Dynamic power matching method of photovoltaic hydrogen production system

Abstract

The invention provides a dynamic power matching method of a photovoltaic hydrogen production system, which comprises the following steps: acquiring load demand data from an electrolytic tank in the hydrogen production system, combining stable power output, finely adjusting a power adjusting range by utilizing a proportional-integral control method, and judging a real-time power matching state when the electrolytic tank runs; after continuous operation parameters are obtained, through conjoint analysis of real-time power data and gas supply parameters, a self-adaptive filtering algorithm is adopted to optimize a dynamic matching process, and the long-term stability of the system under illumination fluctuation is judged; and extracting abnormal characteristics of power instability and gas supply fluctuation from a long-term stability judgment result, updating control parameters of an MPPT algorithm through historical operation data, and determining an optimized topological structure and a gas supply support scheme. According to the invention, the long-term stability of the system under illumination fluctuation is improved, and efficient and stable operation of the photovoltaic hydrogen production system is realized.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a dynamic power matching method for a photovoltaic hydrogen production system. Background Art

[0002] Photovoltaic hydrogen production is a key technology that combines renewable energy and hydrogen energy. By directly using photovoltaic power generation to drive electrolysis to produce hydrogen, it can effectively reduce dependence on fossil energy and improve energy utilization efficiency.

[0003] The power supply topology and control methods of existing photovoltaic hydrogen production systems are typically based on simple maximum power point tracking (MPPT) algorithms and fixed topologies, but these methods show significant limitations when dealing with drastic fluctuations in light intensity. Traditional systems often struggle to match photovoltaic output power with the load requirements of the electrolytic cell in real time, resulting in reduced energy conversion efficiency and even unstable electrolytic cell operation. In addition, the gas supply support system lacks synergy with the power supply control, making the hydrogen production process easily interrupted by abnormal gas pressure or flow, limiting the reliability and continuity of the system. The core challenges lie in how to achieve dynamic matching between the photovoltaic power supply and the electrolytic hydrogen production system, and how to ensure coordinated optimization of the gas supply support system and MPPT control. The power instability caused by light fluctuations requires a higher level of flexibility in the power supply topology, while the pressure stability and flow balance of the gas supply system directly affect the operating environment of the electrolytic cell. Existing technologies are insufficient in integrating real-time power data with the dynamic adjustment of gas supply parameters, making it difficult to maintain the stability and safety of the hydrogen production process under complex operating conditions.

[0004] Therefore, how to design a photovoltaic hydrogen production power supply topology and control method based on MPPT, integrating a multi-stage buffer gas supply support system to achieve dynamic power matching, stable gas supply and fault adaptive switching, becomes the key issue of this study. Summary of the Invention

[0005] To address the deficiencies of the prior art, the present invention aims to provide a dynamic power matching method for a photovoltaic hydrogen production system. By optimizing the dynamic matching process through an adaptive filtering algorithm, the long-term stability of the system under light fluctuations is improved, thereby achieving efficient and stable operation of the photovoltaic hydrogen production system.

[0006] The present invention is implemented as follows: a dynamic power matching method for a photovoltaic hydrogen production system mainly includes:

[0007] The photovoltaic power supply supplies power to the electrolytic cell, and the ambient light fluctuation parameters are obtained through sensors. The changes in the photovoltaic power supply output power caused by the ambient light fluctuation are then stabilized to obtain a stable power output suitable for the hydrogen production system.

[0008] Collect load demand data from the electrolytic cell in the hydrogen production system, combine it with stable power output, and use the proportional-integral control method to fine-tune the power adjustment range to determine the real-time power matching status of the electrolytic cell during operation;

[0009] If the real-time power matching status exceeds the preset load threshold, the gas pressure data is collected through the pressure sensor in the buffer gas supply system, and the gas supply flow is adjusted using the PID control algorithm to determine the gas supply parameters under stable gas supply conditions;

[0010] Based on the gas supply parameters under stable gas supply conditions, the pressure fluctuation and flow balance indicators are extracted from the electrolytic cell operation data. The operational stability of the hydrogen production system is determined by the preset stability threshold, and the trigger condition for fault switching is obtained.

[0011] If the trigger condition for failover is activated, the system switches to backup power mode through the preset redundant topology, adjusts the gas supply rhythm in combination with the multi-stage gas storage unit of the buffer gas supply system, and determines the continuous operation parameters of the hydrogen production system under abnormal working conditions;

[0012] After obtaining the continuous operation parameters, the dynamic matching process is optimized by combining real-time power data with gas supply parameters using an adaptive filtering algorithm to determine the long-term stability of the system under light fluctuations.

[0013] The abnormal characteristics of power instability and gas supply fluctuations are extracted from the long-term stability judgment results, the control parameters of the MPPT algorithm are updated through historical operation data, and the optimized topology structure and gas supply support plan are determined.

[0014] Furthermore, the voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes:

[0015] The real-time light intensity and output power of the photovoltaic power source are collected by a light sensor and a power sensor, and stored as a first data set;

[0016] Extracting the fluctuation frequency and amplitude of the light intensity and output power using a time series analysis method based on the first data set to generate a first fluctuation parameter set;

[0017] determining the illumination fluctuation amplitude, and if the fluctuation amplitude in the first fluctuation parameter set exceeds a preset power threshold, generating a power instability state flag through a logic judgment method to obtain the power instability state flag;

[0018] According to the power instability state identifier, a K-means algorithm is used to perform cluster analysis on the fluctuation frequency and amplitude in the first fluctuation parameter set to generate a first initial fluctuation feature set, thereby obtaining initial fluctuation features under the power instability state.

