catalytic tube edge map by utilizing a linear convolution kernel, and carrying out edge splitting on the catalytic tube edge map by utilizing the primary linear edge set to obtain a primary curve edge set;
Performing linear fitting on each primary linear edge in the primary linear edge set to obtain a linear tube edge set;
and performing curve fitting on each primary curve edge in the primary curve edge set to obtain a curve tube edge set.
6. The method for converting energy of a hydrogen fuel cell based on multi-level implementation as claimed in claim 1, wherein the calculating the analytical conversion rate corresponding to the target test feature data by using a preset catalytic conversion power model includes:
performing position coding on the target test characteristic data by using a coding layer of a preset catalytic conversion power model to obtain a target test characteristic code;
generating a multi-head vector set of the target test feature codes, and calculating target test attention features corresponding to the multi-head vector set by using an attention mechanism of the catalytic conversion power model;
and vector decoding is carried out on the target test attention characteristic by using a decoding layer of the catalytic conversion power model, and analysis conversion rate corresponding to the target test characteristic data is calculated by using a greedy layer of the catalytic conversion power model.
7. A hydrogen fuel cell energy conversion device based on a multi-level implementation, for implementing a hydrogen fuel cell energy conversion method based on a multi-level implementation as claimed in claim 1, characterized in that the device comprises:
The data cleaning module is used for acquiring a historical hydrogen battery test data set, cleaning the data of the historical hydrogen battery test data set to obtain a standard hydrogen battery test data set, extracting catalytic layer data sets in the standard hydrogen battery test data set one by one to serve as target catalytic layer data sets, and taking a catalytic tube model diagram in the target catalytic layer data sets as a target catalytic tube model diagram;
the feature extraction module is configured to perform a primary filtering operation on the target catalytic tube model map to obtain a filtered catalytic tube model map, extract a hierarchical feature and a type feature of the filtered catalytic tube model map, and integrate the hierarchical feature and the type feature into a catalytic tube feature of the target catalytic tube model map, where the extracting the hierarchical feature and the type feature of the filtered catalytic tube model map includes: and carrying out edge enhancement on the filtering catalytic model diagram by using the following edge intensity algorithm to obtain an enhanced catalytic tube model diagram:
；
wherein,means the edge strength of the edge of the catalytic tube, < >>Is the sign of maximum>Refers to the maximum value of the edge intensity, +.>Refers to absolute value symbols, " >Refers to the gray function of the filtered catalytic model map,/->Means that the coordinates in the filtering catalytic model diagram are +.>Gray value of the pixel of +.>Is a gradient operator symbol,>is a circumference rate symbol, < >>Is indicated as differential, ->Is an exponential function; performing edge binarization operation on the reinforced catalytic tube model graph to obtain a catalytic tube edge graph; extracting primary characteristics of a pipeline from the catalytic tube edge map, and extracting a straight tube edge set and a curve tube edge set from the catalytic tube edge map by utilizing an edge detection mode; extracting hierarchical features from the straight line pipe edge set, extracting pipeline bending features from the curve pipe edge set, and fusing the pipeline primary features and the pipeline bending features into type features;
the data updating module is used for updating the target catalytic layer data set by utilizing the catalytic tube characteristics to obtain a target catalytic layer characteristic data set, and updating the standard hydrogen battery test data set by utilizing all the target catalytic layer characteristic data sets to obtain a test characteristic data set;
the model training module is used for selecting test feature data in the test feature data set one by one as target test feature data, calculating analysis conversion rate corresponding to the target test feature data by using a preset catalytic conversion power model, calculating analysis loss values of the catalytic conversion power model according to all the analysis conversion rates and the test feature data set, and performing iterative updating on the catalytic conversion power model by using the analysis loss values to obtain a conversion power analysis model;
The energy conversion module is used for obtaining primary battery parameters of the hydrogen battery to be converted, analyzing target conversion rate corresponding to the primary battery parameters according to the conversion power analysis model, generating a conversion parameter set corresponding to the primary battery parameters according to the target conversion rate, generating a target conversion method according to the conversion parameter set, and carrying out energy conversion on the hydrogen battery to be converted according to the target conversion method.
CN202310399937.XA 2023-04-14 2023-04-14 Multi-level implementation-based hydrogen fuel cell energy conversion method and device Active CN116154239B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310399937.XA CN116154239B (en) 2023-04-14 2023-04-14 Multi-level implementation-based hydrogen fuel cell energy conversion method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310399937.XA CN116154239B (en) 2023-04-14 2023-04-14 Multi-level implementation-based hydrogen fuel cell energy conversion method and device

Publications (2)

Publication Number Publication Date
CN116154239A CN116154239A (en) 2023-05-23
CN116154239B true CN116154239B (en) 2023-11-21

Family

ID=86360281

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310399937.XA Active CN116154239B (en) 2023-04-14 2023-04-14 Multi-level implementation-based hydrogen fuel cell energy conversion method and device

Country Status (1)

