A test method and equipment for an ALK-PEM hybrid hydrogen production system

Abstract

This invention relates to a testing method and equipment for an ALK-PEM hybrid hydrogen production system, belonging to the field of electrolyzer testing technology. The method includes: determining multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve representing a wind and solar fluctuation pattern; generating multiple types of simulated fluctuation power data based on the multiple types of typical power waveform curves; adjusting the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system according to the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; acquiring operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; and generating a test report for the ALK-PEM hybrid hydrogen production system based on the operating data.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of electrolyzer testing technology, specifically to a testing method and equipment for an ALK-PEM hybrid hydrogen production system. Background Technology

[0002] Against the backdrop of the large-scale development of the hydrogen energy industry, the ALK-PEM hybrid hydrogen production system, as the core equipment for green hydrogen production, directly determines the overall efficiency and safety of the hydrogen production system through its performance parameters, operational stability, and adaptability to operating conditions. The ALK-PEM hybrid hydrogen production system combines the high-power stable operation of an alkaline electrolyzer (ALK) with the wide-range power fluctuation adaptability of a proton exchange membrane electrolyzer (PEM). Its structure and operating logic are more complex than those of a single electrolyzer, significantly increasing the requirements for the professionalism, comprehensiveness, and accuracy of testing methods.

[0003] However, existing testing methods are significantly out of sync with actual operating conditions and cannot effectively adapt to the multi-unit integrated coupling characteristics of the alkaline unit and proton exchange membrane unit in the integrated ALK-PEM hybrid hydrogen production system, resulting in insufficient testing accuracy. Summary of the Invention

[0004] The purpose of the embodiments in this specification is to provide a testing method and equipment for an ALK-PEM hybrid hydrogen production system, so as to overcome the problem of insufficient testing accuracy in existing methods.

[0005] To address the aforementioned technical problems, this specification provides, in one aspect, a testing method for an ALK-PEM hybrid hydrogen production system, comprising: Based on wind and solar sample data, typical curves of multiple power waveforms are determined; each type of typical power waveform curve represents a wind and solar fluctuation pattern. Based on the typical curves of the various power waveforms, generate various types of simulated fluctuating power data; Based on the various types of simulated fluctuation power data, the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system is adjusted to perform a test on the ALK-PEM hybrid hydrogen production system. Obtain operational data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; Based on the operational data, a test report for the ALK-PEM hybrid hydrogen production system is generated.

[0006] Furthermore, the determination of typical power waveform curves for various types based on wind and solar sample data includes: The wind and solar sample data are sampled according to multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale; the multiple time scales include a first scale, a second scale and a third scale, and the wind and solar sample waveforms at the first scale, the second scale and the third scale are used to characterize the instantaneous change characteristics, short-term ramp characteristics and long-term trend characteristics of wind and solar power, respectively. Shape similarity clustering is performed on multiple wind and light waveforms at each time scale; Based on the clustering results, the shape features of various types of wind and light sampling waveforms at each time scale are extracted; Based on the shape features, typical curves of multiple power waveforms at each time scale are generated.

[0007] Furthermore, the step of performing shape similarity clustering on multiple wind and light waveforms at each time scale includes: Each wind and solar sampling waveform is divided into multiple waveform segments; the multiple waveform segments include a rising segment, a peak hold segment, and a falling segment. The rising segment is used to characterize the power ramping capability of the PEM unit and the ALK unit, the peak hold segment is used to characterize the overload sustaining capability of the PEM unit and the ALK-PEM hybrid hydrogen production system, and the falling segment is used to characterize the power recovery capability of the PEM unit and the ALK unit. The weights of each waveform segment are determined based on the sensitivity and tolerance of the PEM and ALK units to each waveform segment. Based on the weights, the morphological feature parameters of each waveform segment are fused to obtain the waveform feature vector of each wind and light sampling waveform; Cluster the waveform feature vectors at each time scale.

[0008] Furthermore, the method also includes: Frequency distribution curves are constructed based on the frequency of occurrence of various wind and light sampling waveforms at each time scale; Identify the long tail portion in the frequency distribution curve; the long tail portion is used to characterize the waveform category where the frequency occurs below a preset frequency threshold. Extract rare waveforms in the long tail portion whose waveform fluctuation amplitude exceeds a preset amplitude threshold; Calculate the clustering effectiveness index corresponding to the number of candidate clusters at each time scale; Using the clustering effectiveness index as the optimization objective and preserving the categories of the scarce waveforms as the constraint, the minimum number of clusters that maximizes the clustering effectiveness index and satisfies the constraint is taken as the optimal number of clusters at this time scale. The clustering of waveform feature vectors at each time scale includes: Cluster the waveform feature vectors according to the optimal number of clusters at each time scale.

[0009] Furthermore, the step of generating multiple types of simulated fluctuating power data based on the typical curves of the multiple types of power waveforms includes: Obtain multiple preset wind and solar power output test scenario requirements; For each wind and solar power output test scenario, one or more typical power waveform curves are selected from the multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario. The simulated fluctuation test sequence is interpolated to obtain the simulated fluctuation power data corresponding to the requirements of the wind and solar power output test scenario.

[0010] Furthermore, for each wind and solar power output test scenario, one or more typical power waveform curves are selected from the multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario requirement, including: Based on the requirements of each wind and solar power output test scenario, one type of typical power waveform curves from multiple types of power waveforms at the third scale is selected as the basic trend curve. Calculate the fluctuation energy coupling coefficient between the basic trend curve and the typical power waveform curves of various types at the first and second scales; From the typical power waveform curves at the first and second scales, the class with the highest wave energy coupling coefficient is selected as the superimposed wave curve. Based on the wave energy coupling coefficient, the phase offset and amplitude modulation ratio of the superimposed wave curve relative to the basic trend curve are determined. The superimposed wave curve is superimposed on the basic trend curve according to the phase offset and amplitude modulation ratio to obtain the simulated wave test sequence.

[0011] Furthermore, the step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuating power data, the first operating power of the PEM unit and the ALK unit under multiple preset power levels of the ALK-PEM hybrid hydrogen production system is determined. The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire the first operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system at each preset power level; the first operating data includes at least current, voltage, temperature, pressure, hydrogen production efficiency, gas purity, and water replenishment rate.

[0012] Furthermore, the step of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Based on the current and voltage at each preset power level, the current-voltage characteristic curves are fitted to obtain the electrochemical efficiency of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system. Based on the temperature at each power level, calculate the temperature rise rate and temperature deviation to determine the thermal management efficiency of the ALK-PEM hybrid hydrogen production system; The water consumption for hydrogen production is calculated based on the water replenishment rate and gas purity at each power level in order to determine the water balance efficiency of the ALK-PEM hybrid hydrogen production system. The electrochemical efficiency, thermal management efficiency, and water balance efficiency are used to generate a static performance test report that includes the dominant factors affecting hydrogen production efficiency deviation.

[0013] Furthermore, determining the thermal management efficiency of the ALK-PEM hybrid hydrogen production system includes: The thermal management efficiency of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ; In the formula, Thermal management efficiency of the ALK-PEM hybrid hydrogen production system; and These are the temperature rise rates for the PEM and ALK units, respectively. and These are the actual operating temperatures of the PEM and ALK units, respectively. and These are the theoretical optimal operating temperatures for the PEM and ALK cells, respectively. The operating temperature range of the ALK-PEM hybrid hydrogen production system; , and This is a preset constant.

[0014] Furthermore, the step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the second operating power of the PEM unit and ALK unit under multiple preset wind and solar fluctuation scenarios is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire second operating data for the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system under each preset wind and solar fluctuation scenario; the second operating data includes at least the power response speed, hydrogen production efficiency change rate, and gas production purity stability.

[0015] Furthermore, the step of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined based on the rate of change of hydrogen production efficiency and the power response speed under various preset wind and solar fluctuation scenarios. Time-frequency analysis was performed on the gas production purity under various preset wind and solar fluctuation scenarios to obtain the dominant frequency components of gas production purity fluctuation. Based on the dominant frequency components of the hydrogen production efficiency stability and gas purity fluctuations, a dynamic operating condition test report is generated, including the dominant factors of hydrogen production efficiency deviation.

[0016] Furthermore, determining the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios includes: Based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios, the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ; In the formula, To improve the hydrogen production efficiency and stability of the ALK-PEM hybrid hydrogen production system; The actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at time t; The theoretically optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system; Let be the simulated fluctuation power at time t; T be the total test duration. and These are the power response speed and reference power response speed of the ALK-PEM hybrid hydrogen production system, respectively. and These are the rate of change in hydrogen production efficiency and the rate of change in input power, respectively. , and This is a preset constant.

[0017] Furthermore, the step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the third operating power of the PEM unit and the ALK unit under multiple preset power allocation ratios is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire third operating data for the PEM unit and ALK unit under each preset power allocation ratio; the third operating data includes at least the coupling response speed, power allocation accuracy, and hydrogen production efficiency.

[0018] Furthermore, the step of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Correlation analysis was performed on the coupling response speeds of PEM and ALK units under various preset power allocation ratios to obtain the rate of change matching coefficients during the power handover process between PEM and ALK units. Based on the hydrogen production efficiency under each preset power allocation ratio, calculate the cooperative efficiency loss coefficient between the PEM unit and the ALK unit. Based on the power allocation accuracy, rate of change matching coefficient, and collaborative efficiency damage coefficient, a collaborative operation test report is generated, which includes the dominant factors of hydrogen production efficiency deviation.

[0019] Furthermore, the correlation analysis of the coupling response speeds of the PEM unit and the ALK unit under each preset power allocation ratio, to obtain the rate-of-change matching coefficient during the power handover process between the PEM unit and the ALK unit, includes: The following formula was used to perform a correlation analysis on the coupling response speeds of the PEM and ALK cells under various preset power allocation ratios, and the rate of change matching coefficients during the power handover process between the PEM and ALK cells were obtained: ; In the formula, The rate of change matching coefficient; and , , are the power change rates of the PEM cell and the ALK cell at time t, respectively; T is the duration of the power transfer process; The response lag time of the PEM and ALK elements. and These are the coupling response speeds of the PEM unit and the ALK unit, respectively. For reference response lag time.

[0020] Furthermore, the step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the fourth operating power of the PEM unit and ALK unit under multiple preset start-stop conditions is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire the fourth operating data of the PEM unit and ALK unit under each preset start-stop condition; the fourth operating data includes at least the start-stop response time, energy consumption during start-stop process, electrode temperature change rate, and sealing performance stability.

[0021] Furthermore, the step of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Statistical analysis was performed on the start-stop response times of the PEM unit and the ALK unit under various preset start-stop conditions to obtain the consistency of the start-stop response of the PEM unit and the ALK unit. Peak values ​​of electrode temperature change rates under various preset start-stop conditions were extracted to obtain the thermal stress influence of PEM and ALK units. The sealing performance stability under various preset start-stop conditions is trend-fitted to obtain the sealing reliability of the PEM unit and the ALK unit. Statistical analysis was performed on the energy consumption during the start-up and shutdown process under various preset start-up and shutdown conditions to identify the start-up and shutdown stages where abnormal impact events occurred. Based on the results of start-stop response consistency, thermal stress impact, sealing reliability, and abnormal event identification, a start-stop characteristic test report is generated, including the dominant factors of hydrogen production efficiency deviation.

[0022] Furthermore, the statistical analysis of the start-stop response times of the PEM unit and the ALK unit under various preset start-stop conditions to obtain the consistency of the start-stop responses of the PEM unit and the ALK unit includes: The start-stop response times of the PEM and ALK units under various preset start-stop conditions are statistically analyzed using the following formula to obtain the consistency of the start-stop responses of the PEM and ALK units: ; In the formula, To ensure consistent start / stop responses; and , where are the start / stop response times of the PEM unit and the ALK unit respectively in the i-th start / stop operation; N is the total number of start / stop cycles; and These are the average start / stop response times for the PEM unit and the ALK unit, respectively. and These are the standard deviations of the start / stop response times for the PEM and ALK units, respectively. and Degradation trend coefficient and allowable degradation limit for start and stop response times respectively; , and This is a preset constant.

[0023] Furthermore, embodiments of this specification provide a testing apparatus for an ALK-PEM hybrid hydrogen production system, comprising: The determination module is used to determine typical curves of multiple types of power waveforms based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern. The first generation module is used to generate multiple types of simulated fluctuating power data based on the typical curves of the multiple types of power waveforms. The testing module is used to adjust the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system according to the multi-type simulated fluctuation power data, so as to perform testing on the ALK-PEM hybrid hydrogen production system. The acquisition module is used to acquire operational data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; The second generation module is used to generate a test report for the ALK-PEM hybrid hydrogen production system based on the operating data.

[0024] Furthermore, embodiments of this specification provide a computer device, including: Memory, used to store computer programs; A processor for executing the computer program to implement the test method for the ALK-PEM hybrid hydrogen production system described above.

[0025] Furthermore, embodiments of this specification provide a computer storage medium storing computer program instructions, which, when executed, implement the test method for the ALK-PEM hybrid hydrogen production system described above.

[0026] Furthermore, embodiments of this specification provide a computer program product, including a computer program that, when executed by a processor, implements the aforementioned test method for the ALK-PEM hybrid hydrogen production system.

[0027] As can be seen from the technical solutions provided in the embodiments of this specification above, these embodiments can determine multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern; multiple types of simulated fluctuation power data are generated based on the multiple types of typical power waveform curves; the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system is adjusted based on the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process are obtained; and a test report of the ALK-PEM hybrid hydrogen production system is generated based on the operating data. By extracting typical fluctuation patterns from real wind and solar data, the test conditions closely match actual operating scenarios; by generating multi-scale simulated fluctuation data, the dynamic response test of the electrolyzer under complex operating conditions can be comprehensively covered; the differentiated control of the PEM and ALK units accurately evaluates the collaborative operating performance of the ALK-PEM hybrid hydrogen production system; and the final output test report provides an intuitive basis for equipment optimization and fault diagnosis. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below.

[0029] Figure 1 This is a flowchart of a test method for an ALK-PEM hybrid hydrogen production system provided in the embodiments of this specification; Figure 2 This is a schematic diagram of the structural composition of a test device for an ALK-PEM hybrid hydrogen production system provided in the embodiments of this specification; Figure 3 This is a schematic diagram of the structural composition of a computer device provided in the embodiments of this specification. Detailed Implementation

[0030] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0031] It should be noted that the terms "first," "second," etc., used in this specification, claims, and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0032] In some embodiments, the ALK-PEM hybrid hydrogen production system may include an integrated PEM unit and an ALK unit.

[0033] The ALK-PEM hybrid hydrogen production system adopts an integrated structural design, which integrates a proton exchange membrane electrolyzer unit and an alkaline electrolyzer unit. The proton exchange membrane electrolyzer unit is abbreviated as PEM unit, and the alkaline electrolyzer unit as ALK unit. The PEM unit uses a proton exchange membrane as the electrolyte, featuring fast response speed, high current density, and high gas production pressure, making it suitable for handling the high-frequency components of power fluctuations. The ALK unit uses an alkaline solution as the electrolyte, featuring large single-cell capacity and relatively low cost, making it suitable for handling the low-frequency components and base loads of power fluctuations. The PEM and ALK units are integrated into an integrated ALK-PEM hybrid hydrogen production system through electrical connections and fluid pipelines, sharing some auxiliary systems, including a pure water system, a gas-liquid separation system, and a purification system. In the integrated ALK-PEM hybrid hydrogen production system, the PEM and ALK units can operate collaboratively, dynamically adjusting their respective power sharing ratios according to changes in power input, achieving overall efficiency optimization and improved operational stability.

[0034] This specification provides a test method for an ALK-PEM hybrid hydrogen production system, referring to... Figure 1 As shown, real-world wind and solar sample data can be used as the source. Through multi-timescale analysis and clustering, typical power waveform curves representing different fluctuation modes can be extracted, and simulated fluctuating power data that closely reflects actual operating conditions can be generated. This data drives the ALK-PEM hybrid hydrogen production system to sequentially perform comprehensive tests on static performance, dynamic operating conditions, coordinated operation, and start-stop characteristics, while simultaneously collecting multi-dimensional operating parameters of the PEM unit, ALK unit, and the system. Through decoupling analysis, stratified efficiency indicators such as electrochemistry, thermal management, and water balance are calculated. Combined with dynamic response and coordinated matching coefficients, a standardized test report containing performance bottleneck identification and optimization directions is generated.

[0035] In practice, the following steps may be included: S101: Based on wind and solar sample data, determine typical curves for multiple types of power waveforms; each type of typical power waveform curve represents a wind and solar fluctuation pattern.

[0036] In some embodiments, step S101 may specifically include: sampling the wind and solar sample data at multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale; the multiple time scales include a first scale, a second scale, and a third scale, and the wind and solar sample waveforms at the first scale, the second scale, and the third scale are used to characterize the instantaneous change characteristics, short-term ramp-up characteristics, and long-term trend characteristics of wind and solar power, respectively; performing shape similarity clustering on the multiple wind and solar waveforms at each time scale; extracting the shape features of various types of wind and solar sample waveforms at each time scale based on the clustering results; and generating typical curves of multiple types of power waveforms at each time scale based on the shape features.

[0037] Wind and solar sample data consists of historical output power time series collected from wind power or photovoltaic power plants. This series includes power values ​​at continuous time points, reflecting the fluctuation characteristics of actual wind and solar resources. The wind and solar sample data can be sampled at multiple preset time scales, each corresponding to a specific sampling interval, thereby extracting fluctuation patterns at different time granularities from the raw wind and solar data. These multiple time scales can include an instantaneous abrupt change scale (first scale), a short-term ramp-up scale (second scale), and a long-term trend scale (third scale). The sampling interval for the instantaneous abrupt change scale is shorter than that for the short-term ramp-up scale, and the sampling interval for the short-term ramp-up scale is shorter than that for the long-term trend scale.

[0038] Multiple transient change sampling waveforms are obtained by sampling wind and solar sample data at the transient change scale. Each transient change sampling waveform can be used to characterize the transient change characteristics of wind and solar power in seconds or minutes, such as step changes and spikes.

[0039] Multiple short-time climbing sampling waveforms were obtained by sampling wind and solar sample data at a short-time climbing scale. Each short-time climbing sampling waveform can be used to characterize the climbing characteristics of wind and solar power continuously rising or falling within a time period of minutes to quarter minutes.

[0040] Multiple long-term trend sampling waveforms are obtained by sampling wind and solar power sample data on a long-term trend scale. Each long-term trend sampling waveform can be used to characterize the long-term trend characteristics of wind and solar power, such as slow changes and periodic fluctuations over time periods ranging from hours to days.

[0041] Shape similarity clustering can be performed on multiple wind and light sampled waveforms at each time scale. Shape similarity clustering can be a clustering method that uses the similarity of waveform geometry as the classification criterion. Algorithms including, but not limited to, dynamic time warping algorithms are used to calculate the morphological distance between different waveforms; the smaller the morphological distance, the more similar the waveforms are. Based on the calculated morphological distance matrix, hierarchical clustering or density clustering algorithms are used to group wind and light sampled waveforms with similar shapes into the same category.

[0042] For multiple transient mutation sampling waveforms at the transient mutation scale, clustering is used to obtain multiple transient mutation cluster waveforms, each representing a typical transient mutation mode.

[0043] For multiple short-time ramp sampling waveforms at the short-time ramp scale, multiple short-time ramp cluster waveforms are obtained by clustering, and each cluster represents a typical short-time ramp pattern.

[0044] For multiple long-term trend sampling waveforms at the long-term trend scale, clustering is used to obtain multiple long-term trend cluster waveforms, each representing a typical long-term trend pattern.

[0045] Based on the clustering results, shape features of various wind and light sampling waveforms at each time scale can be extracted. Shape features are used to describe the geometric shape of the waveform, including but not limited to one or more combinations of waveform peak position, peak amplitude, rising edge slope, falling edge slope, waveform duration, and fluctuation amplitude.

[0046] For each instantaneous change cluster waveform at the instantaneous change scale, the average shape feature of all waveforms in that cluster is calculated as the representative shape feature of that instantaneous change waveform. For each short-term climbing cluster waveform at the short-term climbing scale, the representative shape feature is extracted in the same way. For each long-term trend cluster waveform at the long-term trend scale, the representative shape feature is also extracted.

[0047] Based on shape features, typical power waveform curves for various time scales can be generated. These typical power waveform curves are standardized power time series with typical waveform morphology characteristics, and can be used as dynamic input in subsequent electrolytic cell testing.

[0048] For each type of transient change cluster waveform at the transient change scale, based on the representative shape characteristics of that type of transient change waveform, a typical curve of transient change power waveform is generated by waveform synthesis method. This curve reflects the transient change mode corresponding to that type, such as a power step of a specific amplitude or a power spike of a specific duration.

[0049] For each type of short-time ramp cluster waveform at the short-time ramp scale, a corresponding typical curve of short-time ramp power waveform is generated. This curve reflects the short-time ramp pattern corresponding to that category, such as a power increase or decrease at a specific rate of change.

[0050] For each type of long-term trend cluster waveform at the long-term trend scale, a corresponding typical curve of long-term trend power waveform is generated. This curve reflects the long-term trend pattern corresponding to that category, such as slow oscillation or trend change of a specific waveform.

[0051] A library of typical power waveforms is formed, which can be used to cover the main fluctuation patterns of wind and solar power at different time scales.

