Water electrolysis hydrogen production monitoring and control method with dynamic fluctuation adaptability

Abstract

The invention provides a water electrolysis hydrogen production monitoring and control method with dynamic fluctuation adaptability, and relates to the technical field of hydrogen production. Comprising the steps that a multi-dimensional operation parameter set of an electrolytic cell is collected in real time through a multi-source sensor array, an electrolysis process prediction model is established based on the LSTM and the CNN neural network, and therefore the prediction result of the electrolysis state is determined. And according to the prediction result, a dynamic parameter adjustment matrix is constructed, and the electrolytic current density, the electrolyte concentration and other key parameters of the electrolytic cell are adjusted for the first time through fuzzy PID control. After adjustment, the multi-dimensional operation parameter set is updated, the health degree index of the electrolytic cell is calculated, and the relation between the index and a preset health index threshold value is monitored in real time. And second-time adjustment is carried out according to the monitoring result, so that safe and efficient operation of the electrolytic cell under the dynamic fluctuation condition is ensured. The problems that in the prior art, an electrolytic cell is slow in response to load change, insufficient in monitoring capacity, dependent on manual maintenance and the like are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen production, and particularly to a method for monitoring and controlling electrolytic water hydrogen production with dynamic fluctuation adaptability. Background Art

[0002] With the rapid development of renewable energy, especially the increasingly widespread application of wind energy and solar energy, the electrolytic water hydrogen production technology has gradually become an important method for hydrogen production. As a clean energy source, hydrogen is of great significance in reducing greenhouse gas emissions and addressing global climate change. Therefore, improving the production efficiency and stability of hydrogen has a profound impact on the future application of renewable energy.

[0003] Among many electrolytic water hydrogen production technologies, the anion exchange membrane (AEM) electrolytic water hydrogen production technology has attracted much attention due to its high efficiency, simple operation, and good durability. Compared with the traditional proton exchange membrane (PEM) electrolytic water hydrogen production system, the AEM system has lower energy consumption and cost advantages, and is more adaptable to power fluctuations. This enables the AEM electrolytic water hydrogen production technology to have good adaptability in the case of a high proportion of renewable energy, and its application shows an increasingly important role in the peak shaving and load balancing of renewable energy power generation systems.

[0004] However, the existing monitoring technology for hydrogen production by electrolysis of water still has shortcomings. For example, patent CN117071000B discloses a remote safety monitoring system for PEM electrolyzers, including an equipment parameter analysis module, a membrane state analysis module, an impurity abnormality analysis module, an early warning evaluation analysis module, a cleaning treatment analysis module, and a maintenance database. Although the patent collects current, voltage, and temperature parameters in real time through sensors, and generates an equipment operation evaluation index to determine whether the equipment is in normal working condition, there are still significant defects in actual operation. First, when the equipment load changes rapidly, the system lacks the ability to adjust the electrolysis parameters in real time and cannot dynamically respond to the state changes in the electrolyzer. Especially when the output of renewable energy sources such as wind power and solar energy fluctuates, the traditional system reacts slowly, which may lead to a decrease in the stability of the electrolysis process, and ultimately affect the production and purity of hydrogen. Secondly, the existing monitoring technology mainly focuses on single parameter monitoring, lacks comprehensive analysis and optimization of multiple key indicators of the electrolysis process. Therefore, during the operation of the equipment, it is impossible to effectively integrate the information of each module, make intelligent fault warnings and maintenance decisions, and increase the potential risk of equipment failure. Finally, the strong dependence on manual maintenance is also a major problem of the existing monitoring system. Current maintenance work often relies on regular manual inspections, and it is difficult to achieve fully automated monitoring, which not only increases the cost of operation and maintenance, but also makes it impossible to adjust or handle equipment in a timely manner in emergencies, affecting the overall hydrogen production efficiency and safety. In summary, the monitoring and control of anion exchange membrane (AEM) water electrolysis hydrogen production technology is crucial, and the shortcomings of existing monitoring technologies in dynamic adaptability, comprehensive monitoring capabilities, and intelligent decision-making capabilities provide technical challenges that need to be urgently addressed for the development of water electrolysis hydrogen production technology. Summary of the invention