[0019] Furthermore, the voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes:

[0020] According to the initial fluctuation feature set in the power unstable state, the output power of the photovoltaic power source is tracked in real time using the Perturb and Observe algorithm to generate a first power tracking data set, thereby obtaining the first power tracking data set;

[0021] If the output power fluctuation amplitude in the first power tracking data set exceeds a preset threshold, adjusting the voltage parameters in the topology structure through a logic judgment method to generate a first voltage adjustment set;

[0022] According to the first voltage adjustment set, a linear interpolation method is used to optimize the current parameters in the topology structure to generate a first current adjustment set;

[0023] By comparing the first current adjustment set with a preset power adjustment range, an iterative optimization method is used to determine a final power adjustment range required for dynamic matching, and the final power adjustment range is determined.

[0024] Furthermore, the voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes:

[0025] After obtaining the power adjustment range required for dynamic matching, the sensor is used to obtain the ambient light intensity change data in real time, and the fast Fourier transform algorithm is used to analyze the fluctuation frequency and amplitude of the light intensity change data to determine the power adjustment range;

[0026] If the fluctuation frequency is higher than a preset threshold, the topology of the multi-stage DC-DC converter is dynamically adjusted through a control algorithm, and pulse width modulation technology is used to optimize current control to obtain a stable voltage output;

[0027] The stable voltage output is received by the load matching module, and the matching degree between the output power and the load power is adjusted according to the real-time load demand of the hydrogen production system to determine the energy conversion efficiency;

[0028] The energy conversion efficiency data is obtained, voltage and current fluctuations are monitored in real time through a feedback control loop, and the voltage and current fluctuations are corrected using a proportional integral differential algorithm to obtain a stable power output suitable for a hydrogen production system.

[0029] Furthermore, the load demand data is collected from the electrolytic cell in the hydrogen production system, and the power adjustment range is fine-tuned using a proportional-integral control method in combination with the stable power output to determine the real-time power matching status of the electrolytic cell during operation, including:

[0030] Obtaining load demand data from an electrolytic cell sensor, using a high-frequency sampling method to collect the current and voltage of the electrolytic cell in real time, and processing the collected raw data through a data smoothing and filtering algorithm to obtain smoothed load demand data;

[0031] According to the smoothed load demand data, the proportional integral control method is used to calculate the power regulation amount, using the formula u(t)=K p e(t)+K i inte(t)d(t) calculates the control output and determines the power regulation amount;

[0032] Where u(t) is the power regulation amount, e(t) is the deviation between the load demand data and the current power output, K p is the proportionality coefficient, K i is the integration coefficient;

[0033] If the power adjustment amount exceeds the preset power range, the power adjustment amount is clipped by a limiter, and the clipped adjustment amount is distributed using a dynamic adjustment algorithm to obtain an adjusted power output value;

[0034] By comparing the deviation between the adjusted power output value and the load demand data, a preset threshold is used to determine whether the deviation is within an allowable range, thereby obtaining the real-time power matching status of the electrolytic cell.

[0035] Furthermore, if the real-time power matching state exceeds a preset load threshold, gas pressure data is collected through a pressure sensor in the buffer gas supply system, and the gas supply flow is adjusted using a PID control algorithm to determine gas supply parameters under stable gas supply conditions, including:

[0036] If the real-time power exceeds the load threshold, the state monitoring module compares the real-time power with the load threshold to obtain a trigger signal;

[0037] acquiring gas pressure data from a pressure sensor in a buffer gas supply system according to the trigger signal to determine a pressure state of the gas supply system;

[0038] Using a PID control algorithm, based on the error between the gas pressure data and the target pressure value, the adjustment amount of the gas supply flow is calculated to obtain the flow control instruction;

[0039] The valve opening of the buffer gas supply system is adjusted through the flow control instruction to determine whether the gas supply parameters meet the preset stability conditions.

[0040] Furthermore, according to the gas supply parameters under stable gas supply conditions, the pressure fluctuation and flow balance indicators are extracted from the electrolytic cell operation data, and the operation stability of the hydrogen production system is judged by a preset stability threshold to obtain the triggering conditions for fault switching, including:

[0041] Obtaining raw data of pressure fluctuation and flow balance from electrolytic cell operation data, and preprocessing the raw data using a time series analysis tool to obtain pressure fluctuation values ​​and flow balance indicators;

[0042] According to a preset stability threshold, a comparison algorithm is used to determine whether the pressure fluctuation value and the flow balance index exceed the threshold range. If so, it is determined that the system operation is unstable;

[0043] Classifying the unstable operation data by a logistic regression algorithm, obtaining the electrolytic cell parameter characteristics that cause the instability, and obtaining the potential triggering conditions for fault switching;

[0044] In response to the potential triggering conditions, a rule engine is used to generate a fault switching instruction according to the gas supply conditions and the electrolytic cell parameter characteristics to determine the final triggering conditions.

[0045] Furthermore, if the triggering condition for the fault switching is activated, the system switches to the backup power supply mode through the preset redundant topology structure, adjusts the gas supply rhythm in combination with the multi-stage gas storage unit of the buffer gas supply system, and determines the continuous operation parameters of the hydrogen production system under abnormal working conditions, including:

[0046] If the trigger condition is activated, the system switches to a backup power supply mode through a preset redundant topology structure, obtains a stable power supply from the backup power supply, and determines the power input parameters of the hydrogen production system under abnormal operating conditions;

[0047] According to the gas storage capacity of the multi-stage gas storage unit, the gas supply rhythm is adjusted through the buffer gas supply system to obtain a stable gas flow rate and determine the raw material input parameters of the hydrogen production system;

[0048] If the hydrogen production system detects an abnormal operating condition, the logic controller adjusts the operating parameters by combining the power input parameters and the raw material input parameters to obtain a control instruction for continuous operation;

[0049] The control instructions are used to drive the execution unit of the hydrogen production system, and the preset operating parameters are used to adjust the reaction rate to obtain a stable hydrogen production output.

