Control method of high-efficiency networking type energy storage converter applied to hydrogen production system

Abstract

The invention discloses a control method of an efficient networking type energy storage converter applied to a hydrogen production system, and the method comprises the steps: starting an energy storage converter and hydrogen production system equipment, initializing a controller, a sensor and a communication module, and building communication connection; acquiring an active power demand signal and a reactive power demand signal in real time, and measuring an output voltage instantaneous value and an output current instantaneous value; the collected data are preprocessed, the power demand change of the hydrogen production system and the output state change trend of the energy storage converter are predicted, and whether the operation condition of the hydrogen production system exceeds a preset range or not is judged; line impedance parameters and load dynamic characteristics of the hydrogen production system are monitored in real time, and a self-adaptive virtual impedance control algorithm is adopted to dynamically adjust a virtual impedance value; state information of the energy storage converter is transmitted to a monitoring center in real time through an encrypted communication protocol, a control instruction is received, and data traffic is managed by adopting a load balancing algorithm. The stability and efficiency of the hydrogen production system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy and power electronics technology, and particularly to a control method for an efficient grid-forming energy storage converter applied to a hydrogen production system. Background Art

[0002] Under the background of accelerating energy transformation, the combination of hydrogen production systems and energy storage technologies has become a research hotspot. During the operation of a hydrogen production system, the power demand fluctuates greatly with different hydrogen production process stages and load changes, and has strict requirements for power quality. The energy storage converter plays a key role in power regulation and power quality guarantee.

[0003] Currently, there are some similar technical solutions in the related fields. For example, a control method, device, and hydrogen production system with the name of a hydrogen production system mainly obtains the actual power generation power of a power generation device and the actual state of charge of an energy storage device, dynamically calculates the target hydrogen production power of the hydrogen production system; then, according to the target hydrogen production power, intelligently controls the power supply of the hydrogen production system and manages the charge and discharge states of the energy storage device, so as to ensure the coordinated operation among the power generation device, hydrogen production device, and energy storage device.

[0004] However, the existing technical solutions have obvious deficiencies. Their power regulation methods are too rough to accurately adapt to the complex power change curves of hydrogen production systems. Due to relying only on fixed thresholds and step sizes, it is difficult to quickly and accurately respond to power demands when facing the complex multi-stage load changes of the hydrogen production process, which easily leads to large fluctuations in system voltage and frequency, affecting the stable operation of hydrogen production equipment and hydrogen production efficiency. At the same time, this solution lacks an effective dynamic adjustment mechanism in power quality control. For abnormal changes in output voltage and current, it cannot be adjusted precisely in a timely manner, and it is difficult to meet the high-quality power demand of hydrogen production systems. Summary of the Invention

[0005] To solve the above technical problems existing in the prior art, the present invention proposes a control method for an efficient grid-forming energy storage converter applied to a hydrogen production system, which realizes more efficient and stable energy storage conversion control of the hydrogen production system through real-time data acquisition, advanced control algorithms, and fine adjustments in multiple links.

[0006] To achieve the above object, the present invention provides a control method for an efficient grid-forming energy storage converter applied to a hydrogen production system, including:

[0007] Real-time collect the active power demand signal and reactive power demand signal of the hydrogen production system, and real-time measure the instantaneous value of the output voltage and the instantaneous value of the output current of the energy storage converter, and convert the collected active power demand signal, reactive power demand signal, instantaneous value of the output voltage, and instantaneous value of the output current into digital signals and transmit them to the microprocessor unit of the controller;

[0008] Preprocess the collected data based on the moving average filtering algorithm, combine the trend analysis algorithm to predict the change of the power demand of the hydrogen production system and the change trend of the output state of the energy storage converter, and judge whether the operating condition of the hydrogen production system exceeds the preset range through the threshold comparison mechanism. If the operating condition of the hydrogen production system changes, enter the adaptive power adjustment process. If abnormal voltage or current is displayed, enter the power quality adjustment process;

[0009] Real-time monitor the line impedance parameters and load dynamic characteristics of the hydrogen production system, and dynamically adjust the virtual impedance value using the adaptive virtual impedance control algorithm;

[0010] Transmit the status information of the energy storage converter to the monitoring center in real time through the encrypted communication protocol, receive control instructions, and manage the data traffic using the load balancing algorithm.

[0011] Preferably, before real-time collecting the active power demand signal and reactive power demand signal of the hydrogen production system, it further includes:

[0012] Start the energy storage converter and hydrogen production system equipment, perform initialization configuration on the controller, sensor and communication module, set the initial operating parameters, and establish a communication connection, where the initial operating parameters include the output values of active power and reactive power, and the reference thresholds of voltage and current.

[0013] Preferably, the initialization configuration of the controller includes: microprocessor operation mode setting, communication protocol stack parameter configuration, control algorithm loading, clock synchronization calibration and hardware self-check. Among them, the hardware self-check includes sensor connection state detection and power supply stability verification.

[0014] Preferably, the threshold comparison mechanism is to set multi-level voltage / current deviation thresholds. When the voltage / current data exceeds the first-level threshold, start the preliminary adjustment. When the voltage / current data exceeds the second-level threshold, trigger the emergency protection mechanism.

