Electric hydrogen production system temperature optimization control method based on self-adaptive control

Abstract

The invention provides an electric hydrogen production system temperature optimization control method based on self-adaptive control, and relates to the technical field of electric hydrogen production system temperature control. Comprising the following steps: acquiring a current signal of an electrolytic bath at the current moment, a current instruction signal provided by a power grid at a future moment, historical current data and environmental data, and predicting the current signal by using a prediction algorithm to obtain a current signal at the future moment; determining a temperature state set corresponding to the electrolytic bath according to the current signal at the future moment, wherein the temperature state set comprises the bath front temperature of the electrolytic bath, the bath rear temperature of the electrolytic bath and the cooling water return temperature in the electric hydrogen production system; constructing a corresponding opening degree prediction model according to each temperature in the temperature state set to determine a cold water valve regulation and control sub-signal set; determining a final regulation and control signal according to the regulation and control sub-signal set; and determining the opening degree of the cold water valve according to the regulation and control signal so as to optimally control the temperature of the electric hydrogen production system. The problem of low temperature control accuracy in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control of an electrolytic hydrogen production system, and particularly to a method for optimizing the temperature control of an electrolytic hydrogen production system based on adaptive control. Background Art

[0002] The structure of the electrolytic hydrogen production system is as follows Figure 2 As shown, it includes two loops: an alkali solution loop and a cold water loop. In the alkali solution loop, an electrolytic water reaction occurs in the electrolyzer to generate hydrogen gas (H2) and oxygen gas (O2). The product gas is carried out of the electrolyzer by the circulating alkali solution and enters the gas-alkali separator. In the separator, the product gas is separated from the alkali solution, and the gas leaves the electrolysis system from above the separator and is utilized or stored by subsequent processes; the alkali solution returns to the electrolyzer after passing through a filter and a circulation pump.

[0003] In an alkaline electrolytic water hydrogen production system, an electrolytic water reaction occurs in the electrolyzer to generate product hydrogen and oxygen. This electrolytic reaction releases a large amount of heat. To prevent damage to the diaphragm caused by the over-limit temperature of the electrolyzer, it is necessary to use the cooling water loop to cool the alkali solution. The cooling water loop includes a chiller, a cold water valve, and a heat exchanger (such as a heat exchange coil) placed in the separator. The cooling water flowing out of the chiller cools the alkali solution through the heat exchanger in the separator, and then the alkali solution returns to the electrolyzer, thereby indirectly controlling the temperature of the electrolyzer. Therefore, the temperature control of the electrolyzer can be achieved by changing the opening degree of the cold water valve. When the opening degree of the cold water valve increases, the cooling water flow rate increases, and the cooling effect on the electrolyzer is enhanced; when the opening degree of the cold water valve decreases, the cooling water flow rate decreases, and the temperature effect on the electrolyzer, and the cooling effect on the electrolyzer decreases.

[0004] Regarding the temperature control of the electrolytic hydrogen production system, the existing patent CN113930805B discloses a temperature prediction control method and device for an electrolytic hydrogen production system. This patent is based on the current signal to predict the temperature changes before and after the cell. The controller adjusts the opening degree of the valve according to the predicted temperature to keep the system temperature within the set range. This patent relies on the real-time feedback of temperature changes, monitors the current to infer the temperature state, and thus makes corresponding control decisions. However, the method for obtaining the current signal in the current patent only relies on the command signal of the power grid for prediction, and the prediction accuracy is low. Moreover, the current patent does not monitor the opening degree of the cold water valve in real time. When rapid temperature adjustment is required, the actual state of the cold water valve cannot be accurately grasped. This not only affects the control accuracy but also increases the safety hazards of the system in case of emergencies. 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 method for optimizing the temperature control of an electrolytic hydrogen production system based on adaptive control.

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

[0007] A temperature optimization control method for an electrolytic hydrogen production system based on adaptive control, comprising:

[0008] Obtaining the current signal of the electrolytic cell at the current moment, the current command signal of the future moment provided by the power grid, historical current data and environmental data, and using a prediction algorithm to predict the current signal to obtain the current signal of the future moment;

[0009] Determining the temperature state set corresponding to the electrolytic cell according to the current signal of the future moment, wherein the temperature state set includes: the pre-tank temperature of the electrolytic cell, the post-tank temperature of the electrolytic cell, and the return water temperature of the cooling water in the electrolytic hydrogen production system;

[0010] Constructing a corresponding opening prediction model according to each temperature in the temperature state set to determine the cold water valve control sub-signal set; wherein, the control sub-signal set includes: a first control sub-signal, a second control sub-signal, and a third control sub-signal;

[0011] Determining the final control signal according to the control sub-signal set;

[0012] Determining the opening of the cold water valve according to the control signal to optimize the control of the temperature of the electrolytic hydrogen production system.

