Control Method, Device and System for Electrolytic Hydrogen Production Equipment

Abstract

This specification provides a control method, device, and system for an electrolytic hydrogen production device. The operation stage is divided into an intervention control stage and an automatic control stage. In the intervention control stage, manual intervention is used to control the electrolytic hydrogen production device to reach the optimal working state. The correlation relationship between the controllable parameters, input electric energy, and state parameters of the electrolytic hydrogen production device in the case of reaching the optimal working state is learned through a time series network model. When the prediction accuracy reaches the preset requirement, it enters the automatic control stage. Then, the time series network model is used to predict the state parameters of the electrolytic hydrogen production device when it reaches the optimal working state at the next moment, and this prediction result is used as the control target to control the automatic controller to adjust the controllable parameters of the electrolytic hydrogen production device. The time series network model in this solution learns the correlation relationship between the controllable parameters and the state parameters under the interference of input electric energy, so as to be able to adapt to the volatility of the input electric energy of the electrolytic hydrogen production device.

Description

Technical Field

[0001] This application relates to the field of new wind and solar energy technologies, and particularly to a control method, device, and system for electrolytic hydrogen production equipment. Background Art

[0002] Wind power generation and photovoltaic power generation, as clean new energy sources, can continuously generate electrical energy.

[0003] The main principle of a wind-solar hydrogen production system is to use the electrical energy generated by wind power generation and photovoltaic power generation to convert water into hydrogen through an electrolyzer. However, the volatility of wind energy and light energy will directly affect the stability of electrical energy, and unstable electrical energy will affect the efficiency of the electrolysis process and the hydrogen production. When dealing with this change, the existing technology usually adopts a single control strategy, resulting in insufficient adaptability of the system to electrical energy changes, inability to efficiently utilize the volatility of photovoltaic power generation and wind power generation, and low energy efficiency of the hydrogen production system. Summary of the Invention

[0004] This specification provides a control method, device, and system for hydrogen production equipment to improve the adaptability of the hydrogen production equipment to electrical energy changes, effectively utilize the volatility of photovoltaic power generation and wind power generation, and improve the energy efficiency of the hydrogen production system.

[0005] The first aspect of this specification provides a control method for electrolytic hydrogen production equipment, including: in the intervention control stage, recording each controllable parameter and input electrical energy of the electrolytic hydrogen production equipment, and collecting the state parameters of the electrolytic hydrogen production equipment; in the intervention control stage, manually intervening to control the electrolytic hydrogen production equipment to reach the optimal working state; learning the correlation relationship between the controllable parameters, input electrical energy, and state parameters of the electrolytic hydrogen production equipment through a time series network model; entering the automatic control stage when the prediction accuracy rate of the time series network model reaches a preset requirement; in the automatic control stage, predicting each state parameter of each electrolytic hydrogen production equipment through the time series network model; using the prediction result of the time series network model as the control target, and controlling the automatic controller to adjust each controllable parameter of the electrolytic hydrogen production equipment.

[0006] In some embodiments, the state parameters of the electrolytic hydrogen production equipment include the set temperature of the electrolyte in the electrolyzer or the heating power of the heater in the electrolyzer, the set temperature of the electrolyte replenishing liquid in the circulation pipeline or the cooling power of the cooler on the circulation pipeline, and the valve opening degree of the alkali tank communicating with the electrolyzer to replenish the electrolyte.

[0007] In some embodiments, the correlation relationships between the controllable parameters, the input electric energy, and the state parameters of the electrolytic hydrogen production equipment are learned through a time-series network model, including: in the intervention control stage, marking the optimal state parameter group of the electrolytic hydrogen production equipment; screening out the optimal state parameter groups at least continuously for N acquisition moments; determining training data according to the optimal state parameter groups and the values of the controllable parameters and the input electric energy at the corresponding acquisition moments, so as to obtain a first training data set; N is an integer greater than 2; training the time-series network model with the first training data set.

[0008] In some embodiments, after training the time-series network model, it further includes: screening out the optimal state parameters at least continuously for M acquisition moments to form a first verification data set, and the intersection of the first verification data set and the first training data set is empty; M is an integer greater than 2; evaluating the prediction accuracy of the time-series network model through the first verification data set.

[0009] In some embodiments, the method further includes: in the automatic control stage, using the optimal state parameter group of the first automatic sub-stage as training data to form a second training data set, and using the optimal state parameter group of the second automatic sub-stage as verification data to form a second verification data set; the second automatic sub-stage is an operation sub-stage after the first automatic sub-stage; updating the time-series network model through the second training data set; verifying the time-series network model through the second verification data set; in the case where the prediction accuracy of the time-series network model reaches the preset requirement, continuing the automatic control stage; in the case where the prediction accuracy of the time-series network model does not reach the preset requirement, exiting the automatic control stage.

[0010] In some embodiments, in the automatic control stage, when the observed parameters of the actuators, raw materials, and products of the hydrogen production equipment exceed the optimal value range, an alarm message is generated.

[0011] In some embodiments, after generating the alarm message, it further includes: sending the alarm message to the user terminal; receiving the target value of the target controllable parameter sent by the user terminal.

[0012] In some embodiments, the automatic controller includes a PID controller or a LADRC controller.

[0013] The second aspect of this specification provides a control device for an electrolytic hydrogen production device, including: an acquisition unit, configured to record each controllable parameter and input electric energy of the electrolytic hydrogen production device and collect the state parameters of the electrolytic hydrogen production device during the intervention control stage; during the intervention control stage, manually intervene to control the electrolytic hydrogen production device to reach the optimal working state; a training unit, configured to learn the correlation between the controllable parameters, input electric energy, and state parameters of the electrolytic hydrogen production device through a time series network model; a switching unit, configured to enter the automatic control stage when the prediction accuracy rate of the time series network model reaches a preset requirement; a prediction unit, configured to predict each state parameter of each electrolytic hydrogen production device through the time series network model during the automatic control stage; a control unit, configured to use the prediction result of the time series network model as a control target and control an automatic controller to adjust each controllable parameter of the electrolytic hydrogen production device.

[0014] The third aspect of this specification provides a control system for an electrolytic hydrogen production device, including: an electrolytic hydrogen production device, configured to implement hydrogen production by using electrolytic hydrogen production technology; an automatic controller, configured to adjust each controllable parameter of the electrolytic hydrogen production device to adjust the working state of the electrolytic hydrogen production device; a control device for the electrolytic hydrogen production device, configured to execute the control method for the electrolytic hydrogen production device according to any one of the first aspects.

[0015] The control method, device, and system for the electrolytic hydrogen production device provided in this specification divide the operation stage of the electrolytic hydrogen production device into an intervention control stage and an automatic control stage. During the intervention control stage, manually intervene to control the electrolytic hydrogen production device to reach the optimal working state, and learn the correlation between the controllable parameters, input electric energy, and state parameters of the electrolytic hydrogen production device when it reaches the optimal working state through a time series network model. When the prediction accuracy rate of the time series network model reaches a preset requirement, the user can choose to enter the automatic control stage. During the automatic control stage, use the time series network model to predict the state parameters of the electrolytic hydrogen production device when it reaches the optimal working state at the next moment, and use this prediction result as a control target to control the automatic controller to adjust each controllable parameter of the electrolytic hydrogen production device. This solution enables the user to ensure that the electrolytic hydrogen production device is in the optimal working state without intervening in the operation process during the automatic control stage; moreover, the time series network model in this solution learns the correlation between the controllable parameters, "input electric energy", and state parameters of the electrolytic hydrogen production device, and the input electric energy of the electrolytic hydrogen production device has fluctuations. That is to say, the time series network model learns the correlation between the controllable parameters and state parameters under the interference of the input electric energy. Therefore, controlling the hydrogen production device according to the prediction result of the time series network model can not only enable the hydrogen production device to reach the optimal working state during the automatic control stage, but also adapt to the fluctuations of the input electric energy of the electrolytic hydrogen production device. Description of the Drawings

[0016] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings described below are only some embodiments recorded in the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of the usage scenario of an electrolytic hydrogen production device;

[0018] Figure 2 It is a schematic diagram of an electrolytic hydrogen production device using alkaline electrolysis technology;

[0019] Figure 3 It is a schematic flowchart of a control method for the electrolytic hydrogen production device provided in this specification;

[0020] Figure 4 It is a schematic diagram of a method for learning the correlation between controllable parameters, input electric energy, and state parameters of an electrolytic hydrogen production device through a time series network model;

[0021] Figure 5 It is another schematic diagram of a method for learning the correlation between controllable parameters, input electric energy, and state parameters of an electrolytic hydrogen production device through a time series network model;

[0022] Figure 6 It is another schematic flowchart of a control method for the electrolytic hydrogen production device provided in this specification;

[0023] Figure 7 It is a block diagram of the LADRC controller;

[0024] Figure 8 It is a block diagram of a control device for the electrolytic hydrogen production device provided in this specification;

[0025] Figure 9 It is a schematic diagram of the structure of an electronic device provided in this specification. Specific embodiments

[0026] In order to enable those skilled in the art to better understand the technical solutions in the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the protection scope of the present application.

[0027] In order to improve the adaptability of electrolytic hydrogen production equipment to changes in electric energy, effectively utilize the volatility of photovoltaic power generation and wind power generation, and improve the energy efficiency of the hydrogen production system, this specification provides a control method for electrolytic hydrogen production equipment.