[0015] Preferably, the adaptive power adjustment process includes:

[0016] Calculate the power adjustment amount based on the fuzzy logic control algorithm, combine machine learning to optimize the membership function and fuzzy rule base, and dynamically adjust the active power and reactive power commands of the energy storage converter.

[0017] Preferably, the adaptive power adjustment process further includes:

[0018] Optimization of the fuzzy logic control algorithm, including: learning the historical power data and control effect through a machine learning model, and using the backpropagation algorithm to optimize the fuzzy rule base and membership function parameters. Specifically:

[0019] θ new = θ old - η·▽J(θ);

[0020] Where θ new is the updated parameter, θ old is the current parameter, η is the learning rate, and ▽J(θ) is the gradient of the loss function with respect to the parameter.

[0021] Preferably, the power quality adjustment process includes:

[0022] Adopt a sliding mode variable structure control strategy, combine integral control to eliminate the steady-state error, and optimize the sliding mode control parameters through a neural network;

[0023] Among them, the sliding mode variable structure control strategy is: define a sliding mode surface function in the voltage outer loop, and improve the switching function using the boundary layer method; in the current inner loop, combine integral control to accumulate the current error, eliminate the steady-state error, and dynamically adjust the sliding mode surface parameters through a neural network.

[0024] Preferably, the role of the current error in integral control is:

[0025] u(t) = K p ·e(t) + K i ·∫e(t)dt;

[0026] Where u(t) is the controller output; K p is the proportional gain; K i is the integral gain; e(t) is the current error, representing the difference between the desired current and the actual current.

[0027] Preferably, the adaptive virtual impedance control algorithm includes:

[0028] Based on the real-time line impedance parameters and load power changes, dynamically calculate the resistance component and inductance component of the virtual impedance to ensure stable system power distribution and harmonics meeting the standards.

[0029] Preferably, the encryption communication protocol is the AES symmetric encryption algorithm, and the encryption process is:

[0030] C = E k (P);

[0031] Where C is the encrypted ciphertext, E k is the encryption function, P is the plaintext data, and k is the encryption key.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] 1. Precise power regulation and stable operation of the system:

[0034] Through real-time data acquisition, the present invention uses high-precision sensors to obtain accurate power demand and converter output information. For data processing, algorithms such as moving average filtering and trend analysis are adopted to accurately judge the working conditions. In the adaptive power adjustment link, based on a rich and perfect fuzzy logic control algorithm rule base, combined with a dynamically adjusted membership function, and then introducing machine learning optimization, it realizes fine adaptive adjustment of power to ensure the stability of the system. It can accurately respond to the complex and changeable power demand of the hydrogen production system, maintain the stable and efficient operation of the system, and effectively avoid equipment failures and production interruptions caused by power fluctuations. During the hydrogen production process, the power demand varies greatly in different stages. In the prior art, due to lagging or inaccurate adjustment, the system voltage and frequency are often unstable, affecting the hydrogen production efficiency and equipment life. However, the method of the present invention can significantly improve this situation.

[0035] 2. Optimization and improvement of power quality:

[0036] In the power quality adjustment process, an adaptive sliding mode control algorithm is adopted to adjust parameters in real time according to system changes; integral control is combined to eliminate steady-state errors; neural network sliding mode control is used for intelligent optimization, enabling the converter to stably output high-quality electric energy under complex working conditions and overcoming the limitations of traditional single control algorithms. It greatly improves the stability and accuracy of the output voltage and current of the energy storage converter, ensures the power quality of the hydrogen production system, reduces the damage of interference such as harmonics to equipment, and improves the operation reliability of the hydrogen production equipment and product quality. In the prior art, when dealing with power grid fluctuations or load changes, the power quality control is poor, which easily causes misoperation of the hydrogen production equipment or deviation of production parameters.

[0037] 3. Reliable guarantee of system coordination and communication:

[0038] In the communication and interaction link, a real-time monitoring mechanism is established to promptly detect communication faults; equipment is regularly inspected and maintained, and software is updated; firewall, intrusion detection, and encryption technologies are used to ensure network security; load balancing algorithms are used to manage data traffic. With multiple measures, communication stability, security, and efficiency are ensured, promoting system coordination. It realizes high-efficiency coordination control and safe and reliable communication between the energy storage converter and the hydrogen production system equipment, ensures the timely and accurate interaction of information, improves the overall automation and intelligence level of the system, and reduces the need for manual intervention and the risk of misoperation. In the past, there were often communication delays, data loss, or security vulnerabilities in the technology, affecting the system coordination effect and operation safety.

[0039] 4. Enhancement of controller performance and laying a stable foundation:

[0040] Preset the control algorithm parameters according to the hydrogen production process and operating conditions, and optimize the power command output using a simplified PI control algorithm; accurately configure the communication protocol stack parameters to ensure communication; calibrate the clock to ensure time synchronization; strictly perform hardware self-checks to promptly eliminate faults, comprehensively improving the performance of the controller. Initialize and configure the controller comprehensively, covering operation modes, communication protocol stacks, algorithm loading, clock calibration, and hardware self-checks, significantly enhancing the controller's response speed, processing accuracy, and stability, laying a solid foundation for the continuous and stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments and descriptions thereof of this application are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0042] Figure 1 is a flowchart of a control method for an efficient grid-forming energy storage converter applied to a hydrogen production system according to an embodiment of the present invention;

[0043] Figure 2 is a control timing diagram of the energy storage converter in the hydrogen production system according to an embodiment of the present invention;

[0044] Figure 3 is a control class diagram of the energy storage converter in the hydrogen production system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The following will refer to the drawings and combine with the embodiments to detail this application.

