CN114880941A - Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system - Google Patents

Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system Download PDF

Info

Publication number
CN114880941A
CN114880941A CN202210563694.4A CN202210563694A CN114880941A CN 114880941 A CN114880941 A CN 114880941A CN 202210563694 A CN202210563694 A CN 202210563694A CN 114880941 A CN114880941 A CN 114880941A
Authority
CN
China
Prior art keywords
electrolytic cell
preset
model
test array
test
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210563694.4A
Other languages
Chinese (zh)
Inventor
许泽阳
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.)
Sungrow Shanghai Co Ltd
Original Assignee
Sungrow Shanghai Co Ltd
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 Sungrow Shanghai Co Ltd filed Critical Sungrow Shanghai Co Ltd
Priority to CN202210563694.4A priority Critical patent/CN114880941A/en
Publication of CN114880941A publication Critical patent/CN114880941A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/20Design optimisation, verification or simulation
    • G06F30/27Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • CCHEMISTRY; METALLURGY
    • C25ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
    • C25BELECTROLYTIC OR ELECTROPHORETIC PROCESSES FOR THE PRODUCTION OF COMPOUNDS OR NON-METALS; APPARATUS THEREFOR
    • C25B1/00Electrolytic production of inorganic compounds or non-metals
    • C25B1/01Products
    • C25B1/02Hydrogen or oxygen
    • C25B1/04Hydrogen or oxygen by electrolysis of water
    • CCHEMISTRY; METALLURGY
    • C25ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
    • C25BELECTROLYTIC OR ELECTROPHORETIC PROCESSES FOR THE PRODUCTION OF COMPOUNDS OR NON-METALS; APPARATUS THEREFOR
    • C25B15/00Operating or servicing cells
    • C25B15/02Process control or regulation
    • C25B15/023Measuring, analysing or testing during electrolytic production
    • CCHEMISTRY; METALLURGY
    • C25ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
    • C25BELECTROLYTIC OR ELECTROPHORETIC PROCESSES FOR THE PRODUCTION OF COMPOUNDS OR NON-METALS; APPARATUS THEREFOR
    • C25B15/00Operating or servicing cells
    • C25B15/02Process control or regulation
    • C25B15/023Measuring, analysing or testing during electrolytic production
    • C25B15/025Measuring, analysing or testing during electrolytic production of electrolyte parameters
    • C25B15/027Temperature
    • CCHEMISTRY; METALLURGY
    • C25ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
    • C25BELECTROLYTIC OR ELECTROPHORETIC PROCESSES FOR THE PRODUCTION OF COMPOUNDS OR NON-METALS; APPARATUS THEREFOR
    • C25B9/00Cells or assemblies of cells; Constructional parts of cells; Assemblies of constructional parts, e.g. electrode-diaphragm assemblies; Process-related cell features
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • G06N3/084Backpropagation, e.g. using gradient descent
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08Thermal analysis or thermal optimisation
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02EREDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02E60/30Hydrogen technology
    • Y02E60/36Hydrogen production from non-carbon containing sources, e.g. by water electrolysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Materials Engineering (AREA)
  • Electrochemistry (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Metallurgy (AREA)
  • Organic Chemistry (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Automation & Control Theory (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Analytical Chemistry (AREA)
  • Computational Linguistics (AREA)
  • Computer Hardware Design (AREA)
  • Inorganic Chemistry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Geometry (AREA)
  • Electrolytic Production Of Non-Metals, Compounds, Apparatuses Therefor (AREA)

Abstract

The application provides a method and a device for establishing an electrolytic cell model in a water electrolysis hydrogen production system, wherein the method comprises the steps of firstly, obtaining an original test array set required for establishing the electrolytic cell model, wherein the original test array set comprises N groups of electrolytic cell test arrays; secondly, screening each group of electrolytic cell test arrays in the original test array set to obtain a modeling test array set; finally, training a preset neural network model based on the modeling test array set, and using the preset neural network model meeting preset training conditions as an electrolytic cell model, namely the application can train the neural network model by using the screened array set, so that abnormal data interference is eliminated, and the accuracy of the neural network model is improved; and the neural network model meeting the preset training conditions is used as the electrolytic cell model, so that the problems of large fitting voltage error and low model precision of the electrolytic cell model obtained by fitting formula coefficients mainly based on a formula and test data in the existing related scheme are solved.

