A method for actively controlling multi-time scale electrolyzer impedance detection

Abstract

The application provides a kind of active control multi-time scale electrolytic cell impedance detection method, it is related to the field of electrolytic hydrogen production, comprising: determining at least two time scales corresponding to the current detection target;Collecting electrolytic cell working condition information, based on electrolytic cell working condition information, determine the optimal frequency band of the time scale corresponding to the current detection target;Based on electrolytic cell working condition information, determine the signal amplitude;Based on electrolytic cell working condition information, determine the data acquisition interval corresponding to the time scale corresponding to the current detection target;Based on the optimal frequency band, signal amplitude and data acquisition interval corresponding to each time scale, multi-band alternative scanning is carried out on the electrolytic cell, and scanning data is obtained;Based on scanning data, generate electrolytic cell impedance test results, with the advantages of improving the accuracy and efficiency of electrolytic cell impedance measurement under dynamic working conditions.

Description

Technical Field

[0001] This invention relates to the field of electrolytic hydrogen production, and in particular to an actively controlled multi-timescale electrolyzer impedance detection method. Background Technology

[0002] An electrolyzer is a core piece of equipment that uses electrical energy to drive a chemical reaction (electrolysis process), and it is widely used in industrial production, energy conversion, and materials preparation. Its basic principle is to use an external direct current (DC) to cause ions in the electrolyte solution or molten electrolyte to undergo a redox reaction on the electrode surface, thereby achieving the decomposition or synthesis of substances. Electrolytic hydrogen production is the process of decomposing water into hydrogen and oxygen through electrolysis. Specifically, DC is passed through an electrolyzer filled with electrolyte, and water molecules undergo an electrochemical reaction at the electrodes, decomposing into hydrogen and oxygen. Electrolytic hydrogen production can be widely used in various applications requiring hydrogen, including fuel cells, chemical raw materials, and energy storage. By converting other energy sources (such as solar and wind energy) into hydrogen energy for storage, energy storage and transportation problems can be effectively solved. The impedance of the electrolyzer is a key parameter reflecting its electrochemical performance and operating status; impedance measurement is crucial for optimizing processes, ensuring safety, and extending equipment lifespan.

[0003] Existing electrolytic cell impedance measurement technologies require point-by-point frequency sweeping across a wide frequency range (e.g., 0Hz–100kHz), with a single complete measurement taking several minutes or even longer. During this time, the electrolytic cell's operating state may change, leading to decreased accuracy in low-frequency data and failing to capture rapid impedance changes under complex operating conditions such as fluctuating loads and start-up / shutdown transients. Furthermore, impedance analysis based solely on a single time scale (e.g., high or low frequency) cannot simultaneously correlate short-term transient responses with long-term performance degradation patterns.

[0004] Therefore, there is a need to provide an active control method for multi-timescale electrolytic cell impedance detection to improve the accuracy and efficiency of electrolytic cell impedance measurement under dynamic operating conditions. Summary of the Invention

[0005] This invention provides an actively controlled multi-timescale electrolytic cell impedance detection method, comprising: determining at least two timescales corresponding to the current detection target; determining an initial frequency band based on the current electrolytic cell type; collecting electrolytic cell operating condition information; and determining the optimal frequency band, signal amplitude, and data acquisition interval for the timescale corresponding to the current detection target based on the electrolytic cell operating condition information and the initial frequency band; performing multi-frequency band alternating scanning on the electrolytic cell based on the optimal frequency band, signal amplitude, and data acquisition interval for each timescale to acquire scanning data; and generating electrolytic cell impedance test results based on the scanning data.

[0006] Furthermore, based on the current detection target, at least two time scales are determined, including: determining multiple preset detection targets; for each preset detection target, determining at least two time scales corresponding to the preset detection target; and based on the at least two time scales corresponding to each preset detection target, determining at least two time scales corresponding to the current detection target.

[0007] Further, determining the initial frequency band based on the current electrolytic cell type includes: determining multiple preset electrolytic cell types; for each preset electrolytic cell type, determining the initial frequency band corresponding to the preset electrolytic cell type; and determining the initial frequency band based on the initial frequency band corresponding to each preset electrolytic cell type and the current electrolytic cell type.

[0008] Furthermore, the electrolytic cell operating information includes at least the real-time current density, real-time electrolytic cell temperature, real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance.