[0046] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0047] The present invention proposes a control method for an efficient grid-forming energy storage converter applied to a hydrogen production system, as Figure 1 , including:

[0048] Start the energy storage converter and hydrogen production system equipment, initialize and configure the controller, sensors, and communication module, set the initial operating parameters, and establish a communication connection, where the initial operating parameters include active power, reactive power output values, and reference thresholds for voltage and current;

[0049] The active power demand signal and reactive power demand signal of the hydrogen production system are collected in real time through a power sensor, and the instantaneous output voltage value and instantaneous output current value of the energy storage converter are measured in real time. The collected active power demand signal, reactive power demand signal, instantaneous output voltage value and instantaneous output current value are converted into digital signals and transmitted to the microprocessor unit of the controller;

[0050] Based on the moving average filtering algorithm, the collected data is preprocessed. Combining with the trend analysis algorithm, the power demand change of the hydrogen production system and the output state change trend of the energy storage converter are predicted. Through the threshold comparison mechanism, it is judged whether the operating condition of the hydrogen production system exceeds the preset range. If the operating condition of the hydrogen production system changes, the adaptive power adjustment process is entered. If the displayed voltage or current is abnormal, the power quality adjustment process is entered;

[0051] The line impedance parameters and load dynamic characteristics of the hydrogen production system are monitored in real time, and the virtual impedance value is dynamically adjusted by using the adaptive virtual impedance control algorithm;

[0052] The state information of the energy storage converter is transmitted to the monitoring center in real time through an encrypted communication protocol, and control instructions are received, and the load balancing algorithm is used to manage the data flow.

[0053] Specifically, in this embodiment, the system is initialized first. The device is started and the controller, sensors and communication module are configured. The operating parameters are set and the communication connection is established. The microprocessor operation mode, communication protocol stack, algorithm loading, clock calibration and hardware self-check of the controller are all optimized by formulas and strictly operated to lay a foundation for subsequent operation.

[0054] Then, real-time data collection is carried out. Appropriate sensors are selected and reasonably installed according to the hydrogen production and converter requirements. After collection, the data is processed and converted into digital signals, and then the power is calculated using formulas to provide data support for control. Subsequent data processing and analysis adopt algorithms such as moving average filtering and trend analysis, and a threshold comparison mechanism is set. When the working condition changes, the adaptive power adjustment is carried out, and the fuzzy logic algorithm and optimization measures are used to determine the power distribution and adjustment amount; when the power quality is abnormal, the sliding mode variable structure control strategy and improved algorithm are used for adjustment; the system is also optimized by using the adaptive virtual impedance control algorithm based on the monitoring information. Finally, through communication and interaction, the converter state is transmitted to the monitoring center and instructions are received, and a variety of technologies are used to ensure stable and reliable communication to achieve system collaborative optimization.

[0055] Further, start the energy storage converter and related equipment of the hydrogen production system, initialize and configure the controller, sensors, communication module, etc., set the initial operating parameters, including but not limited to the default active power, reactive power output values, reference thresholds of voltage and current, etc. in this embodiment, and establish a communication connection between the energy storage converter and other equipment of the hydrogen production system to ensure the smoothness of the data transmission channel. As Figure 2 is the control timing diagram of the energy storage converter of the hydrogen production system, as Figure 3 is the control class diagram of the energy storage converter of the hydrogen production system.

[0056] The initialization configuration of the controller includes: setting the operation mode of the microprocessor, configuring the communication protocol stack parameters, loading the control algorithm, calibrating the clock synchronization, and performing a hardware self-check. Among them, the hardware self-check includes detecting the connection status of the sensors and verifying the power supply stability.

[0057] Specifically, in the step of initializing the system, in addition to starting the basic energy storage converter and hydrogen production system equipment, it is also necessary to perform a detailed initialization configuration on each functional module inside the controller to ensure the efficient and stable operation of the entire system.

[0058] First, initialize the operation mode of the microprocessor unit inside the controller, which involves how the microprocessor processes input data, executes arithmetic tasks, and its working mode. In this process, it is necessary to clarify the data storage format, arithmetic precision, and interrupt handling mechanism parameters. The selection of the operation mode determines the response speed and processing ability of the controller in different working environments, and the data storage format can ensure the efficiency of data storage and reading. At the same time, the configuration of the interrupt handling mechanism enables the controller to give priority to high-priority tasks when an emergency occurs, ensuring real-time performance and system stability.

[0059] The configuration of the communication protocol stack of the controller is a step to ensure reliable communication between the energy storage converter and other equipment of the hydrogen production system. According to the selected communication method, such as industrial Ethernet, CAN bus, or wireless communication, etc., it is necessary to set corresponding communication rate, data frame format, and verification method and other parameters. When selecting the industrial Ethernet communication method, the controller needs to configure a suitable Ethernet protocol stack, including protocols such as TCP / IP and UDP, and adjust the communication rate and the maximum transmission unit (MTU) of the transmission according to the network load and real-time requirements. The accurate configuration of the communication parameters ensures the reliability of the system and the stability of data transmission, avoiding inaccurate control caused by data loss or delay.