[0052] By setting multiple time scales to capture the instantaneous abrupt changes, short-term ramp-ups, and long-term trend characteristics of wind and solar power, the problem of a single scale being unable to comprehensively characterize the fluctuation features of wind and solar power is solved. Shape similarity clustering of waveforms at each scale automatically extracts several representative fluctuation patterns from massive historical data, avoiding the subjectivity and bias of manually setting typical operating conditions. Extracting the shape features of various waveforms and generating typical curves provides standardized dynamic input signals for subsequent testing, ensuring that the test conditions retain the statistical characteristics of actual wind and solar power while possessing repeatability and comparability. The generated typical power waveform curves cover multi-scale fluctuation characteristics from second-level abrupt changes to hourly trends, meeting the needs of dynamic response testing of electrolyzer systems at different time scales.

[0053] In some embodiments, step S101 may further include: performing time-frequency decomposition on the wind and solar sample data to obtain multiple intrinsic mode components characterizing different fluctuation frequency components; classifying each intrinsic mode component into a first scale, a second scale, and a third scale according to the frequency distribution characteristics of each intrinsic mode component; the intrinsic mode components of the first scale, the second scale, and the third scale are used to characterize the instantaneous change characteristics, short-term ramp-up characteristics, and long-term trend characteristics of wind and solar power, respectively; reconstructing the waveforms of the intrinsic mode components at each time scale to obtain multiple wind and solar sampled waveforms at that time scale; extracting the morphological feature parameters of each wind and solar sampled waveform at each time scale; and clustering the multiple wind and solar sampled waveforms at each time scale based on the morphological feature parameters to generate typical curves of multiple power waveforms at that time scale.

[0054] Time-frequency decomposition can employ empirical mode decomposition (EMD) algorithms to decompose the original non-stationary wind and solar power signal into a series of intrinsic mode function (EMF) components with different characteristic time scales. Each EMF component represents the fluctuation component within a specific frequency range of the original signal. EMF can adaptively handle nonlinear and non-stationary signals without requiring pre-defined basis functions.

[0055] Spectral analysis can be performed on each intrinsic modal component to calculate the center frequency and frequency distribution range of each intrinsic modal component. Based on the frequency distribution characteristics of each intrinsic modal component, they are respectively classified into the instantaneous abrupt change scale (first scale), the short-term ramp scale (second scale), and the long-term trend scale (third scale).

[0056] The instantaneous change scale corresponds to the high-frequency fluctuation component, with a frequency higher than the preset high-frequency threshold. It can be used to characterize the instantaneous change characteristics of wind and solar power, such as second-level power surges.

[0057] The short-term ramp scale corresponds to the mid-frequency fluctuation component, with a frequency between the high-frequency threshold and the low-frequency threshold. It can be used to characterize the short-term ramp characteristics of wind and solar power, such as minute-level power rise and fall.

[0058] Long-term trend scales correspond to low-frequency fluctuation components with frequencies below a preset low-frequency threshold. They can be used to characterize the long-term trend characteristics of wind and solar power, such as slow changes on an hourly scale.

[0059] The high-frequency threshold is higher than the low-frequency threshold, and the specific value can be set according to the fluctuation characteristics of the actual wind and solar resources.

[0060] Waveform reconstruction can be performed on the intrinsic mode components at each time scale to obtain multiple wind and solar sampling waveforms at that time scale. For multiple high-frequency intrinsic mode components belonging to the instantaneous change scale, the multiple high-frequency components at the same moment are linearly superimposed to reconstruct the instantaneous change sampling waveform.

[0061] For multiple mid-frequency intrinsic mode components belonging to the short-time ramp scale, the short-time ramp sampling waveform is reconstructed in the same way.

[0062] For multiple low-frequency intrinsic mode components belonging to the long-term trend scale, the long-term trend sampling waveform is reconstructed in the same way.

[0063] The reconstructed wind and solar sampling waveforms retain the fluctuation characteristics of the corresponding frequency bands in the original signal, while eliminating interference from other frequency bands.

[0064] The morphological feature parameters of each wind and solar sampling waveform at each time scale can be extracted. These morphological feature parameters can be used to describe the waveform's geometry, including but not limited to one or more combinations of the waveform's zero-crossing rate, peak factor, impulse factor, waveform factor, and margin factor. The zero-crossing rate characterizes the frequency with which the waveform crosses zero and can be used to reflect the oscillatory characteristics of the fluctuation. The peak factor, the ratio of the waveform's peak value to its effective value, can be used to reflect the intensity of the impulse component in the waveform. The impulse factor, the ratio of the waveform's peak value to its rectified average value, can also be used to characterize impulse characteristics. The waveform factor, the ratio of the effective value to the rectified average value, can be used to describe the overall shape of the waveform. The margin factor, the ratio of the peak value to the square root amplitude, can be used to reflect the steepness of the waveform. For instantaneous abrupt change sampling waveforms, their morphological feature parameters are calculated to form an feature vector. For short-term ramp sampling waveforms and long-term trend sampling waveforms, the corresponding morphological feature parameters are calculated to form feature vectors respectively.

[0065] Based on morphological feature parameters, multiple wind and light sampling waveforms at each time scale can be clustered. K-means clustering or Gaussian mixture model clustering algorithms are used, with feature vectors composed of morphological feature parameters as input, grouping waveforms with close distances in the feature space into the same category. For transient change sampling waveforms, clustering is performed based on their morphological feature parameters to obtain multiple clusters of transient change waveforms. For short-term upslope sampling waveforms and long-term trend sampling waveforms, clustering is performed separately to obtain multiple clusters of short-term upslope waveforms and multiple clusters of long-term trend waveforms. During the clustering process, the optimal number of clusters is determined using the silhouette coefficient or the Davidson-Burding index.

[0066] For each cluster of waveforms at each time scale, a typical power waveform curve is generated. For each instantaneous change cluster of waveforms at the instantaneous change scale, the mean waveform of all waveforms in that cluster is calculated, and this mean waveform is smoothed and filtered to remove high-frequency noise, resulting in a typical power waveform curve for the instantaneous change. For each short-term ramp cluster of waveforms at the short-term ramp scale, a typical power waveform curve for the short-term ramp is generated in the same way. For each long-term trend cluster of waveforms at the long-term trend scale, a typical power waveform curve for the long-term trend is generated in the same way. The typical power waveform curves of multiple types constitute a complete dynamic test condition input set, covering the typical fluctuation patterns of wind and solar power in different frequency bands.

[0067] By employing empirical mode decomposition (EMD) to perform time-frequency decomposition on wind and solar sample data, non-stationary and nonlinear wind and solar power signals can be adaptively decomposed into intrinsic mode components of different frequency bands. This avoids the limitation of traditional filtering methods requiring preset cutoff frequencies and demonstrates better adaptability to the complex fluctuation components in actual wind and solar data. Based on frequency distribution characteristics, each component is assigned to different time scales, achieving a scientific separation of three fluctuation characteristics: instantaneous abrupt changes, short-term ramp-ups, and long-term trends. This allows waveforms at each scale to independently reflect the fluctuation patterns of specific frequency bands. Morphological feature parameters are extracted from the reconstructed waveforms at each scale and clustered to identify wind and solar fluctuation patterns from both frequency and morphological dimensions, improving the accuracy of typical curves in depicting actual fluctuation characteristics. The generated typical curves can completely preserve the energy distribution and morphological characteristics of the original wind and solar signals in each frequency band, providing more accurate input conditions for testing the multi-time-scale dynamic response characteristics of the electrolyzer system.

[0068] S102: Generate various types of simulated fluctuating power data based on the typical curves of the various power waveforms.

[0069] In some embodiments, step S102 may specifically include: obtaining multiple preset wind and solar power output test scenario requirements; for each wind and solar power output test scenario requirement, selecting one or more typical power waveform curves from the multiple types of typical power waveform curves and combining them to obtain a simulated fluctuation test sequence corresponding to the wind and solar power output test scenario requirement; and performing interpolation processing on the simulated fluctuation test sequence to obtain simulated fluctuation power data corresponding to the wind and solar power output test scenario requirement.

[0070] Typical power waveform curves for each type are standardized power time series with specific morphological characteristics. These curves are derived from the analysis and refinement of historical wind and solar power data and can be used to characterize typical fluctuation patterns of wind and solar power at different time scales. Multiple preset wind and solar power output test scenario requirements are obtained. Each wind and solar power output test scenario requirement is pre-set operating condition description data for dynamic testing of the electrolyzer system. This data includes specific requirements for power fluctuation patterns during the test. Wind and solar power output test scenario requirements may include, but are not limited to, one or more combinations of extreme fluctuation scenario requirements, normal fluctuation scenario requirements, and smooth fluctuation scenario requirements. Extreme fluctuation scenario requirements correspond to operating conditions where power changes dramatically in a short period of time, such as power jumping from a low point to a high point within seconds. Normal fluctuation scenario requirements correspond to operating conditions where the power fluctuation amplitude and rate of change are at a moderate level, such as power exhibiting regular fluctuations within minutes. Smooth fluctuation scenario requirements correspond to operating conditions where power changes slowly and the fluctuation amplitude is small, such as power exhibiting a slow trend change within hours.

[0071] For each wind and solar power output test scenario, one or more typical power waveform curves are selected from multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario.

[0072] For extreme fluctuation scenarios, a type of curve representing rapid step changes is selected from typical power waveform curves at the instantaneous change scale, and a type of curve representing continuous rise is selected from typical power waveform curves at the short-term ramp scale. The two types of curves are spliced ​​and combined according to a preset time sequence relationship to form a simulated fluctuation test sequence that reflects the characteristics of extreme fluctuations.

[0073] For typical fluctuation scenarios, a type of curve representing a moderate rate of change is selected from the typical power waveform curves of short-term ramp scale, and directly used as the simulated fluctuation test sequence.

[0074] For scenarios with gentle fluctuations, a type of curve representing slow changes is selected from typical power waveform curves on a long-term trend scale, and directly used as the simulated fluctuation test sequence.

[0075] During the combination process, a weighted smoothing algorithm can be used to process the connection points between different curves to eliminate power jumps at the splicing points and ensure the continuity and smoothness of the simulated fluctuation test sequence.

[0076] Interpolation processing can be performed on simulated fluctuation test sequences to obtain simulated fluctuation power data corresponding to the requirements of the wind and solar power output test scenario. Interpolation processing involves inserting new data points between adjacent data points in the simulated fluctuation test sequence according to a specific algorithm to improve the temporal resolution of the data. A cubic spline interpolation algorithm is used to interpolate the simulated fluctuation test sequence, encrypting the time intervals of the original sequence to the second or sub-second level, generating high-temporal-resolution simulated fluctuation power data. This simulated fluctuation power data is a discrete power value sequence, with each value corresponding to the power value at a specific time point, which can be used as power input in subsequent electrolytic cell system testing. During the interpolation process, the waveform shape of the original sequence remains unchanged; only the density of data points is increased to ensure that the interpolated data still retains the fluctuation characteristics of the original typical curve.

[0077] By setting up various wind and solar power output test scenarios, targeted testing of the electrolytic cell system's operational performance under different types of fluctuation conditions was achieved. Matching curves were selected and combined from typical power waveform curves of multiple types, ensuring that the generated simulated fluctuation test sequence accurately reflects the typical fluctuation characteristics of various scenarios, avoiding the problem of using a single curve to cover complex operating conditions. Smoothing was applied to the connections between different curves to eliminate power jumps at splicing points, ensuring the continuity of the test input signal and avoiding additional test interference introduced by signal abrupt changes. Interpolation processing was performed on the simulated fluctuation test sequence to generate high-time-resolution power data, meeting the accuracy requirements of the input signal for the dynamic response test of the electrolytic cell system. The generated simulated fluctuation power data retains the morphological characteristics of typical curves and has sufficient time resolution, making it suitable as a standard input for the dynamic testing of the electrolytic cell system.

[0078] In some embodiments, step S102 may further include: acquiring a plurality of preset fluctuation energy characteristic indicators; the fluctuation energy characteristic indicators include at least the fluctuation intensity level and fluctuation duration range at each time scale; selecting matching curves from typical power waveform curves at different time scales as fluctuation components at each scale according to the fluctuation energy characteristic indicators; determining the phase difference and amplitude ratio between the fluctuation components at each scale based on the temporal correlation of the fluctuation components at different time scales; superimposing the fluctuation components at each scale according to the phase difference and amplitude ratio to generate a composite fluctuation test sequence; and interpolating the composite fluctuation test sequence to obtain simulated fluctuation power data corresponding to the fluctuation energy characteristic indicators.

[0079] Typical power waveform curves include those for instantaneous abrupt changes, short-term ramps, and long-term trends. The instantaneous abrupt change scale corresponds to the high-frequency fluctuations in wind and solar power, the short-term ramp scale corresponds to the mid-frequency fluctuations, and the long-term trend scale corresponds to the low-frequency fluctuations.

[0080] Multiple preset fluctuation energy characteristic indicators can be obtained. These indicators are combinations of parameters used to characterize the intensity of wind and solar power fluctuations, including the fluctuation intensity level and duration range at various time scales. The fluctuation intensity level can be divided into weak, medium, and strong levels based on the power change rate or fluctuation amplitude, with each level corresponding to a power change rate range or fluctuation amplitude range. The fluctuation duration range can be the length of time a specific fluctuation intensity lasts, divided into short-term, medium-term, and long-term ranges, each corresponding to a time length interval. The fluctuation energy characteristic indicators can also include the energy ratio between fluctuation components at different time scales, i.e., the proportion of fluctuation components at different scales in the total fluctuation energy.

[0081] Based on the fluctuation energy characteristic index, matching curves can be selected from typical power waveform curves at different time scales as fluctuation components for each scale. For instantaneous change scales, based on the fluctuation intensity level and duration range at that scale, a type of curve matching the level and range of typical instantaneous change power waveform curves is selected as the instantaneous change fluctuation component. For short-term ramp scales, based on the fluctuation intensity level and duration range at that scale, a matching type of curve is selected from typical short-term ramp power waveform curves as the short-term ramp fluctuation component. For long-term trend scales, based on the fluctuation intensity level and duration range at that scale, a matching type of curve is selected from typical long-term trend power waveform curves as the long-term trend fluctuation component.

[0082] Based on the temporal correlation of fluctuation components at different time scales, the phase difference and amplitude ratio between fluctuation components at each scale are determined. The temporal correlation can be the time offset and amplitude coupling relationship between fluctuation components at different time scales in actual wind and solar power, which can be obtained through statistical analysis of a large amount of historical wind and solar data. Through cross-correlation analysis of wind and solar sample data, a phase difference distribution model and an amplitude ratio distribution model between fluctuation components at different scales are established. Based on this distribution model, a set of phase difference values ​​is selected from a pre-defined set of phase difference values ​​as the phase offset between the instantaneous abrupt fluctuation component and the short-term climbing fluctuation component, and between the short-term climbing fluctuation component and the long-term trend fluctuation component. Based on this distribution model, a set of ratio values ​​is selected from a pre-defined set of amplitude ratio values ​​as the amplitude scaling coefficients between the instantaneous abrupt fluctuation component, the short-term climbing fluctuation component, and the long-term trend fluctuation component.

[0083] The fluctuation components at different scales can be superimposed according to phase difference and amplitude ratio to generate a composite fluctuation test sequence. The instantaneous abrupt change fluctuation component can be scaled according to a determined amplitude ratio, the short-term ramp fluctuation component can be scaled according to a determined amplitude ratio and then time-shifted according to phase difference, and the long-term trend fluctuation component can be scaled according to a determined amplitude ratio and then time-shifted according to phase difference. The three scale fluctuation components, after scaling and shifting, are aligned on the time axis and merged using a linear superposition method to obtain the composite fluctuation test sequence. This composite fluctuation test sequence simultaneously includes three fluctuation characteristics: instantaneous abrupt change, short-term ramp, and long-term trend, and the temporal and amplitude relationships between these characteristics conform to the statistical laws of actual wind and solar power. During the superposition process, the values ​​of each component at each time point are added to form the final composite sequence.

[0084] Interpolation processing can be performed on composite wave test sequences to obtain simulated wave power data corresponding to wave energy characteristic indicators. A piecewise cubic Hermitian interpolation algorithm is used to process the composite wave test sequences, improving the time resolution of the data while maintaining the monotonicity of the sequence and the location of extreme points, generating simulated wave power data at the second or sub-second level. This simulated wave power data is a complete power time series and can be used as power input in dynamic testing of electrolytic cell systems. The interpolated data retains all the key characteristics of the original composite waveform, including the peak position, rate of change, and duration of the waveform at each scale.

[0085] By setting fluctuation energy characteristic indicators, quantitative control of the intensity and duration of simulated fluctuating power was achieved, enabling the generated test data to accurately match specific energy characteristic requirements. Matching fluctuation components were selected from typical curves at different time scales, ensuring that the fluctuation characteristics of the simulated data in each frequency band were consistent with the target indicators, avoiding cross-scale feature confusion. The phase difference and amplitude ratio between components at each scale were determined based on actual time-series correlations, solving the problem that simple superposition methods could not reflect the coupling relationships of frequency components in real fluctuations, making the generated composite sequence closer to the fluctuation characteristics of actual wind and solar power. Superimposing the components at each scale according to the determined phase difference and amplitude ratio, the generated composite fluctuation test sequence can realistically reproduce the multi-scale coupled fluctuation characteristics of actual wind and solar power, providing more realistic input conditions for performance testing of the electrolyzer system under complex fluctuation conditions. The generated simulated fluctuating power data can accurately match the preset fluctuation energy characteristic indicators, providing precise input conditions for performance testing of the electrolyzer system under specific fluctuation energy distribution conditions.

[0086] S103: Based on the aforementioned multi-type simulated fluctuation power data, adjust the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system to perform a test on the ALK-PEM hybrid hydrogen production system.

[0087] In some embodiments, the testing of the ALK-PEM hybrid hydrogen production system includes: static performance testing, dynamic operating condition testing, cooperative operation testing, and start-stop characteristic testing.

[0088] In some embodiments, step S103 may specifically include: determining the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system under each test based on the multi-type simulated fluctuation power data, so as to perform a test on the ALK-PEM hybrid hydrogen production system.

[0089] During static performance testing, constant power mode data can be selected from various types of simulated fluctuating power data. This constant power mode data is a sequence where the power value remains unchanged. Based on the constant power mode data, the operating power of the PEM and ALK units under static performance testing is determined. Static performance testing includes multiple steady-state operating points, each corresponding to a fixed power output value. The operating power of the PEM and ALK units is configured according to a preset allocation ratio, with the PEM unit assuming the first portion of the power value and the ALK unit assuming the second portion. The sum of the first and second portions equals the total operating power. The PEM and ALK units are then operated at the configured operating power to perform the static performance test. During the static performance test, the PEM and ALK units maintain a constant operating power for a preset duration to evaluate the performance of the ALK-PEM hybrid hydrogen production system under steady-state conditions.

[0090] By selecting constant power mode data from simulated fluctuating power data, standardized steady-state input conditions are provided for static performance testing. The operating power of the PEM and ALK units is configured according to a preset allocation ratio, enabling the static performance test to cover operating states under different power allocation schemes. The PEM and ALK units maintain constant power during the test, eliminating the interference of power fluctuations on performance evaluation, and ensuring that the acquired test data accurately reflects the true performance of the ALK-PEM hybrid hydrogen production system under steady-state conditions.

[0091] During dynamic operating condition testing, dynamic fluctuation pattern data is selected from multiple types of simulated fluctuating power data. This dynamic fluctuation pattern data is a sequence of power values ​​that continuously change over time. The dynamic fluctuation pattern data includes fluctuation characteristics at multiple time scales, including instantaneous abrupt changes, short-term ramp-up characteristics, and long-term trend characteristics. Based on the dynamic fluctuation pattern data, the operating power of the PEM and ALK units under dynamic operating condition testing is determined. Using the dynamic fluctuation pattern data as the total power input, the power values ​​assigned to each of the PEM and ALK units are calculated according to a preset dynamic power allocation strategy. The dynamic power allocation strategy is based on the dynamic response characteristics of the PEM and ALK units; the PEM unit has a faster response speed and handles the high-frequency fluctuation portion of the power, while the ALK unit has a relatively slower response speed and handles the low-frequency fluctuation portion. The PEM and ALK units are then adjusted to operate according to the calculated power values ​​to execute the dynamic operating condition test. During the dynamic operating condition test, the operating power of the PEM and ALK units changes with the dynamic fluctuation pattern data, which is used to evaluate the dynamic response performance of the ALK-PEM hybrid hydrogen production system under fluctuating conditions.

[0092] By selecting dynamic fluctuation mode data as input, the input conditions for dynamic operating condition testing are made highly consistent with the actual wind and solar fluctuation characteristics. A power allocation strategy is formulated based on the dynamic response characteristics of the PEM and ALK units, fully leveraging the complementary advantages of the fast response speed of the PEM unit and the large capacity of the ALK unit. The operating power of the PEM and ALK units follows the changes in input data, enabling accurate testing of the power tracking capability, response speed, and control accuracy of the ALK-PEM hybrid hydrogen production system under multi-scale fluctuation conditions.