[0028] Electrolytic hydrogen production equipment is a general term for devices used to produce hydrogen, which can convert raw materials such as hydrogen-containing compounds or water into hydrogen through various chemical or physical methods. Figure 1 It is a schematic diagram of the usage scenario of electrolytic hydrogen production equipment. The electric energy from wind power generation and photovoltaic power generation is mainly used to supply the load, and the remaining electric energy used by the load is supplied to the electrolytic hydrogen production equipment.

[0029] Current hydrogen production technologies include traditional alkaline electrolyzers, proton exchange membrane (PEM) electrolyzers, and solid oxide electrolyzers (SOEC).

[0030] A proton exchange membrane (PEM) electrolyzer uses a proton exchange membrane as the electrolyte, has high energy efficiency, can operate at high current densities, has fast startup and response speeds, and high hydrogen purity. However, the cost of proton exchange membranes and catalysts (such as precious metals like platinum) is relatively high, which limits its large-scale application.

[0031] Solid oxide electrolyzers have high energy efficiency, especially when using high-temperature waste heat or high-temperature thermal energy from renewable energy sources. However, they need to operate at high temperatures, which requires high heat resistance of materials and high requirements for the thermal management of equipment.

[0032] Traditional alkaline electrolyzers mainly consist of an anode, a cathode, and an electrolyte (such as sodium hydroxide or potassium hydroxide). The biggest advantage of alkaline electrolysis technology is that the anode and cathode plates do not contain precious metals, so the cost of the electrolyzer is relatively low.

[0033] Alkaline electrolysis technology decomposes water into hydrogen and oxygen through the current flowing through the electrolyte. On the anode (positive electrode), water molecules are oxidized, releasing oxygen, electrons, and hydrogen ions. The reaction formula is:

[0034]

[0035] On the cathode (negative electrode), hydrogen ions accept electrons and are reduced to form hydrogen. The reaction formula is:

[0036]

[0037] Through these two processes, the electrolyzer realizes the electrolysis of water, generating hydrogen and oxygen. The concentration of the electrolyte has a direct impact on the reaction rate and efficiency. Alkaline electrolytes have good electrical conductivity, can effectively reduce the resistance during the electrolysis process, and increase the hydrogen production.

[0038] New energy power sources generated by wind power and solar power provide the necessary direct current for the electrolyzer to drive the electrolysis reaction. The stability of the power source and the precise control of the output voltage directly affect the efficiency of the electrolysis reaction and the hydrogen production.

[0039] Temperature and pressure also have a significant impact on the efficiency of the electrolysis reaction. By adjusting the operating temperature and pressure of the electrolyzer, the reaction conditions can be optimized to improve the hydrogen production efficiency.

[0040] To ensure the effective separation of the generated hydrogen and oxygen, a gas separation device is usually equipped inside the electrolyzer to prevent the mixing of hydrogen and oxygen and ensure the safety and purity of the products.

[0041] Figure 2 It is a schematic diagram of an electrolytic hydrogen production device using alkaline electrolysis technology. The two electrodes in the alkaline electrolyzer form the hydrogen side and the oxygen side respectively. A hydrogen separator, an oxygen separator, a scrubber, and a purification system are provided to separate the electrolytically generated hydrogen and oxygen. Oxygen can be directly discharged into the outdoor air.

[0042] During the electrolytic hydrogen production process, due to factors such as the thermal effect of the electrolysis reaction itself, the ohmic heat generated when each conductive component is energized, and the frictional heat generated by gas diffusion and convection, the temperature inside the electrolyzer will gradually rise. The optimal operating temperature of the electrolytic hydrogen production device is usually 70 - 90 °C. Exceeding this temperature will lead to a decrease in electrolysis efficiency, accelerated material aging, a decrease in hydrogen quality, and even an increase in risks such as hydrogen explosion. Therefore, it is necessary to cool the electrolyte in the electrolyzer. Usually, a circulation pipeline is set for the electrolyte, and the electrolyte in the circulation pipeline is cooled by a cooler.

[0043] The electrolyte in the electrolyzer can be recycled through the circulation pipeline. A cooler and a circulation pump are set in the circulation pipeline, and the first temperature controller is used to control the cooler to cool the electrolyte in the circulation pipeline to the set temperature. The electrolyte supply for the electrolyzer is provided by an alkali tank.

[0044] To improve the efficiency of electrolytic hydrogen production, the second temperature controller can be used to control the heater in the electrolyzer to heat the electrolyte in the electrolyzer to the optimal operating temperature.

[0045] The diaphragm, which is the core part of the electrolyzer, can ensure that hydrogen and oxygen molecules cannot pass through the diaphragm, but allows electrolyte ions to pass through and is resistant to corrosion by high-concentration alkali liquor. However, if the fluctuations of wind power and solar power are too large, resulting in a sudden increase in the input electric energy of the electrolytic hydrogen production device and a sudden change in load reduction, the liquid level will drop. If the liquid level difference is large, it will cause certain damage to the diaphragm. Therefore, it is necessary to control the liquid level of the electrolyzer.