[0060] Next, the controller needs to load the control algorithm program and parameter configuration file pre-stored in its internal memory. The control algorithm program and parameter configuration file are preset with the optimal control parameters for different hydrogen production processes and operating conditions to support subsequent real-time adjustment. These preset parameters include the operating modes, power distribution rules, adjustment strategies, etc. of the energy storage converter under different loads and different conditions, and can provide accurate references for the controller during actual operation. At the same time, the clock system of the controller is calibrated. Clock calibration ensures the time synchronization of the controller during the execution of multiple control links, avoiding control inaccuracies caused by time errors. Especially during multi-task parallel processing, the precise synchronization of the clock can improve control accuracy and system stability.

[0061] During the initialization of the controller, it is also necessary to perform self-checks on the hardware status, which includes checking whether each hardware interface is working properly, ensuring that the sensors are firmly connected, and the power supply is stable, etc. If any hardware faults are found, the controller will issue an alarm through the alarm device to notify the operator for maintenance in a timely manner. The self-check function can greatly improve the reliability of the system and avoid control anomalies caused by hardware faults during subsequent operation.

[0062] During the specific initialization process, the settings of the operation mode, communication protocol stack, clock synchronization, etc. of the microprocessor are constrained and optimized by the following formula:

[0063] P output =K p ·(P demand -P measured )+K i ·∫(P demand -P measured )dt;

[0064] The above formula is a simplified proportional-integral control (PI control) algorithm. Among them, P output is the power command output by the controller, P demand is the power demand of the hydrogen production system, P measured is the actual output power of the energy storage converter, K p is the proportional gain, K i is the integral gain, and the integral term represents the correction of the long-term deviation. The purpose of this formula is to dynamically adjust the output power of the energy storage converter to make it as close as possible to the desired power demand, so as to ensure the stable operation of the system under different loads and conditions.

[0065] In summary, through the initialization configuration of each functional module of the controller, including the operation mode of the microprocessor, the setting of the communication protocol stack, the loading of the control algorithm, the calibration of the clock system, and the hardware self-check, etc., a solid foundation can be laid for the efficient and stable control of the subsequent energy storage converter, ensuring that it can quickly respond and make accurate adjustments during the system operation.

[0066] Furthermore, by using power sensors installed on the connection line between the hydrogen production system and the energy storage converter, the active power demand signal and reactive power demand signal of the hydrogen production system are collected in real time; meanwhile, on the AC output side of the energy storage converter, the instantaneous value of the output voltage and the instantaneous value of the output current are measured in real time through a voltage sensor and a current sensor. After the collected analog signals are amplified, filtered and other processed by the signal conditioning circuit, they are converted into digital signals and transmitted to the microprocessor unit of the controller.

[0067] In the real-time data acquisition step, the selection and installation of sensors directly affect the monitoring accuracy of the system and the reliability of data acquisition. First of all, the selection of power sensors must be matched according to the power range and accuracy requirements of the hydrogen production system. Selecting power sensors with high sensitivity, wide measurement range and good linearity can ensure accurate capture of the active power demand and reactive power demand signals of the hydrogen production system. These power sensors should have a wide frequency response range to adapt to the power fluctuations and load changes that may occur in the hydrogen production system. In addition, the linearity of the sensor is particularly important because it directly relates to the accuracy of the measurement data. Nonlinear response will lead to deviation of the measurement results, thus affecting subsequent data processing and control decisions.

[0068] During the installation process, the power sensor should be connected in series on the connection line between the hydrogen production system and the energy storage converter to ensure that it can monitor the power change in real time. The selection of the installation location needs to ensure that the sensor can directly contact the power transmission line while avoiding the influence of interference sources. For voltage and current sensors, their selection needs to consider the amplitude and frequency characteristics of the output voltage and current of the energy storage converter. The voltage sensor needs to have an appropriate range and bandwidth to ensure accurate measurement of the output voltage of the energy storage converter; the current sensor requires a fast response time to be able to capture the instantaneous change of the current in real time. The voltage sensor is generally connected in parallel at both ends of the AC output side of the energy storage converter, while the current sensor can be installed in a through-hole or split-core manner on the output current line to achieve accurate measurement of the current.

[0069] In the actual data acquisition process, the voltage and current signals collected by the power sensor can be used to calculate the real-time power of the energy storage converter through the following formula:

[0070] P = V measured · I measured · cos(θ);

[0071] where P is the power (active power or reactive power), V measured is the real-time voltage collected by the voltage sensor, I measured is the real-time current collected by the current sensor, and θ is the phase angle between the voltage and the current. The purpose of this formula is to calculate the power through the product of the voltage and the current and their phase difference. When the voltage and the current are completely synchronized, the phase angle is zero, and the calculated power is the active power; while when there is a phase difference between the voltage and the current, the power calculated by the formula may contain a reactive power component. Accurate power calculation is the basis for the energy storage converter and the hydrogen production system to operate efficiently and stably.

[0072] In summary, by reasonably selecting sensors, accurately choosing the installation positions, and performing necessary regular calibrations, it is possible to ensure that the system collects accurate and reliable raw data. These data provide a solid foundation for subsequent data processing and control decisions, ensuring the efficient coordinated operation of the entire hydrogen production system and the energy storage converter.