Description

Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system
Technical Field
The invention relates to the technical field of hydrogen production, in particular to a method and a device for establishing an electrolytic cell model in a water electrolysis hydrogen production system.
Background
The water electrolysis hydrogen production system mainly comprises an electrolytic bath, a gas-liquid separation device and a washer, and realizes hydrogen production by water electrolysis through a purification process flow. In order to perform performance analysis and optimal design on the water electrolysis hydrogen production system, a large number of experimental tests are required.
In the test process, the electrolyte needs to be preheated, and the test takes long time. Although the data amount required by the test can be reduced by modeling the water electrolysis hydrogen production system, so that the test pressure of engineering testers is reduced, the existing method for establishing the electrolytic cell model in the water electrolysis hydrogen production system is mainly obtained by performing a formula coefficient fitting method based on a formula and test data, the fitting voltage error is large, and the model precision is low.
Disclosure of Invention
Therefore, the method and the device for establishing the electrolytic cell model in the water electrolysis hydrogen production system are provided, and the problems that the existing related scheme is large in fitting voltage error and low in model precision due to the electrolytic cell model obtained by fitting formula coefficients mainly based on a formula and test data are solved.
In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions:
the invention discloses a method for establishing an electrolytic cell model in a water electrolysis hydrogen production system, which comprises the following steps:
acquiring an original test array set required for establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell;
screening each group of the electrolytic cell test array in the original test array set to obtain a modeling test array set;
and training a preset neural network model based on the modeling test array set, and taking the preset neural network model meeting preset training conditions as the electrolytic cell model.
Optionally, in the method for establishing an electrolytic cell model in a system for producing hydrogen by electrolyzing water, screening each group of the electrolytic cell test arrays in the original test array set to obtain a modeling test array set, the method includes:
respectively judging whether each group of the electrolytic cell test array in the original test array set meets preset screening conditions to obtain a judgment result of each group of the electrolytic cell test array;
and based on the judgment result, rejecting the electrolytic cell test array which does not meet the preset screening condition to obtain the modeling test array set.
Optionally, in the method for establishing an electrolytic cell model in a system for producing hydrogen by electrolyzing water, determining whether the electrolytic cell test array meets a preset screening condition includes:
determining the point slope of the data points corresponding to the temperature of the electrolytic cell and the target fitting coefficient in the electrolytic cell test array in the target electrolytic cell temperature curve;
judging whether the point slope of the data point meets the requirement of a preset slope or not;
if the judgment result is yes, judging that the electrolytic cell test array meets the preset screening condition;
if the judgment result is negative, the electrolytic cell test array is judged not to meet the preset screening condition.
Optionally, in the method for establishing an electrolytic cell model in a hydrogen production system by electrolyzing water, a point slope of the data point meets a preset slope requirement, including:
the absolute values of the differences between the point slopes of the data points and the point slopes of adjacent data points are all smaller than the preset slope deviation.
Optionally, in the method for establishing an electrolytic cell model in a hydrogen production system by electrolyzing water, the generating process of the target electrolytic cell temperature curve includes:
respectively determining the temperature of the electrolytic cell in each group of electrolytic cell test array and the target fitting coefficient;
and fitting based on the temperature of the electrolytic cell in each group of electrolytic cell test arrays and the target fitting coefficient to generate the target electrolytic cell temperature curve.
Optionally, in the method for establishing an electrolytic cell model in a hydrogen production system by electrolyzing water, training a preset neural network model based on the modeling test array set includes:
carrying out normalization processing on the modeling test array set to obtain a normalized modeling test data set;
and taking the normalized modeling test data set as a training sample, and training the preset neural network model by using a preset transfer function and a preset training function.
Optionally, in the method for establishing an electrolytic cell model in a system for producing hydrogen by electrolyzing water, after taking a preset neural network model satisfying a preset training condition as the electrolytic cell model, the method further includes:
determining the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated;
and taking the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated as the input of the electrolytic cell model to obtain the predicted cell voltage of the electrolytic cell corresponding to the working condition to be simulated.
Optionally, in the method for establishing an electrolytic cell model in the system for producing hydrogen by electrolyzing water, the mode of obtaining the electrolytic cell test array is field test or experimental test.
Optionally, in the method for establishing an electrolytic cell model in a hydrogen production system by electrolyzing water, the step of enabling the preset neural network model to meet the preset training condition includes:
the training error of the preset neural network model is smaller than the target training error, and/or the training times of the preset neural network model are larger than the target training times.
Optionally, in the method for establishing an electrolytic cell model in the hydrogen production system by electrolyzing water, the preset neural network model is a BP neural network model.
The second aspect of the invention discloses an electrolytic tank model establishing device in a water electrolysis hydrogen production system, which comprises:
the acquisition unit is used for acquiring an original test array set required by establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell;
the screening unit is used for screening each group of the electrolytic cell test array in the original test array set to obtain a modeling test array set;
and the training modeling unit is used for training a preset neural network model based on the modeling test array set, and taking the preset neural network model meeting preset training conditions as the electrolytic cell model.