[0050] Furthermore, after obtaining the continuous operation parameters, the dynamic matching process is optimized by using a combined analysis of real-time power data and gas supply parameters, and an adaptive filtering algorithm is used to determine the long-term stability of the system under light fluctuations, including:

[0051] Acquire continuous operating parameters and real-time power data from the operating equipment, and simultaneously acquire gas supply parameters, and perform high-frequency sampling using a data acquisition frequency control module to obtain the raw data set;

[0052] Based on the original data set, feature extraction is performed on the real-time power data and gas supply parameters through a joint analysis module. If the feature value exceeds a preset threshold range, normalization processing is performed to determine the feature set after analysis;

[0053] Adopting an adaptive filtering algorithm to optimize the analyzed feature set, and performing smoothing processing on the data through a Kalman filter to obtain the optimized matching parameters;

[0054] Under the condition of light fluctuation, the long-term stability of the optimized matching parameters is evaluated by the stability judgment module. If the fluctuation amplitude of the parameters is lower than a preset threshold, it is judged that the system has long-term stability.

[0055] Furthermore, the abnormal characteristics of power instability and gas supply fluctuation are extracted from the long-term stability judgment results, the control parameters of the MPPT algorithm are updated through historical operation data, and the optimized topology structure and gas supply support plan are determined, including:

[0056] Acquire operation data from historical records, and use a time series analysis tool to extract abnormal features for power fluctuation and gas supply stability to obtain the abnormal feature data set;

[0057] According to the abnormal characteristic data set, a data update tool is used to adjust the control parameters. If the voltage change exceeds a preset threshold, the parameters of the MPPT algorithm are updated through iterative calculation to determine the optimized control parameter set;

[0058] Using the optimized control parameter set, a topology analysis tool is used to evaluate the topology structure, and the connection mode is adjusted according to the uneven airflow distribution to obtain the optimized topology structure;

[0059] According to the optimized topology, a gas supply scheme is designed using a fluid mechanics simulation tool. If the gas supply fluctuation is less than a preset threshold, a support scheme is generated through parameter mapping to determine the final gas supply support scheme.

[0060] The technical solution provided by the embodiment of the present invention has the following beneficial effects:

[0061] The present invention discloses a dynamic power matching method for a photovoltaic hydrogen production system. By real-time collection of light intensity and output power data, combined with ambient light fluctuation parameters, the initial fluctuation characteristics under unstable power conditions are determined. The Perturb and Observe MPPT algorithm is used to track the output of the photovoltaic power supply in real time, and the topology structure is adjusted through a multi-stage DC-DC converter to achieve voltage stabilization. Combined with the load requirements of the electrolytic cell of the hydrogen production system, the proportional integral control method is used to fine-tune the power, and the PID control algorithm is used to adjust the gas supply flow. The present invention also provides a fault switching mechanism, which switches to the backup power mode under abnormal working conditions to ensure continuous operation of the system. The dynamic matching process is optimized by an adaptive filtering algorithm, the long-term stability of the system under light fluctuations is improved, and efficient and stable operation of the photovoltaic hydrogen production system is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 Schematic diagram of a flow chart of an embodiment of the present invention;

[0063] Figure 2 Schematic diagram of the process of step S102 of the embodiment of the present invention;

[0064] Figure 3 3 is a flow chart of step S109 in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] like Figure 1-Figure 3 In this embodiment, a dynamic power matching method of a photovoltaic hydrogen production system may specifically include:

[0067] In step S101, real-time light intensity data and output power data are collected from the photovoltaic power source, ambient light fluctuation parameters are obtained through sensors, and the light fluctuation amplitude is determined in combination with a preset power threshold to obtain initial fluctuation characteristics under power instability.

[0068] The real-time light intensity and output power of the photovoltaic power source are collected using a light sensor and a power sensor and stored as a first data set, thereby obtaining the first data set. Based on the first data set, a time series analysis method is used to extract the fluctuation frequency and amplitude of the light intensity and output power, generating a first fluctuation parameter set, and determining the first fluctuation parameter set. If the fluctuation amplitude in the first fluctuation parameter set exceeds a preset power threshold, a power instability state indicator is generated using a logical judgment method, thereby obtaining the power instability state indicator. Based on the power instability state indicator, a K-means algorithm is used to perform cluster analysis on the fluctuation frequency and amplitude in the first fluctuation parameter set, generating a first initial fluctuation feature set, thereby obtaining the first initial fluctuation feature set.

[0069] Specifically, the photovoltaic power supply collects ambient light intensity data at a frequency of 10 times per second through the photosensor installed on the surface of the component, and the data range is 0-1500W / m 2 The DC output power is recorded at the same frequency by current and voltage sensors with an accuracy of ±0.5%. A sliding window algorithm is used to process 100 consecutive sampling points (10 seconds of data) in real time. The window slides 50 sampling points each time, and the standard deviation of the light intensity within the window is calculated as the fluctuation parameter. When the standard deviation exceeds the preset threshold of 50W / m 2 For the identified fluctuation state, the fast Fourier transform is used to analyze the data of the last 5 minutes, and the characteristic frequency components with amplitude exceeding 20% ​​of the reference value in the 0.1-1Hz frequency band are extracted. A three-dimensional feature vector (fluctuation amplitude, dominant frequency component, power gradient) is established (with a threshold set to ±5% / s), where ΔP represents the power change and Δt represents the time change. A pre-trained random forest model (consisting of 200 decision trees and a three-node input layer) is used to classify the feature vector. When the output probability exceeds 0.7, an abnormal fluctuation is detected. The Kalman filter algorithm is automatically activated, revising the power prediction value with a 10-second cycle. Filter parameters Q = 0.01 and R = 0.1 keep the prediction error within ±3%. The system also simultaneously records 12-dimensional features, including fluctuation duration and maximum deviation, and stores them in a time series database, providing a data foundation for subsequent fluctuation pattern analysis.