[0005] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a monitoring and control method for hydrogen production by electrolysis of water with dynamic fluctuation adaptability. The present invention solves the problems in the prior art of insufficient dynamic adaptability, comprehensive monitoring capability and intelligent decision-making capability of hydrogen production by electrolysis monitoring technology.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for monitoring and controlling hydrogen production by electrolysis of water with dynamic fluctuation adaptability, comprising:

[0008] The multi-dimensional operating parameter set of the electrolyzer is collected in real time through a multi-source sensor array;

[0009] Based on the LSTM neural network and the CNN neural network, according to the multi-dimensional operating parameter set and the historical data of the electrolytic cell, an electrolysis process prediction model is constructed to determine the electrolysis state prediction result of the electrolytic cell within a preset time;

[0010] Construct a dynamic parameter adjustment matrix for the electrolytic cell according to the predicted result of the electrolytic state;

[0011] Use the dynamic parameter adjustment matrix to perform the first adjustment on the electrolytic current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell through fuzzy PID;

[0012] Update the multi-dimensional operating parameter set according to the adjusted electrolytic cell to obtain an updated parameter set;

[0013] Calculate the health index of the electrolytic cell using the updated parameter set;

[0014] Monitor the health index of the electrolytic cell in real time according to a preset health index threshold, obtain a monitoring result, and perform a second adjustment on the electrolytic cell according to the monitoring result.

[0015] Preferably, the multi-dimensional operating parameter set includes:

[0016] Current density, voltage fluctuation, temperature gradient, pressure change, electrolyte flow rate, hydrogen purity, and anion exchange membrane stress parameters.

[0017] Preferably, the construction method of the electrolysis process prediction model is as follows:

[0018] Obtain the multi-dimensional operating parameter set and perform data preprocessing to obtain a preprocessed multi-dimensional operating parameter set;

[0019] Construct the preprocessed multi-dimensional operating parameter set into three-dimensional input data according to time steps;

[0020] Construct a CNN sub-model and an LSTM sub-model. Among them, the CNN sub-model includes an input layer, a convolutional layer, a pooling layer, and an output layer, and the number of LSTM units in the LSTM sub-model is 128;

[0021] Input the three-dimensional input data into the CNN sub-model to obtain the input data of the LSTM sub-model and input it into the LSTM sub-model to obtain a first output result;

[0022] Add a fully connected layer to the LSTM sub-model and convert the first output result into a second output result;

[0023] Use a linear activation function and the second output result to determine the electrolysis process prediction result;

[0024] Determine an initial electrolysis process prediction model according to the electrolysis process prediction result and the constructed CNN sub-model and LSTM sub-model;

[0025] Train the initial electrolysis process prediction model to obtain the final electrolysis process prediction model.

[0026] Preferably, the expression of the electrolysis process prediction model is:

[0027]

[0028] where X t-N:t is the input multi-dimensional operating parameter set, the time window is t-N:t, is the electrolysis state prediction result at T time steps, σ is the linear activation function, W FC is the fully connected layer, b FC is the bias term.

[0029] Preferably, the expression of the dynamic parameter adjustment matrix is:

[0030]

[0031] where M adjust is the dynamic parameter adjustment matrix, are the predicted current density, concentration change rate, temperature gradient, and pressure fluctuation amplitude respectively, is the time derivative of the predicted parameter, is the short-term fluctuation amount of the predicted parameter, is the spatial gradient of the predicted parameter, α * β * γ * δ * are the first coupling weight coefficient, the second coupling weight coefficient, the third coupling weight coefficient, and the fourth coupling weight coefficient respectively.