[0073] Furthermore, the microprocessor unit of the controller calculates the output power of the energy storage converter in real time based on the collected output voltage and output current signals, and stores data such as the power demand signal, output voltage, current, and power in the memory. These data are analyzed in real time to determine whether the operating conditions of the hydrogen production system have changed.

[0074] If the analysis result shows that the operating conditions of the hydrogen production system have changed significantly, such as a large increase or decrease in power demand beyond the preset normal fluctuation range, it enters the adaptive power adjustment process. First, according to the power demand model established based on different stages of the hydrogen production process and changes in the load, determine the power distribution mode corresponding to the current operating conditions and obtain the corresponding power distribution target value. Then, adopt the fuzzy logic control algorithm, with the power demand signal of the hydrogen production system and the current output power of the energy storage converter as input variables, and calculate the adjustment amounts of the active power and reactive power that the energy storage converter should output under the current operating conditions through steps such as fuzzification, fuzzy reasoning, and defuzzification. Finally, according to the calculated adjustment amounts, dynamically adjust the power command value of the energy storage converter, and consider the capacity limit of the energy storage converter and the charge and discharge characteristics of the battery to perform a limit processing on the power command value to ensure that the energy storage converter operates within a safe and efficient range.

[0075] Specifically, the moving average filtering algorithm is used to filter the collected data such as the power demand signal, output voltage, current, and power, removing high-frequency noise and abnormal fluctuations in the data, making the data smoother and more stable, which is convenient for subsequent analysis and judgment.

[0076] Apply trend analysis algorithms to analyze the data change trend over a period of time, and predict the power demand of the hydrogen production system and the change trend of the output state of the energy storage converter. For example, by analyzing historical power demand data, predict whether the hydrogen production system may have power peaks or valleys in the future, so as to adjust the control strategy of the energy storage converter in advance.

[0077] Judge whether the operating condition of the hydrogen production system exceeds the preset range through a threshold comparison mechanism. If the operating condition of the hydrogen production system changes, enter the adaptive power adjustment process. If abnormal voltage or current is displayed, enter the power quality adjustment process;

[0078] Specifically, the threshold comparison mechanism is to set multiple levels of voltage / current deviation thresholds. When the voltage / current data exceeds the first-level threshold, start the preliminary adjustment. When the voltage / current data exceeds the second-level threshold, trigger the emergency protection mechanism.

[0079] Establish a data threshold comparison mechanism. In addition to judging whether the power demand exceeds the preset normal fluctuation range, multiple thresholds are also set for the amplitude and change rate of the output voltage and current. When the data exceeds different levels of thresholds, different warning and control measures are taken. For example, when the output voltage slightly deviates from the reference value, start the preliminary adjustment mechanism; when the voltage deviation exceeds the severe threshold, immediately trigger the emergency protection measure and stop part of the output of the energy storage converter to protect the safety of the equipment. Through the comprehensive application of various data analysis methods, the operating status of the hydrogen production system and the energy storage converter can be understood more comprehensively and accurately, providing strong support for subsequent judgment and control decisions.

[0080] Furthermore, the adaptive power adjustment process includes:

[0081] Calculate the power adjustment amount based on the fuzzy logic control algorithm, and combine machine learning to optimize the membership function and fuzzy rule base to dynamically adjust the active power and reactive power commands of the energy storage converter.

[0082] Specifically, in the adaptive power adjustment process, the optimization measures of the fuzzy logic control algorithm aim to improve the system's response ability and control accuracy. First of all, enriching and improving the rule base of the fuzzy logic control algorithm is the basis of optimization. By analyzing a large amount of actual operation data and experimental results, summarize the relationship between the power demand signal and the current output power of the energy storage converter under different operating conditions, and different adjustment strategies that should be taken in different operating scenarios can be identified. According to these relationships, add more fuzzy control rules, which can not only cover a wider range of working states more comprehensively, but also improve the adaptability and accuracy of the control algorithm. For example, when the system load changes greatly, some rules may need to be adjusted to more accurately respond to the changing power demand. These new rules and existing rules work together to enhance the controller's ability to handle complex environments.

[0083] Adaptive adjustment of the membership function is another optimization measure, which helps to improve the sensitivity and stability of fuzzy logic control under different operating conditions. The membership function is used to define the membership degree of input variables (such as the difference between power demand and output power) in the fuzzy set, and the control output is determined through fuzzy inference. According to the real-time operating state and control effect of the hydrogen production system, the shape and parameters of the membership function are dynamically adjusted. Specifically, when the system is in a relatively stable operating state, a relatively broad membership function is adopted to improve the smoothness and stability of control and avoid unnecessary fluctuations caused by frequent adjustments. When the system undergoes rapid changes or approaches the critical state, the membership function is adjusted to be more sensitive, so as to quickly respond to changes in working conditions and provide more refined adjustments. This adjustment method enables fuzzy logic control to adapt to the requirements of the system under different working conditions, ensuring that the energy storage converter can efficiently regulate the power output.