[0046] The control method of the electrolytic hydrogen production equipment provided in this specification can be used for Figure 2 the hydrogen production equipment shown

[0047] As Figure 3 shown, the control method of the electrolytic hydrogen production equipment provided in this specification includes the following S10 to S50.

[0048] S10: In the intervention control stage, record the controllable parameters, input electric energy of the electrolytic hydrogen production equipment, and collect the state parameters of the electrolytic hydrogen production equipment; in the intervention control stage, manually intervene to control the electrolytic hydrogen production equipment to reach the optimal working state.

[0049] The controllable parameters refer to the parameters that can be adjusted by users in the electrolytic hydrogen production equipment. By adjusting these controllable parameters, the electrolytic hydrogen production equipment can adapt to the change of electric energy and reach the optimal working state.

[0050] The controllable parameters may include the set temperature of the above first temperature controller and the set temperature of the second temperature controller. Specifically, the controllable parameters may include the set temperature of the electrolyte in the electrolytic cell or the heating power of the heater in the electrolytic cell, the set temperature of the electrolyte replenishing liquid in the circulation pipeline or the cooling power of the cooler on the circulation pipeline, and the valve opening degree of the alkali tank connecting to the electrolytic cell to replenish the electrolyte.

[0051] The input electric energy refers to the electric energy generated by wind power generation and photovoltaic power generation and distributed to the hydrogen production equipment after being used by the load.

[0052] The state parameters refer to the state parameters that can reflect the working state of the hydrogen production equipment. The state parameters may include the measured temperature of the electrolyte in the electrolytic cell, the measured pressure of the electrolytic cell, the measured liquid level of the electrolytic cell, etc.

[0053] The manual intervention stage means that the user can monitor the state parameters of the electrolytic hydrogen production equipment in real time and manually adjust each controllable parameter at any time. In the manual intervention stage, the electrolytic hydrogen production equipment can reach the optimal working state.

[0054] The optimal working state means that the observed parameters of each actuator (such as electrodes, electrolytic cells, heaters, coolers, heating pumps, circulation pumps, exhaust devices, etc.), raw materials (such as electrolytes), and products (such as hydrogen, oxygen) of the hydrogen production equipment are within the optimal value range, and the data within the optimal value range enables the hydrogen production equipment to have high energy efficiency without potential fault hazards and low losses.

[0055] S20: Learn the correlation relationship between the controllable parameters, input electric energy and state parameters of the electrolytic hydrogen production equipment through the time series network model.

[0056] The time series network model can adopt deep learning time series analysis models such as the long short-term memory network (LSTM), and time series network models based on probabilistic graphical models such as the hidden Markov model and the dynamic Bayesian network.

[0057] Take the controllable parameters at the current acquisition moment of the electrolytic hydrogen production equipment and the input electric energy at the current acquisition moment as the input data of the time series network model, and take the state parameters at the next acquisition moment as the output data of the time series network model, so that the time series network model can predict the change of the state parameters of the hydrogen production equipment according to the controllable parameters and input electric energy of the electrolytic hydrogen production equipment. That is to say, take the controllable parameters at the current acquisition moment as the processing actions taken under the state parameters at the current acquisition moment, and take the state parameters at the next acquisition moment as the processing results of the processing actions taken at the current acquisition moment.

[0058] In some embodiments, it is also possible to set a difference between the current acquisition moment of the controllable parameters and input electric energy and the current acquisition moment of the state parameters. For example, t1 and t2 are two successive moments. The controllable parameters and input electric energy at the current acquisition moment are collected at the t1 moment, and the state parameters at the current moment are collected at the t2 moment.

[0059] In some embodiments, as Figure 4 shown, S20 includes the following S21, S22, S23, and S24.

[0060] S21: In the intervention control stage, mark the optimal state parameter group of the electrolytic hydrogen production equipment.

[0061] The state parameter group refers to the set of state parameters (i.e., observation parameters) of each actuator, raw material, and product collected at the same acquisition moment and the corresponding input electric energy. Each state parameter (i.e., observation parameter) in the optimal state parameter group is within the optimal value range.

[0062] The optimal value range of each observation parameter can be set in advance, and the value of each state parameter is compared with its nearest value range to determine whether a set of state parameters collected at the same acquisition moment is the optimal state parameter group.

[0063] It is also possible to present the values of each state parameter collected at each acquisition moment to the user, and the user judges whether each state parameter group is the optimal state parameter group according to his own experience and in combination with the preset optimal value range. That is to say, in the intervention control stage, the user can not only participate in adjusting the controllable parameters but also select the optimal state parameter group.

[0064] S22: Screen out the optimal state parameter groups at least for N consecutive acquisition moments.

[0065] S23: Determine the training data based on the optimal state parameter group, the controllable parameters at the corresponding acquisition times, and the values of the input electric energy, thereby obtaining the first training data set; N is an integer greater than 3.