[0032] Preferably, the first adjustment of the electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell by using the dynamic parameter adjustment matrix through fuzzy PID includes:

[0033] Extract the current electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell according to the multi-dimensional operating parameter set;

[0034] Design a fuzzy PID controller and calculate the output of the PID controller;

[0035] Determine the adjustment factor according to the dynamic parameter adjustment matrix;

[0036] Perform the first adjustment on the current electrolysis current density, electrolyte concentration, working temperature, and pressure according to the output of the PID controller and the adjustment factor.

[0037] Preferably, the design of the fuzzy PID controller and the calculation of the output of the PID controller include:

[0038] Establish fuzzy control rules according to the characteristics of the input variables, where the input variables are the electrolysis current density error, concentration error, temperature error, and pressure error;

[0039] Define fuzzy sets and design corresponding membership functions according to the fuzzy control rules;

[0040] Construct a fuzzy control rule base based on the fuzzy sets;

[0041] Convert the actual error value into a fuzzy value and obtain the fuzzy output integrated through the fuzzy inference mechanism according to the fuzzy control rule base;

[0042] Defuzzify the fuzzy output to obtain the output control signal.

[0043] Preferably, the calculation expression of the health index is:

[0044]

[0045] where HI t is the health index, ξ i,t is the real-time value of the i-th monitoring parameter, ξ i,ref is the reference value of the i-th parameter under standard working conditions, Δξ i,t is the short-term fluctuation of the i-th parameter, ω i is the parameter weight coefficient, λ i is the parameter sensitivity attenuation factor, η is the dynamic control compensation coefficient, ||M adjust || F is the Frobenius norm of the dynamic parameter adjustment matrix, ||M adjust,0 || F is the Frobenius norm of the initial state adjustment matrix.

[0046] A monitoring and control system for electrolytic water hydrogen production with dynamic fluctuation adaptability, comprising:

[0047] A data acquisition module for real-time collecting a multi-dimensional operation parameter set of the electrolytic cell through a multi-source sensor array;

[0048] An electrolysis state prediction module for constructing an electrolysis process prediction model based on the LSTM neural network and the CNN neural network according to the multi-dimensional operation parameter set and the historical data of the electrolytic cell to determine the electrolysis state prediction result of the electrolytic cell within a preset time;

[0049] An adjustment matrix determination module for constructing a dynamic parameter adjustment matrix of the electrolytic cell according to the electrolysis state prediction result;

[0050] The first parameter adjustment module is used to perform the first adjustment on the electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell through fuzzy PID by using the dynamic parameter adjustment matrix;

[0051] The update module is used to update the multi-dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set;

[0052] The health index calculation module is used to calculate the health index of the electrolytic cell by using the updated parameter set;

[0053] The second parameter adjustment module is used to monitor the health index of the electrolytic cell in real time according to a preset health index threshold, obtain a monitoring result, and perform a second adjustment on the electrolytic cell according to the monitoring result.

[0054] The present invention discloses the following technical effects:

[0055] The present invention provides a method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability, which is characterized by including: collecting a multi-dimensional operation parameter set of an electrolytic cell in real time through a multi-source sensor array; based on an LSTM neural network and a CNN neural network, constructing an electrolysis process prediction model according to the multi-dimensional operation parameter set and historical data of the electrolytic cell to determine a prediction result of the electrolysis state of the electrolytic cell within a preset time; constructing a dynamic parameter adjustment matrix according to the electrolysis state prediction result; using the dynamic parameter adjustment matrix to perform a first adjustment on the electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell through fuzzy PID; updating the multi-dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set; calculating a health index of the electrolytic cell by using the updated parameter set; and performing real-time monitoring on the health index of the electrolytic cell according to a preset health index threshold, obtaining a monitoring result and performing a second adjustment on the electrolytic cell according to the monitoring result. By collecting multi-dimensional operation parameters in real time through a multi-source sensor array and combining LSTM and CNN neural networks for data analysis, the present invention realizes immediate and accurate state perception and prediction of the electrolytic cell, and improves the dynamic adjustment ability of the system during load fluctuations. It can respond in a timely manner to changes in external conditions, thereby optimizing the electrolysis process and improving the production stability and efficiency of hydrogen; the electrolysis process prediction model is constructed based on a multi-dimensional operation parameter set and historical data, and through a dynamic parameter adjustment matrix, comprehensive adjustment of the electrolytic cell is performed, solving the problem of single-parameter blind spots in existing monitoring technologies, and realizing real-time monitoring and optimized adjustment of multiple key parameters such as current density, electrolyte concentration, working temperature and pressure, so that the system can more comprehensively and scientifically evaluate the performance of the electrolytic cell during operation; using a dynamic parameter adjustment matrix, intelligent adjustment of the electrolytic cell is performed through fuzzy PID control, and at the same time, a health index is calculated for real-time monitoring, reducing the dependence on manual maintenance and improving the level of maintenance automation. If an abnormality occurs, the system can quickly respond and perform a second adjustment, thereby reducing the risk of equipment failure and improving the safety and efficiency of hydrogen production. Description of the Drawings