[0009] Furthermore, based on the electrolytic cell operating condition information and the initial frequency band, the optimal frequency band for the current detection target corresponding to the time scale is determined, including: determining the polarization state of the electrolytic cell based on the real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance; determining the maximum or minimum frequency threshold based on the polarization state of the electrolytic cell, the real-time current density, and the real-time electrolytic cell temperature; and determining the optimal frequency band for the current detection target corresponding to the time scale based on the maximum or minimum frequency threshold.

[0010] Furthermore, based on the real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance, the polarization state of the electrolytic cell is determined, including: when the ratio of the real-time activation resistance to the real-time ohmic resistance is greater than a first ratio, the polarization state of the electrolytic cell is activation polarization dominant; when the ratio of the real-time mass transfer resistance to the real-time ohmic resistance is greater than a second ratio, the polarization state of the electrolytic cell is mass transfer polarization dominant; when the ratio of the real-time activation resistance to the real-time ohmic resistance is less than or equal to the first ratio, and / or, when the ratio of the real-time mass transfer resistance to the real-time ohmic resistance is greater than or equal to the second ratio, the polarization state of the electrolytic cell is a mixed state.

[0011] Furthermore, based on the polarization state of the electrolytic cell, the real-time current density, and the real-time electrolytic cell temperature, a maximum frequency threshold or a minimum frequency threshold is determined, including: when the polarization state of the electrolytic cell is dominated by activation polarization, the real-time maximum frequency is calculated based on the real-time current density and the real-time electrolytic cell temperature, and the larger value between the preset maximum frequency and the real-time maximum frequency is taken as the maximum frequency threshold; when the polarization state of the electrolytic cell is dominated by mass transfer polarization or a mixed state, the minimum frequency threshold is calculated based on the real-time electrolytic cell temperature.

[0012] Furthermore, based on the electrolytic cell operating condition information and the initial frequency band, the signal amplitude is determined, including: when the polarization state of the electrolytic cell is dominated by activation polarization, the initial signal amplitude is determined based on the maximum amplitude threshold and the basic amplitude, and the signal amplitude is determined based on the real-time electrolytic cell temperature and the initial signal amplitude; when the polarization state of the electrolytic cell is dominated by mass transfer polarization, the signal amplitude is determined based on the basic amplitude, the real-time current density, and the real-time electrolytic cell temperature; when the polarization state of the electrolytic cell is in a mixed state, the signal amplitude is determined based on the real-time electrolytic cell temperature and the basic amplitude.

[0013] Furthermore, based on the electrolytic cell operating condition information and the initial frequency band, the data acquisition interval corresponding to the time scale of the current detection target is determined, including: predicting the impedance change in the future time domain based on the current impedance state and the electrolytic cell operating condition information through model predictive control; and determining the data acquisition interval corresponding to the time scale of the current detection target based on the impedance change in the future time domain.

[0014] Furthermore, based on the scanning data, the electrolytic cell impedance test results are generated, including: determining performance degradation indicators; generating an impedance-time correlation spectrum based on the scanning data; calculating the real-time score of the performance degradation indicator based on the impedance-time correlation spectrum; determining whether to issue a performance degradation warning based on the real-time score of the performance degradation indicator; if not, predicting the periodic score of the performance degradation indicator in the current detection cycle based on the impedance-time correlation spectrum; and determining whether to issue a performance degradation warning based on the periodic score of the performance degradation indicator in the current detection cycle.

[0015] Compared with existing technologies, the active control method for multi-timescale electrolytic cell impedance detection provided by this invention has at least the following advantages: 1. By determining at least two time scales and dynamically adjusting the optimal frequency band and data acquisition interval based on electrolytic cell operating information, multi-frequency band alternating scanning was achieved. Compared with traditional single-time-scale or fixed-frequency-band detection methods, the detection time for a single full-frequency band scan can be reduced by more than 40%.

[0016] 2. The optimal frequency band is dynamically adjusted based on the electrolytic cell operating conditions (such as current, temperature, etc.), avoiding time wasted on ineffective frequency bands and further improving detection efficiency.