[0066] For example, t1, t2, and t3 are three consecutive acquisition times. If the state parameter groups corresponding to the acquisition times t1, t2, and t3 are all optimal state parameter groups, then the state parameter groups corresponding to the acquisition times t1, t2, and t3 constitute a training data in the first training data set. If the state parameter group at the acquisition time t4 after the acquisition time t3 is also an optimal state parameter group, then the state parameter groups corresponding to the acquisition times t1, t2, t3, and t4 constitute a training data in the first training data set. That is to say, the number of acquisition times corresponding to a training data is not fixed, and can be 3, 4, or more, but at least 2.

[0067] S24: Train the time series network model using the first training data set.

[0068] Since the number of acquisition times corresponding to a training data is not fixed, therefore, how to use the training data with different numbers of acquisition times can be specifically determined when training the time series network model.

[0069] The specific method of training the time series network model with the training data is a prior art and will not be elaborated in this specification.

[0070] S30: Enter the automatic control stage when the prediction accuracy rate of the time series network model reaches the preset requirement.

[0071] In some embodiments, after training the time series network model, as Figure 5 shown, S25 and S26 may further be included as follows.

[0072] S25: Screen out the optimal state parameters at at least M consecutive acquisition times to form the first verification data set, and the intersection of the first verification data set and the first training data set is empty; M is an integer greater than 2.

[0073] S26: Evaluate the prediction accuracy rate of the time series network model using the verification data set.

[0074] The first training data set and the first verification data set can be selected from the same data set.

[0075] In some embodiments, the manual intervention stage includes a first intervention sub-stage and a second intervention sub-stage. The first intervention sub-stage is the data acquisition stage, and the first set of the collected data is used to screen data to form the first training data set. The second intervention sub-stage is the trial operation stage of the time series network model, and the state parameter groups in this stage form the second set.

[0076] The best set of state parameters can be screened from the first set to form the first verification data set, or the best set of state parameters can be screened from the second set to form the first verification data set, or the best set of state parameters can be screened from the first set and the second set respectively to form the first verification data set.

[0077] Under normal circumstances, state data at multiple acquisition moments are obtained in one data acquisition stage. The best set of state data is screened from these state data to form the best data set. A part of the best data groups in the best data set is used as training data, and another part of the best data groups is used as verification data. The first training data set and the first verification data set in this way come from the same operation stage of the hydrogen production equipment, and it is difficult to verify the generalization of the time series network model in other operation stages.

[0078] The method provided in this specification for screening the best state data from the second set to form verification data, where the first verification data set and the first training data set come from different operation stages of the hydrogen production equipment, can more effectively verify the generalization of the time series network model in other operation stages.

[0079] In some embodiments, as Figure 6 shown, the control method of the electrolytic hydrogen production equipment further includes the following S61, S62, S63, and S64.

[0080] S61: In the automatic control stage, use the best set of state parameters of the first automatic sub-stage as training data to form the second training data set, and use the best set of state parameters of the second automatic sub-stage as verification data to form the second verification data set; the second automatic sub-stage is the operation sub-stage after the first automatic sub-stage.

[0081] S62: Update the time series network model through the second training data set.

[0082] S63: Verify the time series network model through the second verification data set.

[0083] S64: In the case where the prediction accuracy rate of the time series network model reaches the preset requirement, continue the automatic control stage; in the case where the prediction accuracy rate of the time series network model does not reach the preset requirement, exit the automatic control stage.

[0084] Through the above S61 to S64, the time series network model can be updated in a timely manner according to the operation data in the automatic control stage, and the generalization of the time series network model can be verified, so that the time series network model can adapt to the changes in the electrolytic hydrogen production process, more accurately predict the state parameters of the electrolytic hydrogen production equipment, and thus more accurately determine the control target, preparing for the automatic controller to control the electrolytic hydrogen production equipment.

[0085] S40: In the automatic control stage, predict each state parameter of each hydrogen production device through the time series network model.

[0086] S50: Take the prediction result of the time series network model as the control target, and control the automatic controller to adjust each controllable parameter of the electrolytic hydrogen production device.

[0087] The control method of the electrolytic hydrogen production device provided in this specification divides the operation stage of the electrolytic hydrogen production device into an intervention control stage and an automatic control stage. In the intervention control stage, manually intervene to control the electrolytic hydrogen production device to reach the optimal working state, and learn the correlation relationship between the controllable parameters, input electric energy and state parameters of the electrolytic hydrogen production device when it reaches the optimal working state through the time series network model. When the prediction accuracy rate of the time series network model reaches the preset requirement, the user can choose to enter the automatic control stage. In the automatic control stage, use the time series network model to predict the state parameters of the electrolytic hydrogen production device when it reaches the optimal working state at the next moment, take this prediction result as the control target, and control the automatic controller to adjust each controllable parameter of the electrolytic hydrogen production device. This solution enables the user to ensure that the electrolytic hydrogen production device is in the optimal working state without intervening in the operation process in the automatic control stage; moreover, the time series network model of this solution learns the correlation relationship between the controllable parameters, "input electric energy" and state parameters of the electrolytic hydrogen production device, and the input electric energy of the electrolytic hydrogen production device has volatility, that is to say, the time series network model learns the correlation relationship between the controllable parameters and state parameters under the interference of input electric energy, so as to control the hydrogen production device according to the prediction result of the time series network model, which can not only make the hydrogen production device reach the optimal working state in the automatic control stage, but also adapt to the volatility of the input electric energy of the electrolytic hydrogen production device.