[0084] To further improve the adaptive ability of the fuzzy logic control algorithm, a machine learning algorithm is introduced. By learning historical data through a machine learning model, fuzzy control rules and membership functions can be automatically generated or adjusted. The machine learning method using a neural network is used to learn a large amount of historical power data and control effects. The neural network can optimize the parameters of the control rules and membership functions through the backpropagation algorithm, thereby improving the accuracy of fuzzy control. Specifically, the neural network improves fuzzy logic control through the following optimization process:

[0085] θ new =θ old -η·▽J(θ);

[0086] where θ new is the updated parameter, θ old is the current parameter, η is the learning rate, and ▽J(θ) is the gradient of the loss function J(θ) with respect to the parameter θ. The loss function J(θ) measures the error between the control output and the actual power demand, and the optimization goal is to minimize this error. Through this method, the neural network can automatically adjust the key parameters in the fuzzy control algorithm, enabling it to provide more accurate power adjustment instructions based on historical data and system operation results.

[0087] The fuzzy logic control algorithm optimized by combining machine learning can provide more refined adaptive adjustment when facing the complex and changeable power demand of the hydrogen production system. The system can continuously adjust the control rules and membership functions by learning historical data to cope with different load changes and working condition fluctuations. Finally, this optimization method enables the energy storage converter to better adapt to the complex power demand of the hydrogen production system and achieve efficient and stable power adjustment.

[0088] Furthermore, the power quality adjustment process includes:

[0089] The sliding mode variable structure control strategy is adopted, combined with integral control to eliminate the steady-state error, and the sliding mode control parameters are optimized by neural network;

[0090] Among them, the sliding mode variable structure control strategy is as follows: in the outer voltage loop, the sliding mode surface function is defined, and the switching function is improved by the boundary layer method; in the inner current loop, the integral control is combined to accumulate the current error, eliminate the steady-state error, and the sliding mode surface parameters are dynamically adjusted by neural network.

[0091] Specifically, in the power quality adjustment process, the improvement of the sliding mode control algorithm aims to improve the dynamic response ability and anti-interference ability of the energy storage converter. Adopting the adaptive sliding mode control algorithm is one of the core measures, mainly by adjusting the parameters of the sliding mode control in real time, such as the slope of the sliding mode surface, the gain of the switching function, etc., to cope with the changes of system parameters and the influence of external interference. The sliding mode control depends on the sliding mode surface to determine the change of the control input, and the slope of the sliding mode surface and the gain of the switching function have a direct impact on the response speed and anti-interference ability of the system. When the system is subject to large external interference, automatically increasing the gain of the switching function can improve the resistance of the system to interference and enable the system to quickly return to the desired stable state. For example, when the control system detects changes in the external load or grid fluctuations, by increasing the gain of the switching function, the output of the controller can be adjusted more strongly to quickly eliminate the deviation caused by external interference. When the system parameters change, such as the battery state of the energy storage converter changes or the load conditions change, timely adjusting the slope of the sliding mode surface can ensure that the system still maintains a high dynamic tracking ability and stability under the new working conditions, and avoid excessive oscillation or slow response.

[0092] In order to further improve the control accuracy and system stability, combining other control algorithms, especially the integral control algorithm, to supplement and optimize the sliding mode control is an effective strategy. In the inner current loop control, introducing an integral link can accumulate the current error, thereby eliminating the steady-state error. The integral control continuously adjusts the output through the integral operation of the current error until the current error is zero. After the integral result is combined with the output of the sliding mode control, it can improve the control accuracy while ensuring the system stability.

[0093] Specifically, the role of the current error e(t) in the integral control can be expressed by the following formula:

[0094] u(t) = K p ·e(t) + K i ·∫e(t)dt;

[0095] Among them, u(t) is the controller output, K p is the proportional gain, Ki Let \(K_i\) be the integral gain and \(e(t)\) be the current error, representing the difference between the desired current and the actual current. The proportional term is used to adjust the instantaneous response of the current error, while the integral term can eliminate the steady-state error, enabling the system to maintain an accurate control output after long-term operation. This supplementary strategy helps to make up for the deficiencies that may occur in a single sliding mode control algorithm when facing persistent errors.

[0096] In addition, intelligent control methods such as neural network sliding mode control are adopted to further improve the adaptability and performance of the sliding mode control algorithm. The neural network can learn the uncertainties and nonlinear characteristics of the system and dynamically adjust the parameters of the sliding mode control. Specifically, the neural network can identify and approximate the dynamic behavior of the system under complex operating conditions, and adjust the parameters of the sliding mode surface according to real-time feedback, optimize the gain of the switching function, and even improve the generation process of the control signal. In this way, the neural network can help the sliding mode control algorithm to achieve self-adjustment in a complex environment, further improving the control accuracy and stability of the energy storage converter under nonlinear, time-varying, and complex conditions. The learning process of neural network control usually adopts the backpropagation algorithm to optimize the control parameters by minimizing the loss function:

[0097] \(\theta\) new \(=\theta\) old \(-\eta\cdot\nabla J(\theta)\);

[0098] where \(\theta\) new is the updated parameter, \(\theta\) old is the current parameter, \(\eta\) is the learning rate, and \(\nabla J(\theta)\) is the gradient of the loss function \(J(\theta)\) with respect to the parameter \(\theta\).

[0099] Through this iterative learning process, the neural network can continuously improve the parameters of the sliding mode control to make it more adaptable to the complex power quality requirements in the hydrogen production system.