[0017] 3. Under dynamic operating conditions, by optimizing signal amplitude and data acquisition interval, the impedance measurement error is reduced to ±3%, significantly improving measurement accuracy compared to ±10% of traditional methods. Dynamically adjusting the signal amplitude based on the electrolytic cell's operating conditions ensures a high signal-to-noise ratio measurement signal under various conditions, reducing noise interference. Dynamically adjusting the data acquisition interval based on operating condition information avoids oversampling under stable conditions or undersampling under drastic changes in conditions, ensuring data accuracy and completeness. Attached Figure Description

[0018] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is a flowchart illustrating an active control method for multi-timescale electrolytic cell impedance detection according to some embodiments of this specification. Figure 2 This is a schematic diagram illustrating the process of preprocessing historical measured data from the upstream of the watershed where the forecast section is located, based on some embodiments of this specification. Detailed Implementation

[0019] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0020] Figure 1 This is a schematic flowchart illustrating an active control method for multi-timescale electrolytic cell impedance detection according to some embodiments of this specification, such as... Figure 1 As shown, an active control method for detecting the impedance of an electrolyzer over multiple time scales may include the following steps.

[0021] Step 110: Determine at least two time scales corresponding to the current detection target.

[0022] In some embodiments, step 110 specifically includes: Define multiple preset testing targets, such as dynamic response monitoring, aging assessment, fault diagnosis, and process optimization; For each preset detection target, at least two time scales are determined. For example, the time scales corresponding to dynamic response monitoring are: transient scale and steady-state scale; the time scales corresponding to aging assessment are: steady-state scale and long-term scale; and the time scales corresponding to fault diagnosis are: transient scale, steady-state scale, and long-term scale. Based on at least two time scales corresponding to each preset detection target, determine at least two time scales corresponding to the current detection target. Specifically, the user can specify the current detection target through the operation interface, and the system will automatically determine at least two time scales corresponding to the current detection target.

[0023] For example, time scales can include: The transient scale, with a data acquisition interval ranging from 0.1 to 100 ms, employs high-frequency sweep (1–10 kHz) to capture the impedance response of the electrolytic cell during sudden load changes and start-up / shutdown. Steady-state scale, with data acquisition intervals ranging from 1 to 10 minutes, using a mid-to-low frequency sweep (0.1–100 Hz) to analyze electrode polarization and electrolyte distribution; Over a long time scale, the data acquisition interval ranges from 1 to 24 hours, and low-frequency scanning (0.01–1 Hz) is used to monitor slow variables such as material corrosion and diaphragm aging.

[0024] There is a relationship between the timescale of an electrolyzer and the physical processes involved. Electrolyzers exhibit differentiated electrochemical behavior at different timescales: Transient scale: Corresponding process: Double-layer charging and discharging, activation polarization transient response during load sudden change or start-up / shutdown.

[0025] Frequency band selection: High-frequency signals can capture rapidly changing charge transfer processes.

[0026] A typical phenomenon is the instantaneous fluctuation of the double-layer voltage on the electrode surface during a current step.

[0027] Steady-state scale: Corresponding process: electrode polarization stabilizes, electrolyte concentration gradient is formed.

[0028] Frequency band selection: The mass transfer polarization characteristic frequency of the mid-to-low frequency coverage reflects the diffusion-limiting process.

[0029] Typical phenomenon: The balance between the electrolyte ion migration rate and the reaction rate during steady-state operation.

[0030] Long-term scale: Corresponding processes: slow changes such as material corrosion, membrane micropore blockage, and catalyst deactivation.

[0031] Frequency band selection: Ultra-low frequency corresponds to hourly time constants and is sensitive to changes in material structure.

[0032] Typical phenomenon: The ionic conductivity of the diaphragm gradually decreases with operating time.

[0033] As an example only, the preset detection target is to monitor the dynamic response and long-term stability of the electrolyzer during operation.

[0034] Corresponding time scale one: transient scale Note: At the transient scale, high-frequency sweep frequency can capture the rapid electrochemical behavior of electrolyzers during sudden load changes or start-up and shutdown, such as double-layer charge-discharge and activation polarization transient responses. By monitoring these transient phenomena, the dynamic response capability of the electrolyzer can be evaluated, and potential performance problems during startup or load changes can be detected in a timely manner.

[0035] Corresponding time scale two: steady state scale Note: At a steady-state scale, low-to-medium frequency sweeps can analyze the electrochemical behavior after electrode polarization has stabilized, such as the formation of electrolyte concentration gradients and the balance between ion migration rates and reaction rates. By monitoring these steady-state phenomena, the performance stability of the electrolyzer under normal operating conditions can be assessed, and potential polarization problems or uneven electrolyte distribution can be detected in a timely manner.