[0088] In some embodiments, to ensure the operation safety of the electrolytic hydrogen production device without human intervention in the automatic control stage, in the automatic control stage, when the observed parameters of each actuator, raw material, and product of the hydrogen production device exceed the optimal value range, an alarm message is generated.

[0089] After generating the alarm message, the alarm message can be sent to the user terminal; receive the target value of the target controllable parameter sent by the user terminal. The user can issue an instruction to adjust the target controllable parameter through the user terminal, and the instruction includes the target value of the target controllable parameter. Through this setting, the user can timely obtain the information that the electrolytic hydrogen production device is not in the optimal working state in the automatic control stage, and conveniently manually intervene in the automatic control stage of the electrolytic hydrogen production device through the user terminal.

[0090] It should be noted that the manual intervention in the operation of the electrolytic hydrogen production equipment through the user terminal during the automatic control stage is different from that during the intervention control stage. The difference is that the former only intervenes through the user terminal when an alarm message is received, and the user does not need to monitor the operation status of the electrolytic hydrogen production equipment in real time, while the latter requires the user to monitor and adjust the operation status of the electrolytic hydrogen production equipment at all times. Compared with the latter, the manual intervention during the automatic control stage can reduce the workload of the user while ensuring that the electrolytic hydrogen production equipment is in the best working state.

[0091] The above automatic controller can be a PID controller or a LADRC controller.

[0092] The LADRC (Linear Active Disturbance Rejection Control) controller includes: a Linear Extended State Observer (LESO) for real-time estimation of the internal uncertainties and external disturbances of the system; an object model P for describing the dynamic characteristics of the system; and a Linear State Error Feedback Control (LSEF) module for rapid disturbance compensation. The electrolytic hydrogen production equipment can be converted into the following state-space expression form:

[0093]

[0094] where x(t) is the state vector, y(t) is the output vector, u(t) is the input vector, , and there is

[0095]

[0096] And the expression of the controller is

[0097]

[0098] where z(t) is the control variable, and L o is the observer gain. At this time, the observer gain is set to

[0099]

[0100] Assume that the generalized disturbance is bounded. When is asymptotically stable, (p is an integer) will approximate the system output y(t) and its derivatives of each order, and will also approximate . Therefore, can be used for disturbance rejection control.

[0101] Based on the states and total disturbances obtained by the observer, a control law can be designed to compensate for the total disturbances as follows

[0102]

[0103] In the above formula is the reference signal for tracking, which can specifically be the set values of various state parameters.

[0104]

[0105] is the generalized reference signal, which includes the reference signal and its derivatives of all orders, and the state feedback controller gain is:

[0106]

[0107] In summary, the state-space expression of the linear active disturbance rejection controller can be obtained as:

[0108]

[0109] The block diagram of the LADRC controller mainly consists of the controlled object, the controller (such as Figure 7 shown as Ko in Figure 7 ) and the extended state observer (such as ESO), and can be expressed in the form of

[0110] By taking the difference between the output signal y(t) of each state parameter, etc., and the set reference signal r(t), and then directly inputting the error signal e(t) into the LADRC controller (i.e., Ko) for control, real-time feedback and compensation of disturbances are achieved, that is, the disturbance deviation caused by all factors such as input electric energy, in the tank, and the environment, so as to make the output as close as possible to the set value.

[0111] The parameters of the LADRC control algorithm can be determined by an adaptive adjustment mechanism and user intervention. Specifically, in the adaptive adjustment mechanism, the parameters of the LADRC control algorithm can be adjusted in real time through the adaptive mechanism. The system will dynamically adjust the controllable parameters according to the state parameters monitored in real time to adapt to changes in the external environment and internal disturbances. For example, when it is detected that the internal pressure of the electrolytic cell rises abnormally, the LADRC controller will automatically increase the opening degree of the pressure relief valve to maintain system stability.

[0112] The user intervention method can specifically be to set a user intervention interface. Through a mobile terminal such as a mobile phone, users can view the state parameters in real time and manually adjust the parameters of the LADRC algorithm as needed. For example, users can adjust the sensitivity of pressure control or the response speed of temperature control according to experience or preset optimization goals.