The invention provides a method for establishing an electrolytic cell model in a water electrolysis hydrogen production system, which comprises the steps of firstly obtaining an original test array set required for establishing the electrolytic cell model, wherein the original test array set comprises N groups of electrolytic cell test arrays; secondly, screening each group of the electrolytic cell test array in the original test array set to obtain a modeling test array set; thirdly, training the preset neural network model based on the modeling test array set, and using the preset neural network model meeting the preset training conditions as an electrolytic cell model, namely the application can train the neural network model by using the screened array set, so that abnormal data interference is eliminated, and the accuracy of the neural network model is improved; and the neural network model meeting the preset training conditions is used as the electrolytic cell model, so that the problems of large fitting voltage error and low model precision of the electrolytic cell model obtained by fitting formula coefficients mainly based on a formula and test data in the existing related scheme are solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 to fig. 2 are flow charts of two methods for establishing an electrolytic cell model in a water electrolysis hydrogen production system provided by an embodiment of the present application;
FIG. 3 is a flowchart illustrating a method for determining whether a test array of an electrolytic cell satisfies a predetermined screening condition according to an embodiment of the present disclosure;
FIG. 4 is a flow chart of a target cell temperature profile generation provided in the examples of the present application;
FIG. 5 is a flowchart of training a neural network model according to an embodiment of the present disclosure;
FIG. 6 is a flow chart of another method for establishing an electrolytic cell model in a water electrolysis hydrogen production system according to an embodiment of the present application;
FIG. 7 is a diagram of cell voltage error of an electrolytic cell when a formula obtained by directly fitting with a matlab fitting tool is used as an electrolytic cell model according to an embodiment of the present application;
FIG. 8 is a diagram illustrating a cell voltage error of an electrolytic cell corresponding to a model of the electrolytic cell constructed by a method for modeling the model of the electrolytic cell in a system for producing hydrogen by electrolyzing water according to an embodiment of the present disclosure;
fig. 9a is a test result diagram of the BP neural network modeling after removing abnormal data at 95 ℃ according to the embodiment of the present application;
fig. 9b is a test result diagram of the BP neural network modeling after removing abnormal data at 80 ℃ according to the embodiment of the present application;
fig. 9c is a test result diagram of the BP neural network modeling after removing abnormal data at 65 ℃ according to the embodiment of the present application;
fig. 9d is a test result diagram of the BP neural network modeling after removing abnormal data at 55 ℃ according to the embodiment of the present application;
fig. 9e is a test result diagram of the BP neural network modeling after removing abnormal data at 29 ℃ according to the embodiment of the present application;
FIG. 10a is a graph of test results of modeling by directly using a BP neural network without rejecting abnormal data at 95 ℃ according to an embodiment of the present disclosure;
FIG. 10b is a diagram of a test result of modeling by directly using a BP neural network without rejecting abnormal data at 80 ℃ according to an embodiment of the present disclosure;
FIG. 10c is a diagram of a test result of modeling by directly using a BP neural network without rejecting abnormal data at 65 ℃ according to an embodiment of the present application;
FIG. 10d is a graph of test results obtained by modeling abnormal data at 55 ℃ directly using a BP neural network according to the embodiment of the present application;
fig. 10e is a test result diagram of modeling by directly using a BP neural network without removing abnormal data at 29 ℃ according to the embodiment of the present application;
fig. 11 is a schematic structural diagram of an apparatus for establishing an electrolytic cell model in a system for producing hydrogen by electrolyzing water according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The application provides a method for establishing an electrolytic cell model in a water electrolysis hydrogen production system, which aims to solve the problems of large fitting voltage error and low model precision of the electrolytic cell model obtained by fitting formula coefficients mainly based on a formula and test data in the existing related scheme.
Referring to fig. 1, the method for establishing the electrolytic cell model in the water electrolysis hydrogen production system mainly comprises the following steps:
s100, obtaining an original test array set required by the establishment of the electrolytic cell model.
The set of raw test arrays may include N sets of cell test arrays. N is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell.
In practical application, each group of electrolytic cell test array in the original test array set can be obtained in a field test or experimental test mode.
It should be noted that the specific value of N can be determined according to the specific application environment and the user requirement, and the larger N is, the more representative the data is, and the higher the accuracy of the created electrolytic cell model is.
S102, screening each group of electrolytic cell test arrays in the original test array set to obtain a modeling test array set.
And the number of the groups of the electrolytic cell test arrays in the original test array set is more than or equal to the number of the groups of the electrolytic cell test arrays in the modeling test array set.
In practical application, the specific process of performing step S102 and screening each group of the electrolyzer test arrays in the original test array set to obtain the modeling test array set can be as shown in fig. 2, and mainly includes steps S200 and S202.
S200, respectively judging whether each group of electrolytic cell test array in the original test array set meets preset screening conditions, and obtaining the judgment result of each group of electrolytic cell test array.
The specific process of judging whether the electrolytic cell test array meets the preset screening condition can be shown in fig. 3, and mainly comprises the following steps:
s300, determining the point slope of the data points corresponding to the temperature of the electrolytic cell and the target fitting coefficient in the electrolytic cell test array in the target electrolytic cell temperature curve.
In practical application, the generation process of the target electrolyzer temperature curve can be as shown in fig. 4, and mainly comprises steps S400 and S402:
s400, respectively determining the temperature of the electrolytic cell in each group of electrolytic cell test array and a target fitting coefficient.
Assuming that the temperature of the electrolytic cell is T, the cell voltage of the electrolytic cell is U, and the current density of the electrolytic cell is I, fitting can be performed according to a formula U which is a + b × I + c × log (d × I +1) aiming at the temperature T of the electrolytic cell, the cell voltage U of the electrolytic cell and the current density I of the electrolytic cell in each group of electrolytic cell test array, fitting coefficients a, b, c and d are determined, and the analogy is repeated to obtain the fitting coefficients a, b, c and d corresponding to the temperature T of the electrolytic cell.
Specifically, a fitting curve deviation sum of squares formula can be determined; then, respectively solving a partial derivative function for the fitting coefficients b, c and d and setting zero to obtain a partial derivative matrix. And solving the partial derivative matrix to obtain specific values of the fitting coefficients a, b, c and d. Wherein:
the fitted curve deviation sum of squares formula is:
Figure BDA0003657468140000061
fitting coefficient b the partial derivative function is set to zero:
Figure BDA0003657468140000062