[0070] Step S102 : Based on the initial fluctuation characteristics in the power instability state, the Perturb and Observe MPPT algorithm is used to track the photovoltaic power output in real time, adjust the voltage and current parameters in the topology structure, and determine the power adjustment range required for dynamic matching.

[0071] Based on the initial fluctuation characteristic set, the output power of the photovoltaic power source is tracked in real time using the Perturb and Observe algorithm to generate a first power tracking data set, thereby obtaining the first power tracking data set. If the output power fluctuation amplitude in the first power tracking data set exceeds a preset threshold, the voltage parameters in the topology structure are adjusted using a logical judgment method to generate a first voltage adjustment set, thereby determining the first voltage adjustment set. Based on the first voltage adjustment set, the current parameters in the topology structure are optimized using a linear interpolation method to generate a first current adjustment set, thereby obtaining the first current adjustment set. By comparing the first current adjustment set with a preset power adjustment range, an iterative optimization method is used to determine the final power adjustment range required for dynamic matching, thereby determining the final power adjustment range.

[0072] Specifically, in the process of regulating the power fluctuation of photovoltaic power supply, the Perturb and Observe MPPT algorithm is used to apply a periodic disturbance of ±2V to the operating point voltage. The disturbance interval is set to 0.5 seconds. The maximum power point tracking direction is determined by comparing the power change value ΔP before and after the disturbance (accuracy ±0.2W). When it is detected that the power increment ΔP is less than the threshold value of 0.5W for three consecutive times, it is determined to enter a steady state and the disturbance is suspended. For the Boost converter topology, a PWM control strategy with a duty cycle adjustment step of 0.5% is adopted. The switching frequency (10-20kHz) is dynamically adjusted according to the real-time collected input voltage (range 25-50V, sampling rate 1kHz) to stabilize the output voltage at 100±1V. The least squares method is used to fit the latest 30 sets of voltage-power data points to calculate the dynamic internal resistance. (accuracy ±0.05Ω), where ΔU is the voltage change, ΔI is the current change, when the internal resistance change rate exceeds 5% / s, the impedance matching compensation is triggered, and the PI controller (K P =0.8, K I =0.05) to adjust the inductor current to the optimal operating range (5-8A). For the step-shaped power curve caused by shadow occlusion, the differential evolution algorithm is used to optimize the disturbance amplitude. The population size is set to 20, the mutation factor F = 0.6, and the crossover probability CR = 0.9. The global maximum power point is found within 10 iterations. The system calculates the power regulation efficiency in real time. (sampling period 200ms), where ΔP out is the output power change, ΔP in The system automatically switches to variable-step perturbation mode when η remains below 85% for two consecutive minutes, starting with 1% and decreasing to 0.1% according to the Fibonacci sequence. All adjustment parameters and operating status are written to an in-memory database in JSON format and persisted to solid-state storage every five minutes. The data fields include 23 dimensions, including voltage reference value, current ripple coefficient, and adjustment time.

[0073] In step S103, after obtaining the power adjustment range required for dynamic matching, the topology structure is adjusted through a multi-stage DC-DC converter, and voltage stabilization is performed to address the power instability caused by light fluctuations, thereby obtaining a stable power output suitable for the hydrogen production system.

[0074] The sensor acquires the ambient light intensity change data in real time, and uses the fast Fourier transform algorithm to analyze the fluctuation frequency and amplitude of the light intensity change data to determine the power adjustment range. If the fluctuation frequency is higher than the preset threshold, the topology of the multi-stage DC-DC converter is dynamically adjusted through the control algorithm, and the pulse width modulation technology is used to optimize the current control to obtain a stable voltage output. The stable voltage output is received through the load matching module, and the matching degree between the output power and the load power is adjusted according to the real-time load demand of the hydrogen production system to determine the energy conversion efficiency. The energy conversion efficiency data is obtained, and the voltage and current fluctuations are monitored in real time through the feedback control loop. The proportional integral differential algorithm is used to correct the voltage and current fluctuations to obtain a stable power output suitable for the hydrogen production system.

[0075] Specifically, in dynamic power matching, the output characteristic curve of the photovoltaic array is first collected in real time through the maximum power point tracking algorithm (such as the perturbation observation method), and the voltage sampling interval is set to 0.1 seconds. When the light intensity is detected from 1000W / m 2 Sudden drop to 600W / m 2 When the algorithm adjusts the operating point voltage in 0.5V steps, it locks the new maximum power point voltage of 28.7V after 5 iterations. At this time, the power fluctuation range is determined to be 120W-300W. In view of this fluctuation range, a three-stage staggered parallel Buck-Boost topology is used for power regulation. The first stage DC-DC stabilizes the input voltage at 48±2V, and the duty cycle PID control (proportional coefficient K P =0.8, integration time T i =0.01s) to achieve a dynamic response time of less than 20ms; the second stage adopts dual closed-loop control, with the current loop bandwidth set to 1kHz and the voltage loop using feedforward compensation to control the output voltage ripple within 1%; the third stage uses the Kalman filter algorithm to predict the power change trend, updates the PWM modulation signal with a period of 0.1 seconds, and finally outputs a stable voltage of 72V±0.5V to the electrolyzer. At the algorithm level, a sliding window variance analysis (window width 30 seconds, threshold σ 2 ≤5W 2) monitors output power stability in real time. If three consecutive windows exceed the limit, it automatically triggers topology reconfiguration, switching from a two-phase parallel connection to a four-phase parallel connection, restoring system stability within 150ms. To cope with extreme lighting conditions, a power prediction model based on an LSTM neural network (128 hidden nodes, 10-second prediction window) is deployed. This proactively adjusts the DC-DC gain parameters, ensuring that the hydrogen production system maintains power output fluctuations within ±2% even when the irradiance suddenly changes by 30%.

[0076] Step S104 , collecting load demand data from the electrolytic cell in the hydrogen production system, combining it with stable power output, and using the proportional integral control method to fine-tune the power adjustment range to determine the real-time power matching status of the electrolytic cell during operation.