[0013] Preferably, it further includes:

[0014] Determining the opening rate of the cold water valve within a preset time interval;

[0015] Comparing the opening rate of the cold water valve within the preset time interval with a preset opening rate threshold to obtain a comparison result;

[0016] Determining whether there is an abnormality in the temperature state set according to the comparison result.

[0017] Preferably, obtaining the current signal of the electrolytic cell at the current moment, the current command signal of the future moment provided by the power grid, historical current data and environmental data, and using a prediction algorithm to predict the current signal to obtain the current signal of the future moment, includes:

[0018] Determining the initial current signal of the future moment according to the current command signal of the future moment provided by the power grid;

[0019] Constructing a corresponding knowledge graph data according to the initial current signal of the future moment, historical current data and environmental data;

[0020] Performing feature extraction on the knowledge graph data to obtain a corresponding feature set;

[0021] Inputting the feature set into the constructed current prediction model to obtain the current signal of the future moment;

[0022] The expression of the current prediction model is as follows:

[0023]

[0024] where, is the current signal at a future time, f() is a function modeled based on GBDT, and x t-i represents the current signal at time t-i, and Z t is the feature set extracted from the knowledge graph, and θ is the model parameter set.

[0025] Preferably, the feature extraction of the knowledge graph data to obtain the corresponding feature set includes:

[0026] Performing correlation analysis on each node and edge of the knowledge graph data to obtain a correlation set;

[0027] The expression of the correlation set is: R = {(N i , N j , r ij ) where N i and N j are different nodes in the graph, and r ij represents the correlation between them;

[0028] According to the correlation set, determining the statistical features of the initial future-time current signal and historical current data and the change features of their environmental factors;

[0029] Aggregating the statistical features and the change features to obtain the corresponding feature set.

[0030] Preferably, the statistical features include:

[0031] Current mean, current standard deviation, and current change rate. Among them, the calculation expression of the current mean is:

[0032]

[0033] The expression of the current standard deviation is:

[0034]

[0035] The expression of the current change rate is:

[0036]

[0037] where N is the number of samples, I i is the current magnitude observed for the initial future-time current signal and historical current data, is the initial future-time current signal.

[0038] Preferably, determining the temperature state set corresponding to the electrolytic cell according to the current signal at the future moment includes:

[0039] Obtaining the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and the cooling water flow rate thereof;

[0040] Directly measuring the return water temperature of the cooling water by using a temperature sensor;

[0041] Calculating the pre-tank temperature of the electrolytic cell according to the current signal at the future moment, the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and the cooling water flow rate;

[0042] Calculating the post-tank temperature of the electrolytic cell according to the pre-tank temperature of the electrolytic cell;

[0043] Among them, the expressions of the pre-tank temperature and the post-tank temperature are respectively:

[0044]

[0045] T 槽后 (t) = T 槽前 (t) - ΔT loss (t)

[0046] Among them, T env , R cell is the equivalent resistance of the electrolytic cell, Q cool (t) is the cooling water flow rate, α and β are respectively the heat capacity and the heat dissipation coefficient of the electrolytic cell, and ΔT loss (t) is the temperature loss during the caustic soda circulation process.

[0047] Preferably, constructing a corresponding opening prediction model according to each temperature in the temperature state set to determine the cold water valve control sub-signal set includes:

[0048] Determining a first control sub-signal according to the pre-tank temperature;

[0049] Determining a second control sub-signal according to the post-tank temperature;

[0050] Determining a third control sub-signal according to the return water temperature of the cooling water;

[0051] Among them, the expression of the first control sub-signal is:

[0052]

[0053] The expression of the second control sub-signal is:

[0054]

[0055] The expression of the third control sub-signal is:

[0056]

[0057] Among them, K P , K I , K D are the first PID control parameter, the second PID control parameter, and the third PID control parameter respectively, and a is the heat exchange efficiency coefficient; Q max is the maximum flow rate of cooling water, e(t) is the deviation between the temperature in front of the tank and the set temperature, and λ1, λ2, and λ3 are the first weight coefficient, the second weight coefficient, and the third weight coefficient respectively.