the fitting coefficient c is a partial derivative function and is set to zero:
Figure BDA0003657468140000071
the fitting coefficient d is a partial derivative function and is set to zero:
Figure BDA0003657468140000072
the formula U ═ a + b × I + c × log (d × I +1) is obtained by adjusting the formula U ═ a + (r1 × T + r2) × I + s × log ((E/T ^2+ F/T + G) × I +1) according to the conventional formula.
In addition, the values of a, b, c and d can be obtained by fitting based on test data by using a Matlab or Python fitting tool.
For example, using a Matlab fitting tool, one can fit to obtain: u-0.0131 XT-1.2168 + (1 e) -6 ×T-1e -6 ) xI + (-0.0012 XT +0.2592) xlog ((0.2527 XT 2-41.697 XT +1823.1) xI + 1); using Pytrhon fitting toolFitting can result in:
U=(0.0181345×T-2.44413727)+(-4.65974016e -4 +6.17311687e -6 ×T)×I+(2.52031111e -1 -5.06355998e -4 ×T-5.4153663e -6 T 2 )×log((-19079.53711162+2511176.65734026/T-5495842.47215084/T 2 ) X I + 1); u in the formula represents the cell voltage of the electrolyzer. As can be seen from the above fitting results, the formula U ═ a + b × I + c × log (d × I +1) is adjusted relative to the conventional formula, for example, a in the formula is no longer a constant, but a function of temperature T, and the like.
S402, fitting is carried out on the basis of the temperature of the electrolytic cell in each group of electrolytic cell test array and the target fitting coefficient, and a target electrolytic cell temperature curve is generated.
In practical application, one of the fitting coefficients a, b, c and d can be selected as a target fitting coefficient, and a target electrolytic tank temperature curve can be generated according to the electrolytic tank temperature in each group of electrolytic tank test arrays and the corresponding target fitting coefficient.
Of course, the temperature curve of the electrolytic cell corresponding to the fitting coefficients a, b, c and d can be generated according to the fitting coefficients a, b, c and d corresponding to the temperature of the electrolytic cell in each group of electrolytic cell test array and the temperature of each electrolytic cell obtained by solving. And then, according to the actual application condition and the user requirement, selecting one curve from the electrolytic bath temperature curves corresponding to the fitting coefficients a, b, c and d as a target electrolytic bath temperature curve, and determining according to the specific application environment and the user requirement, wherein the curve is within the protection range of the application.
It should be noted that the target fitting coefficient may be any one of the fitting coefficients a, b, c, and d, and may be determined according to the specific application environment and the user requirement, which are all within the protection scope of the present application.
S302, judging whether the point slope of the data point meets the requirement of a preset slope.
In practical application, the specific process of determining whether the point slope of the data point meets the preset slope requirement is as follows: and judging whether the absolute value of the difference value between the point slope of the data point and the point slope of the adjacent data point is smaller than the preset slope deviation or not. And if the absolute value of the difference value between the point slope of the data point and the point slope of the adjacent data point is judged to be smaller than the preset slope deviation, namely the judgment result is yes, the point slope of the data point is judged to meet the preset slope requirement. And if the absolute value of the difference value between the point slope of the data point and the point slope of the adjacent data point is judged to be not less than the preset slope deviation, namely the judgment result is negative, judging that the point slope of the data point does not meet the preset slope requirement.
It should be noted that, if a data point is a boundary data point, that is, a minimum value point or a maximum value point, the number of adjacent data points is 1; if the data point is not a boundary data point, i.e., a non-maximum point, the number of neighboring data points is 2.
If the point slope of the data point is determined to satisfy the preset slope requirement, that is, if the determination result is yes, executing step S304; if the point slope of the data point is determined not to satisfy the predetermined slope requirement, i.e., if the determination result is no, step S306 is executed.
S304, judging that the electrolytic cell test array meets the preset screening condition.
In practical application, after the point slope of the data point is judged to meet the requirement of the preset slope, the electrolytic cell test array can be judged to meet the preset screening condition.
S306, judging that the electrolytic cell test array does not meet the preset screening condition.
In practical application, after the point slope of the data point is judged not to meet the requirement of the preset slope, the electrolytic cell test array can be judged not to meet the preset screening condition.
Assuming that the number of groups of the electrolytic cell test array is i, i is a positive integer, the target fitting coefficient is a fitting coefficient b, and the temperature T of each electrolytic cell i The corresponding fitting coefficient b is b Ti Then each T can be obtained i And b Ti Point slope K of the corresponding temperature curve i And obtaining the point slope of each data point. Wherein Ki ═ b Ti -b Ti-1 /(T i -T i-1 ). Then, if a certain data point K i+1 -K i And K i -K i-1 If the temperature data corresponding to the Ki points are all larger than the set slope deviation Kmax which is equal to 0.1, the temperature data corresponding to the Ki points are abnormal numbersAccordingly, it is possible to eliminate. In other words, for each data point, it is determined whether the absolute value of the difference between the point slopes of the data point and the adjacent data point is smaller than the preset slope deviation, and if the difference between the point slopes of the data point and the adjacent data point is not smaller than the set slope deviation, it is determined that the data at the data point is abnormal data and should be rejected.
S202, based on the judgment result, eliminating the electrolytic cell test arrays which do not meet the preset screening condition to obtain a modeling test array set.
In practical application, if a certain group of the electrolytic cell test array does not meet the preset screening condition, the group of the electrolytic cell test array is indicated as abnormal data, and the group of the electrolytic cell test array is rejected in order to ensure the accuracy of the obtained electrolytic cell model.
It can be understood that the abnormal electrolytic cell test array in the original test array set can be removed by the mode of judging the fitting parameters of the formula, so as to obtain the modeling test array set.
And S104, training the preset neural network model based on the modeling test array set, and taking the preset neural network model meeting the preset training conditions as an electrolytic cell model.
In practical applications, the specific process of executing the modeling test array set in step S104 to train the preset neural network model may be as shown in fig. 5, and mainly includes steps S500 and S502.
And S500, carrying out normalization processing on the modeling test array set to obtain a normalized modeling test array set.
It should be noted that, the normalization processing may be performed on each group of the electrolyzer test array in the modeling test array set by using the existing normalization processing method, so as to obtain the normalized modeling test array set.