[0113] This specification provides a control device for an electrolytic hydrogen production device, which can be used to implement the control method of the above-mentioned electrolytic hydrogen production device. As Figure 8 shown, the device includes an acquisition unit 10, a training unit 20, a switching unit 30, a prediction unit 40, and a control unit 50.

[0114] The acquisition unit 10 is configured to record each controllable parameter and input electric energy of the electrolytic hydrogen production device during the intervention control stage, and collect the state parameters of the electrolytic hydrogen production device; during the intervention control stage, the electrolytic hydrogen production device is manually intervened to reach the optimal working state.

[0115] The training unit 20 is configured to learn the correlation between the controllable parameters, input electric energy, and state parameters of the electrolytic hydrogen production device through a time series network model.

[0116] The switching unit 30 is configured to enter the automatic control stage when the prediction accuracy of the time series network model reaches a preset requirement.

[0117] The prediction unit 40 is configured to predict each state parameter of each electrolytic hydrogen production device through the time series network model during the automatic control stage.

[0118] The control unit 50 is configured to use the prediction result of the time series network model as a control target, and control the automatic controller to adjust each controllable parameter of the electrolytic hydrogen production device.

[0119] The descriptions and functions of the above devices can be understood by referring to the content of the control method part of the electrolytic hydrogen production device, and will not be elaborated here.

[0120] An embodiment of the present invention further provides an electronic device. As Figure 9 shown, the electronic device may include a processor 901 and a memory 902, where the processor 901 and the memory 902 may be connected through a bus or other means. Figure 9 Taking the connection through the bus as an example.

[0121] The processor 901 may be a central processing unit (CPU). The processor 901 may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. chips, or a combination of the above types of chips.

[0122] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the control method of the electrolytic hydrogen production device in the embodiments of the present invention (for example Figure 8 the acquisition unit 10, training unit 20, switching unit 30, prediction unit 40, and control unit 50 shown in

[0123] The memory 902 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created by the processor 901, etc. In addition, the memory 902 may include high-speed random access memory and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 902 may optionally include a memory remotely provided relative to the processor 901, and these remote memories can be connected to the processor 901 through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and combinations thereof.

[0124] The one or more modules are stored in the memory 902 and, when executed by the processor 901, execute the control method of the electrolytic hydrogen production device in the embodiments shown in Figure 4 and Figure 5 shown.

[0125] Specific details of the above electronic device can be understood by referring to the corresponding related descriptions and effects in the method embodiments, and will not be elaborated here.

[0126] This specification also provides a computer storage medium, which stores computer program instructions, and when the computer program instructions are executed, the steps of the control method of the above electrolytic hydrogen production device are implemented.

[0127] This specification also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the control method of the above electrolytic hydrogen production device are implemented.

[0128] Those skilled in the art can understand that to implement all or part of the processes in the above-described embodiment methods, it can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-described method embodiments. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (abbreviation: HDD), or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memories.

[0129] Each embodiment in this specification is described in a progressive manner. For the same or similar parts between each embodiment, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

[0130] The systems, devices, modules, or units described in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions.

[0131] For the convenience of description, when describing the above device, it is divided into various units according to functions and described separately. Of course, when implementing the present application, the functions of each unit can be realized in the same or multiple software and / or hardware.

[0132] From the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of certain parts of each embodiment of the present application.

[0133] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, small computers, large computers, distributed computing environments including any of the above systems or devices, and so on.

[0134] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

[0135] Although the present application has been depicted by way of embodiments, those of ordinary skill in the art will recognize that the present application has many modifications and variations without departing from the spirit of the present application, and it is intended that the appended claims include such modifications and variations without departing from the spirit of the present application.

Claims

1. A control method for electrolytic hydrogen production equipment, characterized in that: include: In the intervention control stage, the controllable parameters and input electric energy of the electrolytic hydrogen production equipment are recorded, and the state parameters of the electrolytic hydrogen production equipment are collected, and the electrolytic hydrogen production equipment is controlled to achieve the best working state through manual intervention; The controllable parameters of the electrolytic hydrogen production equipment to achieve the best working state, the correlation between the input electric energy and the state parameters are learned through the time series network model; the input electric energy of the electrolytic hydrogen production equipment is the electric energy of wind power generation and / or photovoltaic power generation with fluctuation; When the prediction accuracy of the time series network model reaches the preset requirement, the automatic control stage is entered; In the automatic control stage, the state parameters of each electrolytic hydrogen production equipment are predicted by the timing network model; Using the prediction result of the timing network model as the control target, controlling the automatic controller to adjust various controllable parameters of the electrolytic hydrogen production equipment; Among them, the relationship between the controllable parameters of the electrolytic hydrogen production equipment to achieve the optimal working state, the input power and the state parameters is learned through the time series network model, including: In the intervention control stage, the optimal state parameter group of the electrolytic hydrogen production equipment is marked; the state parameter group refers to the collection of state parameters of each actuator, raw material, and product and the corresponding input electric energy collected at the same collection time, and the optimal state parameter group refers to each state parameter at a time being within the optimal value range; Filter out the best state parameter group of at least N consecutive acquisition moments as a training data, thereby obtaining a first training data set; N is an integer greater than or equal to 3; When the first training data set is used to train the timing network model, the controllable parameters of the electrolytic hydrogen production equipment at the current collection time and the input electric energy at the current collection time in the best state parameter group of at least N consecutive collection times are used as the input data of the timing network model, and the state parameters at the next collection time are used as the output data of the timing network model to learn the correlation between the controllable parameters, input electric energy and state parameters.