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings without creative efforts.

[0057] Figure 1 Flowchart of a method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability provided by an embodiment of the present invention;

[0058] Figure 2Schematic diagram of a monitoring and control system for electrolytic water hydrogen production with dynamic fluctuation adaptability provided by an embodiment of the present invention.

[0059] Explanation of reference numerals:

[0060] 1 - Data acquisition module, 2 - Electrolysis state prediction module, 3 - Adjustment matrix determination module, 4 - First parameter adjustment module, 5 - Update module, 6 - Health index calculation module, 7 - Second parameter adjustment module. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0062] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0063] As Figure 1 shown, the present invention provides a method for monitoring and controlling electrolytic water hydrogen production with dynamic fluctuation adaptability, including:

[0064] Step 100: Real - time collect a multi - dimensional operation parameter set of the electrolytic cell through a multi - source sensor array;

[0065] Step 200: Based on the LSTM neural network and the CNN neural network, construct an electrolysis process prediction model according to the multi - dimensional operation parameter set and the historical data of the electrolytic cell to determine the electrolysis state prediction result of the electrolytic cell within a preset time;

[0066] Step 300: Construct a dynamic parameter adjustment matrix of the electrolytic cell according to the electrolysis state prediction result;

[0067] Step 400: Use the dynamic parameter adjustment matrix to perform the first adjustment on the electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell through fuzzy PID;

[0068] Step 500: Update the multi - dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set;

[0069] Step 600: Calculate the health index of the electrolytic cell using the updated parameter set;

[0070] Step 700: Monitor the health index of the electrolytic cell in real time according to a preset health index threshold, obtain the monitoring result, and make a second adjustment to the electrolytic cell according to the monitoring result.

[0071] Further, the multi-dimensional operation parameter set includes:

[0072] Current density, voltage fluctuation, temperature gradient, pressure change, electrolyte flow rate, hydrogen purity, and anion exchange membrane stress parameter.

[0073] Specifically, the selection of sensors includes:

[0074] Current sensor: Used to measure the current density of the electrolytic cell.

[0075] Voltage sensor: Used to monitor the voltage fluctuation of the electrolytic cell.

[0076] Temperature sensor: Used to detect the temperature distribution inside the electrolytic cell.

[0077] Pressure sensor: Used to monitor the pressure change inside the electrolytic cell.

[0078] Flow sensor: Used to measure the flow rates of the electrolyte and hydrogen.

[0079] Gas analyzer: Used to detect the purity of hydrogen.

[0080] Strain sensor: Used to monitor the stress of the anion exchange membrane.

[0081] According to the structure of the electrolytic cell, arrange the sensors reasonably to ensure that each sensor can accurately monitor the parameters it needs. For the electrolytic cell, follow the following arrangement principles:

[0082] Uniform distribution: Ensure that each sensor covers different areas of the electrolytic cell to obtain comprehensive monitoring data.

[0083] Close to key components: The gas analyzer should be close to the hydrogen outlet, while the current and voltage sensors should be close to the electrodes.