[0077] The load demand data is obtained from the electrolytic cell sensor, and the current and voltage of the electrolytic cell are collected in real time using a high-frequency sampling method. The collected raw data is processed by a data smoothing and filtering algorithm to obtain smoothed load demand data. Based on the smoothed load demand data, the proportional integral control method is used to calculate the power regulation amount, using the formula u(t)=K p e(t)+K i int e(t)d(t) calculates the control output and determines the power adjustment amount, where u(t) is the power adjustment amount, e(t) is the deviation between the load demand data and the current power output, K p is the proportionality coefficient, K i is the integral coefficient. If the power adjustment amount exceeds the preset power range, the power adjustment amount is clipped by a limiter, and the clipped adjustment amount is distributed using a dynamic adjustment algorithm to obtain an adjusted power output value. By comparing the deviation of the adjusted power output value with the load demand data, a preset threshold is used to determine whether the deviation is within the allowable range, thereby obtaining the real-time power matching status of the electrolytic cell.

[0078] Specifically, in the hydrogen production system, the load demand data of the electrolytic cell is collected in real time through sensors. For example, the current sensor measures the current as 200A, and the voltage sensor measures the voltage as 400V, thereby calculating the real-time power demand as 80kW. Combined with the stable power output, the system sets the target power to 85kW and uses the proportional-integral control method for fine-tuning. The proportional coefficient K p Set to 0.5, the integral coefficient K i Set to 0.1, through the algorithm P out =K p *(P target -P actual )+K i *∫(P target -P actual )dt, where Pout is the output power, P target is the target power, P actual The system calculates the actual power to be 2.5kW, adjusting the electrolytic cell's power from 80kW to 82.5kW. The system continuously monitors power matching. When the actual power deviates from the target power by more than ±1kW, it recalculates the adjusted power to ensure optimal electrolytic cell operation. This allows the system to dynamically adjust power output, improve hydrogen production efficiency, and reduce energy consumption.

[0079] In step S105, if the real-time power matching state exceeds the preset load threshold, the gas pressure data is collected through the pressure sensor in the buffer gas supply system, and the gas supply flow is adjusted using the PID control algorithm to determine the gas supply parameters under stable gas supply conditions.

[0080] If the real-time power exceeds the load threshold, the state monitoring module compares the real-time power with the load threshold to generate a trigger signal. Based on the trigger signal, gas pressure data is obtained from the pressure sensor in the buffer gas supply system to determine the pressure state of the gas supply system. A PID control algorithm is used to calculate the adjustment amount of the supply gas flow rate based on the error between the gas pressure data and the target pressure value, and the flow control instruction is obtained. The valve opening of the buffer gas supply system is adjusted using the flow control instruction to ensure that the gas supply parameters meet the preset stability conditions.

[0081] Specifically, when the real-time power matching state exceeds the preset load threshold (for example, a sudden increase in load causes the power demand to exceed 10% of the rated value), the system triggers the buffer gas supply mechanism, and collects gas pressure data in real time through the pressure sensor installed on the gas supply pipeline (range 0-1MPa, accuracy ±0.5% FS). The sampling frequency is set to 100Hz to ensure dynamic response. The collected pressure signal is processed by Kalman filtering and compared with the set pressure value (such as 0.6MPa). If the deviation exceeds the allowable range (±0.05MPa), the PID control algorithm is started for adjustment. The PID controller uses an incremental algorithm, and the parameter is set to the proportional coefficient K p =2.5, integration time T i =0.1s, differential time T d =0.02s, through the discretization formula Calculate the adjustment amount, Δu(k) is the gas pressure adjustment amount, e(k) is the instantaneous pressure deviation, and output 4-20mA signal to drive the electric control valve opening. During the adjustment process, the pressure change rate is monitored in real time (such as the target is 0.01MPa / s). When the pressure fluctuation standard deviation is less than 0.01MPa and lasts for 5 seconds, it is determined to be stable. At this time, the valve opening (such as 45%), gas flow rate (120m 3Parameters such as rpm ( / h) are used as baseline values ​​under stable operating conditions, and the system parameter database is updated for subsequent optimization. The entire control cycle is completed within 200ms, ensuring a balance between dynamic response and steady-state accuracy.

[0082] Step S106 , based on the gas supply parameters under stable gas supply conditions, extract pressure fluctuation and flow balance indicators from the electrolytic cell operation data, judge the operation stability of the hydrogen production system through a preset stability threshold, and obtain the triggering condition for fault switching.

[0083] Raw data on pressure fluctuation and flow balance are obtained from electrolytic cell operating data. This raw data is preprocessed using a time series analysis tool to obtain pressure fluctuation values ​​and flow balance indicators. Based on preset stability thresholds, a comparison algorithm is used to determine whether these pressure fluctuation values ​​and flow balance indicators exceed the threshold range. If so, system operation is determined to be unstable. A logistic regression algorithm is used to classify the unstable data, identifying the electrolytic cell parameter characteristics that cause instability and determining potential trigger conditions for fault switching. Based on these potential trigger conditions, a rules engine is used to generate fault switching instructions based on gas supply conditions and the electrolytic cell parameter characteristics to determine the final trigger condition.

[0084] Specifically, under stable gas supply conditions, real-time data from the electrolytic cell operation is collected to extract pressure fluctuation and flow balance indicators. First, the pressure fluctuation indicator is evaluated by calculating the standard deviation of the pressure values ​​per minute.

[0085] For example, when the pressure value fluctuates between 1.0MPa and 1.2MPa, the standard deviation is 0.05MPa. If the preset stability threshold is 0.1MPa, the current pressure fluctuation is within the acceptable range. Secondly, the flow balance index is evaluated by calculating the coefficient of variation of the hourly flow value.