[0093] During collaborative operation testing, collaborative test mode data can be selected from multiple types of simulated fluctuating power data. This collaborative test mode data consists of a stepped change sequence including multiple power steps. The collaborative test mode data includes multiple different power levels, each with a preset duration, and power steps are set between adjacent power levels. Based on the collaborative test mode data, the operating power of the PEM and ALK units under collaborative operation testing is determined. Collaborative operation testing is used to evaluate the collaborative performance of the PEM and ALK units during power changes. The operating power of the PEM and ALK units is configured according to various power allocation schemes, each corresponding to a different power sharing ratio. The power allocation schemes include a first allocation scheme, a second allocation scheme, and a third allocation scheme. In the first allocation scheme, the PEM unit bears a higher proportion; in the second allocation scheme, the PEM and ALK units bear the same proportion; and in the third allocation scheme, the ALK unit bears a higher proportion. Under each power allocation scheme, the PEM and ALK units are controlled to operate according to the power step sequence of the collaborative test mode data to execute the collaborative operation test. During the collaborative operation test, the response process, power allocation accuracy, and operational stability parameters of the PEM unit and ALK unit at the power step moment were recorded.

[0094] By setting up collaborative test mode data including multiple power steps, a standardized power change sequence is provided for collaborative operation testing. Multiple power allocation schemes are employed for testing, comprehensively covering the collaborative operation of PEM and ALK units under different power sharing ratios. The response process and allocation accuracy at power step moments are recorded, enabling accurate evaluation of the power transfer capability, load balancing performance, and coupled operational stability between PEM and ALK units.

[0095] During start-stop characteristic testing, start-stop test mode data is selected from multiple types of simulated fluctuating power data. This start-stop test mode data consists of a sequence of switches including start and stop commands. The start-stop test mode data includes light-load start command, rated start command, temporary stop command, and long-term stop command, each command corresponding to a specific power change trajectory. Based on the start-stop test mode data, the operating power of the PEM and ALK units under start-stop characteristic testing is determined. Under the light-load start command, the PEM and ALK units start operating at a preset percentage of the rated power. Under the rated start command, the PEM and ALK units start operating at the rated power. Under the temporary stop command, the PEM and ALK units execute a rapid stop procedure, restarting after a preset short stop duration. Under the long-term stop command, the PEM and ALK units execute a complete stop procedure, restarting after a preset long stop duration. The start-stop characteristic test is performed by controlling the PEM and ALK units to run according to the command sequence of the start-stop test mode data. Each start-stop command is executed a preset number of times to evaluate the reliability and durability of the ALK-PEM hybrid hydrogen production system during repeated start-stop processes.

[0096] By setting up start-stop test mode data including various start-stop commands, a standardized start-stop sequence is provided for start-stop characteristic testing. Two modes, light-load start and rated start, cover start-up conditions under different load conditions, while two modes, temporary shutdown and long-term shutdown, cover restart conditions under different shutdown durations. Each command is executed multiple times in a loop, effectively evaluating the performance degradation trend, sealing reliability, and thermal stability of the ALK-PEM hybrid hydrogen production system during repeated start-stop processes.

[0097] S104: Obtain operational data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process.

[0098] In some embodiments, step S104 may specifically include: acquiring the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system in each test process.

[0099] During each test, operational data for the PEM unit, the ALK unit, and the overall ALK-PEM hybrid hydrogen production system can be acquired. PEM unit operational data includes input current, input voltage, operating temperature, operating pressure, hydrogen production, oxygen production, hydrogen production efficiency, and gas purity. ALK unit operational data includes input current, input voltage, operating temperature, operating pressure, hydrogen production, oxygen production, hydrogen production efficiency, and gas purity. The overall ALK-PEM hybrid hydrogen production system operational data includes total input power, total hydrogen production, total hydrogen production efficiency, separation system liquid level, system pressure fluctuation, water replenishment rate, and power wastage rate. Operational data can be acquired using a multi-channel synchronous acquisition method, with all data channels sampling according to a unified time base and a sampling frequency not lower than a preset frequency threshold. The acquired operational data is stored according to test items, with a separate dataset created for each test item, including operational parameter values ​​at all acquisition times for that test item.

[0100] By synchronously acquiring operational data from the PEM unit, ALK unit, and the ALK-PEM hybrid hydrogen production system, the temporal correspondence of data at each level was ensured, providing a data foundation for subsequent analysis of the correlation between unit performance and overall performance. The multi-channel synchronous acquisition method eliminated analytical errors caused by inconsistent data acquisition times. Data was stored according to test items, facilitating comparative analysis and trend mining of operational data under different test conditions.

[0101] S105: Based on the operating data, generate a test report for the ALK-PEM hybrid hydrogen production system.

[0102] In some embodiments, step S105 may specifically include: calculating the performance indicators, stability indicators, and synergy indicators of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system based on the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system in each test process; and generating a test report of the ALK-PEM hybrid hydrogen production system based on the calculated performance indicators, stability indicators, and synergy indicators.

[0103] Based on operational data, the performance, stability, and synergy indicators of the PEM unit can be calculated. The PEM unit's performance indicators include hydrogen production efficiency at rated power, gas purity, and power response speed. The PEM unit's stability indicators include the temperature fluctuation range, pressure fluctuation range, and voltage fluctuation range during operation. The PEM unit's synergy indicators include the deviation between the actual power supplied and the commanded power during power allocation, and the response delay time during power adjustment. Based on operational data, the performance, stability, and synergy indicators of the ALK unit can also be calculated. The ALK unit's performance indicators include hydrogen production efficiency at rated power, gas purity, and power response speed. The ALK unit's stability indicators include the temperature fluctuation range, pressure fluctuation range, voltage fluctuation range, and liquid level fluctuation range during operation. The ALK unit's synergy indicators include the deviation between the actual power supplied and the commanded power during power allocation, and the response delay time during power adjustment. Based on operational data, the overall performance, stability, and synergy indicators of the ALK-PEM hybrid hydrogen production system can be calculated. The overall performance indicators of the ALK-PEM hybrid hydrogen production system include comprehensive hydrogen production efficiency, power curtailment rate, and gas purity compliance rate. The overall stability indicators of the ALK-PEM hybrid hydrogen production system include the system pressure fluctuation range, the absolute value of the liquid level difference in the separation system, and the number of times parameters exceed their range. The overall synergy indicators of the ALK-PEM hybrid hydrogen production system include the power distribution accuracy between the PEM and ALK units and the smoothness of the coupling response. Based on the calculated performance, stability, and synergy indicators, a test report for the ALK-PEM hybrid hydrogen production system is generated. The test report may include a test condition description section, a test data summary section, an indicator calculation result display section, and a conclusion judgment section. The test condition description section records the source of the simulated fluctuating power data used in the test, the test environment parameters, and the test time. The test data summary section displays the raw data statistics for each test in tabular form. The indicator calculation result display section displays the calculation results of the performance, stability, and synergy indicators in graphical form. The conclusion judgment section compares the results with the preset pass / fail standards, providing a judgment on whether each indicator is qualified, and indicating the performance shortcomings and improvement directions of the ALK-PEM hybrid hydrogen production system.

[0104] A comprehensive performance evaluation system was constructed by calculating multi-dimensional indicators for the PEM unit, ALK unit, and the ALK-PEM hybrid hydrogen production system as a whole. Performance indicators assess the efficiency and gas production quality of the electrolyzer, stability indicators assess the operational stability of the electrolyzer, and synergy indicators assess the coordination effect between units. The test report includes a complete information chain from raw data to final conclusions, providing direct evidence for the design optimization, parameter debugging, and troubleshooting of the ALK-PEM hybrid hydrogen production system. The standardized test report format facilitates comparison of results between different test batches, meeting the testing needs for the large-scale R&D and industrial application of the ALK-PEM hybrid hydrogen production system.

[0105] In some embodiments, the above-mentioned shape similarity clustering of multiple wind and solar waveforms at each time scale may specifically include: dividing each wind and solar sampling waveform into multiple waveform segments; the multiple waveform segments include a rising segment, a peak hold segment, and a falling segment, wherein the rising segment is used to characterize the power ramping capability of the PEM unit and the ALK unit, the peak hold segment is used to characterize the overload sustaining capability of the PEM unit and the ALK-PEM hybrid hydrogen production system, and the falling segment is used to characterize the power recovery capability of the PEM unit and the ALK unit; determining the weight of each waveform segment based on the sensitivity and tolerance of the PEM unit and the ALK unit to each waveform segment; fusing the morphological feature parameters of each waveform segment based on the weight to obtain the waveform feature vector of each wind and solar sampling waveform; and clustering the waveform feature vectors at each time scale.

[0106] Multiple wind and solar power sampling waveforms can be acquired at each time scale, with each waveform representing a curve of power variation over time. Each waveform is divided into multiple segments, based on the first and second derivative characteristics of the waveform. These segments include rising segments, peak hold segments, and falling segments.

[0107] The rising segment is the part of the waveform where the power value continuously increases. It starts at a local minimum point and ends at a local maximum point. The rising segment can be used to characterize the ramp-up capability of PEM and ALK units during the power increase process. Specifically, it can reflect the ramp-up rate, overshoot, and settling time during the ramp-up process.

[0108] The peak hold segment is the part of the waveform where the power value is maintained near the peak value. The starting point is the end of the rising segment and the ending point is the starting point of the falling segment. The peak hold segment can be used to characterize the continuous operation capability of PEM unit and ALK-PEM hybrid hydrogen production system under overload conditions. Specifically, it can reflect the overload amplitude, overload duration and overload efficiency.

[0109] The falling segment is the part of the waveform where the power value continuously decreases. It starts at the end of the peak holding segment and ends at the next local minimum point. The falling segment can be used to characterize the recovery capability of PEM and ALK units during the power reduction process. Specifically, it can reflect the recovery rate, the fluctuation amplitude during the recovery process, and the adjustment accuracy.

[0110] The weights of each waveform segment are determined based on the sensitivity and tolerance of the PEM and ALK units to each waveform segment. Sensitivity can be the degree of drastic response of the PEM or ALK unit to the operating condition change represented by the waveform segment, while tolerance can be the ability of the PEM or ALK unit to maintain stable operation under the operating condition represented by the waveform segment.

[0111] For the rising segment, PEM cells are more sensitive to rapid power increases than ALK cells. Therefore, the rising segment can be assigned a higher weight in the characteristic analysis of PEM cells and a lower weight in the characteristic analysis of ALK cells.

[0112] For the peak sustaining phase, the overall tolerance of the ALK-PEM hybrid hydrogen production system to overload conditions depends on the combined effect of the PEM and ALK units. Therefore, the peak sustaining phase can be assigned a higher weight in the overall characteristic analysis of the ALK-PEM hybrid hydrogen production system.

[0113] For the falling segment, the ALK cell may experience reverse current risk due to rapid power decrease; therefore, the falling segment can be assigned a higher weight in the safety analysis of the ALK cell. The weights can be determined using the analytic hierarchy process (AHP) or expert scoring, with the sum of the weights for each waveform segment being 1.

[0114] Based on weights, the morphological feature parameters of each waveform segment can be fused to obtain the waveform feature vector of each wind and solar sampling waveform. The morphological feature parameters include the duration, rate of change, fluctuation amplitude, integral area, and extreme point value of each waveform segment. For the rising segment, the rising duration, maximum rising rate, and integral area of ​​the rising segment are extracted as morphological feature parameters. For the peak holding segment, the holding duration, peak fluctuation amplitude, and overload integral area are extracted as morphological feature parameters. For the falling segment, the falling duration, maximum falling rate, and integral area of ​​the falling segment are extracted as morphological feature parameters. The morphological feature parameters of each waveform segment are weighted and combined according to their corresponding weights to form a multi-dimensional feature vector, which is the waveform feature vector of the wind and solar sampling waveform. Each dimension in the waveform feature vector corresponds to a weighted morphological feature parameter, completely preserving the feature information in the original waveform that is of great significance for electrolyzer testing.

[0115] Clustering can be performed on waveform feature vectors at each time scale. Density clustering or spectral clustering algorithms can be used, employing Euclidean or Mahalanobis distance between waveform feature vectors as a similarity measure to divide the vectors in the feature space into multiple categories. The clustering results in high similarity between waveform feature vectors within the same category and low similarity between waveform feature vectors in different categories. After clustering, each category corresponds to a set of wind and light sampling waveforms with similar morphological features.

[0116] By segmenting the wind and solar sampling waveform into rising, peak-holding, and falling segments, a refined decomposition of the waveform morphology is achieved. Each waveform segment corresponds to a different performance dimension that needs to be focused on during electrolyzer testing. Weights are determined based on the sensitivity and tolerance of each waveform segment to the PEM and ALK units, making the extracted waveform features more closely match the actual response characteristics of the electrolyzer and avoiding the problem of key features being obscured by uniform weighting. Clustering based on the weighted fusion waveform feature vectors generates typical curves that better reflect the actual testing requirements of the electrolyzer under different operating conditions, improving the relevance and effectiveness of subsequent tests.

[0117] In some embodiments, the testing method for the ALK-PEM hybrid hydrogen production system may further include: constructing a frequency distribution curve based on the occurrence frequency of various wind and solar sampling waveforms at each time scale; identifying the long tail portion of the frequency distribution curve; the long tail portion being used to characterize waveform categories with occurrence frequencies below a preset frequency threshold; extracting rare waveforms in the long tail portion whose waveform fluctuation amplitude exceeds a preset amplitude threshold; calculating the clustering effectiveness index corresponding to the number of candidate clusters at each time scale; using the clustering effectiveness index as the optimization objective and retaining the categories of the rare waveforms as the constraint, and determining the minimum number of clusters that maximizes the clustering effectiveness index and satisfies the constraint as the optimal number of clusters at that time scale. Based on this, the above-mentioned clustering of waveform feature vectors at each time scale may specifically include: clustering the waveform feature vectors according to the optimal number of clusters at each time scale.

[0118] Frequency of occurrence can be the proportion of times a particular waveform appears in historical landscape data out of the total number of waveforms. The waveforms are sorted by frequency of occurrence from highest to lowest, and a frequency distribution curve is plotted with waveform category on the x-axis and frequency of occurrence on the y-axis. The frequency distribution curve shows a decreasing trend, with high-frequency categories concentrated at the beginning of the curve and low-frequency categories distributed at the end.

[0119] It can identify the long tail portion of the frequency distribution curve. The long tail portion can be a region at the tail end of the frequency distribution curve, where the frequency of waveform categories occurring is below a preset frequency threshold. The preset frequency threshold is set based on the statistical significance of the waveform categories; categories below this threshold are considered low-probability events. The long tail portion can be used to characterize waveform categories that occur less frequently but may be important for electrolyzer testing. These categories correspond to extreme operating conditions that are less common but fluctuate dramatically in actual wind and solar power generation.

[0120] Rare waveforms whose amplitude exceeds a preset amplitude threshold can be extracted from the long-tail portion of the waveform. The amplitude fluctuation can be the ratio of the difference between the maximum and minimum power values ​​in the waveform to the rated power. The preset amplitude threshold is set based on the safe operating range of the electrolyzer; fluctuations exceeding this threshold may impact the electrolyzer. From the waveform categories in the long tail portion, categories with amplitude fluctuations exceeding the preset threshold are selected, and representative waveforms in these categories are marked as rare waveforms. Although rare waveforms occur infrequently, their dramatic fluctuations place higher demands on the electrolyzer's resilience and protection mechanisms, and therefore must be covered in the test.

[0121] The clustering effectiveness index can be calculated for each number of candidate clusters at each time scale. The number of candidate clusters ranges from 2 to a preset maximum number of clusters, which is set empirically or determined based on the total number of waveforms. The clustering effectiveness index uses the silhouette coefficient or the Davidson-Burding index to evaluate the compactness and separability of the clustering results. For each number of candidate clusters, the clustering algorithm is executed, and the corresponding clustering effectiveness index value is calculated. The larger the clustering effectiveness index value, the better the clustering effect at that number of clusters.

[0122] An optimization problem is constructed using the clustering effectiveness index as the optimization objective and the retention of rare waveform categories as a constraint. Retaining rare waveform categories means that the category containing the rare waveform cannot be merged into other categories in the clustering results; that is, the rare waveform must exist as an independent category. Under this constraint, the number of clusters that maximizes the clustering effectiveness index is selected from all candidate cluster numbers. If multiple candidate cluster numbers satisfy the constraint and their clustering effectiveness indices are similar, the smallest cluster number is selected as the optimal cluster number. The minimum cluster number principle ensures that the clustering results are as concise as possible while meeting testing requirements, avoiding over-subdivision.

[0123] Based on the optimal number of clusters, waveform feature vectors at each time scale can be clustered. Following the determined optimal number of clusters for each time scale, K-means clustering or hierarchical clustering algorithms are used to cluster the waveform feature vectors, yielding the final clustering result for that time scale.

[0124] By constructing frequency distribution curves and identifying the long-tail portion, low-frequency extreme operating conditions can be identified and labeled, avoiding the problem of traditional clustering methods focusing only on high-frequency waveforms and ignoring extreme conditions. Extracting rare waveforms with fluctuation amplitudes exceeding a threshold from the long-tail portion ensures that extreme operating conditions that significantly impact the safety of the electrolyzer are retained within the testing range. With the goal of maximizing the clustering effectiveness index and the constraint of retaining rare waveforms, the optimal number of clusters is determined, achieving a balance between clustering effectiveness and test coverage. Clustering with the optimal number of clusters generates typical curves that represent most routine operating conditions while also covering the extreme conditions crucial for electrolyzer testing, improving the comprehensiveness and safety of the testing.

[0125] In some embodiments, the above-mentioned selection of one or more typical power waveform curves from the multiple types of typical power waveform curves to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario requirement for each wind and solar power output test scenario requirement may further include: selecting one type of typical power waveform curve from the multiple types of typical power waveform curves at the third scale as a basic trend curve according to the requirements of each wind and solar power output test scenario; calculating the fluctuation energy coupling coefficient between the basic trend curve and the typical power waveform curves at the first and second scales; selecting the type with the highest fluctuation energy coupling coefficient from the typical power waveform curves at the first and second scales respectively as the superimposed fluctuation curve; determining the phase offset and amplitude modulation ratio of the superimposed fluctuation curve relative to the basic trend curve based on the fluctuation energy coupling coefficient; and superimposing the superimposed fluctuation curve onto the basic trend curve according to the phase offset and amplitude modulation ratio to obtain the simulated fluctuation test sequence.

[0126] Multiple wind and solar power output test scenario requirements can be obtained, each including specific requirements for power fluctuation patterns. For each wind and solar power output test scenario requirement, one type of typical power waveform curve is selected from multiple types of curves on a long-term trend scale as the basic trend curve. The long-term trend scale corresponds to the slow change characteristics of wind and solar power, and the basic trend curve can be used to characterize the overall trend and long-term change characteristics of power fluctuations under the test scenario. The selection criterion can be that the change trend of the basic trend curve is consistent with the direction of long-term power change in the test scenario requirements, specifically including a continuous upward trend, a continuous downward trend, or a trend of rising first and then falling.

[0127] This method can calculate the fluctuation energy coupling coefficient between the basic trend curve and typical power waveform curves at the instantaneous abrupt change scale and the short-term ramp scale. The instantaneous abrupt change scale corresponds to the high-frequency fluctuation characteristics of wind and solar power, while the short-term ramp scale corresponds to the mid-frequency fluctuation characteristics. The fluctuation energy coupling coefficient is used to characterize the energy transfer relationship and morphological correlation between waveforms at different time scales. The local fluctuation energy distribution of the basic trend curve is calculated, and the instantaneous rate of change at each time point in the basic trend curve is extracted as the energy change characteristic. The energy concentration frequency band and energy release mode of the typical power waveform curves at each instantaneous abrupt change scale are calculated, and the energy ramp rate and energy duration of the typical power waveform curves at each short-term ramp scale are calculated. Using cross-correlation analysis, the correlation coefficient between the energy change characteristics of the basic trend curve and the energy characteristics of each instantaneous abrupt change scale curve is calculated as the first coupling coefficient. The correlation coefficient between the basic trend curve and each short-term ramp scale curve is calculated using the same method as the second coupling coefficient. The first and second coupling coefficients constitute the fluctuation energy coupling coefficient.

[0128] The class with the highest fluctuation energy coupling coefficient can be selected from typical power waveform curves at both the instantaneous change scale and the short-term ramp scale as the superimposed fluctuation curve. For the instantaneous change scale, the class with the highest coupling coefficient to the basic trend curve is selected from multiple typical instantaneous change power waveform curves as the first superimposed fluctuation curve. For the short-term ramp scale, the class with the highest coupling coefficient to the basic trend curve is selected from multiple typical short-term ramp power waveform curves as the second superimposed fluctuation curve. Selecting the curve with the highest coupling coefficient ensures that the superimposed fluctuation has the highest coordination and consistency with the basic trend curve in terms of energy distribution.

[0129] Based on the wave energy coupling coefficient, the phase offset and amplitude modulation ratio of the superimposed wave curve relative to the basic trend curve are determined. The phase offset can be the amount by which the superimposed wave curve needs to be shifted on the time axis to align the peaks and troughs of the superimposed wave curve with the corresponding energy change points in the basic trend curve. The phase offset is determined based on the time delay that maximizes the correlation coefficient in cross-correlation analysis. The amplitude modulation ratio can be the scaling ratio of the amplitude of the superimposed wave curve to match the wave amplitude of the superimposed wave curve with the energy change intensity at the corresponding time point in the basic trend curve. The amplitude modulation ratio is determined based on the ratio of the local energy change amplitude of the basic trend curve to the original amplitude of the superimposed wave curve. The first and second superimposed wave curves each have their own phase offset and amplitude modulation ratio.