[0036] Another example is the preset testing objective: to assess the aging degree of the electrolytic cell and predict maintenance needs.

[0037] Corresponding time scale one: steady-state scale Note: Although steady-state metrics are primarily used to monitor the steady-state operating status of electrolyzers, long-term, continuous steady-state monitoring data can also reflect the aging trend of electrolyzers. For example, by monitoring long-term changes in electrolyte ion migration rates and reaction rates, the degree of aging of the electrolyzer can be assessed.

[0038] Corresponding time scale two: Long-term scale Note: Over a long timescale, low-frequency scanning can monitor changes in slow variables such as material corrosion and diaphragm aging in electrolytic cells. These slow variables are the main manifestations of electrolytic cell aging. Through long-term, sparse monitoring, the aging trend of the electrolytic cell can be grasped, maintenance needs can be predicted, and repair or replacement plans can be arranged in advance.

[0039] Understandably, optimizing electrolyzer detection across various time scales allows for rapid response to load surges at transient scales; coverage of major polarization processes at steady-state scales; and sparse monitoring of aging trends at long-term scales. Compared to traditional methods requiring several hours for continuous scanning across the entire frequency band (0.01Hz–10kHz), this significantly reduces overall detection time and improves efficiency. Furthermore, by precisely allocating detection resources based on the electrolyzer's behavioral characteristics at different time scales, sufficient data is obtained at critical time scales while avoiding wasted resources at unnecessary scales. Targeted monitoring of the electrolyzer's behavioral characteristics at different time scales enables more accurate capture of key information and improves the accuracy of monitoring results.

[0040] Step 120: Collect electrolytic cell operating condition information, and based on the electrolytic cell operating condition information, determine the optimal frequency band for the time scale corresponding to the current detection target.

[0041] In some embodiments, determining the optimal frequency band for the time scale corresponding to the current detection target based on electrolytic cell operating condition information includes: The polarization state of the electrolyzer is determined based on the real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance. The maximum or minimum frequency threshold is determined based on the polarization state of the electrolytic cell, the real-time current density, and the real-time temperature of the electrolytic cell. The optimal frequency band for the current detection target is determined based on the maximum or minimum frequency threshold.

[0042] In some embodiments, the polarization state of the electrolyzer is determined based on the real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance, including: When the ratio of the real-time activation resistance to the real-time ohmic resistance is greater than the first ratio (e.g., 3), the polarization state of the electrolyzer is dominated by activation polarization. When the ratio of real-time mass transfer resistance to real-time ohmic resistance is greater than a second ratio (e.g., 2), the polarization state of the electrolyzer is dominated by mass transfer polarization. When the ratio of real-time activation resistance to real-time ohmic resistance is less than or equal to a first ratio, and / or the ratio of real-time mass transfer resistance to real-time ohmic resistance is greater than or equal to a second ratio, the polarization state of the electrolytic cell is a mixed state.

[0043] In some embodiments, determining the maximum or minimum frequency threshold based on the polarization state of the electrolyzer, real-time current density, and real-time electrolyzer temperature includes: When the polarization state of the electrolyzer is dominated by activation polarization, the real-time maximum frequency is calculated based on the real-time current density and the real-time electrolyzer temperature. The larger value between the preset maximum frequency and the real-time maximum frequency is taken as the maximum frequency threshold. When the polarization state of the electrolyzer is dominated by mass transfer polarization or a mixed state, the minimum frequency threshold is calculated based on the real-time electrolyzer temperature.

[0044] Specifically, when the polarization state of the electrolyzer is dominated by activation polarization, the maximum frequency threshold can be determined according to the following formula: , in, The maximum frequency threshold, For real-time current density, This refers to the real-time temperature of the electrolytic cell.

[0045] When the polarization state of the electrolyzer is dominated by mass transfer polarization or a mixed state, the minimum frequency threshold can be calculated according to the following formula: , in, This is the minimum frequency threshold.

[0046] For each time scale corresponding to the current detection target, if the maximum value of the frequency band corresponding to the time scale is greater than the maximum frequency threshold, then the maximum value of the frequency band corresponding to the time scale is replaced with the maximum frequency threshold; if the minimum value of the frequency band corresponding to the time scale is less than the minimum frequency threshold, then the minimum value of the frequency band corresponding to the time scale is replaced with the minimum frequency threshold.