S502, taking the normalized modeling test array set as a training sample, and training the preset neural network model by using a preset transfer function and a preset training function.
The preset transfer function may be a Purelin function, and the preset training function may be a gradient descent training function. Of course, the present invention is not limited to this, and the existing transfer function or training function may be selected, and the present invention is not limited to this, and is within the scope of the present invention.
It should be noted that the preset neural network model may be a BP neural network model. Specifically, the preset neural network model has 2 input layer nodes, 5 hidden layer nodes and 1 output layer node.
In practical application, a modeling test array set can be used as a training sample, a Purelin function is used as a preset transfer function, and a gradient descent function is used as a preset training function to train a preset neural network model.
It should be noted that the preset neural network model satisfying the preset training condition may be that the training error of the preset neural network model is smaller than the target training error, and/or the training times of the preset neural network model are greater than the target training times. The target training error is a minimum training error which needs to be met by a preset neural network model, and the target training times are preset training times of the preset neural network model. Specifically, the specific values of the target training error and the target training times can be set according to the actual application situation and the user requirement, and the method is not specifically limited and is within the protection range of the method.
In the training process, whether the preset neural network model meets the preset training condition or not can be judged in real time, and if yes, the preset neural network model corresponding to the condition that the preset neural network model meets the preset training condition is judged to be used as the electrolytic cell model.
It should be further noted that, in the training process, the learning rate of the preset neural network model may also be set. In addition, a batch updating method can be adopted to update the weight and the bias in the preset neural network model. The specific update process may be as follows:
(1) all layers 2 ≦ l ≦ 3 are given Δ w (l) 0 and Δ b (l) 0. Wherein, Δ W (l) Representing an all-zero matrix, Δ b (l) Representing an all zero vector.
(2) And i is 1: m, and m is the number of training samples. Wherein, a gradient matrix of neuron weights and biases of each layer can be calculated using a back propagation algorithm
Figure BDA0003657468140000101
And
Figure BDA0003657468140000102
then, calculate
Figure BDA0003657468140000103
Computing
Figure BDA0003657468140000104
(3) And updating the weight and the bias according to the calculation result. Wherein, can make W (l) =W (l) +1/mΔW (l) ,b (l) =b (l) +1/mΔb (l)
Based on the principle, the method for establishing the electrolytic cell model in the hydrogen production system by electrolyzing water, provided by the embodiment, comprises the steps of firstly obtaining an original test array set required for establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; secondly, screening each group of electrolytic cell test arrays in the original test array set to obtain a modeling test array set; finally, training a preset neural network model based on the modeling test array set, and using the preset neural network model meeting preset training conditions as an electrolytic cell model, namely the application can train the neural network model by using the screened array set, so that abnormal data interference is eliminated, and the accuracy of the neural network model is improved; and the neural network model meeting the preset training conditions is used as the electrolytic cell model, so that the problems of large fitting voltage error and low model precision of the electrolytic cell model obtained by fitting formula coefficients mainly based on a formula and test data in the existing related scheme are solved.
Alternatively, in another embodiment provided by the present application, after the preset neural network model satisfying the preset training condition is used as the electrolytic cell model in step S104, as shown in fig. 6, the method for establishing the electrolytic cell model in the water electrolysis hydrogen production system may further include the following steps:
s600, determining the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated.
The temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated can be determined through field detection or other existing modes. Of course, the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated can be determined according to the actual application environment and the user requirements, and the determination mode is not specifically limited in the application and belongs to the protection scope of the application.
S602, taking the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated as the input of an electrolytic cell model, and obtaining the predicted cell voltage of the electrolytic cell corresponding to the working condition to be simulated.
In practical application, in order to improve the prediction accuracy of the electrolytic cell model, the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated can be normalized first and then used as the input of the electrolytic cell model.
It should be noted that the specific way of normalization processing is the same as the way of normalization processing of the modeling test array set, and reference may be made to this way, and details are not repeated here.
The cell model may be predicted from the input cell temperature and cell density to obtain a predicted cell voltage corresponding to the cell temperature.
In this embodiment, can be based on the electrolysis trough temperature and the electrolysis trough current density that wait to simulate the operating mode and correspond, utilize the electrolysis trough model to treat the electrolysis trough cell voltage that the operating mode corresponds and predict, obtain prediction electrolysis trough cell voltage, because the voltage fitting precision of the electrolysis trough model that this application provided is higher, the prediction electrolysis trough cell voltage that consequently obtains is more close to actual value.
Based on the method for establishing the electrolytic cell model in the hydrogen production system by electrolyzing water provided by the embodiment, the following describes the accuracy of the electrolytic cell model in the hydrogen production system by comparing the data obtained in the present application:
FIG. 7 shows the cell voltage error diagram of the electrolyzer when the equation obtained by fitting directly using the matlab fitting tool was used as the model of the electrolyzer. In the figure, the range of 0-30 of the abscissa corresponds to the test result at the temperature of 53 ℃, and the range of 30-90 corresponds to the test result at the temperature of 63 ℃. Specifically, the abscissa represents the number of test points (i.e., the first and second test points …) and the ordinate represents the percentage error between the measured cell voltage and the cell voltage calculated using the matlab fitting tool fitting equation (i.e., the equation described above) as the cell model.
The corresponding cell voltage error graph of the electrolytic cell when the electrolytic cell model is constructed by the method provided by the application (namely, data screening is firstly carried out and then modeling is carried out by using the BP neural network) is shown in FIG. 8. Wherein the abscissa has the same meaning as that of FIG. 7, and the ordinate represents the error percentage between the measured cell voltage of the electrolyzer and the cell voltage of the electrolyzer predicted according to the model of the electrolyzer provided in the present case, and also the test results corresponding to 53 ℃ and 63 ℃.