[0084] Further, the construction method of the electrolysis process prediction model is:

[0085] Obtain the multi-dimensional operation parameter set and perform data preprocessing to obtain the preprocessed multi-dimensional operation parameter set;

[0086] Construct the preprocessed multi-dimensional operation parameter set into three-dimensional input data according to time steps;

[0087] Specifically, the time series data is used to construct three-dimensional input data according to a window of a fixed length (for example, the window length is 10, that is, every 10 time points are used as an input sample); the shape of the three-dimensional data is (number of samples, time steps, number of features). For example, if there are 1000 time point data and a window of length 10 is used: the shape of the input data = (991, 10, number of features). Assuming the number of features is 7, the final input shape is (991, 10, 7).

[0088] Construct a CNN sub-model and an LSTM sub-model. Among them, the CNN sub-model includes: an input layer, a convolutional layer, a pooling layer, and an output layer. The number of LSTM units in the LSTM sub-model is 128;

[0089] Input the three-dimensional input data into the CNN sub-model to obtain the input data of the LSTM sub-model and input it into the LSTM sub-model to obtain a first output result;

[0090] Specifically, the constructed three-dimensional input data is passed into the CNN sub-model. The CNN will extract higher-level features through convolutional and pooling operations, and the feature output from the CNN is passed into the LSTM sub-model. The LSTM will use the historical information of the time series for further analysis to capture the dynamic changes in the electrolysis process.

[0091] Add a fully connected layer to the LSTM sub-model and convert the first output result into a second output result;

[0092] Specifically, a fully connected layer is added after the output of the LSTM sub-model, and this layer maps the output of the LSTM to the final prediction result.

[0093] Use a linear activation function and the second output result to determine the prediction result of the electrolysis process;

[0094] According to the prediction result of the electrolysis process and the constructed CNN sub-model and LSTM sub-model, determine an initial electrolysis process prediction model;

[0095] Train the initial electrolysis process prediction model to obtain a final electrolysis process prediction model.

[0096] Furthermore, the expression of the electrolysis process prediction model is:

[0097]

[0098] Among them, X t-N:t is the input multi-dimensional operating parameter set, the time window is t - N:t, is the prediction result of the electrolysis state at T time steps, σ is the linear activation function, W FC is the fully connected layer, bFC is the bias term.

[0099] Expanding the above formula gives:

[0100] Input feature extraction (CNN part)

[0101] F CNN = ReLU(MaxPool(Conv3D(X t-N:t , K conv )));

[0102] Temporal feature modeling (LSTM part)

[0103] H t = LSTM θ (F CNN , H t-1 );

[0104] Prediction output (fully connected layer)

[0105]

[0106] Among them, F CNN is the high-dimensional feature tensor output by the CNN, capturing the spatial coupling relationship of multiple parameters, and K conv is the CNN convolution kernel parameter, used to extract spatial local features (such as the spatial correlation between temperature gradient and current density), and H t is the hidden state of the LSTM, encoding temporal dynamic features (such as the hysteresis effect of pressure fluctuations on membrane stress).

[0107] In the above formula, the CNN is used to extract the spatial correlation of multiple parameters (such as the local coupling of "temperature gradient - membrane stress"), and the LSTM captures the temporal evolution law (such as the voltage fluctuation trend), overcoming the limitation that traditional single models only focus on time series or space. The prediction error is reduced by about 18% compared with the single LSTM model (verified by simulation data), especially in high-fluctuation scenarios (such as sudden changes in wind speed), the prediction stability is significantly improved.

[0108] Furthermore, the expression of the dynamic parameter adjustment matrix is:

[0109]

[0110] Among them, M adjust is the dynamic parameter adjustment matrix, are the predicted value of current density, the rate of change of concentration, the temperature gradient, and the amplitude of pressure fluctuation respectively, is the time derivative of the predicted parameter, is the short-term fluctuation amount of the predicted parameter, is the spatial gradient of the predicted parameter, and α * and β *, γ * , δ * are the first coupling weight coefficient, the second coupling weight coefficient, the third coupling weight coefficient, and the fourth coupling weight coefficient, respectively.