[0047] Step 130: Determine the signal amplitude based on the electrolytic cell operating condition information.

[0048] Specifically, it includes: When the polarization state of the electrolyzer is dominated by activation polarization, the initial signal amplitude is determined based on the maximum amplitude threshold and the basic amplitude, and the signal amplitude is determined based on the real-time electrolyzer temperature and the initial signal amplitude. When the polarization state of the electrolyzer is dominated by mass transfer polarization, the signal amplitude is determined based on the base amplitude, real-time current density, and real-time electrolyzer temperature. When the polarization state of the electrolytic cell is mixed, the signal amplitude is determined based on the real-time electrolytic cell temperature and the basic amplitude.

[0049] Specifically, the formula for calculating the basic amplitude is as follows: , in, The base amplitude.

[0050] The formula for calculating the maximum amplitude threshold is: , in, This is the threshold for the maximum amplitude.

[0051] When the polarization state of the electrolyzer is dominated by activation polarization, the initial signal amplitude can be determined according to the following formula: , in, This is the initial signal amplitude.

[0052] The signal amplitude can be determined based on the base amplitude, real-time current density, and real-time electrolytic cell temperature using the following formula: , in, This represents the signal amplitude.

[0053] When the polarization state of the electrolytic cell is dominated by mass transfer polarization, the signal amplitude can be determined according to the following formula: , Reduce the amplitude to avoid electrolyte turbulence interfering with the diffusion process.

[0054] When the polarization state of the electrolytic cell is mixed, the signal amplitude can be calculated using the following formula: , The conductivity of the electrolyte increases with increasing temperature, so the amplitude needs to be appropriately increased to maintain the signal-to-noise ratio.

[0055] Step 140: Based on the electrolytic cell operating condition information, determine the data acquisition interval corresponding to the time scale of the current detection target. In some embodiments, determining the data acquisition interval corresponding to the time scale of the current detection target based on the electrolytic cell operating condition information and the initial frequency band includes: Model Predictive Control (MPC) predicts future impedance changes in the time domain based on the current impedance state and electrolyzer operating conditions. Based on the impedance changes in the future time domain, the data acquisition interval corresponding to the time scale of the current detection target is determined.

[0056] Specifically, it may include the following steps: 1. Prediction Model Construction: Impedance Response Prediction Based on Equivalent Circuit Model (ECM) (1) Discretization and State-Space Representation of ECM Model State variables: Electric double-layer voltage (EDV) represents the potential difference between the electrode surface and the electrolyte, and is an important indicator of the electrochemical behavior of an electrolytic cell.

[0057] Diffusion capacitance reflects the dynamic characteristics of the ion diffusion process in the electrolyte.

[0058] Input variables: Alternating current disturbance is used to stimulate the impedance response of the electrolytic cell.

[0059] Scanning frequency, i.e., the frequency value currently used for detection.

[0060] Output variables: The predicted impedance value is calculated using the Randles equivalent circuit model and includes parameters such as ohmic resistance Rohm, activation polarization resistance Ract, double-layer capacitance Cdl, and Warburg impedance (expressed as σ / jω).

[0061] By converting the equivalent circuit model of the electrolytic cell into discrete state-space equations, MPC can predict future impedance responses based on the current state and inputs.

[0062] For each time scale, a prediction model corresponding to that time scale can be established.

[0063] (2) Frequency domain response prediction Discrete Fourier Transform: Time-domain perturbation signal The current amplitude and phase information at different frequencies are obtained by converting the excitation to the frequency domain.

[0064] Impedance response prediction: The impedance response at each frequency point is predicted using a state-space model, including the amplitude and phase of the impedance.

[0065] MPC assesses the impedance response at different frequencies to determine the information value of each frequency band, such as whether it covers the characteristic frequency of the electrolyzer and whether the signal-to-noise ratio is high enough.

[0066] 2. Rolling optimization mechanism MPC performs the following steps in each control cycle, balancing measurement efficiency and data quality through an objective function: Dynamic condition constraints: Frequency range constraint: Limit the range of scanning frequencies to ensure that detection is performed within the effective frequency band and avoid invalid or redundant scanning. This is dynamically adjusted through step 120.