[0058] Preferably, determining the final control signal according to the set of regulator signals includes:

[0059] Determining the dynamic weight coefficient corresponding to the set of regulator signals according to the dynamic weight distribution strategy;

[0060] Determining the final control signal according to the dynamic weight coefficient.

[0061] The present invention discloses the following technical effects:

[0062] The present invention provides a temperature optimization control method for an electrolytic hydrogen production system based on adaptive control, including: obtaining the current signal of the electrolytic cell at the current moment, the current command signal of the power grid at the future moment, historical current data, and environmental data, and using a prediction algorithm to predict the current signal to obtain the current signal at the future moment; determining the corresponding temperature state set of the electrolytic cell according to the current signal at the future moment, where the temperature state set includes: the temperature in front of the electrolytic cell, the temperature behind the electrolytic cell, and the return water temperature of the cooling water in the electrolytic hydrogen production system; constructing a corresponding opening prediction model according to each temperature in the temperature state set to determine the set of cold water valve regulator signals; among them, the set of regulator signals includes: the first regulator signal, the second regulator signal, and the third regulator signal; determining the final control signal according to the set of regulator signals; determining the opening of the cold water valve according to the control signal to optimize the temperature control of the electrolytic hydrogen production system. By combining the adaptive control algorithm and the instant prediction of multiple temperature state sets, the present invention optimizes the temperature control strategy of the electrolytic hydrogen production system. The specific beneficial effects include: improving stability: By real-time predicting and dynamically adjusting the opening of the cold water valve, the stability of the temperature in the electrolytic cell is ensured, and the influence on the hydrogen production efficiency caused by overheating or overcooling is avoided; improving energy efficiency: The optimized temperature control strategy effectively reduces energy consumption, improves the system operation efficiency, reduces the usage amount of cooling water, and thus realizes the energy-saving effect. Enhancing the system response ability: This method has strong adaptability and can quickly respond to different environmental conditions and load changes to maintain the high-efficiency operation of the system. Description of the Drawings

[0063] 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. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0064] Figure 1 Flowchart of a temperature optimization control method for an electrolytic hydrogen production system based on adaptive control provided by an embodiment of the present invention;

[0065] Figure 2 Schematic diagram of the structure of the electrolytic hydrogen production system provided by an embodiment of the present invention. Specific embodiments

[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0067] 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 drawings and specific embodiments.

[0068] As Figure 1 shown, the present invention provides a temperature optimization control method for an electrolytic hydrogen production system based on adaptive control, including:

[0069] Step 100: Obtain the current current signal of the electrolyzer at the current moment, the future current command signal provided by the power grid, historical current data, and environmental data, and use a prediction algorithm to predict the current signal to obtain the future current signal;

[0070] Step 200: Determine the corresponding temperature state set of the electrolyzer according to the future current signal, where the temperature state set includes: the pre-tank temperature of the electrolyzer, the post-tank temperature of the electrolyzer, and the return water temperature of the cooling water in the electrolytic hydrogen production system;

[0071] Step 300: Construct a corresponding opening prediction model according to each temperature in the temperature state set to determine the cold water valve regulation sub-signal set; where the regulation sub-signal set includes: a first regulation sub-signal, a second regulation sub-signal, and a third regulation sub-signal;

[0072] Step 400: Determine the final regulation signal according to the regulation sub-signal set;

[0073] Step 500: Determine the opening degree of the cold water valve according to the regulation signal to optimize the control of the temperature of the electrolytic hydrogen production system.

[0074] Furthermore, it also includes:

[0075] Determine the opening degree rate of the cold water valve within a preset time interval;

[0076] Compare the opening degree rate of the cold water valve within the preset time interval with a preset opening degree rate threshold to obtain a comparison result;

[0077] Determine whether there is an abnormality in the temperature state set according to the comparison result.

[0078] Specifically, the definition of the opening degree rate: within a specific time interval, the system records the current opening degree of the cold water valve. To determine the opening degree rate, the system will compare the opening degree changes of the cold water valve at the current moment and the future moment. Specifically, the system will calculate the change amount of the opening degree during the period from the current moment to the future moment, and then compare the change amount with the time interval to obtain the opening degree rate of the cold water valve.

[0079] In the system, a preset allowable opening degree rate threshold is set, which defines the maximum opening degree amplitude that the cold water valve can change per unit time. In this embodiment, it can be set that the valve is allowed to change at most 5% per second. In each control cycle, the system will compare the calculated opening degree rate of the cold water valve with the preset threshold. If the opening degree rate is within the allowable range, the system is normal; if the opening degree rate exceeds the set threshold, it indicates an abnormal situation.