Comparing the test results shown in fig. 7 and fig. 8, it can be found that the error of the electrolytic cell model provided by the present invention is within 2%, and the error of the electrolytic cell model using the formula is within 5%, so that the modeling accuracy of the model provided by the present invention is higher than that of the formula method.
In addition, the test results obtained by eliminating the abnormal data and then modeling by using the BP neural network may be as shown in fig. 9a to 9e, while the test results obtained by directly modeling by using the BP neural network without eliminating the abnormal data may be as shown in fig. 10a to 10 e. Wherein, FIGS. 9a to 9e are the test results corresponding to 95 deg.C, 80 deg.C, 65 deg.C, 55 deg.C and 29 deg.C, respectively, and FIGS. 10a to 10e are the test results corresponding to 95 deg.C, 80 deg.C, 65 deg.C, 55 deg.C and 29 deg.C, respectively. Specifically, the abscissa in the figure represents the current density, and the ordinate represents the cell voltage. Comparing the test results corresponding to the abnormal data removed and the abnormal data not removed at the same temperature respectively shows that the accuracy of modeling is higher when the abnormal data is removed by screening and then modeling than when the abnormal data is not removed by screening, for example: as a result of the test at 29 ℃, it can be seen that the error of FIG. 10e without rejecting abnormal data is very large, especially at low current density, such as 0-10 Acm-2, the error is over 100% at most, while the error in FIG. 9e is only within 4% under the same conditions.
Based on the method for establishing the electrolytic cell model in the hydrogen production system by electrolyzing water provided by the above embodiment, another embodiment of the present application further provides an apparatus for establishing the electrolytic cell model in the hydrogen production system by electrolyzing water, please refer to fig. 11, which mainly includes:
an obtaining unit 100, configured to obtain an original test array set required for establishing an electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell.
And the screening unit 102 is used for screening each group of electrolytic cell test arrays in the original test array set to obtain a modeling test array set.
And the training modeling unit 104 is used for training the preset neural network model based on the modeling test array set, and taking the preset neural network model meeting the preset training conditions as the electrolytic cell model.
Optionally, the screening unit 102 includes:
and the judging unit is used for respectively judging whether each group of electrolytic cell test array in the original test array set meets the preset screening condition or not to obtain the judgment result of each group of electrolytic cell test array.
And the rejecting unit is used for rejecting the electrolytic cell test array which does not meet the preset screening condition based on the judgment result to obtain a modeling test array set.
Optionally, the determining unit is specifically configured to, when determining whether the cell test array satisfies the preset screening condition:
and determining the point slope of the data points corresponding to the temperature of the electrolytic cell and the target fitting coefficient in the electrolytic cell test array in the target electrolytic cell temperature curve.
And judging whether the point slope of the data point meets the preset slope requirement or not.
If the judgment result is yes, judging that the electrolytic cell test array meets the preset screening condition;
if the judgment result is negative, the electrolytic cell test array is judged not to meet the preset screening condition.
Optionally, the point slope of the data point satisfies a preset slope requirement, including:
the absolute values of the differences between the point slopes of the data points and the point slopes of the adjacent data points are all smaller than the preset slope deviation.
Optionally, the generating of the target electrolyzer temperature profile comprises:
and respectively determining the temperature of the electrolytic cell and a target fitting coefficient in each group of electrolytic cell test arrays.
And fitting based on the temperature of the electrolytic cell in each group of electrolytic cell test arrays and the target fitting coefficient to generate a target electrolytic cell temperature curve.
Optionally, the training modeling unit 104 is configured to, when training the preset neural network model based on the modeling test array set, specifically:
and carrying out normalization processing on the modeling test array set to obtain a normalized modeling test data set.
And taking the normalized modeling test data set as a training sample, and training a preset neural network model by using a preset transfer function and a preset training function.
Optionally, after the training modeling unit is configured to use a preset neural network model satisfying a preset training condition as the electrolyzer model, the apparatus further includes:
and the determining unit is used for determining the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated.
And the prediction unit is used for taking the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated as the input of the electrolytic cell model to obtain the predicted cell voltage of the electrolytic cell corresponding to the working condition to be simulated.
Alternatively, the acquisition unit 100 acquires the cell test array in a field test or a test.
Optionally, the preset neural network model satisfies a preset training condition, including:
the training error of the preset neural network model is smaller than the target training error, and/or the training times of the preset neural network model are larger than the target training times.
Optionally, the preset neural network model is a BP neural network model.
Based on the above, the obtaining unit 100 in the device for establishing an electrolytic cell model in the system for producing hydrogen by electrolyzing water provided by this embodiment is used to obtain an original test array set required for establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell; the screening unit 102 is used for screening each group of electrolytic cell test arrays in the original test array set to obtain a modeling test array set; the training modeling unit 104 is used for training the preset neural network model based on the modeling test array set, and using the preset neural network model meeting the preset training condition as the electrolytic cell model, namely the application can train the neural network model by using the screened array set, so that abnormal data interference is eliminated, and the accuracy of the neural network model is improved; in addition, the neural network model meeting the preset training conditions is used as the electrolytic cell model, and the problems that the fitting voltage error is large and the model precision is low in the existing related scheme which is mainly based on a formula and test data and is obtained by fitting formula coefficients are solved.
Features described in the embodiments in the present specification may be replaced with or combined with each other, and the same and similar portions among the embodiments may be referred to each other, and each embodiment is described with emphasis on differences from other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