[0067] Disturbance amplitude constraint: Control the amplitude of AC current disturbance to prevent excessive interference to the electrolytic cell. This is dynamically adjusted through step 130.

[0068] Scan time constraint: Limit the time of each scan to ensure that the detection process is not too long and affects the normal operation of the electrolytic cell.

[0069] Objective function design: The objective function consists of two parts: one part is the data value term (such as signal-to-noise ratio, characteristic frequency coverage, impedance characteristic capture efficiency, etc.), which is used to measure the contribution of the collected data to the electrolyzer state assessment. The weight of the data value in the objective function is dynamically adjusted according to the data quality. If the data quality is low and the interference is large, the weight of the data value in the objective function is increased. The other part is the acquisition cost term, which is used to measure the resource consumption (such as time, energy consumption, etc.) in the data acquisition process.

[0070] MPC achieves a balance between measurement efficiency and data quality by finding the minimum value of the objective function to obtain the optimal test interval sequence.

[0071] 3. Solution aspect Quadratic Programming (QP) Solver: The data acquisition time interval is optimized online using a quadratic programming solver such as IPOPT (Interior Point OPTimizer).

[0072] The QP (Quadratic Programming Solver) solver can transform rolling optimization problems into standard quadratic programming problems and obtain the optimal solution through iterative solving.

[0073] Online optimization means that the solver can adjust the data acquisition time interval in real time during the operation of the electrolyzer to adapt to changes in the state of the electrolyzer.

[0074] Traditional model predictive control (MPC) is commonly used in industrial control for real-time optimization of dynamic systems. However, its design is mostly for general scenarios and is not customized for the multi-timescale characteristics and dynamic operating conditions of electrolytic cells, resulting in the following shortcomings: Fixed time scale: The data acquisition interval is based on fixed rules and cannot be dynamically adjusted according to different time scales (transient, steady state, long-term) of the electrolyzer.

[0075] Single optimization objective: The objective function focuses on controlling stability or energy consumption, without taking into account the unique data value of the electrolyzer (such as impedance characteristic capture efficiency).

[0076] Static constraints: parameters such as frequency band and amplitude are fixed and cannot be adjusted in real time according to polarization state or temperature.

[0077] Step 140 has undergone the following targeted improvements: 1. Dynamic modeling at multiple time scales The impedance response of the electrolyzer is decomposed into three time scales: transient, steady-state, and long-term. An independent prediction model is established for each scale to accurately match the dynamic characteristics of the electrolyzer.

[0078] 2. Dynamically adjust the data acquisition interval to achieve adaptive operation. Based on real-time operating conditions, the constraints of MPC, such as frequency range and disturbance amplitude, are dynamically adjusted according to polarization state, temperature, and current density.

[0079] 3. Customized design of the objective function The MPC objective function incorporates data value metrics specific to electrolytic cells (such as signal-to-noise ratio, characteristic frequency coverage, impedance feature capture efficiency) and acquisition cost metrics.

[0080] 4. Coordinated optimization of frequency band and acquisition interval By jointly optimizing frequency band selection and data acquisition intervals, rather than processing them independently, resource allocation becomes more efficient and redundant detection is avoided.

[0081] 5. Online real-time adjustment and feedback closed loop By dynamically adjusting the parameters of the MPC model through real-time data feedback, a closed-loop optimization is formed, enhancing robustness.

[0082] Step 150: Based on the optimal frequency band, signal amplitude, and data acquisition interval corresponding to each time scale, perform multi-frequency band alternating scanning on the electrolytic cell to acquire scanning data.

[0083] Specifically, it can integrate a multi-channel signal generator and a high-speed data acquisition device to support multi-point synchronous measurement and acquire scanning data. Through impedance transformer coupling with a DC power supply, seamless superposition of AC disturbance signals and DC power supply can be achieved.

[0084] This is just an example of transient response capture in a wind power-to-hydrogen scenario. A 2MW PEM electrolyzer participates in grid frequency regulation; the scan parameters are: Transient scaling: Performs a 1–10 kHz frequency sweep every 100 ms to capture dynamic changes in Ract.

[0085] Steady-state scale: A frequency sweep of 0.1–100 Hz is performed every 5 minutes to analyze the electrolyte mass transfer state.