2. The method according to claim 1, characterized in that The controllable parameters of the electrolytic hydrogen production equipment include the set temperature of the electrolyte in the electrolytic cell or the heating power of the heater in the electrolytic cell, the set temperature of the electrolyte replenisher in the circulation pipeline or the cooling power of the cooler on the circulation pipeline, and the valve opening of the alkali box connected to the electrolytic cell to replenish the electrolyte.

3. The method according to claim 1, characterized in that After training the temporal network model, the method further includes: Filter out the best state parameters of at least M consecutive acquisition moments to form a first validation data set, where the intersection of the first validation data set and the first training data set is empty; M is an integer greater than 2; The prediction accuracy of the temporal network model is evaluated using the first validation data set.

4. The method according to claim 1, characterized in that: The method further comprises: In the automatic control stage, the optimal state parameter group of the first automatic sub-stage is used as training data to form a second training data set, and the optimal state parameter group of the second automatic sub-stage is used as verification data to form a second verification data set; the second automatic sub-stage is the operation sub-stage after the first automatic sub-stage; Updating the temporal network model using the second training data set; Verifying the temporal network model using the second verification data set; When the prediction accuracy of the timing network model reaches the preset requirement, the automatic control stage continues; when the prediction accuracy of the timing network model does not reach the preset requirement, the automatic control stage is exited.

5. The method according to claim 1, characterized in that In the automatic control stage, when the status parameters of the actuators, raw materials and products of the hydrogen production equipment exceed the optimal value range, an alarm message is generated.

6. The method according to claim 1, characterized in that After the alarm information is generated, it also includes: Sending the alarm information to a user terminal; Receive a target value of a target controllable parameter sent by a user terminal.

7. The method according to claim 1, characterized in that The automatic controller includes a PID controller or a LADRC controller.

8. A control device for electrolytic hydrogen production equipment, characterized in that: include: The acquisition unit is used to record various controllable parameters and input electric energy of the electrolytic hydrogen production equipment and collect state parameters of the electrolytic hydrogen production equipment during the intervention control stage, so as to control the electrolytic hydrogen production equipment to achieve the best working state through manual intervention; A training unit is used to learn the controllable parameters of the electrolytic hydrogen production equipment to achieve the best working state, the correlation between the input electric energy and the state parameters through a time series network model; the input electric energy of the electrolytic hydrogen production equipment is the electric energy of wind power generation and / or photovoltaic power generation with fluctuation; A switching unit, used to enter the automatic control stage when the prediction accuracy of the time series network model reaches a preset requirement; A prediction unit, used for predicting various state parameters of various electrolytic hydrogen production equipment through the timing network model during the automatic control stage; A control unit, used to control the automatic controller to adjust various controllable parameters of the electrolytic hydrogen production equipment by taking the prediction result of the timing network model as a control target; Among them, the relationship between the controllable parameters of the electrolytic hydrogen production equipment to achieve the optimal working state, the input power and the state parameters is learned through the time series network model, including: In the intervention control stage, the optimal state parameter group of the electrolytic hydrogen production equipment is marked; the state parameter group refers to the collection of state parameters of each actuator, raw material, and product and the corresponding input electric energy collected at the same collection time, and the optimal state parameter group refers to each state parameter at a time being within the optimal value range; Filter out the best state parameter group of at least N consecutive acquisition moments as a training data, thereby obtaining a first training data set; N is an integer greater than or equal to 3; When the first training data set is used to train the timing network model, the controllable parameters of the electrolytic hydrogen production equipment at the current collection time and the input electric energy at the current collection time in the best state parameter group of at least N consecutive collection times are used as the input data of the timing network model, and the state parameters at the next collection time are used as the output data of the timing network model to learn the correlation between the controllable parameters, input electric energy and state parameters.

9. A control system for an electrolytic hydrogen production device, characterized in that: include: Electrolytic hydrogen production equipment, used to prepare hydrogen using electrolytic hydrogen production technology; An automatic controller, used to adjust various controllable parameters of the electrolytic hydrogen production equipment to adjust the working state of the electrolytic hydrogen production equipment; A control device for hydrogen production equipment by electrolysis, used for executing the control method for hydrogen production equipment by electrolysis according to any one of claims 1 to 7.

Images