[0111] Furthermore, the first adjustment of the electrolysis current density, the electrolyte concentration, the working temperature, and the pressure of the electrolytic cell by using the dynamic parameter adjustment matrix through fuzzy PID includes:

[0112] Extract the current electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell according to the multi-dimensional operating parameter set;

[0113] Design a fuzzy PID controller and calculate the output of the PID controller;

[0114] Determine the adjustment factor according to the dynamic parameter adjustment matrix;

[0115] Perform the first adjustment of the current electrolysis current density, electrolyte concentration, working temperature, and pressure according to the output of the PID controller and the adjustment factor.

[0116] Specifically, perform the first adjustment of the electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell according to the calculated output of the PID controller and the value of the dynamic adjustment factor.

[0117] Electrolysis current density: Adjust the electrolytic cell current to ensure production efficiency.

[0118] Electrolyte concentration: Adjust the composition of the electrolyte as needed to optimize the reaction efficiency.

[0119] Working temperature: Adjust the temperature inside the electrolytic cell through cooling or heating measures.

[0120] Pressure: Dynamically increase or decrease gas emissions or inputs according to electrolysis requirements and safety standards.

[0121] Furthermore, the design method of the fuzzy PID controller includes:

[0122] Establish fuzzy control rules according to the characteristics of the input variables, where the input variables are the electrolysis current density error, concentration error, temperature error, and pressure error;

[0123] Define fuzzy sets and design corresponding membership functions, for example:

[0124] The membership functions set for the electrolysis current density error include "Negative Big" (NB), "Negative Small" (NS), "Zero" (ZE), "Positive Small" (PS), "Positive Big" (PB), corresponding to different error ranges respectively;

[0125] Define the membership function similarly for the output quantities (such as the adjustment quantities of electrolysis current density, concentration, temperature, and pressure) to ensure that there are corresponding fuzzy sets for different output quantities;

[0126] Construct a fuzzy control rule base. For example:

[0127] If the current density error is "negative large" and the concentration error is "negative small", then the adjustment quantity of the electrolysis current density is "positive large";

[0128] If the temperature error is "positive small" and the pressure error is "zero", then the adjustment quantity of the working temperature is "negative small";

[0129] In the process of fuzzy inference, use the fuzzy control rules to fuzzify the input, and convert the actual error value into a fuzzy value;

[0130] The fuzzy output obtained through the integration of the fuzzy inference mechanism is finally obtained as a specific control output signal through a defuzzification method (such as the centroid method), and this signal is used to guide the parameter adjustment of the electrolytic cell.

[0131] Furthermore, the health index is a comprehensive index calculated by comprehensively considering various operating parameters of the electrolytic cell (such as current density, temperature, pressure, hydrogen purity, etc.), and is used to evaluate the current state and overall health level of the electrolytic cell.

[0132] Compare the calculated health index with a preset threshold to classify the electrolytic cell into a healthy state or an alarm state.

[0133] If the monitoring result shows that the health index is abnormal, first analyze the key parameters that cause the change of the health index (for example: too low electrolyte concentration, too high temperature, etc.).

[0134] Adjustment measures:

[0135] Determine the adjustment strategy according to the analysis result. The adjustment measures include:

[0136] Electrolysis current density: Increase or decrease the electrolysis current density in response to the health state.

[0137] Electrolyte concentration: Adjust the ratio of the electrolyte to ensure that its concentration remains in the ideal range.

[0138] Temperature regulation: Adjust the temperature in the cell by increasing cooling or heating to restore it to the appropriate range.

[0139] Pressure adjustment: Increase or decrease the system pressure as needed to stabilize the electrolysis reaction.