[0086] The corresponding scanning process is as follows: In actual operation, transient and steady-state scale scans are executed alternately. That is, after a transient scale scan is performed every 100ms, the system waits until the next 5-minute steady-state scale scan trigger time before executing the steady-state scan. This ensures that the two scans are synchronized in time; that is, the transient scale scan does not interfere with the execution of the steady-state scale scan, and vice versa. The data obtained from the transient and steady-state scale scans are integrated to form a complete dataset for subsequent comprehensive analysis.

[0087] Another example is the monitoring of diaphragm aging in alkaline electrolyzers. In a 10MW alkaline electrolyzer operating continuously, it is necessary to monitor the low-frequency impedance drift caused by diaphragm blockage.

[0088] Long-term scale: 0.01–1 Hz frequency sweep is performed every 1 hour.

[0089] Steady-state scale: A frequency sweep of 0.1–100 Hz is performed every 30 minutes to verify the stability of mass transfer polarization.

[0090] Step 160: Based on the scan data, generate the electrolytic cell impedance test results.

[0091] Specifically, it includes: Determine performance degradation indicators; Based on the scan data, an impedance-time correlation spectrum is generated; Based on the impedance-time correlation spectrum, the real-time score of the performance degradation index is calculated; Based on the real-time score of the performance degradation index, determine whether to issue a performance degradation warning; If not, based on the impedance-time correlation spectrum, predict the period score of the performance degradation index in the current detection period; Based on the periodic score of the performance degradation index in the current detection period, determine whether to issue a performance degradation warning.

[0092] Specifically, impedance features can be extracted from the scan data, and impedance-time correlation plots can be generated based on the extracted impedance features. For example, transient-scale feature extraction: extract the Ract change curve over time from high-frequency impedance data. Steady-state-scale feature extraction: extract ohmic resistance from low-frequency impedance data, reflecting the ion conduction characteristics of the electrolyte. Since the time intervals between transient and steady-state scans differ significantly, it is recommended to display them separately using dual time axes or subplots. The main plot shows the Ract change curve over time at the transient scale, with time axis intervals of 100ms. The subplots show the key impedance parameters (such as Rohm and Rdiff) change over time at the steady-state scale, with time axis intervals of 5min.

[0093] The real-time score of the performance degradation index can be calculated using the following formula: , in, This is the real-time score for the performance degradation metric. To activate the polarization resistor in real time, The activation polarization resistance in the initial state. This is a real-time diffusion coefficient related term, reflecting the dynamic changes in the diffusion capacity of ions or reactants in the electrolyte. The diffusion coefficient is the baseline value under the initial conditions. This represents the real-time ohmic impedance of the electrolytic cell. This represents the ohmic impedance of the electrolytic cell in its initial state. and As weight, and Greater than 0, .

[0094] if If it is less than 1.1, then it is considered normal and no response measures are needed; if If the value is between 1.1 and 1.3, the warning level is Level 1, and measures such as logging and increasing the frequency of long-term scanning are adopted.

[0095] The period score of the performance degradation index in the current detection period is calculated using the following formula: , in, These are predicted values ​​for performance degradation metrics at future points in time. This represents the real-time score for the performance degradation metric, along with the time window length for historical data. For example, p=24 indicates that data from the past 24 detection periods will be used. For the first Weighting coefficients for each historical time point; For historical time points arrive Changes in performance degradation indicators; The random error term, assumed to follow a normal distribution with a mean of 0, is used to characterize random fluctuations not explained by the model, such as noise interference and unmodeled factors.

[0096] when When the value exceeds the threshold, a performance degradation warning will be issued.