[0080] Judge the abnormality according to the comparison result:

[0081] When the system detects that the opening degree rate exceeds the preset threshold, this situation will be marked as "abnormal".

[0082] If the opening degree rate is within the allowable range, the system state is considered "normal".

[0083] Furthermore, obtain the current current signal of the electrolytic cell, the future current command signal provided by the power grid, historical current data, and environmental data at the current moment, and use a prediction algorithm to predict the current signal to obtain the future current signal, including:

[0084] Determine the initial future current signal according to the future current command signal provided by the power grid;

[0085] Construct corresponding knowledge graph data according to the initial future current signal, historical current data, and environmental data;

[0086] Extract features from the knowledge graph data to obtain the corresponding feature set;

[0087] Input the feature set into the constructed current prediction model to obtain the current signal at a future moment;

[0088] The expression of the current prediction model is:

[0089]

[0090] where, is the current signal at a future moment, f() is a function modeled based on GBDT, x t-i represents the current signal at time t-i, Z t is the feature set extracted from the knowledge graph, and θ is the model parameter set.

[0091] Specifically, the system combines the initial current signal at a future moment with historical current data and environmental data to construct a knowledge graph. This graph can show the relationships between different variables, such as the mutual influence between the current signal and temperature, humidity.

[0092] Further, perform feature extraction on the knowledge graph data to obtain the corresponding feature set, including:

[0093] Perform correlation analysis on each node and edge of the knowledge graph data to obtain a correlation set;

[0094] The expression of the correlation set is: R = {(N i , N j , r ij ) where, N i and N j are different nodes in the graph, and r ij represents the correlation between them;

[0095] According to the correlation set, determine the statistical characteristics of the initial current signal at a future moment and historical current data and the change characteristics of their environmental factors;

[0096] Aggregate the statistical characteristics and the change characteristics to obtain the corresponding feature set.

[0097] Specifically, the system analyzes the nodes and edges in the knowledge graph, and the features related to the current signal will be extracted. These features include the current mean value, change amplitude, and the correlation with environmental factors. Each node in the graph represents a variable (such as current, temperature, humidity), and the edge represents the relationship between different variables.

[0098] Correlation analysis: The system first analyzes each node (variable) and its connected edges (relationships) in the knowledge graph: The system will evaluate the correlation between the current signal and temperature, humidity, and other influencing factors.

[0099] Pearson correlation coefficient is used in the correlation analysis to determine which factors have a significant impact on the current change.

[0100] Furthermore, the statistical features include:

[0101] The mean current, the standard deviation of the current, and the current change rate. Among them, the calculation expression of the mean current is:

[0102]

[0103] The expression of the standard deviation of the current is:

[0104]

[0105] The expression of the current change rate is:

[0106]

[0107] Among them, N is the number of samples, I i is the current magnitude obtained by observing the current signal at the initial future moment and the historical current data, is the current signal at the initial future moment.

[0108] Even further, the determination of the temperature state set corresponding to the electrolytic cell according to the current signal at the future moment includes:

[0109] Obtain the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and its cooling water flow rate;

[0110] Use a temperature sensor to directly measure the return water temperature of the cooling water;

[0111] Calculate the pre-tank temperature of the electrolytic cell according to the current signal at the future moment, the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and the cooling water flow rate;

[0112] Calculate the post-tank temperature of the electrolytic cell according to the pre-tank temperature of the electrolytic cell;

[0113] Among them, the expressions of the pre-tank temperature and the post-tank temperature are respectively:

[0114]

[0115] T 槽后 (t) = T 槽前 (t) - ΔT loss (t)

[0116] Among them, T env , R cell is the equivalent resistance of the electrolytic cell, Q cool (t) is the cooling water flow rate, α and β are the heat capacity and heat dissipation coefficient of the electrolytic cell respectively, and ΔT loss (t) is the temperature loss during the caustic solution circulation process.

[0117] Specifically, the thermodynamic characteristics of the electrolytic cell are quantified through physical parameters α, β, and R cell to avoid pure empirical models.

[0118] 1 - e -β·t simulates the process of the temperature approaching the steady state over time, which is more in line with the law of thermal inertia than the linear prediction of existing patents.