[0086] For example, when the flow rate is 100m 3 / h to 120m 3 When the pressure fluctuation standard deviation exceeds 0.1 MPa or the flow rate coefficient of variation exceeds 0.1, the system triggers a fault switching mechanism.

[0087] For example, when the standard deviation of pressure fluctuations reaches 0.12 MPa, the system automatically switches to a backup gas source to ensure the continuity and stability of the hydrogen production process. Through the above-mentioned algorithm and analysis process, the system can monitor and adjust the operating status in real time, ensuring the efficient operation of the hydrogen production system under stable gas supply conditions.

[0088] In step S107, if the triggering condition for fault switching is activated, the system switches to the backup power supply mode through the preset redundant topology structure, adjusts the gas supply rhythm in combination with the multi-stage gas storage unit of the buffer gas supply system, and determines the continuous operation parameters of the hydrogen production system under abnormal working conditions.

[0089] If the trigger condition is activated, the system switches to the backup power supply mode through the preset redundant topology structure, obtains a stable power supply from the backup power supply, and determines the power input parameters of the hydrogen production system under abnormal operating conditions. According to the gas storage capacity of the multi-stage gas storage unit, the gas supply rhythm is adjusted through the buffer gas supply system to obtain a stable gas flow rate and determine the raw material input parameters of the hydrogen production system. If the hydrogen production system detects an abnormal operating condition, the logic controller combines the power input parameters and the raw material input parameters to adjust the operating parameters and obtain a control instruction for continuous operation. The execution unit of the hydrogen production system is driven by the control instruction, and the reaction rate is adjusted using the preset operating parameters to obtain a stable hydrogen production output.

[0090] Specifically, when the triggering conditions for fault switching are activated, the system will automatically switch to the backup power mode through the preset redundant topology to ensure the continuous operation of the hydrogen production system.

[0091] For example, when the main power supply voltage is lower than 220V, the system will immediately start the backup power supply, and the output voltage of the backup power supply will be stabilized at 230V to ensure the normal operation of the equipment. During the switching process, the system will adjust the gas supply rhythm through the multi-stage gas storage unit of the buffer gas supply system to ensure the stability of the gas supply. Specifically, the system will dynamically adjust the valve opening of the gas storage unit through the PID control algorithm according to the current gas pressure value to maintain the gas pressure between 0.8MPa and 1.2MPa. At the same time, the system will monitor the operating parameters of the hydrogen production system in real time, such as temperature, pressure and flow, and optimize the operating parameters through the fuzzy control algorithm to ensure the stable operation of the system under abnormal working conditions.

[0092] For example, if the system detects a temperature exceeding 80°C, it automatically activates the cooling system to control the temperature between 70°C and 75°C. Furthermore, the system uses data analysis and machine learning algorithms to predict potential failures and take preventive measures in advance, further improving system reliability and stability.

[0093] Step S108, after obtaining the continuous operation parameters, the dynamic matching process is optimized by using an adaptive filtering algorithm through a joint analysis of the real-time power data and the gas supply parameters to determine the long-term stability of the system under light fluctuations.

[0094] Continuous operating parameters and real-time power data are acquired from the operating equipment, and gas supply parameters are collected at the same time. High-frequency sampling is performed using a data acquisition frequency control module to obtain the original data set. Based on the original data set, feature extraction is performed on the real-time power data and gas supply parameters using a joint analysis module. If the feature value exceeds a preset threshold range, standardization is performed to determine the analyzed feature set. An adaptive filtering algorithm is used to optimize the analyzed feature set, and the data is smoothed using a Kalman filter to obtain optimized matching parameters. Under light fluctuation conditions, the optimized matching parameters are evaluated for long-term stability using a stability judgment module. If the parameter fluctuation amplitude is lower than a preset threshold, the hydrogen production system is judged to have long-term stability.

[0095] Specifically, when obtaining continuous operating parameters, the system collects the output power data of the photovoltaic array (such as the current power value is 1250W) and the gas supply parameters of the hydrogen energy system (such as the hydrogen flow rate is 2.3L / min) in real time at a sampling interval of 1 second. By establishing a power-flow joint analysis model and using the least squares method to fit the dynamic relationship between the two, the transfer function of power fluctuation and gas demand is obtained. Based on this, an adaptive Kalman filter is designed, whose state equation is set as x(k+1)=0.92x(k)+0.08u(k)+w(k), and observation equation z(k)=1.05x(k)+v(k), where x(k) is the system state at time k, u(k) is the external input, and the covariance matrices of process noise w(k) and observation noise v(k) are initialized to Q=0.01 and R=0.1 respectively. 2 Sudden drop to 600W / m 2 When the filter converges the estimated error to within ±2% within 3 seconds, the system continuously records the power standard deviation σ over 72 hours to assess long-term stability. When σ exceeds the threshold of 15W, the parameter update mechanism is triggered, automatically adjusting the filter gain coefficient K from 0.78 to 0.82. A sliding window algorithm (with a window width set to 30 minutes) is also used to calculate the power fluctuation rate η. When η exceeds 0.25 five times in a row, the system is deemed unstable. At this point, the gas supply system is activated to increase the buffer tank pressure from 0.8MPa to 1.2MPa to maintain stable output.

[0096] Step S109 , extracting abnormal features of power instability and gas supply fluctuation from the long-term stability judgment result, updating the control parameters of the MPPT algorithm through historical operation data, and determining the optimized topology structure and gas supply support plan.

[0097] The operating data is obtained from the historical records. The timing analysis tool is used to extract abnormal features for power fluctuations and air supply stability to obtain an abnormal feature data set. Based on the abnormal feature data set, the control parameters are adjusted using the data update tool. If the voltage change exceeds the preset threshold, the parameters of the MPPT algorithm are updated through iterative calculation to determine the optimized control parameter set. Based on the optimized control parameter set, the topology structure is evaluated using the topology analysis tool. The connection method is adjusted for the uneven distribution of airflow to obtain the optimized topology structure. Based on the optimized topology structure, the fluid mechanics simulation tool is used to design the air supply plan. If the air supply fluctuation is less than the preset threshold, the support plan is generated through parameter mapping to determine the final air supply support plan.