[0100] Through these improvements to the sliding mode control algorithm, the stability and accuracy of the output voltage and current of the energy storage converter can be significantly improved, ensuring that the hydrogen production system can provide high-quality electric energy. Through adaptive parameter adjustment, the introduction of integral control, and the application of intelligent control, the system can flexibly cope with various factors such as external disturbances, equipment aging, or load changes, and continuously maintain efficient power quality control.

[0101] Furthermore, the adaptive virtual impedance control algorithm includes:

[0102] Based on the real-time line impedance parameters and load power changes, dynamically calculate the resistance component and inductance component of the virtual impedance to ensure stable system power distribution and harmonic compliance with standards.

[0103] Specifically, the line impedance parameters of the hydrogen production system are monitored in real time. The current line resistance and line inductance values of the system are obtained through on-line measurement or model-based estimation methods. At the same time, a load monitoring device is used to collect the active power and reactive power change information of the load in real time, and the dynamic characteristics of the load are analyzed. Based on the real-time monitored line impedance parameters and load change information, an adaptive virtual impedance control algorithm is adopted to calculate the virtual impedance value that the energy storage converter should set under the current working conditions, and the resistance component and inductance component of the virtual impedance are dynamically adjusted to optimize the power distribution and stability of the system. At the same time, it is ensured that the impact of the virtual impedance on the power quality of the system is within the allowable range, and the voltage deviation and current harmonics of the system meet the relevant standard requirements.

[0104] Furthermore, the status information of the energy storage converter is transmitted to the monitoring center in real time through an encrypted communication protocol, and control instructions are received, and a load balancing algorithm is used to manage the data traffic.

[0105] The encrypted communication protocol is the AES symmetric encryption algorithm, and the encryption process is as follows:

[0106] C = E k (P);

[0107] Where C is the encrypted ciphertext, E k is the encryption function, P is the plaintext data, and k is the encryption key.

[0108] Specifically, through the established communication network, the operation status information of the energy storage converter, such as output voltage, current, power, battery power, etc., is transmitted to the monitoring center of the hydrogen production system in real time. At the same time, control instructions and parameter setting information from the monitoring center of the hydrogen production system or other devices are received, and according to the received instructions, the control strategy and operation parameters of the energy storage converter are adjusted to achieve the coordinated control and optimized operation of the system. During the communication process, reliable data transmission protocols and encryption technologies are adopted to ensure the accuracy, integrity and security of data transmission, perform reasonable packaging and compression processing on communication data, and set up a data caching mechanism to avoid data loss and communication congestion.

[0109] The management and maintenance measures of the communication network are crucial for ensuring stable, reliable, and secure communication between the energy storage converter and other devices in the hydrogen production system. First of all, establishing a real-time monitoring mechanism for the communication network is a key link. By regularly detecting parameters such as the connection status, data transmission rate, and bit error rate of the communication link, faults or performance degradation in the communication link can be discovered in a timely manner. The monitoring of the connection status can ensure the stability of the physical layer by checking the connection between network devices; while the detection of the data transmission rate and bit error rate helps to evaluate the transmission performance and reliability of the network. For example, when the data transmission rate is too low or the bit error rate is too high, it may mean that there is a network bottleneck or interference. When an anomaly is discovered, through fault diagnosis and location means, the root cause of the problem can be quickly determined, and the communication can be restored by using a backup communication link or an automatic repair mechanism. This mechanism not only ensures the high availability of the network but also enables quick recovery in case of communication failures, avoiding affecting the overall operation of the system.

[0110] In addition, regularly inspecting and maintaining the devices in the communication network to ensure the normal operation of the device hardware and software is another important measure to guarantee network stability. During device inspections, it is necessary to check the hardware status of network devices to ensure that hardware such as power supplies, connection ports, and network interfaces are working properly; at the same time, timely update the software patches of communication devices to repair known security vulnerabilities and performance issues, improving the security and stability of the devices. Version management of device software is also important. Over time, new versions of software may provide more efficient algorithms and functions, and updating the software version can improve the performance and response speed of the entire system.

[0111] To prevent the destruction of the communication network by external malicious attacks, it is essential to adopt network security protection measures. Security devices such as firewalls and intrusion detection systems (IDS) can monitor network traffic in real time, detect, and prevent unauthorized access or malicious attacks. Network security protection also includes data encryption technology to ensure the security and confidentiality of data during transmission. To ensure data confidentiality, symmetric encryption or asymmetric encryption algorithms are usually used to encrypt data. For example, in symmetric encryption, a shared key is used to encrypt and decrypt data, while in asymmetric encryption, a pair of public and private keys are used for encryption and decryption. These encryption algorithms can ensure that even if the data is intercepted during transmission, it cannot be read by unauthorized third parties.

[0112] Common symmetric encryption algorithms include AES (Advanced Encryption Standard), and its encryption process can be expressed as:

[0113] C = E k (P);

[0114] where C is the encrypted ciphertext, E kLet \(E\) be the encryption function, \(P\) be the plaintext data, and \(k\) be the encryption key.

[0115] During decryption, the same key \(k\) is used to restore the plaintext data:

[0116] \(P = D\) k (C);

[0117] where \(D\) k is the decryption function used to convert the ciphertext \(C\) into the plaintext data \(P\).

[0118] This encryption process effectively ensures the security of the data and defends against external attacks during communication.