[0119] Furthermore, constructing a corresponding opening prediction model based on each temperature in the temperature state set to determine the cold water valve control sub-signal set includes:

[0120] Determining the first control sub-signal according to the temperature in front of the cell;

[0121] Determining the second control sub-signal according to the temperature behind the cell;

[0122] Determining the third control sub-signal according to the cooling water return temperature;

[0123] Among them, the expression of the first control sub-signal is:

[0124]

[0125] The expression of the second control sub-signal is:

[0126]

[0127] The expression of the third control sub-signal is:

[0128]

[0129] Among them, K P , K I , K D are the first PID control parameter, the second PID control parameter, and the third PID control parameter respectively, a is the heat exchange efficiency coefficient; Q max is the maximum cooling water flow rate, e(t) is the deviation between the temperature in front of the cell and the set temperature, and λ1, λ2, and λ3 are the first weight coefficient, the second weight coefficient, and the third weight coefficient respectively.

[0130] Specifically, by accurately analyzing and regulating the temperature before the cell, the temperature after the cell, and the return water temperature of the cooling water, the system can maintain the temperature of the electrolytic cell within the set range during operation, improving the production efficiency of hydrogen; the regulation sub-signal adopts the PID control method, which can effectively respond to the influence brought by temperature changes, realize the rapid and precise regulation of the electrolytic cell temperature, and ensure the stability of the system under different working conditions.

[0131] Further, determining the final regulation signal according to the set of regulation sub-signals includes:

[0132] Determining the dynamic weight coefficient corresponding to the set of regulation sub-signals according to the dynamic weight distribution strategy;

[0133] Determining the final regulation signal according to the dynamic weight coefficient.

[0134] Specifically, real-time monitoring: The system monitors the operating state of the electrolytic cell in real time, especially the changes in the current temperature before the cell, the temperature after the cell, and the return water temperature of the cooling water.

[0135] Basis for weight adjustment: According to the differences between the temperature before the cell, the temperature after the cell, and the return water temperature of the cooling water and the target set temperature, calculate the importance of each regulation sub-signal to the final regulation signal. The greater the temperature deviation, the higher the weight of the relevant regulation sub-signal.

[0136] Weight distribution rule: Assume that the current temperature deviation before the cell is large and the temperature deviation after the cell is small, then the weight of the first regulation sub-signal will increase relatively, and vice versa. The weight distribution can use normalization processing to ensure that the sum of all weights is 1.

[0137] Calculating the final regulation signal according to the dynamic weight coefficient:

[0138] Once the dynamic weight coefficient is determined, the system multiplies each regulation sub-signal by its corresponding dynamic weight coefficient and then accumulates to obtain the final regulation signal.

[0139] If the value of the first regulation sub-signal is U1, the weight coefficient is W1; the value of the second regulation sub-signal is U2, the weight coefficient is W2; the value of the third regulation sub-signal is U3, the weight coefficient is W3, then the final regulation signal Y can be calculated by the following method:

[0140] Final regulation signal Y = (U1 × W1) + (U2 × W2) + (U3 × W3).

[0141] The system effectively utilizes the dynamic weight distribution strategy to adjust the weights of the regulation sub-signals in a timely manner according to the actual operating state of the current electrolytic cell. This method ensures the accuracy of the final regulation signal, improves the temperature control accuracy of the hydrogen production system by electrolysis, and enhances the response ability and stability of the system.

[0142] In the present specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments, and the same or similar parts among the various embodiments can be referred to each other.

[0143] 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 to the present invention.

Claims

1. An optimal temperature control method for an electrolytic hydrogen production system based on adaptive control, characterized in that, Including: Obtain the current signal of the electrolytic cell at the current moment, the current command signal at a future moment provided by the power grid, historical current data, and environmental data, and use a prediction algorithm to predict the current signal to obtain the current signal at the future moment; Determine the corresponding temperature state set of the electrolytic cell according to the current signal at the future moment, where the temperature state set includes: the front temperature of the electrolytic cell, the rear temperature of the electrolytic cell, and the return water temperature of the cooling water in the hydrogen production system by electrolysis; Construct a corresponding opening prediction model according to each temperature in the temperature state set to determine the cold water valve regulation sub-signal set; wherein, the regulation sub-signal set includes: a first regulation sub-signal, a second regulation sub-signal, and a third regulation sub-signal; Determine the final regulation signal according to the regulation sub-signal set; Determine the opening of the cold water valve according to the regulation signal to optimize the control of the temperature of the hydrogen production system by electrolysis.