[0098] Specifically, in the long-term stability judgment results, by analyzing historical operating data, abnormal characteristics of power instability and gas supply fluctuations are extracted.

[0099] For example, if system power fluctuates by more than ±5% within 10 minutes, it is considered power instability; if gas supply pressure fluctuates by more than ±0.2 MPa within 5 minutes, it is considered gas supply fluctuation. These abnormal characteristics are classified and labeled using data mining algorithms (such as K-means clustering) to form an abnormal feature library. Based on these characteristics, the control parameters of the MPPT algorithm are updated and optimized using the adaptive particle swarm optimization (APSO) algorithm, reducing the MPPT algorithm's response time from 2 seconds to 1 second, improving the system's dynamic response capability. Furthermore, through topology optimization using a genetic algorithm (GA), the system's efficiency was increased from 92% to 95%. In the gas supply support solution, a fuzzy control algorithm is introduced to dynamically adjust the gas supply valve opening based on real-time gas supply pressure and flow data, ensuring that the gas supply pressure remains stable within the range of 0.5 MPa ± 0.05 MPa. Through these methods, the system demonstrates greater stability and efficiency in long-term operation, meeting operational requirements under complex operating conditions.

[0100] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the concept of this application. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A dynamic power matching method for a photovoltaic hydrogen production system, characterized in that: The method comprises: The photovoltaic power supply supplies power to the electrolytic cell, and the ambient light fluctuation parameters are obtained through sensors. The changes in the photovoltaic power supply output power caused by the ambient light fluctuation are then stabilized to obtain a stable power output suitable for the hydrogen production system. Collect load demand data from the electrolytic cell in the hydrogen production system, combine it with stable power output, and use the proportional-integral control method to fine-tune the power adjustment range to determine the real-time power matching status of the electrolytic cell during operation; If the real-time power matching status exceeds the preset load threshold, the gas pressure data is collected through the pressure sensor in the buffer gas supply system, and the gas supply flow is adjusted using the PID control algorithm to determine the gas supply parameters under stable gas supply conditions; Based on the gas supply parameters under stable gas supply conditions, the pressure fluctuation and flow balance indicators are extracted from the electrolytic cell operation data. The operational stability of the hydrogen production system is determined by the preset stability threshold, and the trigger condition for fault switching is obtained. If the trigger condition for failover is activated, the system switches to backup power mode through the preset redundant topology, adjusts the gas supply rhythm in combination with the multi-stage gas storage unit of the buffer gas supply system, and determines the continuous operation parameters of the hydrogen production system under abnormal working conditions; After obtaining the continuous operation parameters, the dynamic matching process is optimized by combining real-time power data with gas supply parameters using an adaptive filtering algorithm to determine the long-term stability of the system under light fluctuations. The abnormal characteristics of power instability and gas supply fluctuations are extracted from the long-term stability judgment results, the control parameters of the MPPT algorithm are updated through historical operation data, and the optimized topology structure and gas supply support plan are determined.

2. The method according to claim 1, characterized in that The voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes: The real-time light intensity and output power of the photovoltaic power source are collected by a light sensor and a power sensor, and stored as a first data set; Extracting the fluctuation frequency and amplitude of the light intensity and output power using a time series analysis method based on the first data set to generate a first fluctuation parameter set; determining the illumination fluctuation amplitude, and if the fluctuation amplitude in the first fluctuation parameter set exceeds a preset power threshold, generating a power instability state flag through a logic judgment method to obtain the power instability state flag; According to the power instability state identifier, a K-means algorithm is used to perform cluster analysis on the fluctuation frequency and amplitude in the first fluctuation parameter set to generate a first initial fluctuation feature set, thereby obtaining initial fluctuation features under the power instability state.

3. The method according to claim 2, characterized in that The voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes: According to the initial fluctuation feature set in the power unstable state, the output power of the photovoltaic power source is tracked in real time using the Perturb and Observe algorithm to generate a first power tracking data set, thereby obtaining the first power tracking data set; If the output power fluctuation amplitude in the first power tracking data set exceeds a preset threshold, adjusting the voltage parameters in the topology structure through a logic judgment method to generate a first voltage adjustment set; According to the first voltage adjustment set, a linear interpolation method is used to optimize the current parameters in the topology structure to generate a first current adjustment set; By comparing the first current adjustment set with a preset power adjustment range, an iterative optimization method is used to determine a final power adjustment range required for dynamic matching, and the final power adjustment range is determined.

4. The method according to claim 3, characterized in that The voltage stabilization process for the photovoltaic power supply output power variation caused by ambient light fluctuations includes: After obtaining the power adjustment range required for dynamic matching, the sensor is used to obtain the ambient light intensity change data in real time, and the fast Fourier transform algorithm is used to analyze the fluctuation frequency and amplitude of the light intensity change data to determine the power adjustment range; If the fluctuation frequency is higher than a preset threshold, the topology of the multi-stage DC-DC converter is dynamically adjusted through a control algorithm, and pulse width modulation technology is used to optimize current control to obtain a stable voltage output; The stable voltage output is received by the load matching module, and the matching degree between the output power and the load power is adjusted according to the real-time load demand of the hydrogen production system to determine the energy conversion efficiency; The energy conversion efficiency data is obtained, voltage and current fluctuations are monitored in real time through a feedback control loop, and the voltage and current fluctuations are corrected using a proportional integral differential algorithm to obtain a stable power output suitable for a hydrogen production system.