[0140] Furthermore, the calculation expression of the health index is:

[0141]

[0142] Among them, HI t is the health index, ξ i,t is the real-time value of the i-th monitoring parameter, ξ i,ref is the reference value of the i-th parameter under standard working conditions, Δξ i,t is the short-term fluctuation of the i-th parameter, ω i is the parameter weight coefficient, λ i is the parameter sensitivity attenuation factor, η is the dynamic control compensation coefficient, ||M adjust || F is the Frobenius norm of the dynamic parameter adjustment matrix, ||M adjust,0 || F is the Frobenius norm of the initial state adjustment matrix.

[0143] As Figure 2 shown, the present invention also provides a monitoring and control system for electrolytic water hydrogen production with dynamic fluctuation adaptability, including:

[0144] A data acquisition module 1, configured to collect a multi-dimensional operation parameter set of the electrolytic cell in real time through a multi-source sensor array;

[0145] An electrolysis state prediction module 2, configured to construct an electrolysis process prediction model based on an LSTM neural network and a CNN neural network according to the multi-dimensional operation parameter set and the historical data of the electrolytic cell to determine an electrolysis state prediction result of the electrolytic cell within a preset time;

[0146] An adjustment matrix determination module 3, configured to construct a dynamic parameter adjustment matrix of the electrolytic cell according to the electrolysis state prediction result;

[0147] A first parameter adjustment module 4, configured to perform a first adjustment on the electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell by using the dynamic parameter adjustment matrix through fuzzy PID;

[0148] An update module 5, configured to update the multi-dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set;

[0149] A health index calculation module 6, configured to calculate the health index of the electrolytic cell by using the updated parameter set;

[0150] A second parameter adjustment module 7, configured to perform real-time monitoring on the health index of the electrolytic cell according to a preset health index threshold, obtain a monitoring result and perform a second adjustment on the electrolytic cell according to the monitoring result.

[0151] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other.

[0152] In this article, specific examples are used to elaborate on the principles and implementation manners of the present invention. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present invention. At the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation on the present invention.

Claims

1. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability, characterized in that, Including: Real-time collecting a multi-dimensional operation parameter set of an electrolytic cell through a multi-source sensor array; Based on an LSTM neural network and a CNN neural network, constructing an electrolysis process prediction model according to the multi-dimensional operation parameter set and historical data of the electrolytic cell to determine a prediction result of the electrolysis state of the electrolytic cell within a preset time; Constructing a dynamic parameter adjustment matrix of the electrolytic cell according to the electrolysis state prediction result; Using the dynamic parameter adjustment matrix to perform a first adjustment on the electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell through fuzzy PID; Updating the multi-dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set; Calculating a health index of the electrolytic cell by using the updated parameter set; Performing real-time monitoring on the health index of the electrolytic cell according to a preset health index threshold, obtaining a monitoring result and performing a second adjustment on the electrolytic cell according to the monitoring result.

2. The electrolyzed water hydrogen production monitoring and control method with dynamic fluctuation adaptability according to claim 1, characterized in that, The multi-dimensional operation parameter set includes: Current density, voltage fluctuation, temperature gradient, pressure change, electrolyte flow rate, hydrogen purity and anion exchange membrane stress parameter.

3. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 1, characterized in that, The construction method of the electrolysis process prediction model is: Obtaining the multi-dimensional operation parameter set and performing data preprocessing to obtain a preprocessed multi-dimensional operation parameter set; Constructing the preprocessed multi-dimensional operation parameter set into three-dimensional input data according to time steps; Constructing a CNN sub-model and an LSTM sub-model, wherein the CNN sub-model includes: an input layer, a convolutional layer, a pooling layer and an output layer, and the number of LSTM units of the LSTM sub-model is 128; Inputting the three-dimensional input data into the CNN sub-model to obtain input data of the LSTM sub-model and inputting the input data into the LSTM sub-model to obtain a first output result; Adding a fully connected layer to the LSTM sub-model and converting the first output result into a second output result; Using a linear activation function and the second output result to determine an electrolysis process prediction result; Determining an initial electrolysis process prediction model according to the electrolysis process prediction result and the constructed CNN sub-model and LSTM sub-model; Training the initial electrolysis process prediction model to obtain a final electrolysis process prediction model.

4. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 3, characterized in that, The expression of the electrolysis process prediction model is: Among them, X t-N:t is the input multi-dimensional operating parameter set, and the time window is t-N:t, is the predicted result of the electrolysis state at T time steps, σ is the linear activation function, W FC is the fully connected layer, b FC is the bias term.

5. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 3, characterized in that The expression of the dynamic parameter adjustment matrix is: Among them, M adjust is the dynamic parameter adjustment matrix, are respectively the predicted value of current density, the rate of change of concentration, the temperature gradient, and the amplitude of pressure fluctuation, is the time derivative of the predicted parameter, is the short-term fluctuation amount of the predicted parameter, is the spatial gradient of the predicted parameter, α * , β * , γ * , δ * are respectively the first coupling weight coefficient, the second coupling weight coefficient, the third coupling weight coefficient, and the fourth coupling weight coefficient.

6. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 2, characterized in that, The performing the first adjustment on the electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell through fuzzy PID by using the dynamic parameter adjustment matrix includes: Extracting the current electrolysis current density, electrolyte concentration, working temperature and pressure of the electrolytic cell according to the multi-dimensional operation parameter set; Designing a fuzzy PID controller and calculating the output of the PID controller; Determining an adjustment factor according to the dynamic parameter adjustment matrix; Performing a first adjustment on the current electrolysis current density, electrolyte concentration, working temperature and pressure according to the output of the PID controller and the adjustment factor.

7. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 6, characterized in that, The designing a fuzzy PID controller and calculating the output of the PID controller includes: Fuzzy control rules are established according to the characteristics of the input variables, where the input variables are the electrolysis current density error, concentration error, temperature error, and pressure error; According to the fuzzy control rules, fuzzy sets are defined and corresponding membership functions are designed; Based on the fuzzy sets, a fuzzy control rule base is constructed; The actual error value is converted into a fuzzy value and the fuzzy output obtained by integrating through the fuzzy inference mechanism according to the fuzzy control rule base; The fuzzy output is defuzzified to obtain the output control signal.

8. A method for monitoring and controlling hydrogen production by electrolyzing water with dynamic fluctuation adaptability according to claim 6, characterized in that The calculation expression of the health index is: Among them, HI t is the health index, ξ i,t is the real-time value of the i-th monitoring parameter, ξ i,ref is the reference value of the i-th parameter under standard working conditions, Δξ i,t is the short-term fluctuation of the i-th parameter, ω i is the parameter weight coefficient, λ i is the parameter sensitivity attenuation factor, η is the dynamic control compensation coefficient, ||M adjust || F is the Frobenius norm of the dynamic parameter adjustment matrix, ||M adjust,0 || F is the Frobenius norm of the initial state adjustment matrix.

9. A monitoring and control system for electrolytic water hydrogen production with dynamic fluctuation adaptability, characterized in that, Including: A data acquisition module for real-time collecting a multi-dimensional operation parameter set of the electrolytic cell through a multi-source sensor array; An electrolysis state prediction module for constructing an electrolysis process prediction model based on an LSTM neural network and a CNN neural network according to the multi-dimensional operation parameter set and the historical data of the electrolytic cell to determine the electrolysis state prediction result of the electrolytic cell within a preset time; An adjustment matrix determination module for constructing a dynamic parameter adjustment matrix of the electrolytic cell according to the electrolysis state prediction result; A first parameter adjustment module for using the dynamic parameter adjustment matrix to perform the first adjustment on the electrolysis current density, electrolyte concentration, working temperature, and pressure of the electrolytic cell through fuzzy PID; An update module for updating the multi-dimensional operation parameter set according to the adjusted electrolytic cell to obtain an updated parameter set; A health index calculation module for calculating the health index of the electrolytic cell by using the updated parameter set; A second parameter adjustment module for real-time monitoring the health index of the electrolytic cell according to a preset health index threshold, obtaining a monitoring result and performing a second adjustment on the electrolytic cell according to the monitoring result.

Images