Claims (11)

1. A method for establishing an electrolytic cell model in a water electrolysis hydrogen production system is characterized by comprising the following steps:
acquiring an original test array set required for establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell;
screening each group of the electrolytic cell test array in the original test array set to obtain a modeling test array set;
and training a preset neural network model based on the modeling test array set, and taking the preset neural network model meeting preset training conditions as the electrolytic cell model.
2. The method for building the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in claim 1, wherein screening each group of the electrolytic cell test arrays in the original test array set to obtain a modeling test array set comprises:
respectively judging whether each group of the electrolytic cell test array in the original test array set meets preset screening conditions to obtain a judgment result of each group of the electrolytic cell test array;
and based on the judgment result, rejecting the electrolytic cell test array which does not meet the preset screening condition to obtain the modeling test array set.
3. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in claim 2, wherein the step of judging whether the electrolytic cell test array satisfies the preset screening condition comprises the steps of:
determining the point slope of the data points corresponding to the temperature of the electrolytic cell and the target fitting coefficient in the electrolytic cell test array in the target electrolytic cell temperature curve;
judging whether the point slope of the data point meets the requirement of a preset slope or not;
if the judgment result is yes, judging that the electrolytic cell test array meets the preset screening condition;
if the judgment result is negative, the electrolytic cell test array is judged not to meet the preset screening condition.
4. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in claim 3, wherein the point slope of the data point satisfies the preset slope requirement, comprising:
the absolute values of the differences between the point slopes of the data points and the point slopes of adjacent data points are all smaller than the preset slope deviation.
5. The method for establishing the electrolytic cell model in the water electrolysis hydrogen production system according to claim 3, wherein the generation process of the target electrolytic cell temperature curve comprises the following steps:
respectively determining the temperature of the electrolytic cell in each group of electrolytic cell test array and the target fitting coefficient;
and fitting based on the temperature of the electrolytic cell in each group of electrolytic cell test arrays and the target fitting coefficient to generate the target electrolytic cell temperature curve.
6. The method for establishing the electrolytic cell model in the water electrolysis hydrogen production system according to any one of claims 1 to 5, wherein the training of the preset neural network model based on the modeling test array set comprises:
carrying out normalization processing on the modeling test array set to obtain a normalized modeling test data set;
and taking the normalized modeling test data set as a training sample, and training the preset neural network model by using a preset transfer function and a preset training function.
7. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in any one of claims 1 to 5, wherein after the model of the electrolytic cell is the preset neural network model satisfying the preset training condition, the method further comprises:
determining the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated;
and taking the temperature of the electrolytic cell and the current density of the electrolytic cell corresponding to the working condition to be simulated as the input of the electrolytic cell model to obtain the predicted cell voltage of the electrolytic cell corresponding to the working condition to be simulated.
8. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in any one of claims 1 to 5, wherein the manner of obtaining the electrolytic cell test array is field test or experimental test.
9. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in any one of claims 1 to 5, wherein the preset neural network model satisfies preset training conditions comprising:
the training error of the preset neural network model is smaller than the target training error, and/or the training times of the preset neural network model are larger than the target training times.
10. The method for establishing the electrolytic cell model in the system for producing hydrogen by electrolyzing water as claimed in any one of claims 1 to 5, wherein the predetermined neural network model is a BP neural network model.
11. An electrolytic tank model establishing device in a water electrolysis hydrogen production system is characterized by comprising:
the acquisition unit is used for acquiring an original test array set required by establishing the electrolytic cell model; the original test array set comprises N groups of electrolytic cell test arrays; n is a positive integer, and each group of the electrolytic cell test arrays comprises the temperature of the electrolytic cell, the cell voltage of the electrolytic cell and the corresponding current density of the electrolytic cell;
the screening unit is used for screening each group of the electrolytic cell test array in the original test array set to obtain a modeling test array set;
and the training modeling unit is used for training a preset neural network model based on the modeling test array set, and taking the preset neural network model meeting preset training conditions as the electrolytic cell model.
CN202210563694.4A 2022-05-23 2022-05-23 Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system Pending CN114880941A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210563694.4A CN114880941A (en) 2022-05-23 2022-05-23 Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210563694.4A CN114880941A (en) 2022-05-23 2022-05-23 Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system