5. The method according to claim 1, wherein The method collects load demand data from the electrolytic cell in the hydrogen production system, combines it with stable power output, and uses a proportional-integral control method to fine-tune the power adjustment range to determine the real-time power matching status of the electrolytic cell during operation, including: Obtaining load demand data from an electrolytic cell sensor, using a high-frequency sampling method to collect the current and voltage of the electrolytic cell in real time, and processing the collected raw data through a data smoothing and filtering algorithm to obtain smoothed load demand data; According to the smoothed load demand data, the proportional integral control method is used to calculate the power regulation amount, using the formula u(t)=K p e(t)+K i int e(t)d(t) calculates the control output and determines the power adjustment amount; Where u(t) is the power regulation amount, e(t) is the deviation between the load demand data and the current power output, K p is the proportionality coefficient, K i is the integration coefficient; If the power adjustment amount exceeds the preset power range, the power adjustment amount is clipped by a limiter, and the clipped adjustment amount is distributed using a dynamic adjustment algorithm to obtain an adjusted power output value; By comparing the deviation between the adjusted power output value and the load demand data, a preset threshold is used to determine whether the deviation is within an allowable range, thereby obtaining the real-time power matching status of the electrolytic cell.

6. The method according to claim 1, characterized in that If the real-time power matching state exceeds the preset load threshold, the gas pressure data is collected through the pressure sensor in the buffer gas supply system, and the gas supply flow is adjusted using the PID control algorithm to determine the gas supply parameters under the gas supply stable condition, including: If the real-time power exceeds the load threshold, the state monitoring module compares the real-time power with the load threshold to obtain a trigger signal; acquiring gas pressure data from a pressure sensor in a buffer gas supply system according to the trigger signal to determine a pressure state of the gas supply system; Using a PID control algorithm, based on the error between the gas pressure data and the target pressure value, the adjustment amount of the gas supply flow rate is calculated to obtain the flow control instruction; The valve opening of the buffer gas supply system is adjusted through the flow control instruction to determine whether the gas supply parameters meet the preset stability conditions.

7. The method according to claim 1, characterized in that The method extracts pressure fluctuation and flow balance indicators from electrolytic cell operation data based on gas supply parameters under stable gas supply conditions, determines the operational stability of the hydrogen production system through a preset stability threshold, and obtains the triggering conditions for fault switching, including: Obtaining raw data of pressure fluctuation and flow balance from electrolytic cell operation data, and preprocessing the raw data using a time series analysis tool to obtain pressure fluctuation values ​​and flow balance indicators; According to a preset stability threshold, a comparison algorithm is used to determine whether the pressure fluctuation value and the flow balance index exceed the threshold range. If so, it is determined that the system operation is unstable; Classifying the unstable operation data by a logistic regression algorithm, obtaining the electrolytic cell parameter characteristics that cause the instability, and obtaining the potential triggering conditions for fault switching; In response to the potential triggering conditions, a rule engine is used to generate a fault switching instruction according to the gas supply conditions and the electrolytic cell parameter characteristics to determine the final triggering conditions.

8. The method according to claim 1, characterized in that If the trigger condition for the fault switch is activated, the system switches to the backup power supply mode through the preset redundant topology structure, adjusts the gas supply rhythm in combination with the multi-stage gas storage unit of the buffer gas supply system, and determines the continuous operation parameters of the hydrogen production system under abnormal working conditions, including: If the trigger condition is activated, the system switches to a backup power supply mode through a preset redundant topology structure, obtains a stable power supply from the backup power supply, and determines the power input parameters of the hydrogen production system under abnormal operating conditions; According to the gas storage capacity of the multi-stage gas storage unit, the gas supply rhythm is adjusted through the buffer gas supply system to obtain a stable gas flow rate and determine the raw material input parameters of the hydrogen production system; If the hydrogen production system detects an abnormal operating condition, the logic controller adjusts the operating parameters by combining the power input parameters and the raw material input parameters to obtain a control instruction for continuous operation; The control instructions are used to drive the execution unit of the hydrogen production system, and the preset operating parameters are used to adjust the reaction rate to obtain a stable hydrogen production output.

9. The method according to claim 1, characterized in that After obtaining the continuous operation parameters, the dynamic matching process is optimized by combining real-time power data with gas supply parameters using an adaptive filtering algorithm to determine the long-term stability of the system under light fluctuations, including: Acquire continuous operating parameters and real-time power data from the operating equipment, and simultaneously acquire gas supply parameters, and perform high-frequency sampling using a data acquisition frequency control module to obtain the raw data set; Based on the original data set, feature extraction is performed on the real-time power data and gas supply parameters through a joint analysis module. If the feature value exceeds a preset threshold range, normalization processing is performed to determine the feature set after analysis; Adopting an adaptive filtering algorithm to optimize the analyzed feature set, and performing smoothing processing on the data through a Kalman filter to obtain the optimized matching parameters; Under the condition of light fluctuation, the long-term stability of the optimized matching parameters is evaluated by the stability judgment module. If the fluctuation amplitude of the parameters is lower than a preset threshold, it is judged that the system has long-term stability.

10. The method according to claim 1, characterized in that The method extracts abnormal features of power instability and gas supply fluctuation from the long-term stability judgment results, updates the control parameters of the MPPT algorithm based on historical operation data, and determines the optimized topology and gas supply support solution, including: Acquire operation data from historical records, and use a time series analysis tool to extract abnormal features for power fluctuation and gas supply stability to obtain the abnormal feature data set; According to the abnormal characteristic data set, a data update tool is used to adjust the control parameters. If the voltage change exceeds a preset threshold, the parameters of the MPPT algorithm are updated through iterative calculation to determine the optimized control parameter set; Using the optimized control parameter set, a topology analysis tool is used to evaluate the topology structure, and the connection mode is adjusted according to the uneven airflow distribution to obtain the optimized topology structure; According to the optimized topology, a gas supply scheme is designed using a fluid mechanics simulation tool. If the gas supply fluctuation is less than a preset threshold, a support scheme is generated through parameter mapping to determine the final gas supply support scheme.

Images