[0130] The simulated wave test sequence can be obtained by superimposing the superimposed wave curves onto the base trend curve according to the phase offset and amplitude modulation ratio. The first superimposed wave curve is scaled according to its amplitude modulation ratio and shifted over time according to its phase offset to obtain the processed first superimposed curve. The second superimposed wave curve is scaled according to its amplitude modulation ratio and shifted over time according to its phase offset to obtain the processed second superimposed curve. The processed first and second superimposed curves are then superimposed point-by-point onto the base trend curve to obtain the final simulated wave test sequence. This simulated wave test sequence simultaneously includes the long-term trend characteristics of the base trend curve, the instantaneous abrupt change characteristics of the first superimposed curve, and the short-term ramp characteristics of the second superimposed curve. Furthermore, the temporal and amplitude relationships between these characteristics conform to the coupling law of multi-scale wave components in actual wind and solar power.

[0131] By selecting a base trend curve from a long-term trend scale, the simulated wave test sequence is ensured to meet the long-term power variation requirements of the test scenario at the macroscopic level. The wave energy coupling coefficient is calculated, and the curve with the highest coupling degree is selected as the superimposed wave curve, resolving the energy distribution inconsistency caused by arbitrarily superimposing waveforms from different time scales. The phase offset and amplitude modulation ratio are determined based on the coupling coefficient, ensuring that the superimposed composite waveform also conforms to the physical coupling relationship between frequency band components in actual waves at the microscopic level. By superimposing curves at various scales according to the determined parameters, the generated simulated wave test sequence retains the morphological characteristics of various typical curves and reproduces the real coupling relationship between multi-scale waves, providing input conditions closer to actual operating conditions for the testing of the ALK-PEM hybrid hydrogen production system.

[0132] In some embodiments, step S103 may further include: determining the first operating power of the PEM unit and the ALK unit under multiple preset power levels of the ALK-PEM hybrid hydrogen production system based on the multiple types of simulated fluctuating power data. Based on this, step S104 may further include: acquiring the first operating data of the PEM unit, the ALK unit, and the ALK-PEM hybrid hydrogen production system under each preset power level; the first operating data includes at least current, voltage, temperature, pressure, hydrogen production efficiency, gas purity, and water replenishment rate. Accordingly, step S105 may further include: fitting the current-voltage characteristic curves based on the current and voltage at each preset power level to obtain the electrochemical efficiency of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system; calculating the temperature rise rate and temperature deviation based on the temperature at each power level to determine the thermal management efficiency of the ALK-PEM hybrid hydrogen production system; calculating the hydrogen production water consumption based on the water replenishment rate and gas purity at each power level to determine the water balance efficiency of the ALK-PEM hybrid hydrogen production system; and generating a static performance test report including the dominant factors of hydrogen production efficiency deviation from the electrochemical efficiency, thermal management efficiency, and water balance efficiency.

[0133] Constant power mode data can be selected from various types of simulated fluctuating power data. This constant power mode data is a sequence where the power value remains unchanged. Based on the constant power mode data, the operating power of the PEM and ALK units under multiple preset power levels of the ALK-PEM hybrid hydrogen production system is determined. The preset power levels are multiple fixed power output points set in the static performance test, specifically including 30%, 50%, 70%, and 100% of the rated power of the ALK-PEM hybrid hydrogen production system. Each preset power level corresponds to a total power value. For each preset power level, the sub-power undertaken by the PEM unit and the sub-power undertaken by the ALK unit are determined according to a preset power allocation ratio. The sum of the sub-powers of the PEM unit and the ALK unit equals the preset power level value. The sub-powers of the PEM unit and the ALK unit under each preset power level are taken as the operating power at that level.

[0134] The PEM and ALK units can be adjusted to operate at preset power levels. The PEM unit operates according to its sub-power, and the ALK unit operates according to its sub-power. At each preset power level, the PEM and ALK units maintain a constant power and operate continuously for a preset duration to ensure that the system reaches thermal and water balance.

[0135] During static performance testing, operating data for the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system at each preset power level can be acquired. This operating data constitutes a set of parameters characterizing the steady-state performance of the ALK-PEM hybrid hydrogen production system, including at least current, voltage, temperature, pressure, hydrogen production efficiency, gas purity, and water replenishment rate. Current includes the input current of the PEM unit, the input current of the ALK unit, and the total input current of the ALK-PEM hybrid hydrogen production system. Voltage includes the terminal voltage of the PEM unit, the terminal voltage of the ALK unit, and the total input voltage of the ALK-PEM hybrid hydrogen production system. Temperature includes the electrolysis chamber temperature of the PEM unit, the electrolyte temperature of the ALK unit, and the outlet hydrogen temperature of the ALK-PEM hybrid hydrogen production system. Pressure includes the anode and cathode pressures of the PEM unit, the hydrogen-side and oxygen-side pressures of the ALK unit, and the system pressure of the ALK-PEM hybrid hydrogen production system. Hydrogen production efficiency includes the hydrogen production efficiency of the PEM unit, the hydrogen production efficiency of the ALK unit, and the overall hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system. Gas purity includes the hydrogen purity produced by the PEM unit, the hydrogen purity produced by the ALK unit, and the final hydrogen purity after purification by the ALK-PEM hybrid hydrogen production system. Water replenishment rate includes the water replenishment flow rate of the PEM unit, the water replenishment flow rate of the ALK unit, and the total water replenishment flow rate of the ALK-PEM hybrid hydrogen production system. This operational data is collected using a multi-point synchronous sampling method, with continuous data collection at the end of the stable operation phase of each preset power level. The average value within the collection period is taken as the representative value for that power level.

[0136] Based on the current and voltage at each preset power level, volt-ampere characteristic curves are fitted to obtain the electrochemical efficiency of the PEM unit, the ALK unit, and the ALK-PEM hybrid hydrogen production system. The volt-ampere characteristic curves describe the relationship between the electrolyzer voltage and current density. For the PEM unit, with current density as the abscissa and voltage as the ordinate, the current density and voltage values ​​at each power level are used as data points, and the least squares method is used to fit the polarization curve of the PEM unit. Based on the polarization curves, the voltage efficiency of the PEM unit at each power level is calculated; the voltage efficiency is the ratio of the theoretical decomposition voltage to the actual voltage. The voltage efficiency at each power level is weighted and averaged to obtain the overall electrochemical efficiency of the PEM unit. For the ALK unit, the same method is used to fit the polarization curve of the ALK unit based on its current density and voltage values ​​at each power level, and the overall electrochemical efficiency of the ALK unit is calculated. For the ALK-PEM hybrid hydrogen production system as a whole, with the total current as the abscissa and the total voltage as the ordinate, the overall volt-ampere characteristic curve is fitted, and the overall comprehensive electrochemical efficiency is calculated. Electrochemical efficiency is used to evaluate the electrical performance of an electrolyzer in converting electrical energy into chemical energy, specifically reflecting the activation overpotential level and ohmic overpotential level of the electrolyzer.

[0137] Based on the temperature at each power level, the temperature rise rate and temperature deviation can be calculated to determine the thermal management efficiency of the ALK-PEM hybrid hydrogen production system. The temperature rise rate is the rate of temperature change over time during the electrolyzer's transition from cold start-up to stable operation. For each preset power level, the temperature change curve from start-up to stabilization is obtained, and the average slope of the curve is calculated as the temperature rise rate at that level. The temperature deviation is the ratio of the absolute value of the difference between the actual operating temperature and the preset target temperature to the preset target temperature. For each preset power level, the average temperature during stable operation is obtained, and the difference between this temperature and the preset operating temperature is calculated. This difference is then divided by the preset operating temperature to obtain the temperature deviation. A comprehensive evaluation of the temperature rise rate and temperature deviation at each level indicates higher thermal management efficiency. The greater the number of levels with a temperature rise rate below a preset threshold and a temperature deviation within a preset range, the higher the thermal management efficiency. Thermal management efficiency is used to evaluate the electrolyzer's ability to balance heat generation and heat removal, specifically reflecting the rationality of the electrolyzer's heat dissipation design and the accuracy of temperature control.

[0138] Based on the water replenishment rate and gas purity at each power level, the water consumption for hydrogen production can be calculated to determine the water balance efficiency of the ALK-PEM hybrid hydrogen production system. Hydrogen production water consumption is the amount of pure water consumed per unit of hydrogen production. For each preset power level, the stable water replenishment rate and stable hydrogen production at that level are obtained. The water replenishment rate is divided by the hydrogen production to obtain the hydrogen production water consumption at that level. The hydrogen production water consumption at each level is compared with the theoretical water consumption, which is calculated based on the stoichiometric ratio of the water electrolysis chemical reaction. The portion of hydrogen production water consumption exceeding the theoretical water consumption represents additional water loss, specifically including water vapor loss carried out by hydrogen, water vapor loss carried out by oxygen, and system leakage losses. Combined with gas purity data, the specific reasons for the high hydrogen production water consumption are analyzed. If the hydrogen purity meets the standard but the water consumption is high, it indicates insufficient gas-liquid separation efficiency or severe gas-liquid carryover. If the hydrogen purity does not meet the standard and the water consumption is high, it indicates cross-contamination in the electrolyzer or membrane permeation problems. Water balance efficiency is used to evaluate the water management capability of an electrolyzer, specifically reflecting the electrolyzer's gas-liquid separation efficiency, sealing performance, and water transport characteristics of the membrane electrode.

[0139] The system can comprehensively analyze electrochemical efficiency, thermal management efficiency, and water balance efficiency to generate a static performance test report that includes the main factors contributing to hydrogen production efficiency deviations. The static performance test report includes a summary table of raw data at each preset power level, volt-ampere characteristic curves, temperature change curves, and water replenishment rate change curves. The report quantitatively evaluates electrochemical efficiency, thermal management efficiency, and water balance efficiency, providing scores and grades for each efficiency indicator. The report further analyzes the deviation between hydrogen production efficiency and theoretical efficiency, identifying the main factors causing the efficiency deviation. If the electrochemical efficiency is significantly lower than the benchmark value, and the volt-ampere characteristic curve shows an excessively rapid voltage rise in the high current density region, the main factor is determined to be excessive electrochemical polarization loss, and the optimization direction is to improve catalyst activity or optimize electrode structure. If the thermal management efficiency is low, and the temperature deviation is large, the main factor is determined to be thermal management imbalance, and the optimization direction is to improve heat dissipation design or optimize temperature control strategy. If the water balance efficiency is low, and the water consumption for hydrogen production is significantly higher than the theoretical value, the main factor is determined to be water balance imbalance, and the optimization direction is to improve gas-liquid separator design or optimize water replenishment control logic. After the static performance test report is output, it is used to guide the design optimization and operation parameter debugging of the ALK-PEM hybrid hydrogen production system.

[0140] Static performance tests were conducted by setting multiple preset power levels, comprehensively covering the ALK-PEM hybrid hydrogen production system's operating range from low to full load. Multi-dimensional operational data, including current, voltage, temperature, and water replenishment rate, were acquired at each power level, providing a complete data foundation for subsequent efficiency analysis. Electrochemical efficiency was calculated by fitting volt-ampere characteristic curves, achieving a quantitative assessment of the electrolyzer's electrical performance. Thermal management efficiency was assessed by calculating the temperature rise rate and temperature deviation, achieving a quantitative assessment of the electrolyzer's thermal balance state. Water consumption for hydrogen production was calculated to assess water balance efficiency, achieving a quantitative assessment of the electrolyzer's water management capability. A comprehensive analysis of electrochemical efficiency, thermal management efficiency, and water balance efficiency was performed, identifying the dominant factors contributing to hydrogen production efficiency deviations and achieving precise localization of static performance issues. The generated static performance test report not only provides performance evaluation conclusions but also clearly identifies performance shortcomings and improvement directions, providing direct technical guidance for the iterative optimization of the ALK-PEM hybrid hydrogen production system.

[0141] In some embodiments, step S103 may further include: determining the second operating power of the PEM unit and ALK unit under multiple preset wind and solar fluctuation scenarios based on the multi-type simulated fluctuation power data. Based on this, step S104 may further include: acquiring the second operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system under each preset wind and solar fluctuation scenario; the second operating data includes at least power response speed, hydrogen production efficiency change rate, and gas production purity stability. Correspondingly, step S105 may further include: determining the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system based on the hydrogen production efficiency change rate and power response speed under each preset wind and solar fluctuation scenario; performing time-frequency analysis on the gas production purity under each preset wind and solar fluctuation scenario to obtain the dominant frequency components of gas production purity fluctuation; and generating a dynamic operating condition test report including the dominant factors of hydrogen production efficiency deviation based on the hydrogen production efficiency stability and the dominant frequency components of gas production purity fluctuation.

[0142] Dynamic fluctuation pattern data can be selected from various types of simulated fluctuating power data. This dynamic fluctuation pattern data is a sequence of power values ​​that change continuously over time, specifically including typical power waveform curves and their derived sequences generated according to the wind and solar data processing steps. Based on the dynamic fluctuation pattern data, the operating power of PEM and ALK units under multiple preset wind and solar fluctuation scenarios is determined. The preset wind and solar fluctuation scenarios are typical fluctuation conditions pre-set for the dynamic response test of electrolyzers, specifically including instantaneous change scenarios, short-term ramp scenarios, long-term trend scenarios, and composite fluctuation scenarios. Instantaneous change scenarios correspond to a step change in power within seconds, which can be used to test the rapid response capability of electrolyzers. Short-term ramp scenarios correspond to a continuous increase or decrease in power within minutes, which can be used to test the continuous tracking capability of electrolyzers. Long-term trend scenarios correspond to a slow change in power within hours, which can be used to test the long-term adaptability of electrolyzers. Composite fluctuation scenarios include superimposed fluctuations at multiple time scales, which can be used to test the comprehensive response capability of electrolyzers under multi-scale fluctuation coupling conditions. Under each preset wind and solar fluctuation scenario, the power values ​​undertaken by the PEM unit and the ALK unit are calculated based on the simulated fluctuation power data corresponding to that scenario, and these values ​​are taken as the operating power under that scenario. The PEM unit undertakes the high-frequency fluctuation component, and the ALK unit undertakes the low-frequency fluctuation component. The division between high and low frequencies is determined based on the preset cutoff frequency.

[0143] The PEM and ALK units are controlled to operate at their respective operating power levels under each preset wind and solar fluctuation scenario. The PEM unit adjusts its operating state according to the high-frequency power component it is responsible for, and the ALK unit adjusts its operating state according to the low-frequency power component it is responsible for. Under each preset wind and solar fluctuation scenario, the PEM and ALK units operate continuously according to the fluctuation sequence corresponding to that scenario, with an operating time of not less than the complete cycle of the fluctuation sequence for that scenario, to ensure that dynamic response characteristics are fully captured.

[0144] During dynamic operating condition testing, operational data can be acquired for the PEM unit, the ALK unit, and the ALK-PEM hybrid hydrogen production system under each preset wind and solar fluctuation scenario. This operational data constitutes a set of parameters characterizing the dynamic response performance of the ALK-PEM hybrid hydrogen production system, including at least power response speed, hydrogen production efficiency change rate, and gas purity. Power response speed refers to the rate at which the actual output power of the electrolyzer follows the commanded power change, specifically including the rise response speed and the fall response speed. For the PEM unit, the time interval from the commanded power step change to the actual power reaching 90% of the target value is calculated as the rise response time, and its reciprocal is the rise response speed. The same method is used to calculate the response speed for the ALK unit. For the ALK-PEM hybrid hydrogen production system as a whole, the total power response speed is calculated. The hydrogen production efficiency change rate represents the drastic change in hydrogen production efficiency over time during fluctuations. The efficiency deviation rate at that moment is obtained by dividing the difference between the instantaneous hydrogen production efficiency and the baseline efficiency at each sampling moment by the baseline efficiency. Calculate the maximum, average, and standard deviation of the efficiency deviation rate throughout the entire fluctuation process. The maximum value represents the peak amplitude of the efficiency fluctuation, the average value represents the overall deviation level of efficiency, and the standard deviation represents the dispersion of the efficiency fluctuation. The produced gas purity is the specific numerical sequence of the produced hydrogen purity during the fluctuation process. Hydrogen purity data is continuously collected during the fluctuation process, and the maximum, minimum, average, and standard deviation of the purity are recorded.

[0145] The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system was determined based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios. Hydrogen production efficiency stability refers to the electrolyzer's ability to maintain high-efficiency operation under power fluctuation conditions. For each preset wind and solar fluctuation scenario, a correlation analysis was performed between the rate of change of hydrogen production efficiency and the power response speed. A scatter plot of response speed versus efficiency change rate was plotted with power response speed on the x-axis and hydrogen production efficiency change rate on the y-axis. The slope of the fitted curve of the scatter plot was calculated; a smaller absolute value of the slope indicates lower sensitivity of efficiency to response speed and higher hydrogen production efficiency stability. The slopes of the fitted curves for each scenario were compared to identify the scenario type with the worst hydrogen production efficiency stability. If the absolute value of the slope in the instantaneous change scenario is significantly greater than in other scenarios, it indicates that the electrolyzer's ability to maintain efficiency under rapid power changes is insufficient. If the average efficiency change rate is significantly higher in the short-term ramp scenario, it indicates that the efficiency degradation of the electrolyzer is more severe during continuous tracking. If the standard deviation of the efficiency change rate is significantly larger in the combined fluctuation scenario, it indicates that the efficiency fluctuation of the electrolyzer is more severe under multi-scale coupled fluctuations.

[0146] Time-frequency analysis can be performed on the gas production purity under various preset wind and solar fluctuation scenarios to obtain the dominant frequency components of the gas production purity fluctuations. The time-frequency analysis uses short-time Fourier transform or wavelet transform methods to convert the gas production purity signal in the time domain into an energy distribution in the frequency domain. A Fourier transform is performed on the gas production purity time series for each scenario to obtain the spectrum of the purity signal. The horizontal axis of the spectrum represents frequency, and the vertical axis represents the amplitude of the corresponding frequency component. One or more frequency components with the largest amplitude in the spectrum are identified as the dominant frequency components of the gas production purity fluctuations. The dominant frequency components reflect the main sources of disturbance causing the gas production purity fluctuations. If the dominant frequency components are concentrated in the high-frequency band, it indicates that the gas production purity fluctuations are mainly affected by rapid power changes, which may be related to the fast response characteristics of the PEM unit. If the dominant frequency components are concentrated in the mid-frequency band, it indicates that the gas production purity fluctuations are mainly affected by short-term power ramp-ups, which may be related to power allocation strategies. If the dominant frequency components are concentrated in the low-frequency band, it indicates that the gas production purity fluctuations are mainly affected by long-term power trends, which may be related to the slow changes in the system's thermal and water balances. By comparing the dominant frequency components under different scenarios, the influence of different fluctuation scenarios on gas production purity is analyzed.

[0147] Based on the dominant frequency components of hydrogen production efficiency stability and gas purity fluctuations, a dynamic operating condition test report can be generated, including the dominant factors causing hydrogen production efficiency deviations. The dynamic operating condition test report includes raw data curves for each preset wind and solar power fluctuation scenario, a power response rate statistics table, a hydrogen production efficiency change rate statistics table, and gas purity analysis charts. The report quantitatively assesses the stability of hydrogen production efficiency under each scenario, provides the correlation analysis results of response rate and efficiency change rate, and marks the scenario type with the worst hydrogen production efficiency stability. The report displays the time-frequency analysis results of gas purity fluctuations, marking the dominant frequency components and their amplitudes for each scenario. The report further analyzes the deviation between hydrogen production efficiency and the baseline efficiency, identifying the dominant factors causing efficiency deviations under dynamic operating conditions. If the hydrogen production efficiency stability analysis shows a significant decrease in efficiency under instantaneous change scenarios, and the dominant frequency components of gas purity fluctuations match the power change frequency, the dominant factor is determined to be PEM unit response lag, and the optimization direction is to improve the control algorithm of the PEM unit or enhance the dynamic response capability of the PEM unit. If the hydrogen production efficiency stability analysis shows that the efficiency continuously declines under short-term ramp-up scenarios, and the dominant frequency component of the gas purity fluctuation is related to the ramp-up rate, then the dominant factor is identified as the accumulation of tracking error in the ALK unit. The optimization direction is to optimize the power allocation strategy or improve the regulation characteristics of the ALK unit. If the hydrogen production efficiency stability analysis shows that the efficiency fluctuates drastically under complex fluctuation scenarios, and the gas purity fluctuation has multiple dominant frequency components, then the dominant factor is identified as multi-scale coupled control misalignment. The optimization direction is to develop a multi-timescale coordinated control strategy. After the dynamic operating condition test report is output, it is used to guide the optimization of the control strategy and adjustment of operating parameters of the ALK-PEM hybrid hydrogen production system under fluctuating operating conditions.