Publications (1)

Publication Number Publication Date
CN114880941A true CN114880941A (en) 2022-08-09

Family

ID=82676845

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210563694.4A Pending CN114880941A (en) 2022-05-23 2022-05-23 Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system

Country Status (1)

Country Link
CN (1) CN114880941A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115522227A (en) * 2022-09-30 2022-12-27 阳光电源股份有限公司 Electrolytic bath working state monitoring method, system, controller and medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115522227A (en) * 2022-09-30 2022-12-27 阳光电源股份有限公司 Electrolytic bath working state monitoring method, system, controller and medium

Similar Documents

Publication Publication Date Title
Zhang et al. An effective LS-SVM-based approach for surface roughness prediction in machined surfaces
JP4388033B2 (en) Information processing apparatus, information processing method, and program
CN109635337B (en) Industrial process soft measurement modeling method based on block increment random configuration network
CN109884892A (en) Process industry system prediction model based on crosscorrelation time lag grey correlation analysis
CN114880941A (en) Method and device for establishing electrolytic cell model in water electrolysis hydrogen production system
CN108595892A (en) Soft-measuring modeling method based on time difference model
CN107729988B (en) Blue algae bloom prediction method based on dynamic deep belief network
CN114841062A (en) Shallow sea earth sound parameter inversion method based on feedback neural network model
CN110807510B (en) Parallel learning soft measurement modeling method for industrial big data
CN110221540B (en) Continuous stirred reactor system control method based on Hammerstein model
CN111105005A (en) Wind power prediction method
CN114777192A (en) Secondary network heat supply autonomous optimization regulation and control method based on data association and deep learning
CN101285816A (en) Copper matte air refining procedure parameter soft sensing instrument and its soft sensing method
Mao et al. Coating thickness modeling and prediction for hot-dip galvanized steel strip based on GA-BP neural network
Colorado-Garrido et al. Neural networks for Nyquist plots prediction during corrosion inhibition of a pipeline steel
CN113268925A (en) Dynamic soft measurement method based on differential evolution algorithm time delay estimation
CN110794676A (en) CSTR process nonlinear control method based on Hammerstein-Wiener model
CN116595880A (en) Intelligent mixed modeling method for boiling bed residual oil hydrogenation device
Roeva Improvement of genetic algorithm performance for identification of cultivation process models
CN111612227A (en) Load prediction method based on K-means clustering and bat optimization neural network
CN110909492A (en) Sewage treatment process soft measurement method based on extreme gradient lifting algorithm
CN116522747A (en) Two-stage optimized extrusion casting process parameter optimization design method
CN115619028A (en) Clustering algorithm fusion-based power load accurate prediction method
CN114066250A (en) Method, device, equipment and storage medium for measuring and calculating repair cost of power transmission project
CN115034140A (en) Surface water quality change trend prediction method based on key control factors

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