[0097] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for actively controlling multi-timescale electrolytic cell impedance detection, characterized in that, include: Determine at least two time scales corresponding to the current detection target, wherein the time scales are transient, steady-state, or long-term; Collect electrolytic cell operating condition information, and based on the electrolytic cell operating condition information, determine the optimal frequency band of the time scale corresponding to the current detection target. The electrolytic cell operating condition information includes real-time current density, real-time electrolytic cell temperature, real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance. The signal amplitude is determined based on the electrolytic cell operating condition information; Based on the electrolytic cell operating condition information, determine the data acquisition interval corresponding to the time scale of the current detection target; Based on the optimal frequency band, signal amplitude, and data acquisition interval corresponding to each time scale, the electrolytic cell is scanned alternately in multiple frequency bands to obtain scanning data. Based on the scan data, generate electrolytic cell impedance test results; Among these, determining the optimal frequency band for the time scale corresponding to the current detection target based on electrolytic cell operating condition information includes: The polarization state of the electrolytic cell is determined based on the real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance, wherein the polarization state of the electrolytic cell is either activation polarization-dominated, mass transfer polarization-dominated, or a mixed state. The maximum or minimum frequency threshold is determined based on the polarization state of the electrolytic cell, the real-time current density, and the real-time temperature of the electrolytic cell. The optimal frequency band for the current detection target is determined based on the maximum or minimum frequency threshold. Based on the electrolyzer operating condition information, the signal amplitude is determined, including: When the polarization state of the electrolyzer is dominated by activation polarization, the initial signal amplitude is determined based on the maximum amplitude threshold and the basic amplitude, and the signal amplitude is determined based on the real-time electrolyzer temperature and the initial signal amplitude. When the polarization state of the electrolyzer is dominated by mass transfer polarization, the signal amplitude is determined based on the base amplitude, real-time current density, and real-time electrolyzer temperature. When the polarization state of the electrolytic cell is mixed, the signal amplitude is determined based on the real-time electrolytic cell temperature and the basic amplitude.

2. The active control method for detecting the impedance of an electrolytic cell across multiple time scales according to claim 1, characterized in that, Based on the current detection objectives, at least two time scales are determined, including: Identify multiple preset detection targets; For each preset detection target, determine at least two time scales corresponding to the preset detection target; Based on at least two time scales corresponding to each preset detection target, determine at least two time scales corresponding to the current detection target.

3. The active control method for detecting the impedance of an electrolytic cell across multiple time scales according to claim 1, characterized in that, The polarization state of the electrolyzer is determined based on real-time activation resistance, real-time mass transfer resistance, and real-time ohmic resistance, including: When the ratio of the real-time activation resistance to the real-time ohmic resistance is greater than the first ratio, the polarization state of the electrolytic cell is dominated by activation polarization. When the ratio of real-time mass transfer resistance to real-time ohmic resistance is greater than the second ratio, the polarization state of the electrolytic cell is dominated by mass transfer polarization. When the ratio of real-time activation resistance to real-time ohmic resistance is less than or equal to a first ratio, and / or the ratio of real-time mass transfer resistance to real-time ohmic resistance is greater than or equal to a second ratio, the polarization state of the electrolytic cell is a mixed state.

4. The active control method for detecting the impedance of an electrolytic cell across multiple time scales according to claim 3, characterized in that, The maximum frequency threshold is determined based on the polarization state of the electrolyzer, real-time current density, and real-time electrolyzer temperature, including: When the polarization state of the electrolytic cell is dominated by activation polarization, the real-time maximum frequency is calculated based on the real-time current density and the real-time electrolytic cell temperature. The larger value between the preset maximum frequency and the real-time maximum frequency is taken as the maximum frequency threshold.

5. The active control method for detecting the impedance of an electrolytic cell across multiple time scales according to claim 4, characterized in that, The minimum frequency threshold is determined based on the polarization state of the electrolyzer, real-time current density, and real-time electrolyzer temperature, including: When the polarization state of the electrolyzer is dominated by mass transfer polarization or a mixed state, the minimum frequency threshold is calculated based on the real-time electrolyzer temperature.

6. The active control method for detecting the impedance of an electrolyzer across multiple time scales according to claim 3, characterized in that, Based on the electrolyzer operating condition information, determine the data acquisition interval corresponding to the time scale of the current detection target, including: By using model predictive control, the impedance changes in the future time domain are predicted based on the current impedance state and electrolyzer operating information. Based on the impedance changes in the future time domain, the data acquisition interval corresponding to the time scale of the current detection target is determined.

7. The active control method for detecting the impedance of an electrolytic cell across multiple time scales according to claim 1 or 2, characterized in that, Based on the scan data, the electrolytic cell impedance test results are generated, including: Determine performance degradation indicators; Based on the scan data, an impedance-time correlation spectrum is generated; Based on the impedance-time correlation spectrum, the real-time score of the performance degradation index is calculated; Based on the real-time score of the performance degradation index, determine whether to issue a performance degradation warning; If not, based on the impedance-time correlation spectrum, predict the period score of the performance degradation index in the current detection period; Based on the periodic score of the performance degradation index in the current detection period, determine whether to issue a performance degradation warning.

Images