[0119] Meanwhile, it is also very important to monitor and manage the data traffic in the communication network. This management measure can prevent communication congestion caused by excessive data traffic, thus affecting the real-time performance and reliability of the system. Reasonably allocating network resources to ensure appropriate bandwidth allocation among different communication tasks can improve data transmission efficiency and avoid adverse effects of high-traffic tasks on system performance. Traffic control is usually optimized through network load balancing algorithms to ensure reasonable utilization of the bandwidth of each communication link. A commonly used load balancing algorithm is the Round Robin algorithm, which sequentially assigns each request to different servers or links until the loads of all links or servers tend to be balanced, avoiding overloading of a single link or device. In this way, the efficient operation of the communication network can be ensured, reducing latency and packet loss rate and improving the real-time response ability of the system.

[0120] Through these perfect communication network management and maintenance measures, the stability, security, and efficiency of communication between the energy storage converter and other devices of the hydrogen production system can be guaranteed. This not only improves the reliability of the entire system but also provides a solid foundation for realizing efficient and stable energy storage and hydrogen production control.

[0121] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A control method for a high-efficiency grid-type energy storage converter applied to a hydrogen production system, characterized in that: include: Collecting the active power demand signal and reactive power demand signal of the hydrogen production system in real time, and measuring the instantaneous value of the output voltage and the instantaneous value of the output current of the energy storage converter in real time, converting the collected active power demand signal and reactive power demand signal, the instantaneous value of the output voltage and the instantaneous value of the output current into digital signals and transmitting them to the microprocessor unit of the controller; Preprocess the collected data based on the moving average filtering algorithm, and predict the power demand change of the hydrogen production system and the output state change trend of the energy storage converter in combination with the trend analysis algorithm. Judge whether the operating condition of the hydrogen production system exceeds the preset range through the threshold comparison mechanism. If the operating condition of the hydrogen production system changes, enter the adaptive power adjustment process. If the displayed voltage or current is abnormal, enter the power quality adjustment process. Real-time monitoring of line impedance parameters and load dynamic characteristics of the hydrogen production system, and dynamic adjustment of virtual impedance values ​​using an adaptive virtual impedance control algorithm; The state information of the energy storage converter is transmitted to the monitoring center in real time through an encrypted communication protocol, and control instructions are received, and a load balancing algorithm is used to manage data traffic.

2. The control method according to claim 1, characterized in that: Before real-time acquisition of active power demand signal and reactive power demand signal of hydrogen production system, it also includes: Start the energy storage converter and hydrogen production system equipment, initialize the configuration of the controller, sensor and communication module, set the initial operating parameters, and establish a communication connection, wherein the initial operating parameters include active power, reactive power output values, and reference thresholds of voltage and current.

3. The control method according to claim 2, characterized in that: The initialization configuration of the controller includes: microprocessor operation mode setting, communication protocol stack parameter configuration, control algorithm loading, clock synchronization calibration and hardware self-test, wherein the hardware self-test includes sensor connection status detection and power supply stability verification.

4. The control method according to claim 1, characterized in that: The threshold comparison mechanism is to set multiple levels of voltage / current deviation thresholds, start preliminary adjustment when the voltage / current data exceeds the first level threshold, and trigger an emergency protection mechanism when the voltage / current data exceeds the second level threshold.

5. The control method according to claim 1, characterized in that: The adaptive power adjustment process includes: The power adjustment amount is calculated based on the fuzzy logic control algorithm, and the active power and reactive power instructions of the energy storage converter are dynamically adjusted by combining machine learning to optimize the membership function and fuzzy rule base.

6. The control method according to claim 5, characterized in that: The adaptive power adjustment process also includes: The optimization of fuzzy logic control algorithm includes: learning historical power data and control effects through machine learning model, and optimizing fuzzy rule base and membership function parameters by back propagation algorithm, specifically: In the formula, θ new is the updated parameter, θ old is the current parameter, η is the learning rate, is the gradient of the loss function with respect to the parameters.

7. The control method according to claim 1, characterized in that: The power quality adjustment process includes: The sliding mode variable structure control strategy is adopted, combined with integral control to eliminate steady-state errors, and the sliding mode control parameters are optimized through neural networks; Among them, the sliding mode variable structure control strategy is: defining the sliding surface function in the voltage outer loop, using the boundary layer method to improve the switching function; combining integral control in the current inner loop to accumulate the current error, eliminate the steady-state error, and dynamically adjust the sliding surface parameters through the neural network.

8. The control method according to claim 7, characterized in that: The role of the current error in integral control is: u(t)=K p ·e(t)+K i ·∫e(t)dt; Where u(t) is the controller output; K p is the proportional gain; K i is the integral gain; e(t) is the current error, which represents the difference between the expected current and the actual current.

9. The control method according to claim 1, characterized in that: The adaptive virtual impedance control algorithm comprises: Based on real-time line impedance parameters and load power changes, the resistance and inductance components of the virtual impedance are dynamically calculated to ensure that the system power distribution is stable and the harmonics meet the standards.

10. The control method according to claim 1, characterized in that: The encryption communication protocol is the AES symmetric encryption algorithm, and the encryption process is: C=E k (P); In the formula, C is the encrypted ciphertext, E k is the encryption function, P is the plaintext data, and k is the encryption key.

Images