2. The temperature optimization control method of an electrolytic hydrogen production system based on adaptive control according to claim 1, wherein Also including: Determine the opening rate of the cold water valve within a preset time interval; Compare the opening rate of the cold water valve within the preset time interval with a preset opening rate threshold to obtain a comparison result; Determine whether there is an abnormality in the temperature state set according to the comparison result.

3. A temperature optimization control method for an electrolytic hydrogen production system based on adaptive control according to claim 1, characterized in that, Obtain the current signal of the electrolytic cell at the current moment, the current command signal at a future moment provided by the power grid, historical current data, and environmental data, and use a prediction algorithm to predict the current signal to obtain the current signal at the future moment, including: Determine the initial current signal at the future moment according to the current command signal at the future moment provided by the power grid; Construct corresponding knowledge graph data according to the initial current signal at the future moment, historical current data, and environmental data; Extract features from the knowledge graph data to obtain a corresponding feature set; Input the feature set into the constructed current prediction model to obtain the current signal at the future moment; The expression of the current prediction model is: Among them, is the current signal at a future moment, f() is a function modeled based on GBDT, and x t-i represents the current signal at time t-i, and Z t is the feature set extracted by the knowledge graph, and θ is the model parameter set.

4. The temperature optimization control method of an electrolytic hydrogen production system based on adaptive control according to claim 3, characterized in that, The extracting features from the knowledge graph data to obtain a corresponding feature set includes: Perform correlation analysis on each node and edge of the knowledge graph data to obtain a correlation set; The expression of the correlation set is: R = {(N i , N j , r ij ) where N i and N j are different nodes in the graph, and r ij represents the correlation between them; According to the correlation set, determine the statistical features of the initial current signal at the future moment and historical current data and the change features of its environmental factors; Aggregate the statistical features and the change features to obtain a corresponding feature set.

5. The temperature optimization control method of an electrolytic hydrogen production system based on adaptive control according to claim 4, characterized in that, The statistical features include: Current mean, current standard deviation, and current change rate, where the calculation expression of the current mean is: The expression of the current standard deviation is: The expression of the current change rate is: Where N is the number of samples, and I i is the magnitude of the current obtained by observing the initial future - moment current signal and historical current data, is the initial future - moment current signal.

6. The temperature optimization control method of an electrolytic hydrogen production system based on adaptive control according to claim 1, characterized in that, The determining the corresponding temperature state set of the electrolytic cell according to the current signal at the future moment includes: Obtain the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and its cooling water flow rate; Directly measure the return water temperature of the cooling water using a temperature sensor; Calculate the front temperature of the electrolytic cell according to the current signal at the future moment, the equivalent resistance of the electrolytic cell, the heat capacity of the electrolytic cell, the heat dissipation coefficient, and the cooling water flow rate; Calculate the rear temperature of the electrolytic cell according to the front temperature of the electrolytic cell; Wherein, the expressions of the front temperature and the rear temperature are respectively: Among them, T env , R cell is the equivalent resistance of the electrolytic cell, Q cool (t) is the cooling water flow rate, α and β are the heat capacity and heat dissipation coefficient of the electrolytic cell respectively, and ΔT loss (t) is the temperature loss during the caustic soda solution circulation process.

7. A temperature optimization control method for an electrolytic hydrogen production system based on adaptive control according to claim 6, characterized in that, Constructing a corresponding opening prediction model based on each temperature according to the temperature state to determine the cold water valve regulation sub-signal set, including: Determining a first regulation sub-signal according to the temperature in front of the tank; Determining a second regulation sub-signal according to the temperature behind the tank; Determining a third regulation sub-signal according to the return water temperature of the cooling water; Wherein, the expression of the first regulation sub-signal is: The expression of the second regulation sub-signal is: The expression of the third regulation sub-signal is: Among them, K P , K I , K D are the first PID control parameter, the second PID control parameter and the third PID control parameter respectively, and a is the heat exchange efficiency coefficient; Q max is the maximum flow rate of cooling water, e(t) is the deviation between the temperature in front of the tank and the set temperature, and λ1, λ2 and λ3 are the first weight coefficient, the second weight coefficient and the third weight coefficient respectively.

8. A temperature optimization control method for an electrolytic hydrogen production system based on adaptive control according to claim 1, characterized in that, Determining the final regulation signal according to the regulation sub-signal set, including: Determining the dynamic weight coefficient corresponding to the regulation sub-signal set according to the dynamic weight allocation strategy; Determining the final regulation signal according to the dynamic weight coefficient.

Images