[0148] Dynamic operating condition tests were conducted by setting up multiple preset wind and solar fluctuation scenarios, comprehensively covering various fluctuation conditions from instantaneous changes to long-term trends. Multi-dimensional dynamic response data, including power response speed, hydrogen production efficiency change rate, and gas purity, were acquired under each scenario, providing a complete data foundation for subsequent analysis. Correlation analysis between the hydrogen production efficiency change rate and power response speed determined the stability of hydrogen production efficiency, achieving a quantitative assessment of the electrolyzer's ability to maintain dynamic efficiency. Time-frequency analysis of gas purity yielded the dominant frequency components, enabling frequency domain localization of the sources of gas quality fluctuations. Comprehensive analysis of the dominant frequency components of hydrogen production efficiency stability and gas purity fluctuations identified the dominant factors of efficiency deviations under dynamic operating conditions, achieving precise tracing of dynamic performance issues. The generated dynamic operating condition test report not only provides dynamic performance evaluation conclusions but also clearly identifies performance shortcomings and improvement directions, providing direct technical guidance for the optimized operation of the ALK-PEM hybrid hydrogen production system under wind and solar fluctuation scenarios.

[0149] In some embodiments, step S103 may further include: determining the third operating power of the PEM unit and ALK unit under multiple preset power allocation ratios based on the multiple types of simulated fluctuating power data. Based on this, step S104 may further include: acquiring third operating data of the PEM unit and ALK unit under each preset power allocation ratio; the third operating data includes at least coupling response speed, power allocation accuracy, and hydrogen production efficiency. Correspondingly, step S105 may further include: performing correlation analysis on the coupling response speed of the PEM unit and ALK unit under each preset power allocation ratio to obtain the rate of change matching coefficient during the power handover process between the PEM unit and ALK unit; calculating the cooperative efficiency loss coefficient of the PEM unit and ALK unit based on the hydrogen production efficiency under each preset power allocation ratio; and generating a cooperative operation test report including the dominant factors of hydrogen production efficiency deviation based on the power allocation accuracy, rate of change matching coefficient, and cooperative efficiency damage coefficient.

[0150] Collaborative test mode data can be selected from various types of simulated fluctuating power data. This collaborative test mode data includes a stepped change sequence with multiple power steps or a ramp sequence with continuous power changes. Based on the collaborative test mode data, the operating power of the PEM unit and ALK unit under multiple preset power allocation ratios is determined. The preset power allocation ratios are different ways in which the PEM unit and ALK unit share the total power during collaborative operation testing. Specifically, these can include an allocation method where the PEM unit bears a higher power ratio, an allocation method where the PEM unit and ALK unit bear a balanced power ratio, and an allocation method where the ALK unit bears a higher power ratio. Under each preset power allocation ratio, multiple total power levels are set, including 30%, 50%, 70%, and 100% of the rated power of the ALK-PEM hybrid hydrogen production system. For each total power level under each preset power allocation ratio, the sub-power that the PEM unit should bear and the sub-power that the ALK unit should bear are calculated according to the allocation ratio. The sum of the sub-power of the PEM unit and the sub-power of the ALK unit equals the total power level value. The PEM unit sub-power and ALK unit sub-power corresponding to each total power level under each preset power allocation ratio are used as the operating power in the collaborative operation test.

[0151] The PEM and ALK units operate at their respective preset power allocation ratios. Under each preset power allocation ratio, the PEM unit operates sequentially according to the corresponding total power levels, and the ALK unit operates synchronously according to the corresponding total power level sequence. At each total power level, the PEM and ALK units maintain a constant power level and operate continuously for a preset duration. During the transition from one total power level to another, the PEM and ALK units adjust their power according to a preset power handover strategy. This strategy specifies either a synchronization relationship between a decrease in PEM unit power and an increase in ALK unit power, or a synchronization relationship between an increase in PEM unit power and a decrease in ALK unit power.

[0152] During the collaborative operation test, operational data for the PEM and ALK units under each preset power allocation ratio was acquired. This operational data is a set of parameters characterizing the collaborative performance of the PEM and ALK units, including at least coupling response speed, power allocation accuracy, and hydrogen production efficiency. Coupling response speed refers to the speed at which the PEM and ALK units cooperate to complete power transfer during power handover. For each power handover event, the actual power of the PEM and ALK units at the start of the handover and at the end of the handover were recorded. The time interval from the start to the end of the handover was calculated as the coupling response time, and its reciprocal is the coupling response speed. If power overshoot or power oscillation occurs during the power handover, the overshoot amount and the number of oscillations are recorded as auxiliary evaluation indicators of coupling response quality. Power allocation accuracy refers to the degree of closeness between the actual power undertaken by the PEM and ALK units and the commanded power. For each total power level under each preset power allocation ratio, the deviation between the actual power of the PEM unit and the commanded power of the PEM unit under that ratio is calculated, and the deviation between the actual power of the ALK unit and the commanded power of the ALK unit under that ratio is also calculated. The deviation is divided by the corresponding commanded power to obtain the relative deviation percentage, which serves as a quantitative indicator of power allocation accuracy. Power allocation accuracy includes steady-state accuracy and dynamic accuracy. Steady-state accuracy can be the deviation during a constant power phase, while dynamic accuracy can be the maximum instantaneous deviation during power transfer. Hydrogen production efficiency includes the instantaneous hydrogen production efficiency of the PEM unit, the instantaneous hydrogen production efficiency of the ALK unit, and the overall hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system. During the stable operation phase of each total power level, the hydrogen production and input power of each unit are continuously collected, and the average hydrogen production efficiency of each unit and the overall average hydrogen production efficiency are calculated.

[0153] Correlation analysis was performed on the coupling response speeds of PEM and ALK units under various preset power allocation ratios to obtain the rate-of-change matching coefficient during the power handover process. Correlation analysis characterizes the degree of synchronization between the power change rates of the PEM and ALK units. For each power handover event, the rate-of-change curves of the PEM and ALK unit power over time were extracted. The Pearson correlation coefficient between the two rate-of-change curves during the handover period was calculated as the rate-of-change matching coefficient. The rate-of-change matching coefficient ranges from -1 to +1. The closer the absolute value of the coefficient is to +1, the higher the correlation between the power change rates of the PEM and ALK units. In the power handover scenario, when one unit's power decreases while the other unit's power increases, their rates of change should be negatively correlated. Therefore, a correlation coefficient close to -1 indicates good synchronization, and the closer the absolute value of the coefficient is to +1 (it can be an absolute value), the better the synchronization. The further the absolute value of the coefficient deviates from +1, the more it indicates a rate mismatch during the power handover process, which may lead to total power fluctuations or system oscillations. The rate-of-change matching coefficients for each power handover event under each preset power allocation ratio are summarized, and the average matching coefficient and standard deviation of the matching coefficient are calculated for each allocation ratio. The average matching coefficient represents the overall synchronization level under that allocation ratio, and the standard deviation of the matching coefficient represents the stability of the synchronization level.

[0154] Based on the hydrogen production efficiency under each preset power allocation ratio, the synergistic efficiency loss coefficient of the PEM and ALK units is calculated. The synergistic efficiency loss coefficient represents the degree of decrease in overall efficiency due to the synergistic operation of the PEM and ALK units relative to the weighted sum of the individual unit efficiencies. For each total power level under each preset power allocation ratio, the baseline efficiency of the PEM unit operating independently at that power level and the baseline efficiency of the ALK unit operating independently at that power level are obtained. A weighted sum of the baseline efficiencies of the PEM and ALK units is calculated based on this allocation ratio, with the weights representing the power proportion undertaken by each unit. The overall hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system actually measured under this allocation ratio is compared with the aforementioned weighted sum, and the difference is calculated. This difference is divided by the weighted sum to obtain the relative efficiency loss value. The relative efficiency loss values ​​under each total power level are averaged to obtain the synergistic efficiency loss coefficient under that preset power allocation ratio. A positive synergistic efficiency loss coefficient indicates that synergistic operation leads to a decrease in efficiency; a larger coefficient indicates a more severe synergistic efficiency loss. A negative collaborative efficiency loss coefficient indicates that collaborative operation actually improves overall efficiency, suggesting a positive coupling effect between the PEM and ALK units. By comparing the collaborative efficiency loss coefficients under various preset power allocation ratios, the allocation ratios with the minimum and maximum efficiency loss are identified.

[0155] Based on power allocation accuracy, rate of change matching coefficient, and collaborative efficiency loss coefficient, a collaborative operation test report can be generated, including the dominant factors causing hydrogen production efficiency deviation. The collaborative operation test report includes raw operating data curves for each preset power allocation ratio, a power allocation accuracy statistics table, a coupling response speed statistics table, a rate of change matching coefficient analysis chart, and a hydrogen production efficiency comparison chart. The report quantitatively evaluates the power allocation accuracy for each allocation ratio, providing specific values ​​for steady-state and dynamic accuracy, and marking the power levels and handover events where accuracy exceeds tolerance. The report displays the rate of change matching coefficient for each allocation ratio, marking scenarios with low matching coefficients and poor power handover smoothness. The report compares the collaborative efficiency loss coefficients for each allocation ratio, marking the allocation ratios with the minimum and maximum efficiency loss. The report further analyzes the deviation between hydrogen production efficiency and theoretical efficiency, identifying the dominant factors causing efficiency deviations under collaborative operation conditions. If the power allocation accuracy analysis shows that the dynamic accuracy is significantly low at a certain allocation ratio, and the rate of change matching coefficient analysis shows that the absolute value of the matching coefficient deviates significantly from positive one at the same ratio, then the dominant factor is determined to be the dynamic tracking error in the power handover process. The optimization direction is to improve the dynamic response parameters of the power allocation controller or optimize the feedforward compensation algorithm. If the collaborative efficiency loss coefficient analysis shows that the loss coefficient is significantly high at a certain allocation ratio, but the power allocation accuracy and rate of change matching coefficient are both within the normal range, then the dominant factor is determined to be the coupling loss caused by the mismatch of unit characteristics. The optimization direction is to adjust the operating range of the PEM unit and the ALK unit so that they operate in their respective high-efficiency regions as much as possible. If the collaborative efficiency loss coefficient is high at all allocation ratios, and the rate of change matching coefficient fluctuates greatly at different allocation ratios, then the dominant factor is determined to be the unreasonable design of the power handover strategy. The optimization direction is to redesign the rate curve and timing logic of the power handover and develop a collaborative optimization algorithm based on model predictive control. After the collaborative operation test report is output, it is used to guide the optimization of the power allocation strategy and the tuning of control parameters of the ALK-PEM hybrid hydrogen production system in the multi-unit collaborative operation mode.

[0156] By setting multiple preset power allocation ratios for collaborative operation testing, the system comprehensively covered various cooperation modes between the PEM and ALK units, from light to heavy loads and from balanced to biased power sharing. Multi-dimensional collaborative operation data, including coupling response speed, power allocation accuracy, and hydrogen production efficiency, were acquired under each allocation ratio, providing a complete data foundation for subsequent analysis. Correlation analysis of the coupling response speed yielded a rate-of-change matching coefficient, enabling a quantitative assessment of synchronization during power handover. A collaborative efficiency loss coefficient was calculated based on hydrogen production efficiency, achieving a quantitative assessment of efficiency losses caused by collaborative operation. A comprehensive analysis of power allocation accuracy, rate-of-change matching coefficient, and collaborative efficiency loss coefficient identified the dominant factors contributing to efficiency deviations under collaborative operation conditions, enabling precise localization of collaborative performance issues. The generated collaborative operation test report not only provides collaborative performance evaluation conclusions but also clearly identifies performance shortcomings and improvement directions, offering direct technical guidance for the optimized operation of the ALK-PEM hybrid hydrogen production system in unit collaborative mode.

[0157] In some embodiments, step S103 may further include: determining the fourth operating power of the PEM unit and ALK unit under multiple preset start-stop conditions based on the multiple types of simulated fluctuation power data. Based on this, step S104 may further include: acquiring the fourth operating data of the PEM unit and ALK unit under each preset start-stop condition; the fourth operating data includes at least start-stop response time, start-stop process energy consumption, electrode temperature change rate, and sealing performance stability. Accordingly, step S105 may further include: statistically analyzing the start-stop response times of the PEM unit and ALK unit under each preset start-stop condition to obtain the consistency of the start-stop response of the PEM unit and ALK unit; extracting the peak value of the electrode temperature change rate under each preset start-stop condition to obtain the thermal stress influence of the PEM unit and ALK unit; performing trend fitting on the sealing performance stability under each preset start-stop condition to obtain the sealing reliability of the PEM unit and ALK unit; statistically analyzing the energy consumption of the start-stop process under each preset start-stop condition to identify the start-stop stage where abnormal impact events occur; and generating a start-stop characteristic test report including the dominant factors of hydrogen production efficiency deviation based on the start-stop response consistency, thermal stress influence, sealing reliability, and abnormal event identification results.

[0158] Start-up and shutdown test mode data can be selected from various types of simulated fluctuating power data. This start-up and shutdown test mode data consists of a sequence of switches including start-up and shutdown commands. Based on the start-up and shutdown test mode data, the operating power of the PEM and ALK units under multiple preset start-up and shutdown conditions is determined. The preset start-up and shutdown conditions are test conditions specifically set for the start-up and shutdown processes of the electrolyzer, specifically including light-load start-up condition, rated start-up condition, temporary shutdown condition, and long-term shutdown condition. The light-load start-up condition corresponds to the operation process of the ALK-PEM mixed hydrogen production system from a shutdown state to 30% of rated power. The rated start-up condition corresponds to the operation process of the ALK-PEM mixed hydrogen production system from a shutdown state to 100% of rated power. The temporary shutdown condition corresponds to the process of the ALK-PEM mixed hydrogen production system rapidly shutting down from rated power and maintaining the shutdown state for 5 minutes before restarting. The long-term shutdown condition corresponds to the process of the ALK-PEM mixed hydrogen production system executing a complete shutdown procedure from rated power and maintaining the shutdown state for 30 minutes before restarting. Each preset start-stop condition includes some or all of the following stages: startup, stable operation, shutdown, shutdown hold, and restart. For each preset start-stop condition, based on the start-stop test mode data corresponding to that condition, the power change trajectories of the PEM and ALK units throughout the entire start-stop process are determined as the operating power under that condition. The power change trajectory of the PEM unit includes the power ramp-up curve during startup, the power drop curve during shutdown, and the zero-power state during the shutdown hold phase. The power change trajectory of the ALK unit also includes the corresponding ramp-up curve, drop curve, and zero-power state. Each preset start-stop condition is executed a preset number of times, with a minimum of three cycles, to evaluate the repeatability of the start-stop process and the performance degradation trend.

[0159] The PEM and ALK units operate at their respective power levels according to preset start-up and shutdown conditions. In each cycle of each preset start-up and shutdown condition, the PEM unit performs start-up, stable operation, and shutdown operations according to its power change trajectory, while the ALK unit synchronously performs the corresponding operation according to its power change trajectory. During start-up, the PEM and ALK units gradually increase power at a preset start-up rate until the target power value is reached. During shutdown, the PEM and ALK units gradually decrease power at a preset shutdown rate until the power drops to zero. During the shutdown hold phase, the PEM and ALK units maintain a zero-power state, and the system enters standby mode. During restart, the PEM and ALK units again increase power at the start-up rate to restore the target power value.

[0160] During the start-stop characteristic test, operating data were acquired for the PEM unit, the ALK unit, and the ALK-PEM hybrid hydrogen production system under each preset start-stop condition. This operating data is a set of parameters characterizing the performance and reliability changes of the ALK-PEM hybrid hydrogen production system during start-stop, including at least start-stop response time, start-stop energy consumption, electrode temperature change rate, and sealing performance stability. Start-stop response time is the time required for the electrolyzer to complete a start-up or shutdown operation. Start-up response time can be the time interval from receiving the start-up command to the moment the actual power reaches 90% of the target power. Shutdown response time can be the time interval from receiving the shutdown command to the moment the actual power drops to 10% of the rated power. For each cycle of each start-stop condition, the start-up response time of the PEM unit, the shutdown response time of the PEM unit, the start-up response time of the ALK unit, and the shutdown response time of the ALK unit were recorded. Start-stop energy consumption is the total electrical energy consumed or recovered by the electrolyzer during the start-up and shutdown processes. Start-up energy consumption is calculated by integrating the instantaneous power during the start-up phase over time, with the integration interval from the start-up command to the power stabilization point. Shutdown energy consumption is also calculated by integrating the instantaneous power during the shutdown phase over time, with the integration interval from the shutdown command to the power returning to zero. Shutdown energy consumption includes two parts: the electrical energy consumed by the electrolyzer during continued hydrogen production as the power decreases, and the electrical energy consumed by the auxiliary systems to maintain operation. The electrode temperature change rate is the rate at which the temperature of the electrolyzer electrode region changes over time during start-up and shutdown. Temperature sensors are placed at the proton exchange membrane electrode in the PEM unit and the electrode in the ALK unit to continuously collect temperature data throughout the start-up and shutdown process. The average rate of temperature rise from the initial value to the stable value during the start-up phase is calculated, and the average rate of temperature drop from the stable value to the cooling value during the shutdown phase is calculated. The maximum instantaneous values ​​of the temperature rise rate and the maximum instantaneous value of the temperature drop rate are extracted as evaluation indicators of thermal shock intensity. Sealing performance stability is the ability of the electrolyzer to maintain airtightness and liquid tightness during start-up and shutdown. Throughout the start-up and shutdown process, the pressure difference between the hydrogen and oxygen sides of the PEM unit and the ALK unit, as well as the leakage detection signals at each sealing flange connection and the liquid level changes in the separation system, are continuously monitored. The pressure difference fluctuation amplitude, the number of leakage signal triggers, and the amplitude of abnormal liquid level fluctuations are recorded during start-up and shutdown. After each start-up and shutdown cycle, a static pressure holding test is performed to measure the rate of pressure drop within a preset time, which serves as a quantitative evaluation indicator of sealing performance.

[0161] Statistical analysis was performed on the start-stop response times of the PEM and ALK units under various preset start-stop conditions to obtain the consistency of their start-stop responses. Start-stop response consistency refers to the degree of matching between the response speeds of the PEM and ALK units during startup and shutdown. For each preset start-stop condition, the average and standard deviation of the PEM unit's start-up response time were calculated over multiple cycles, as well as the average and standard deviation of the ALK unit's start-up response time. The average start-up response times of the PEM and ALK units were compared, and their ratio was calculated as the start-up response consistency coefficient; a ratio closer to 1 indicates better start-up response consistency. The shutdown response consistency coefficient was calculated using the same method. The start-up and shutdown response consistency coefficients for each condition were summarized and analyzed to identify the types of conditions with poor response consistency. If the start-up response consistency is good under light-load start-up conditions but significantly worsens under rated start-up conditions, it indicates that the difference in response characteristics between the PEM and ALK units is amplified during high-power startup. If the shutdown response is consistent under temporary shutdown conditions, but the startup response consistency deteriorates after long-term shutdown conditions, it indicates that long-term shutdown has a greater impact on the startup performance of a certain unit.

[0162] Peak values ​​of electrode temperature change rates under various preset start-up and shutdown conditions were extracted to determine the thermal stress impact on the PEM and ALK cells. Thermal stress impact refers to the degree of thermal shock to the electrolyzer electrodes caused by drastic temperature changes during start-up and shutdown. For each cycle of each preset start-up and shutdown condition, the maximum temperature rise rate and maximum temperature fall rate were extracted from the PEM cell electrode temperature curve, and the maximum temperature rise rate and maximum temperature fall rate were extracted from the ALK cell electrode temperature curve. The maximum temperature rise rate for each cycle was averaged to obtain the average peak temperature rise rate for the PEM cell and the average peak temperature rise rate for the ALK cell under that condition. A higher temperature rise rate indicates a stronger thermal shock experienced by the electrodes during startup, which may lead to increased thermal expansion stress on the membrane electrode or a greater risk of catalyst layer detachment. Comparing the peak temperature rise rates under each condition identified the type of condition with the most severe thermal stress impact. If the peak temperature rise rate under the rated start-up condition is significantly higher than that under the light load start-up condition, it indicates that rapid high-power startup will bring a stronger thermal shock. If the peak temperature rise rate after a long-term shutdown is higher than that after a temporary shutdown, it indicates that the electrode temperature tends to match the ambient temperature after prolonged cooling, and the larger temperature difference at startup leads to enhanced thermal shock. The peak temperature rise rate is compared with the allowable temperature change rate threshold of the electrolytic cell design to assess whether the thermal stress is within a safe range.

[0163] The sealing performance stability under various preset start-stop conditions was trend-fitted to obtain the sealing reliability of the PEM and ALK units. Sealing reliability refers to the ability of the electrolytic cell to maintain and decrease its sealing performance during repeated start-stop cycles. For each preset start-stop condition, the static pressure holding test results of multiple cycles were sorted by the number of cycles, and a sealing performance change curve was plotted with the number of cycles as the x-axis and the rate of pressure drop as the y-axis. The curve was fitted using linear regression or exponential regression to obtain the trend equation of sealing performance change with the number of start-stop cycles. The slope or exponential coefficient of the fitted curve represents the amount of sealing performance decay caused by each start-stop cycle; a larger slope indicates faster sealing performance decay and poorer sealing reliability. The number of cycles in which the sealing performance suddenly decreases was identified, and this sudden point may correspond to the moment of damage or failure of the sealing structure. The fitted curves under various conditions were compared to analyze the differences in the impact of different start-stop modes on sealing performance. If the sealing performance decays slowly under temporary shutdown conditions, but significantly accelerates under long-term shutdown conditions, it indicates that temperature or pressure changes during long-term shutdown have a greater impact on the sealing structure. If the sealing performance degradation curve under rated starting conditions shows an early abrupt change point, it indicates that the thermal shock and pressure shock during high-power starting process accelerate the sealing failure process.

[0164] Statistical analysis of energy consumption during start-up and shutdown under various preset start-up and shutdown conditions is performed to identify the start-up and shutdown stages where abnormal impact events occur. This statistical analysis of energy consumption during start-up and shutdown is used to detect abnormal energy consumption phenomena exceeding the normal range during the start-up and shutdown process. For multiple cycles of each preset start-up and shutdown condition, the average and standard deviation of energy consumption during start-up and shutdown are calculated. The start-up energy consumption of each cycle is compared with the average, and cycles exceeding three times the standard deviation of the average are identified as abnormal start-up energy consumption events. Abnormal shutdown energy consumption events are identified using the same method. For identified abnormal events, detailed operating data for that cycle is replayed to pinpoint the time stage of the abnormality. If the abnormal start-up energy consumption occurs in the initial start-up phase, the possible cause is a delay in the start-up of the PEM or ALK unit, leading to prolonged idling of the auxiliary system. If the abnormal start-up energy consumption occurs during the power ramp-up phase, the possible cause is oscillation in the power regulation system causing the actual power trajectory to deviate from the command trajectory. If the abnormal shutdown energy consumption occurs during the power decline phase, the possible cause is that the electrolyzer operates in the low-power range for too long or the shutdown procedure is not fully executed. If abnormal power consumption during shutdown occurs during the standby phase after power has returned to zero, the possible causes are that the auxiliary system was not shut down according to procedure or that there are abnormally power-consuming devices. Record the stage of the abnormal event and its possible causes as an important basis for evaluating the reliability of the start-up and shutdown process.

[0165] Based on the results of start-stop response consistency, thermal stress impact, sealing reliability, and abnormal event identification, a start-stop characteristic test report is generated, including the dominant factors contributing to hydrogen production efficiency deviation. The report includes raw operating data curves for each preset start-stop condition, start-stop response time statistics, start-stop process energy consumption statistics, electrode temperature change curves, and sealing performance degradation curves. The report quantitatively assesses the start-stop response consistency under each condition, providing start-up response consistency coefficients and shutdown response consistency coefficients, and marking the types of conditions with poor response matching. The report demonstrates the thermal stress impact under each condition, providing peak temperature rise rate and peak temperature fall rate, and marking conditions where thermal shock intensity exceeds the design threshold. The report demonstrates sealing reliability under each condition, providing sealing performance degradation trend equations and degradation rates, and marking abrupt changes in sealing performance. The report summarizes the identification results of abnormal shock events, marking the start-stop stage and frequency of occurrence of abnormal events. The report further analyzes the impact of the start-stop process on the long-term operating performance of the ALK-PEM hybrid hydrogen production system, identifying the dominant factors leading to hydrogen production efficiency deviation. If the start-stop response consistency analysis shows a severe mismatch between the PEM and ALK unit responses, and the anomaly identification shows frequent abnormal energy consumption during the power ramp-up phase, the dominant factor is determined to be power handover misalignment during the start-stop process. The optimization direction is to optimize the start-stop timing control of the PEM and ALK units to make their power changes more synchronized. If the thermal stress impact analysis shows a significantly higher peak temperature rise rate, and the seal reliability analysis shows an early abrupt change in the seal performance decay curve, the dominant factor is determined to be damage to the seal structure caused by thermal shock. The optimization direction is to optimize the power ramp-up rate during the start-stop process, add preheating and precooling stages, and reduce the temperature change rate. If the anomaly identification shows frequent abnormal energy consumption during the shutdown phase, and the start-stop response consistency analysis shows poor shutdown response consistency, the dominant factor is determined to be incomplete shutdown procedure execution or defects in the auxiliary system control logic. The optimization direction is to optimize the shutdown procedure flow to ensure that the auxiliary system shuts down at the appropriate time and reduces standby energy consumption. If the seal reliability analysis under all operating conditions shows that the seal performance decreases linearly with the number of start-stop cycles, and the rate of decrease exceeds the design expectation, then the dominant factor is determined to be a deficiency in the selection of sealing materials or the design of the sealing structure. The optimization direction is to select sealing materials with better temperature resistance or improve the sealing structure design. After the start-stop characteristic test report is output, it is used to guide the optimization of the operation strategy and the improvement of the structural design of the ALK-PEM hybrid hydrogen production system in frequent start-stop application scenarios.

[0166] Start-stop characteristic tests were conducted by setting multiple preset start-stop conditions, comprehensively covering various start-stop modes that may be encountered in actual operation, such as light-load start-up, rated start-up, temporary shutdown, and long-term shutdown. Multi-dimensional start-stop process data, including start-stop response time, energy consumption, electrode temperature change rate, and sealing performance stability, were acquired under each condition, providing a complete data foundation for subsequent analysis. Statistical analysis of the start-stop response time yielded start-stop response consistency, enabling a quantitative assessment of the synergy between the PEM and ALK units during start-stop. Peak value extraction of the electrode temperature change rate revealed the impact of thermal stress, enabling a quantitative assessment of the thermal shock intensity during start-stop. Trend fitting of sealing performance stability yielded sealing reliability, enabling a quantitative assessment of the sealing performance degradation under repeated start-stop conditions. Statistical analysis of start-stop energy consumption identified abnormal impact events, achieving precise capture of abnormal operating conditions during start-stop. Comprehensive analysis of start-stop response consistency, thermal stress impact, sealing reliability, and abnormal event identification results identified the dominant factors causing performance deviations during start-stop, achieving a comprehensive diagnosis of start-stop reliability issues. The generated start-stop characteristic test report not only provides start-stop performance evaluation conclusions, but also clearly marks reliability shortcomings and improvement directions, providing direct technical guidance for the optimized design and operation and maintenance of the ALK-PEM hybrid hydrogen production system in frequent start-stop application scenarios.

[0167] In some embodiments, the method further includes: identifying multiple extreme points of the wind and light sample data; determining multiple key waveform inflection points based on the multiple extreme points, the key waveform inflection points including peak inflection points and trough inflection points; determining the filtering window width of the time interval between each time-series adjacent key waveform inflection points according to the time interval between each time-series adjacent key waveform inflection points; fitting the data points within the filtering window width; and using the fitted value of the center point of the window as the filtering output value of that point to obtain the wind and light filtered data. Based on this, the above-mentioned sampling of the wind and light sample data according to multiple preset time scales to obtain multiple wind and light sampled waveforms at each time scale can specifically include: sampling the wind and light filtered data according to multiple preset time scales to obtain multiple wind and light sampled waveforms at each time scale.

[0168] Extreme point identification can be performed on wind and solar sample data to identify multiple extreme points within the data. Extreme points include local maxima and local minima. A local maximum is defined as a point whose power value is greater than the power values ​​of the two adjacent time points, while a local minimum is defined as a point whose power value is less than the power values ​​of the two adjacent time points. Extreme point identification is achieved by traversing the entire wind and solar sample data sequence and comparing the numerical relationships between the current point and its adjacent points.

[0169] Based on the identified extreme points, several key waveform inflection points are determined. These key inflection points are extreme points that decisively influence the overall waveform shape, specifically including peak inflection points and trough inflection points. Peak inflection points are selected from local maxima based on the criteria that the amplitude of the maximum exceeds a preset amplitude threshold and the waveforms on both sides of the point exhibit a clear upward-downward characteristic. Trough inflection points are selected from local minima based on the criteria that the amplitude of the minimum is below a preset amplitude threshold and the waveforms on both sides of the point exhibit a clear downward-upward characteristic. These key waveform inflection points characterize the main change nodes in the wind and solar power waveform, and adjacent peak and trough inflection points constitute a complete fluctuation range.

[0170] Based on the time interval between key inflection points of adjacent waveforms, the filter window width of the time interval between these key inflection points can be determined. Adjacent key inflection points can be two waveforms that are sequentially adjacent, specifically a peak inflection point and its immediate trough inflection point, or a trough inflection point and its immediate peak inflection point. The time interval is the length of time between these two key inflection points, measured in data points. The filter window width is positively correlated with the time interval. A larger time interval indicates a relatively gentle waveform change within that interval, and the filter window width increases accordingly; a smaller time interval indicates a more drastic waveform change within that interval, and the filter window width decreases accordingly. The filter window width and the time interval can be mapped using a linear proportional relationship or a non-linear function relationship to ensure that the filter window width does not exceed half the length of the time interval.

[0171] Within the width of the filtering window, data points within the window are fitted. The fitting method employs, but is not limited to, polynomial least squares fitting, using the time index of each data point within the window as the independent variable and the power value as the dependent variable to establish a polynomial regression model. The order of the polynomial is determined based on the width of the filtering window; a lower-order polynomial is used when the filtering window is small, and a higher-order polynomial can be used when the filtering window is large. The fitting process only utilizes data points within the current window and does not involve data outside the window.

[0172] The fitted value at the center point of the window is used as the filtered output value for that point, yielding the filtered power value. The center point of the window is the currently processed data point, located in the center of the filtering window. The original power value at that point is replaced with the predicted value from the fitted model, completing the filtering process for that point. This process is repeated for all data points, defining the filtering window, fitting data within the window, and outputting the fitted value at the center point, to obtain complete wind and solar filtered data. The wind and solar filtered data is a power time series after filtering, which retains the main waveform shape of the original wind and solar sample data while eliminating high-frequency noise and abnormal spikes.

[0173] Based on this, wind and solar sample data are sampled at multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale. Specifically, the filtered wind and solar data are sampled at multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale. The preset multiple time scales include instantaneous change scale, short-term ramp scale, and long-term trend scale, each time scale corresponding to a specific sampling interval. Using the filtered wind and solar data as the basis for subsequent sampling ensures that the sampled waveform sequences have undergone noise filtering, resulting in clearer and more reliable waveform shapes.

[0174] By identifying extreme points and screening key inflection points in the waveform, the main change nodes of the wind and solar power waveform were accurately located, providing a basis for determining the subsequent adaptive filtering window. The width of the filtering window was dynamically adjusted according to the time interval between adjacent inflection points, enabling the filtering window to adaptively match the local change characteristics of the waveform. A wider window was used in smooth waveform regions to achieve sufficient smoothing, while a narrower window was used in regions of drastic waveform change to retain detailed features. Polynomial fitting was performed within the adaptive window, and the fitted value at the center point was used as the filtered output. This utilized the statistical information of the local data while ensuring the temporal correspondence between the filtered result and the original data. By traversing all data points and processing them point by point, complete wind and solar power filtered data was obtained, providing a high-quality data foundation for subsequent multi-scale sampling and waveform analysis. The generated filtered data effectively preserved the physically meaningful inflection point information in the original waveform while eliminating measurement noise and random disturbances, improving the accuracy and reliability of subsequent waveform clustering and typical curve generation.

[0175] In some embodiments, determining the thermal management efficiency of the ALK-PEM hybrid hydrogen production system may further include: determining the thermal management efficiency of the ALK-PEM hybrid hydrogen production system using the following formula: ; In the formula, Thermal management efficiency of the ALK-PEM hybrid hydrogen production system; and These are the temperature rise rates for the PEM and ALK units, respectively. and These are the actual operating temperatures of the PEM and ALK units, respectively. and These are the theoretical optimal operating temperatures for the PEM and ALK cells, respectively. The operating temperature range of the ALK-PEM hybrid hydrogen production system; , and This is a preset constant.

[0176] Both the PEM and ALK units generate heat during hydrogen production, and their operating temperatures directly affect electrochemical reaction efficiency, membrane electrode lifetime, and sealing performance. Temperature data must include at least the temperature rise rate of the PEM unit, the temperature rise rate of the ALK unit, the actual operating temperature of the PEM unit, and the actual operating temperature of the ALK unit. The temperature rise rate is the change in temperature per unit time, reflecting the unit's thermal response characteristics and heat dissipation capacity under power input. The actual operating temperature is the temperature value of the unit during steady-state operation, reflecting the thermal equilibrium state of the unit at a specific power level.

[0177] The theoretical optimal operating temperature for both the PEM and ALK units can be obtained. The theoretical optimal operating temperature is a preferred temperature value within the operating temperature range that maximizes the unit's hydrogen production efficiency, determined based on equipment design parameters or factory calibration. The theoretical optimal operating temperature can be used to assess whether the actual operating temperature falls within the high-efficiency operating range.

[0178] The operating temperature range of the ALK-PEM hybrid hydrogen production system can be obtained. The operating temperature range is the difference between the equipment's minimum and maximum permissible operating temperatures, determined by equipment design specifications or safe operating procedures. The operating temperature range can be used to normalize temperature deviations, making temperature deviation indicators comparable across different equipment or operating conditions.

[0179] The thermal management efficiency of the ALK-PEM hybrid hydrogen production system can be determined using the above formula. Where, This refers to the thermal management efficiency of the ALK-PEM hybrid hydrogen production system. The value ranges from 0 to 1, with a higher value indicating better thermal management performance. and These are the temperature rise rates for the PEM and ALK units, respectively, and the unit can be degrees Celsius per minute. This is a very small positive number used to avoid the denominator being zero. The first term of the formula is the temperature rise rate matching term, where the numerator is the absolute difference between the temperature rise rates of the PEM unit and the ALK unit, and the denominator is the sum of their temperature rise rates. When the temperature rise rates of the two units are similar, this term approaches 0, indicating good thermal response matching; when the temperature rise rates of the two units differ significantly, this term increases, indicating poor thermal response matching, which may lead to thermal stress or thermal coupling losses within the system. and These are the actual operating temperatures of the PEM and ALK units, respectively, and the unit can be degrees Celsius. and These are the theoretical optimal operating temperatures for the PEM and ALK units, respectively, in degrees Celsius. This represents the operating temperature range of the ALK-PEM hybrid hydrogen production system, expressed in degrees Celsius. The second term in the formula is the temperature deviation term, where the numerator is the sum of the absolute deviations of the actual operating temperatures of the PEM and ALK units from their theoretical optimal operating temperatures, and the denominator is the system's operating temperature range. When the actual operating temperatures of each unit are close to the theoretical optimal operating temperatures, this term approaches 0, indicating good thermal stability; when the actual operating temperatures deviate from the theoretical optimal operating temperatures, this term increases, indicating poor thermal stability, which may affect electrochemical efficiency and equipment lifespan. , All are preset constants. Weights are assigned to the rate of temperature rise. The weighting is based on temperature deviation. , The sum is 1. , The value can be preset according to the requirements of the test scenario. For example, it can be increased in start-stop feature testing. To highlight the importance of temperature rise rate matching, it can increase the efficiency of long-term steady-state operation tests. To highlight the importance of temperature deviation.

[0180] By introducing a temperature rise rate matching term and a temperature deviation term, the thermal management efficiency is elevated from a single-unit temperature indicator to a system-level indicator of the thermal coupling and synergy between the PEM and ALK units. The temperature rise rate matching term quantifies the degree of coordination between the two units in terms of thermal response speed, reflecting the internal thermal stress risk caused by differences in thermal inertia and heat dissipation capacity among different units in the ALK-PEM hybrid hydrogen production system. The temperature deviation term quantifies the degree of deviation of the actual operating temperature of each unit from the optimal efficiency temperature, reflecting the temperature control accuracy and stability of the thermal management system. Through the weighted combination of these two terms, the thermal management performance of the ALK-PEM hybrid hydrogen production system under steady-state operation can be comprehensively evaluated, accurately identifying whether thermal management shortcomings stem from thermal response mismatch or excessive temperature deviation, providing a quantitative basis for heat dissipation system design and thermal control strategy optimization.

[0181] In some embodiments, the determination of the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system based on the rate of change of hydrogen production efficiency and the power response speed under each preset wind and solar fluctuation scenario may further include: determining the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system using the following formula based on the rate of change of hydrogen production efficiency and the power response speed under each preset wind and solar fluctuation scenario: ; In the formula, To improve the hydrogen production efficiency and stability of the ALK-PEM hybrid hydrogen production system; The actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at time t; The theoretically optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system; Let be the simulated fluctuation power at time t; T be the total test duration. and These are the power response speed and reference power response speed of the ALK-PEM hybrid hydrogen production system, respectively. and These are the rate of change in hydrogen production efficiency and the rate of change in input power, respectively. , and This is a preset constant.

[0182] Real-time operational data of the ALK-PEM hybrid hydrogen production system under various preset wind and solar power fluctuation scenarios can be obtained. These scenarios can include instantaneous change scenarios, short-term ramp-up scenarios, long-term trend scenarios, and multi-scale composite fluctuation scenarios, each corresponding to different power fluctuation characteristics. Real-time operational data includes at least the actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at each moment, the input power at each moment, and the system's power response speed under various power change conditions. The actual hydrogen production efficiency can be calculated from the real-time collected hydrogen production flow rate and input power, reflecting the system's energy conversion capability under specific power input. The input power is the power value at the corresponding moment in the simulated fluctuating power data, used to characterize the intensity of external disturbances the system can withstand. The power response speed can be the time delay from a power command change to the actual power response, reflecting the system's ability to track fluctuating power.

[0183] The theoretical optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system can be obtained. The theoretical optimal hydrogen production efficiency is the ideal efficiency value that the system can achieve at the current power level, without considering dynamic response losses. It can be calculated by weighting the efficiency characteristic curves of the PEM and ALK units and their power allocation ratio. The theoretical optimal hydrogen production efficiency can be used to measure the difference between the actual efficiency and the ideal efficiency, reflecting the efficiency loss caused by the dynamic response process.

[0184] A reference power response rate can be obtained for an ALK-PEM hybrid hydrogen production system. This reference power response rate can be preset based on the rated response characteristics of the PEM and ALK units, such as a weighted average of the second-level response rate of the PEM unit and the minute-level response rate of the ALK unit, or determined according to the target response rate in the equipment design specifications. The reference power response rate can be used to normalize the actual response rate, making the response rate indicators of different devices comparable.

[0185] The total test duration for each preset landscape fluctuation scene can be obtained. The total test duration can be the duration of the fluctuation scene from start to finish, and can be used to perform time integral averaging on various indicators.

[0186] The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system can be determined using the above formula. Where, The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is indicated by a value between 0 and 1, with a higher value indicating better efficiency stability. Let t be the actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system. The theoretically optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at the current power level. This represents the relative deviation of efficiency, reflecting the relative difference between the actual efficiency and the theoretical optimal efficiency at time t. The actual power response speed of the ALK-PEM hybrid hydrogen production system under the current power variation conditions. The reference power response speed is given. T represents the total test duration, which is calculated by time integration and averaging over the entire test period, allowing the stability index to reflect the overall performance of the entire dynamic process. Let be the rate of change of hydrogen production efficiency at time t. Let be the rate of change of input power at time t. The efficiency fluctuation sensitivity coefficient represents the degree of efficiency fluctuation caused by unit power fluctuation. It is a very small positive number, used to avoid the denominator being zero. , All are preset constants. The efficiency deviation weight is used to adjust the proportion of efficiency deviation items in the stability assessment. This is the efficiency fluctuation weight, used to adjust the proportion of the efficiency fluctuation sensitivity item in the stability assessment. , The sum is 1. , The value can be preset according to the requirements of the test scenario. For example, it can be increased when assessing the system's anti-disturbance capability. To highlight the impact of efficiency fluctuation sensitivity, the tracking accuracy of the evaluation system can be increased. To highlight the impact of delayed response.

[0187] By introducing an efficiency deviation term and an efficiency fluctuation sensitivity term, the hydrogen production efficiency stability is elevated from a single efficiency change rate statistic to a multi-dimensional dynamic indicator integrating response speed, efficiency deviation, and fluctuation sensitivity. The response hysteresis penalty factor in the efficiency deviation term quantifies the power response speed as a direct impact on efficiency stability, reflecting the synergistic effect of the PEM unit's second-level response and the ALK unit's minute-level response. The efficiency fluctuation sensitivity term quantifies the system's sensitivity to disturbances by measuring the degree of efficiency fluctuation caused by unit power fluctuations, and is weighted in conjunction with the power change rate to highlight stability assessment under strong disturbance scenarios. Through the weighted combination of these two terms, the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system under dynamic fluctuation scenarios can be comprehensively evaluated, accurately identifying whether stability shortcomings stem from excessive response hysteresis or excessive efficiency fluctuation sensitivity, providing a quantitative basis for control system parameter tuning and power allocation strategy optimization.

[0188] In some embodiments, the above-mentioned correlation analysis of the coupling response speeds of the PEM unit and the ALK unit under each preset power allocation ratio to obtain the rate-of-change matching coefficient during the power handover process of the PEM unit and the ALK unit may further include: using the following formula to perform correlation analysis on the coupling response speeds of the PEM unit and the ALK unit under each preset power allocation ratio to obtain the rate-of-change matching coefficient during the power handover process of the PEM unit and the ALK unit: ; In the formula, The rate of change matching coefficient; and , , are the power change rates of the PEM cell and the ALK cell at time t, respectively; T is the duration of the power transfer process; The response lag time of the PEM and ALK elements. and These are the coupling response speeds of the PEM unit and the ALK unit, respectively. For reference response lag time.

[0189] Real-time operating data of the ALK-PEM hybrid hydrogen production system under various preset power allocation ratios can be obtained. Preset power allocation ratios can include multiple configurations where the PEM and ALK units distribute the total input power in different proportions, such as a 20% PEM unit and an 80% ALK unit power allocation ratio, or a 50% PEM unit and a 50% ALK unit power allocation ratio. Real-time operating data includes at least the power change rate of the PEM unit at each time point, the power change rate of the ALK unit at each time point, and the coupling response speed of the two units during power transfer.

[0190] The power change rate of a PEM unit can be the rate at which its actual output power changes over time. It can be obtained by differentiating the PEM unit's power signal over time, reflecting the dynamic response characteristics of the PEM unit during power adjustment. Similarly, the power change rate of an ALK unit can be the rate at which its actual output power changes over time, reflecting the dynamic response characteristics of the ALK unit during power adjustment. The sign of the power change rate indicates whether the unit is in an increased or decreased load state, and the absolute value of the power change rate indicates the severity of the unit's power adjustment.

[0191] The coupling response speed of PEM and ALK cells during power handover can be obtained. The coupling response speed is the delay time between the change in power command and the start of actual power response for both cells during power handover. The coupling response speeds of the PEM and ALK cells can be obtained separately. The coupling response speed of the PEM cell typically exhibits a delay in the order of seconds, reflecting its fast response characteristics; the coupling response speed of the ALK cell typically exhibits a delay in the order of minutes, reflecting its slow response characteristics.

[0192] The response lag time of the PEM and ALK cells can be obtained. The response lag time is the absolute difference between the response start-up time of the PEM and ALK cells during power handover, and can be calculated from the absolute difference in the coupling response speed of the two cells. The response lag time quantifies the degree of asynchrony between the two cells during power handover.

[0193] A reference response lag time can be obtained. This reference response lag time can be preset based on the rated response characteristics of the PEM and ALK units, for example, determined by the difference in target response speeds between the two units in the equipment design specifications, or by the response lag time of the two units under optimal performance in historical test data. The reference response lag time can be used to normalize the actual response lag time.

[0194] The duration of the power handover process can be obtained. The duration of the power handover process can be the length of the time interval from the start of the power allocation command change to the time interval from when the power of both units reaches a stable state again. It can be used to perform time integration processing on various indicators.

[0195] The above formula can be used to perform correlation analysis on the coupling response speeds of PEM and ALK units under various preset power allocation ratios, obtaining the rate-of-change matching coefficient during the power handover process between the PEM and ALK units. Where, This is the rate of change matching coefficient, which can range from -1 to 1. The larger the value, the higher the dynamic coordination between the two units during the power transfer process. Let be the power change rate of the PEM element at time t. Let be the power change rate of the ALK cell at time t. The first part of the formula is the cross-correlation normalization coefficient of the power change rate. The numerator is the time integral of the product of the power change rates of the two cells, and the denominator is the square root of the product of the square integrals of the power change rates of the two cells. When the power change rates of the two cells change in the same direction and synchronously, this part approaches 1, indicating good coordination; when the power change rates of the two cells change in opposite directions and are time-matched, this part approaches -1, indicating that power transfer is in progress and in the correct direction; when the power change rates of the two cells fluctuate randomly or there is a phase shift, this part approaches 0, indicating that coordination has failed. T is the duration of the power transfer process, used to limit the integration interval to ensure that the complete power transfer process is covered. The response lag time of the PEM unit and the ALK unit. The second part of the formula is the response lag time, used as a reference. The penalty factor for response lag is calculated when the response lag time... When the value is zero, the factor is 1 and no penalty is imposed; when the response lag time increases, the factor decreases, reducing the calculation results of the first part; when the response lag time is much greater than the reference response lag time, the factor approaches 0, making the rate of change matching coefficient approach 0, indicating that the severe asynchronous response leads to collaborative failure.

[0196] By introducing a cross-correlation normalization coefficient and a response lag penalty factor, the rate-of-change matching coefficient is elevated from a single measure of power rate-of-change correlation to a multi-dimensional indicator integrating timing synchronization and response lag. The cross-correlation normalization coefficient quantifies the degree of directional coordination and timing matching of power changes between the PEM and ALK units during power handover through the integral correlation of the power rate-of-change sequences. The response lag penalty factor quantifies the degree of temporal coordination in the response startup of the two units by the ratio of the response lag time to the reference lag time, effectively reducing scenarios of asynchronous response. The combination of these two factors comprehensively reflects the dynamic coordination quality of the PEM and ALK units during power handover, accurately identifying whether matching shortcomings in coordinated operation stem from asynchronous power rate-of-change or excessive response startup delay, providing a quantitative basis for power allocation strategy optimization and coupling control parameter tuning.

[0197] In some embodiments, the above-mentioned statistical analysis of the start-stop response times of the PEM unit and the ALK unit under each preset start-stop condition to obtain the consistency of the start-stop response of the PEM unit and the ALK unit may further include: using the following formula to statistically analyze the start-stop response times of the PEM unit and the ALK unit under each preset start-stop condition to obtain the consistency of the start-stop response of the PEM unit and the ALK unit: ; In the formula, To ensure consistent start / stop responses; and , where are the start / stop response times of the PEM unit and the ALK unit respectively in the i-th start / stop operation; N is the total number of start / stop cycles; and These are the average start / stop response times for the PEM unit and the ALK unit, respectively. and These are the standard deviations of the start / stop response times for the PEM and ALK units, respectively. and Degradation trend coefficient and allowable degradation limit for start and stop response times respectively; , and This is a preset constant.

[0198] Real-time operational data of the ALK-PEM hybrid hydrogen production system can be acquired during multiple cycles of start-up and shutdown operations under various preset start-up and shutdown conditions. Preset start-up and shutdown conditions can include different types such as light-load start-up, rated start-up, temporary shutdown, and long-term shutdown, each corresponding to different start-up and shutdown conditions and operational requirements. Real-time operational data includes at least the start-up and shutdown response times of the PEM unit and the ALK unit in each start-up and shutdown operation. The start-up and shutdown response time can be the time delay from issuing the start-up / shutdown command to the corresponding unit actually starting to respond, reflecting the unit's dynamic response capability and state transition efficiency during the start-up and shutdown process. For start-up operations, the start-up and shutdown response time can be the time from issuing the start-up command to the unit starting hydrogen production; for shutdown operations, the start-up and shutdown response time can be the time from issuing the shutdown command to the unit power dropping to zero.

[0199] The total number of start-stop cycles can be obtained. The total number of start-stop cycles can be the number of complete start-stop operations completed under the current test conditions, and can be used to average various statistical indicators.

[0200] The average start-stop response time of PEM and ALK cells during each start-stop operation can be obtained. The average start-stop response time can be obtained by summing the response times of each start-stop operation and dividing by the total number of cycles, reflecting the average response speed of the cell across multiple start-stop operations. The average start-stop response time of PEM cells is typically expressed in seconds, while that of ALK cells is typically expressed in minutes.

[0201] The standard deviation of the start-stop response time for both the PEM and ALK units can be obtained. The standard deviation is calculated by taking the square root of the sum of squared deviations of the response time from the average for each start-stop operation, divided by the total number of cycles. It reflects the degree of fluctuation in the unit's response time across multiple start-stop cycles. A larger standard deviation indicates poorer repeatability of the start-stop process, potentially indicating instability in the control logic or changes in equipment status.

[0202] The degradation trend coefficient of start-stop response time can be obtained. This coefficient is obtained by comparing the relative change in average response time between the first few start-stop cycles and the subsequent few start-stop cycles. Specifically, the average response time of the first m start-stop cycles is taken as the initial baseline, and the average response time of the next m start-stop cycles is taken as the later sample. The relative rate of change of the later cycle relative to the initial cycle is then calculated. The degradation trend coefficient reflects whether the response time gradually slows down with increasing start-stop cycles, and is used to assess the performance degradation of the equipment.

[0203] The allowable upper limit of start-stop response time can be obtained. This allowable upper limit can be preset according to equipment design specifications or testing requirements. It is used to normalize the degradation trend coefficient, making degradation indicators comparable across different equipment or operating conditions. The allowable upper limit can be set as the maximum allowable extension of the equipment's response time, such as 10% or 20%.

[0204] The above formula can be used to statistically analyze the start-stop response times of the PEM and ALK units under various preset start-stop conditions, thus obtaining the consistency of the start-stop responses of the PEM and ALK units. Where, For consistency of start-stop response, this value is between 0 and 1, with a larger value indicating better consistency of start-stop response. and These are the start / stop response times of the PEM unit and the ALK unit during the i-th start / stop operation, respectively, in seconds. The relative difference in response time between two units during the i-th start-up and shutdown is normalized by dividing by the maximum of the two values. This normalization ensures that the difference index is unaffected by the absolute response time of the units and reflects only the relative synchronization degree of the two units during a single start-up and shutdown. The start-up and shutdown asynchrony index is obtained by taking the arithmetic mean of the relative differences over N start-ups and shutdowns. and These are the average start / stop response times for the PEM unit and the ALK unit, respectively. and These are the standard deviations of the start / stop response times of the PEM and ALK units, respectively. The sum of the coefficients of variation of the two units reflects the relative fluctuation of response time during multiple start-ups and shutdowns. The larger the coefficient of variation, the worse the repeatability of the start-up and shutdown process. The degradation trend coefficient of start-stop response time reflects the change trend of response time with the increase of start-stop frequency. The larger the coefficient, the more significant the performance degradation. This is the upper limit of the allowable degradation of the start / stop response time, used to normalize the degradation trend coefficient. , and These are all preset constants, namely the weight for start-stop asynchrony, the weight for start-stop fluctuation, and the weight for degradation trend, with a sum of 1. These can be preset according to the needs of the test scenario. For example, they can be increased when assessing the initial consistency of the equipment. , When assessing the long-term durability of equipment, it can be increased .

[0205] By introducing start-stop asynchrony, start-stop fluctuation, and degradation trend terms, the consistency of start-stop response is elevated from a single unit's statistical dispersion to a multi-dimensional dynamic indicator that integrates inter-unit synchronicity, intra-unit stability, and time series trends. The start-stop asynchrony term quantifies the relative synchronicity between the PEM and ALK units during start-stop processes by calculating and averaging the normalized difference in response times between the two units during each start-stop cycle, identifying start-stop asynchrony issues caused by inherent differences in unit response speeds. The start-stop fluctuation term quantifies the degree of response time fluctuation across multiple start-stop cycles by calculating the sum of the coefficients of variation of the two units, identifying poor start-stop repeatability issues caused by unstable control logic or changes in equipment status. The degradation trend term quantifies the degradation trend of equipment performance by calculating the relative change in response time with increasing number of start-stop cycles, identifying start-stop performance degradation caused by component aging or wear. By combining the three factors in a weighted manner, the collaborative quality of the ALK-PEM hybrid hydrogen production system during start-up and shutdown can be comprehensively evaluated. This allows for the accurate identification of whether the shortcomings in start-up and shutdown performance stem from poor inter-unit synchronization, large intra-unit fluctuations, or significant long-term degradation trends, providing a quantitative basis for optimizing start-up and shutdown control logic and predicting equipment lifespan.

[0206] As can be seen from the technical solutions provided in the embodiments of this specification above, these embodiments can determine multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern; multiple types of simulated fluctuation power data are generated based on the multiple types of typical power waveform curves; the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system is adjusted based on the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process are obtained; and a test report of the ALK-PEM hybrid hydrogen production system is generated based on the operating data. By extracting typical fluctuation patterns from real wind and solar data, the test conditions closely match actual operating scenarios; by generating multi-scale simulated fluctuation data, the dynamic response test of the electrolyzer under complex operating conditions can be comprehensively covered; the differentiated control of the PEM and ALK units accurately evaluates the collaborative operating performance of the ALK-PEM hybrid hydrogen production system; and the final output test report provides an intuitive basis for equipment optimization and fault diagnosis.

[0207] The following is a specific embodiment of this specification: A comprehensive test was conducted on the 5MW ALK+1MW PEM integrated ALK-PEM hybrid hydrogen production system. The test system is a dedicated test platform consisting of a wind and solar power analysis module, a wind and solar power simulation output module, a data acquisition module, an operating condition control module, and a result analysis module. The wind and solar power simulation accuracy is ≤±1%, the parameter acquisition accuracy is ≤±0.5%, the preset hydrogen production efficiency benchmark value is ≥78%, the purified outlet hydrogen purity benchmark value is ≥99.999%, the dew point is ≤-70℃, and the curtailment rate is ≤10%.

[0208] 1. Pre-test preparation: Complete the hardware docking of the 5MW ALK+1MW PEM integrated ALK-PEM hybrid hydrogen production system with the test system to ensure smooth communication between modules; calibrate each module of the test system to ensure the simulation accuracy of wind and solar power is ±0.8% and the parameter acquisition accuracy is ±0.3%; check the sealing performance of the ALK-PEM hybrid hydrogen production system to ensure no leakage and reliable pipeline connection; preset the basic test parameters: ALK unit operating temperature 80~90℃, PEM unit operating temperature 40~50℃, operating pressure 0.1~0.3MPa, absolute value of liquid level difference in the separation system ≤5mm, pressure fluctuation ≤0.05MPa.

[0209] 2. Wind and Solar Data Processing: Obtain second-level raw wind and solar data for the target project site, clean the raw data using statistical methods to remove outliers, filter the data using a Savitzky-Golay filter to remove data noise, and perform waveform clustering analysis at time scales of 0.5min, 1min, 5min, and 15min respectively to obtain typical power waveform curves and extract the limit values ​​of change for dynamic operating condition testing.

[0210] 3. Module-by-module testing.

[0211] 3.1 Static Performance Test: The operating condition control module was set to a fixed power output for the wind and solar power simulation output module. Inputs were made at four levels: 30% (1.8MW), 50% (3MW), 70% (4.2MW), and 100% (6MW) of the rated power of the ALK-PEM hybrid hydrogen production system. Each level was operated stably for 30 minutes. The data acquisition module simultaneously collected parameters such as current, voltage, temperature, and pressure from each unit and the entire system. The recorded hydrogen production efficiencies at each level were 78.5%, 80.2%, 81.5%, and 80.8%, respectively. The hydrogen purity in the separated oxygen outlet was ≤1.2%, the oxygen purity in the separated hydrogen outlet was ≤0.15%, the purified hydrogen outlet purity was 99.9992%, and the dew point was -72℃, all meeting the preset benchmark requirements.

[0212] 3.2 Dynamic Operating Condition Test: The operating condition control module sets the wind and solar power simulation output module to dynamic power output mode. Based on the typical curves obtained from wind and solar data processing, it generates a 24-hour wind and solar fluctuation curve and second-level power data, which are then sent to the data acquisition module. The data acquisition module sends the power data to the operating condition control module, which adjusts the operating power of the ALK and PEM units to simulate the actual wind and solar fluctuations at the project site. After 60 minutes of test operation, the power response speed of each unit was ≤8s, the hydrogen production efficiency change rate was ≤±4.5%, the gas production purity was stable, and there were no parameters exceeding the range.

[0213] 3.3 Collaborative Operation Test: With 75% power input to the ALK unit and 25% power input to the PEM unit, the system was stably operated for 20 minutes at total power levels of 2MW, 3MW, 4MW, 5MW, and 6MW respectively. The data acquisition module synchronously collected parameters such as coupling response speed and power allocation accuracy. The recorded power allocation accuracy was ±2%, coupling response speed was ≤7s, overall hydrogen production efficiency was 80.1%, and power curtailment rate was 8.2%, meeting the collaborative performance requirements.

[0214] 3.4 Start-stop characteristics test: Complete 3 cycles each of light load start, rated start, temporary stop, and long-term stop; the start-stop response time is ≤15s, the energy consumption during the start-stop process is 85% of the energy consumption of 5 minutes of rated power operation, the electrode temperature change rate is ≤3℃ / min, the absolute value of the liquid level difference in the separation system is ≤3mm, the pressure fluctuation is ≤0.03MPa, there is no leakage, and the start-stop stability is good.

[0215] 4. Data Processing and Judgment: The result analysis module performs noise reduction and normalization on the test data, removes outliers, and calculates the core indicator parameters. The results are compared with the standardized judgment criteria. The performance indicators, stability indicators, and synergy indicators all meet the qualification requirements. The result analysis module outputs a test qualification report, indicating that the hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system under large power fluctuations can be further optimized.

[0216] This embodiment verifies that the test method can achieve accurate testing of the integrated ALK-PEM hybrid hydrogen production system in all dimensions. The test conditions closely match the actual wind and solar operation scenarios at the project site, and the test results truly reflect the actual operating performance of the equipment, providing clear guidance for equipment optimization.

[0217] Based on the aforementioned testing method for the ALK-PEM hybrid hydrogen production system, this specification also proposes an embodiment of a testing device for the ALK-PEM hybrid hydrogen production system. This testing device can use real wind and solar sample data as its source, and through multi-timescale analysis and clustering, extract typical power waveform curves characterizing different fluctuation modes, thereby generating simulated fluctuating power data that closely reflects actual operating conditions. This data drives the ALK-PEM hybrid hydrogen production system to sequentially perform comprehensive tests on static performance, dynamic operating conditions, coordinated operation, and start-stop characteristics, simultaneously acquiring multi-dimensional operating parameters of the PEM unit, ALK unit, and the system. Finally, through decoupling analysis, stratified efficiency indicators such as electrochemistry, thermal management, and water balance are calculated, and combined with dynamic response and coordinated matching coefficients, a standardized test report containing performance bottleneck identification and optimization guidance is generated. Figure 2 As shown, the test device 200 of the ALK-PEM hybrid hydrogen production system may specifically include the following modules: The determination module 201 is used to determine typical curves of multiple types of power waveforms based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern. The first generation module 202 is used to generate multiple types of simulated fluctuating power data based on the typical curves of the multiple types of power waveforms. Test module 203 is used to adjust the operating power of PEM unit and ALK unit in ALK-PEM hybrid hydrogen production system according to the multi-type simulated fluctuation power data, so as to perform test on ALK-PEM hybrid hydrogen production system. The acquisition module 204 is used to acquire the operating data of the PEM unit, ALK unit and ALK-PEM hybrid hydrogen production system during the test process; The second generation module 205 is used to generate a test report for the ALK-PEM hybrid hydrogen production system based on the operating data.

[0218] In some embodiments, the determining module 201 may also be used for: The wind and solar sample data are sampled according to multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale; the multiple time scales include a first scale, a second scale and a third scale, and the wind and solar sample waveforms at the first scale, the second scale and the third scale are used to characterize the instantaneous change characteristics, short-term ramp characteristics and long-term trend characteristics of wind and solar power, respectively. Shape similarity clustering is performed on multiple wind and light waveforms at each time scale; Based on the clustering results, the shape features of various types of wind and light sampling waveforms at each time scale are extracted; Based on the shape features, typical curves of multiple power waveforms at each time scale are generated.

[0219] In some embodiments, the determining module 201 may also be used for: Each wind and solar sampling waveform is divided into multiple waveform segments; the multiple waveform segments include a rising segment, a peak hold segment, and a falling segment. The rising segment is used to characterize the power ramping capability of the PEM unit and the ALK unit, the peak hold segment is used to characterize the overload sustaining capability of the PEM unit and the ALK-PEM hybrid hydrogen production system, and the falling segment is used to characterize the power recovery capability of the PEM unit and the ALK unit. The weights of each waveform segment are determined based on the sensitivity and tolerance of the PEM and ALK units to each waveform segment. Based on the weights, the morphological feature parameters of each waveform segment are fused to obtain the waveform feature vector of each wind and light sampling waveform; Cluster the waveform feature vectors at each time scale.

[0220] In some embodiments, the testing apparatus for the ALK-PEM hybrid hydrogen production system described above can also be used for: Frequency distribution curves are constructed based on the frequency of occurrence of various wind and light sampling waveforms at each time scale; Identify the long tail portion in the frequency distribution curve; the long tail portion is used to characterize the waveform category where the frequency occurs below a preset frequency threshold. Extract rare waveforms in the long tail portion whose waveform fluctuation amplitude exceeds a preset amplitude threshold; Calculate the clustering effectiveness index corresponding to the number of candidate clusters at each time scale; Using the clustering effectiveness index as the optimization objective and preserving the categories of the scarce waveforms as the constraint, the minimum number of clusters that maximizes the clustering effectiveness index and satisfies the constraint is taken as the optimal number of clusters at this time scale.

[0221] Based on this, the aforementioned determining module 201 can also be used for: Cluster the waveform feature vectors according to the optimal number of clusters at each time scale.

[0222] In some embodiments, the first generation module 202 described above can also be used for: Obtain multiple preset wind and solar power output test scenario requirements; For each wind and solar power output test scenario, one or more typical power waveform curves are selected from the multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario. The simulated fluctuation test sequence is interpolated to obtain the simulated fluctuation power data corresponding to the requirements of the wind and solar power output test scenario.

[0223] In some embodiments, the first generation module 202 described above can also be used for: Based on the requirements of each wind and solar power output test scenario, one type of typical power waveform curves from multiple types of power waveforms at the third scale is selected as the basic trend curve. Calculate the fluctuation energy coupling coefficient between the basic trend curve and the typical power waveform curves of various types at the first and second scales; From the typical power waveform curves at the first and second scales, the class with the highest wave energy coupling coefficient is selected as the superimposed wave curve. Based on the wave energy coupling coefficient, the phase offset and amplitude modulation ratio of the superimposed wave curve relative to the basic trend curve are determined. The superimposed wave curve is superimposed on the basic trend curve according to the phase offset and amplitude modulation ratio to obtain the simulated wave test sequence.

[0224] In some embodiments, the test module 203 described above can also be used for: Based on the various types of simulated fluctuating power data, the first operating power of the PEM unit and the ALK unit under multiple preset power levels in the ALK-PEM hybrid hydrogen production system is determined.

[0225] Based on this, the aforementioned acquisition module 204 can also be used for: Acquire the first operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system at each preset power level; the first operating data includes at least current, voltage, temperature, pressure, hydrogen production efficiency, gas purity, and water replenishment rate.

[0226] In some embodiments, the second generation module 205 described above can also be used for: Based on the current and voltage at each preset power level, the current-voltage characteristic curves are fitted to obtain the electrochemical efficiency of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system. Based on the temperature at each power level, calculate the temperature rise rate and temperature deviation to determine the thermal management efficiency of the ALK-PEM hybrid hydrogen production system; The water consumption for hydrogen production is calculated based on the water replenishment rate and gas purity at each power level in order to determine the water balance efficiency of the ALK-PEM hybrid hydrogen production system. The electrochemical efficiency, thermal management efficiency, and water balance efficiency are used to generate a static performance test report that includes the dominant factors affecting hydrogen production efficiency deviation.

[0227] In some embodiments, the second generation module 205 described above can also be used for: The thermal management efficiency of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ; In the formula, Thermal management efficiency of the ALK-PEM hybrid hydrogen production system; and These are the temperature rise rates for the PEM and ALK units, respectively. and These are the actual operating temperatures of the PEM and ALK units, respectively. and These are the theoretical optimal operating temperatures for the PEM and ALK cells, respectively. The operating temperature range of the ALK-PEM hybrid hydrogen production system; , and This is a preset constant.

[0228] In some embodiments, the test module 203 described above can also be used for: Based on the various types of simulated fluctuation power data, the second operating power of the PEM unit and ALK unit under multiple preset wind and solar fluctuation scenarios is determined.

[0229] Based on this, the aforementioned acquisition module 204 can also be used for: Acquire second operating data for the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system under each preset wind and solar fluctuation scenario; the second operating data includes at least the power response speed, hydrogen production efficiency change rate, and gas production purity stability.

[0230] In some embodiments, the second generation module 205 described above can also be used for: The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined based on the rate of change of hydrogen production efficiency and the power response speed under various preset wind and solar fluctuation scenarios. Time-frequency analysis was performed on the gas production purity under various preset wind and solar fluctuation scenarios to obtain the dominant frequency components of gas production purity fluctuation. Based on the dominant frequency components of the hydrogen production efficiency stability and gas purity fluctuations, a dynamic operating condition test report is generated, including the dominant factors of hydrogen production efficiency deviation.

[0231] In some embodiments, the second generation module 205 described above can also be used for: Based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios, the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ; In the formula, To improve the hydrogen production efficiency and stability of the ALK-PEM hybrid hydrogen production system; The actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at time t; The theoretically optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system; Let be the simulated fluctuation power at time t; T be the total test duration. and These are the power response speed and reference power response speed of the ALK-PEM hybrid hydrogen production system, respectively. and These are the rate of change in hydrogen production efficiency and the rate of change in input power, respectively. , and This is a preset constant.

[0232] In some embodiments, the test module 203 described above can also be used for: Based on the various types of simulated fluctuation power data, the third operating power of the PEM unit and the ALK unit under multiple preset power allocation ratios is determined; Based on this, the aforementioned acquisition module 204 can also be used for: Acquire third operating data for the PEM unit and ALK unit under each preset power allocation ratio; the third operating data includes at least the coupling response speed, power allocation accuracy, and hydrogen production efficiency.

[0233] In some embodiments, the second generation module 205 described above can also be used for: Correlation analysis was performed on the coupling response speeds of PEM and ALK units under various preset power allocation ratios to obtain the rate of change matching coefficients during the power handover process between PEM and ALK units. Based on the hydrogen production efficiency under each preset power allocation ratio, calculate the cooperative efficiency loss coefficient between the PEM unit and the ALK unit. Based on the power allocation accuracy, rate of change matching coefficient, and collaborative efficiency damage coefficient, a collaborative operation test report is generated, which includes the dominant factors of hydrogen production efficiency deviation.

[0234] In some embodiments, the second generation module 205 described above can also be used for: The following formula was used to perform a correlation analysis on the coupling response speeds of the PEM and ALK cells under various preset power allocation ratios, and the rate of change matching coefficients during the power handover process between the PEM and ALK cells were obtained: ; In the formula, The rate of change matching coefficient; and , , are the power change rates of the PEM cell and the ALK cell at time t, respectively; T is the duration of the power transfer process; The response lag time of the PEM and ALK elements. and These are the coupling response speeds of the PEM unit and the ALK unit, respectively. For reference response lag time.

[0235] In some embodiments, the test module 203 described above can also be used for: Based on the various types of simulated fluctuation power data, the fourth operating power of the PEM unit and ALK unit under multiple preset start-stop conditions is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire the fourth operating data of the PEM unit and ALK unit under each preset start-stop condition; the fourth operating data includes at least the start-stop response time, energy consumption during start-stop process, electrode temperature change rate, and sealing performance stability.

[0236] Based on this, the aforementioned acquisition module 204 can also be used for: Statistical analysis was performed on the start-stop response times of the PEM unit and the ALK unit under various preset start-stop conditions to obtain the consistency of the start-stop response of the PEM unit and the ALK unit. Peak values ​​of electrode temperature change rates under various preset start-stop conditions were extracted to obtain the thermal stress influence of PEM and ALK units. The sealing performance stability under various preset start-stop conditions is trend-fitted to obtain the sealing reliability of the PEM unit and the ALK unit. Statistical analysis was performed on the energy consumption during the start-up and shutdown process under various preset start-up and shutdown conditions to identify the start-up and shutdown stages where abnormal impact events occurred. Based on the results of start-stop response consistency, thermal stress impact, sealing reliability, and abnormal event identification, a start-stop characteristic test report is generated, including the dominant factors of hydrogen production efficiency deviation.

[0237] In some embodiments, the second generation module 205 described above can also be used for: The start-stop response times of the PEM and ALK units under various preset start-stop conditions are statistically analyzed using the following formula to obtain the consistency of the start-stop responses of the PEM and ALK units: ; In the formula, To ensure consistent start / stop responses; and , where are the start / stop response times of the PEM unit and the ALK unit respectively in the i-th start / stop operation; N is the total number of start / stop cycles; and These are the average start / stop response times for the PEM unit and the ALK unit, respectively. and These are the standard deviations of the start / stop response times for the PEM and ALK units, respectively. and Degradation trend coefficient and allowable degradation limit for start and stop response times respectively; , and This is a preset constant.

[0238] As can be seen from the technical solutions provided in the embodiments of this specification above, these embodiments can determine multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern; multiple types of simulated fluctuation power data are generated based on the multiple types of typical power waveform curves; the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system is adjusted based on the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process are obtained; and a test report of the ALK-PEM hybrid hydrogen production system is generated based on the operating data. By extracting typical fluctuation patterns from real wind and solar data, the test conditions closely match actual operating scenarios; by generating multi-scale simulated fluctuation data, the dynamic response test of the electrolyzer under complex operating conditions can be comprehensively covered; the differentiated control of the PEM and ALK units accurately evaluates the collaborative operating performance of the ALK-PEM hybrid hydrogen production system; and the final output test report provides an intuitive basis for equipment optimization and fault diagnosis.

[0239] This specification also provides a computer device for testing an ALK-PEM hybrid hydrogen production system. This device uses real wind and solar sample data as its source, and through multi-timescale analysis and clustering, extracts typical power waveform curves characterizing different fluctuation modes, generating simulated fluctuating power data that closely reflects actual operating conditions. This data drives the ALK-PEM hybrid hydrogen production system to sequentially perform comprehensive tests on static performance, dynamic operating conditions, coordinated operation, and start-stop characteristics, simultaneously acquiring multi-dimensional operating parameters of the PEM unit, ALK unit, and the system. Finally, through decoupling analysis, it calculates stratified efficiency indicators such as electrochemistry, thermal management, and water balance, and combines dynamic response and coordinated matching coefficients to generate a standardized test report containing performance bottleneck identification and optimization guidance. The computer device includes a processor and a memory for storing processor-executable instructions. Specifically, the processor can perform the following tasks according to the instructions: determining multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation pattern; generating multiple types of simulated fluctuation power data based on the multiple types of typical power waveform curves; adjusting the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system based on the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; acquiring the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; and generating a test report for the ALK-PEM hybrid hydrogen production system based on the operating data.

[0240] To execute the above instructions more accurately, please refer to... Figure 3 As shown in the embodiments of this specification, another specific computer device 300 is also provided, wherein the computer device 300 includes a network communication port 301, a processor 302 and a memory 303, and the above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0241] The processor 302 can be specifically used to: determine multiple types of typical power waveform curves based on wind and solar sample data; each type of typical power waveform curve represents a wind and solar fluctuation mode; generate multiple types of simulated fluctuation power data based on the multiple types of typical power waveform curves; adjust the operating power of the PEM unit and ALK unit in the ALK-PEM hybrid hydrogen production system based on the multiple types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system; acquire the operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; and generate a test report for the ALK-PEM hybrid hydrogen production system based on the operating data.

[0242] The memory 303 can be used to store the corresponding instruction program.

[0243] In this embodiment, the network communication port 301 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0244] In this embodiment, the processor 302 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0245] In this embodiment, the memory 303 includes volatile memory and non-volatile memory. The memory 303 can include multiple layers. In digital systems, anything that can store binary data can be a memory; in integrated circuits, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0246] Furthermore, embodiments of this specification also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described... Figure 1 Instructions for the test method of the ALK-PEM hybrid hydrogen production system shown.

[0247] Furthermore, embodiments of this specification provide a computer program product comprising a computer program that, when executed by a processor, implements the above-described... Figure 1 The test method for the ALK-PEM hybrid hydrogen production system is shown.

[0248] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0249] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this specification generally indicates that the preceding and following related objects have an "or" relationship.

[0250] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0251] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0252] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0253] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational tasks to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The task is a function specified in one or more boxes.

[0254] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A test method for an ALK-PEM hybrid hydrogen production system, characterized in that, include: Based on wind and solar sample data, typical curves of various power waveforms were determined; The typical power waveform curve for each category represents a wind and solar fluctuation pattern; Based on the typical curves of the various power waveforms, generate various types of simulated fluctuating power data; Based on the various types of simulated fluctuation power data, the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system is adjusted to perform a test on the ALK-PEM hybrid hydrogen production system. Obtain operational data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process; Based on the operational data, a test report for the ALK-PEM hybrid hydrogen production system is generated.

2. The method according to claim 1, characterized in that, The process of determining typical curves of multiple power waveforms based on wind and solar sample data includes: The wind and solar sample data are sampled according to multiple preset time scales to obtain multiple wind and solar sample waveforms at each time scale; the multiple time scales include a first scale, a second scale and a third scale, and the wind and solar sample waveforms at the first scale, the second scale and the third scale are used to characterize the instantaneous change characteristics, short-term ramp characteristics and long-term trend characteristics of wind and solar power, respectively. Shape similarity clustering is performed on multiple wind and light waveforms at each time scale; Based on the clustering results, the shape features of various types of wind and light sampling waveforms at each time scale are extracted; Based on the shape features, typical curves of multiple power waveforms at each time scale are generated.

3. The method according to claim 2, characterized in that, The method of clustering multiple wind and light waveforms at each time scale based on shape similarity includes: Each wind and solar sampling waveform is divided into multiple waveform segments; the multiple waveform segments include a rising segment, a peak hold segment, and a falling segment. The rising segment is used to characterize the power ramping capability of the PEM unit and the ALK unit, the peak hold segment is used to characterize the overload sustaining capability of the PEM unit and the ALK-PEM hybrid hydrogen production system, and the falling segment is used to characterize the power recovery capability of the PEM unit and the ALK unit. The weights of each waveform segment are determined based on the sensitivity and tolerance of the PEM and ALK units to each waveform segment. Based on the weights, the morphological feature parameters of each waveform segment are fused to obtain the waveform feature vector of each wind and light sampling waveform; Cluster the waveform feature vectors at each time scale.

4. The method according to claim 3, characterized in that, The method further includes: Frequency distribution curves are constructed based on the frequency of occurrence of various wind and light sampling waveforms at each time scale; Identify the long tail portion in the frequency distribution curve; the long tail portion is used to characterize the waveform category where the frequency occurs below a preset frequency threshold. Extract rare waveforms in the long tail portion whose waveform fluctuation amplitude exceeds a preset amplitude threshold; Calculate the clustering effectiveness index corresponding to the number of candidate clusters at each time scale; Using the clustering effectiveness index as the optimization objective and preserving the categories of the scarce waveforms as the constraint, the minimum number of clusters that maximizes the clustering effectiveness index and satisfies the constraint is taken as the optimal number of clusters at this time scale. The clustering of waveform feature vectors at each time scale includes: Cluster the waveform feature vectors according to the optimal number of clusters at each time scale.

5. The method according to claim 2, characterized in that, The step of generating multiple types of simulated fluctuating power data based on the typical curves of the multiple types of power waveforms includes: Obtain multiple preset wind and solar power output test scenario requirements; For each wind and solar power output test scenario, one or more typical power waveform curves are selected from the multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario. The simulated fluctuation test sequence is interpolated to obtain the simulated fluctuation power data corresponding to the requirements of the wind and solar power output test scenario.

6. The method according to claim 5, characterized in that, For each wind and solar power output test scenario, one or more typical power waveform curves are selected from the multiple types of typical power waveform curves and combined to obtain the simulated fluctuation test sequence corresponding to the wind and solar power output test scenario requirement, including: Based on the requirements of each wind and solar power output test scenario, one type of typical power waveform curves from multiple types of power waveforms at the third scale is selected as the basic trend curve. Calculate the fluctuation energy coupling coefficient between the basic trend curve and the typical power waveform curves of various types at the first and second scales; From the typical power waveform curves at the first and second scales, the class with the highest wave energy coupling coefficient is selected as the superimposed wave curve. Based on the wave energy coupling coefficient, the phase offset and amplitude modulation ratio of the superimposed wave curve relative to the basic trend curve are determined. The superimposed wave curve is superimposed on the basic trend curve according to the phase offset and amplitude modulation ratio to obtain the simulated wave test sequence.

7. The method according to claim 1, characterized in that, The step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuating power data, the first operating power of the PEM unit and the ALK unit under multiple preset power levels of the ALK-PEM hybrid hydrogen production system is determined. The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire the first operating data of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system at each preset power level; the first operating data includes at least current, voltage, temperature, pressure, hydrogen production efficiency, gas purity, and water replenishment rate.

8. The method according to claim 7, characterized in that, The process of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Based on the current and voltage at each preset power level, the current-voltage characteristic curves are fitted to obtain the electrochemical efficiency of the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system. Based on the temperature at each power level, calculate the temperature rise rate and temperature deviation to determine the thermal management efficiency of the ALK-PEM hybrid hydrogen production system; The water consumption for hydrogen production is calculated based on the water replenishment rate and gas purity at each power level in order to determine the water balance efficiency of the ALK-PEM hybrid hydrogen production system. The electrochemical efficiency, thermal management efficiency, and water balance efficiency are used to generate a static performance test report that includes the dominant factors affecting hydrogen production efficiency deviation.

9. The method according to claim 8, characterized in that, Determining the thermal management efficiency of the ALK-PEM hybrid hydrogen production system includes: The thermal management efficiency of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ； In the formula, Thermal management efficiency of the ALK-PEM hybrid hydrogen production system; and These are the temperature rise rates for the PEM and ALK units, respectively. and These are the actual operating temperatures of the PEM and ALK units, respectively. and These are the theoretical optimal operating temperatures for the PEM and ALK cells, respectively. The operating temperature range of the ALK-PEM hybrid hydrogen production system; , and This is a preset constant.

10. The method according to claim 1, characterized in that, The step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the second operating power of the PEM unit and ALK unit under multiple preset wind and solar fluctuation scenarios is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire second operating data for the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system under each preset wind and solar fluctuation scenario; the second operating data includes at least the power response speed, hydrogen production efficiency change rate, and gas production purity stability.

11. The method according to claim 10, characterized in that, The process of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: The hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined based on the rate of change of hydrogen production efficiency and the power response speed under various preset wind and solar fluctuation scenarios. Time-frequency analysis was performed on the gas production purity under various preset wind and solar fluctuation scenarios to obtain the dominant frequency components of gas production purity fluctuation. Based on the dominant frequency components of the hydrogen production efficiency stability and gas purity fluctuations, a dynamic operating condition test report is generated, including the dominant factors of hydrogen production efficiency deviation.

12. The method according to claim 11, characterized in that, The determination of the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios includes: Based on the rate of change of hydrogen production efficiency and power response speed under various preset wind and solar fluctuation scenarios, the hydrogen production efficiency stability of the ALK-PEM hybrid hydrogen production system is determined using the following formula: ； In the formula, To improve the hydrogen production efficiency and stability of the ALK-PEM hybrid hydrogen production system; The actual hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system at time t; The theoretically optimal hydrogen production efficiency of the ALK-PEM hybrid hydrogen production system; Let be the simulated fluctuation power at time t; T be the total test duration. and These are the power response speed and reference power response speed of the ALK-PEM hybrid hydrogen production system, respectively. and These are the rate of change in hydrogen production efficiency and the rate of change in input power, respectively. , and This is a preset constant.

13. The method according to claim 1, characterized in that, The step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the third operating power of the PEM unit and the ALK unit under multiple preset power allocation ratios is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire third operating data for the PEM unit and ALK unit under each preset power allocation ratio; the third operating data includes at least the coupling response speed, power allocation accuracy, and hydrogen production efficiency.

14. The method according to claim 13, characterized in that, The process of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Correlation analysis was performed on the coupling response speeds of PEM and ALK units under various preset power allocation ratios to obtain the rate of change matching coefficients during the power handover process between PEM and ALK units. Based on the hydrogen production efficiency under each preset power allocation ratio, calculate the cooperative efficiency loss coefficient between the PEM unit and the ALK unit. Based on the power allocation accuracy, rate of change matching coefficient, and collaborative efficiency damage coefficient, a collaborative operation test report is generated, which includes the dominant factors of hydrogen production efficiency deviation.

15. The method according to claim 14, characterized in that, The correlation analysis of the coupling response speeds of the PEM unit and the ALK unit under each preset power allocation ratio, to obtain the rate-of-change matching coefficient during the power handover process between the PEM unit and the ALK unit, includes: The following formula was used to perform a correlation analysis on the coupling response speeds of the PEM and ALK cells under various preset power allocation ratios, and the rate of change matching coefficients during the power handover process between the PEM and ALK cells were obtained: ； In the formula, The rate of change matching coefficient; and , , are the power change rates of the PEM cell and the ALK cell at time t, respectively; T is the duration of the power transfer process; The response lag time of the PEM and ALK elements. and These are the coupling response speeds of the PEM unit and the ALK unit, respectively. For reference response lag time.

16. The method according to claim 1, characterized in that, The step of adjusting the operating power of the PEM unit and the ALK unit in the ALK-PEM hybrid hydrogen production system based on the various types of simulated fluctuation power data to perform testing on the ALK-PEM hybrid hydrogen production system includes: Based on the various types of simulated fluctuation power data, the fourth operating power of the PEM unit and ALK unit under multiple preset start-stop conditions is determined; The acquisition of operational data from the PEM unit, ALK unit, and ALK-PEM hybrid hydrogen production system during the testing process includes: Acquire the fourth operating data of the PEM unit and ALK unit under each preset start-stop condition; the fourth operating data includes at least the start-stop response time, energy consumption during start-stop process, electrode temperature change rate, and sealing performance stability.

17. The method according to claim 16, characterized in that, The process of generating a test report for the ALK-PEM hybrid hydrogen production system based on the operational data includes: Statistical analysis was performed on the start-stop response times of the PEM unit and the ALK unit under various preset start-stop conditions to obtain the consistency of the start-stop response of the PEM unit and the ALK unit. Peak values ​​of electrode temperature change rates under various preset start-stop conditions were extracted to obtain the thermal stress influence of PEM and ALK units. The sealing performance stability under various preset start-stop conditions is trend-fitted to obtain the sealing reliability of the PEM unit and the ALK unit. Statistical analysis was performed on the energy consumption during the start-up and shutdown process under various preset start-up and shutdown conditions to identify the start-up and shutdown stages where abnormal impact events occurred. Based on the results of start-stop response consistency, thermal stress impact, sealing reliability, and abnormal event identification, a start-stop characteristic test report is generated, including the dominant factors of hydrogen production efficiency deviation.

18. The method according to claim 17, characterized in that, The statistical analysis of the start-stop response times of the PEM unit and the ALK unit under various preset start-stop conditions to obtain the consistency of the start-stop response of the PEM unit and the ALK unit includes: The start-stop response times of the PEM and ALK units under various preset start-stop conditions are statistically analyzed using the following formula to obtain the consistency of the start-stop responses of the PEM and ALK units: ； In the formula, To ensure consistent start / stop responses; and , where are the start / stop response times of the PEM unit and the ALK unit respectively in the i-th start / stop operation; N is the total number of start / stop cycles; and These are the average start / stop response times for the PEM unit and the ALK unit, respectively. and These are the standard deviations of the start / stop response times for the PEM and ALK units, respectively. and Degradation trend coefficient and allowable degradation limit for start and stop response times respectively; , and This is a preset constant.

19. A computer device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method of any one of claims 1-18.

Images