Country Link
CN (1) CN116154239B (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2742611A1 (en) * 2008-11-04 2010-05-14 Nissan Motor Co., Ltd. Current limiting power generation control device and method
CN110243762A (en) * 2019-06-18 2019-09-17 深圳大雷汽车检测股份有限公司 Telemetering motor vehicle tail and supervisory systems and self study high emitter decision algorithm
CN111812535A (en) * 2020-06-30 2020-10-23 南京林业大学 Power battery fault diagnosis method and system based on data driving
DE102020121626A1 (en) * 2019-08-22 2021-02-25 Toyota Jidosha Kabushiki Kaisha CONTROL DEVICE FOR A VEHICLE, LEARNING SYSTEM FOR A VEHICLE AND CONTROL METHOD FOR A VEHICLE
CN112721743A (en) * 2021-02-04 2021-04-30 武汉理工大学 Fuel cell automobile remaining mileage prediction method based on charging-hydrogen using mode
CN114742297A (en) * 2022-04-11 2022-07-12 中国第一汽车股份有限公司 Power battery treatment method
CN115358647A (en) * 2022-10-24 2022-11-18 齐鲁云商数字科技股份有限公司 Hydrogen energy industry chain risk monitoring system and monitoring method based on big data
CN115758255A (en) * 2023-01-10 2023-03-07 佰聆数据股份有限公司 Electricity consumption abnormal behavior analysis method and device under fusion model
CN115759885A (en) * 2023-01-09 2023-03-07 佰聆数据股份有限公司 Material sampling inspection method and device based on distributed material supply
CN115809795A (en) * 2023-02-03 2023-03-17 佰聆数据股份有限公司 Digitalized production team bearing capacity evaluation method and device

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2742611A1 (en) * 2008-11-04 2010-05-14 Nissan Motor Co., Ltd. Current limiting power generation control device and method
CN110243762A (en) * 2019-06-18 2019-09-17 深圳大雷汽车检测股份有限公司 Telemetering motor vehicle tail and supervisory systems and self study high emitter decision algorithm
DE102020121626A1 (en) * 2019-08-22 2021-02-25 Toyota Jidosha Kabushiki Kaisha CONTROL DEVICE FOR A VEHICLE, LEARNING SYSTEM FOR A VEHICLE AND CONTROL METHOD FOR A VEHICLE
CN111812535A (en) * 2020-06-30 2020-10-23 南京林业大学 Power battery fault diagnosis method and system based on data driving
CN112721743A (en) * 2021-02-04 2021-04-30 武汉理工大学 Fuel cell automobile remaining mileage prediction method based on charging-hydrogen using mode
CN114742297A (en) * 2022-04-11 2022-07-12 中国第一汽车股份有限公司 Power battery treatment method
CN115358647A (en) * 2022-10-24 2022-11-18 齐鲁云商数字科技股份有限公司 Hydrogen energy industry chain risk monitoring system and monitoring method based on big data
CN115759885A (en) * 2023-01-09 2023-03-07 佰聆数据股份有限公司 Material sampling inspection method and device based on distributed material supply
CN115758255A (en) * 2023-01-10 2023-03-07 佰聆数据股份有限公司 Electricity consumption abnormal behavior analysis method and device under fusion model
CN115809795A (en) * 2023-02-03 2023-03-17 佰聆数据股份有限公司 Digitalized production team bearing capacity evaluation method and device

Also Published As

Publication number Publication date
CN116154239A (en) 2023-05-23

Similar Documents

Publication Publication Date Title
CN106529719B (en) Wind power prediction method based on particle swarm optimization algorithm wind speed fusion
CN112668611B (en) Kmeans and CEEMD-PE-LSTM-based short-term photovoltaic power generation power prediction method
CN114792156A (en) Photovoltaic output power prediction method and system based on curve characteristic index clustering
CN115795329B (en) Power utilization abnormal behavior analysis method and device based on big data grid
CN116027217A (en) AE-CFSFDP energy storage power station multi-source battery data anomaly detection method
CN115759365A (en) Photovoltaic power generation power prediction method and related equipment
CN111709454A (en) Multi-wind-field output clustering evaluation method based on optimal copula model
CN116154239B (en) Multi-level implementation-based hydrogen fuel cell energy conversion method and device
CN110956304A (en) Distributed photovoltaic power generation capacity short-term prediction method based on GA-RBM
CN116577843A (en) Wind power prediction method for distributed sodar
CN115620038A (en) Common knowledge constrained remote sensing sample migration method
CN114065621A (en) Gas compressor characteristic equation coefficient fitting method based on genetic algorithm
CN112926761A (en) Wind power plant power prediction method
CN110955811A (en) Power data classification method and system based on naive Bayes algorithm
CN112215383A (en) Distributed photovoltaic power generation power prediction method and system
CN116756622B (en) Power consumption analysis method and system based on intelligent kettle
CN117170979B (en) Energy consumption data processing method, system, equipment and medium for large-scale equipment
CN116226687B (en) Reservoir daily initial water level estimation method and device based on data mining technology
CN117131783B (en) Multi-mode learning-based power transmission line risk prediction model, method and system
CN112734093B (en) Hydrogen production device capacity optimal configuration method based on computer
CN114418194B (en) Tower damage prediction method and device based on data driving and model driving
CN117691565A (en) Method, device, equipment and medium for predicting generated energy of wind power station
CN116578927A (en) Wind turbine generator data anomaly detection method based on nuclear density estimation and shannon entropy
CN117895508A (en) Wind power prediction method based on deep learning
CN117713238A (en) Random optimization operation strategy combining photovoltaic power generation and energy storage micro-grid

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant