Fuzzy adaptive power allocation method for electric-hydrogen hybrid energy storage

Abstract

The present invention relates to the field of electric energy storage technology, and aims to provide a fuzzy adaptive electric-hydrogen hybrid energy storage power distribution method. The method is to use a Butterworth low-pass filter with adaptively adjusted parameters to filter the power to be absorbed in a microgrid system including an electric-hydrogen hybrid energy storage system, and respectively allocate the low-frequency power and the filtered power to the hydrogen production electrolyzer and the energy storage battery to realize the joint absorption of electric-hydrogen hybrid energy storage; in the filtering process, the electrolyzer temperature and the battery SOC are used as input variables and the cutoff frequency of the filter is obtained based on the fuzzy inference rule, and the parameters of the filter are adaptively adjusted to maintain the electrolyzer absorption power and the battery SOC within a preset range. The present invention can reduce the energy waste caused by response delay and improve the stability of the system; reduce the risk of battery overcharge and over-discharge, and extend the battery life compared to the traditional power distribution method; enhance the adaptability of the system and improve the overall operating efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of electric energy storage technology, and specifically relates to an electric-hydrogen hybrid energy storage power distribution technology, and more particularly to a fuzzy adaptive electric-hydrogen hybrid energy storage power distribution method based on the dynamic response characteristics of an electrolyzer and the state of charge of a battery. Background Art

[0002] With the growth of renewable energy capacity and the diversification of energy demand on the power load side, the role of energy storage in power systems is becoming increasingly prominent. Hybrid energy storage systems based on hydrogen electrolyzers and lithium batteries are becoming a research hotspot, as they combine the fast response characteristics of batteries with the long-term energy storage capabilities of electrolyzers.

[0003] For hybrid energy storage systems, the commonly used power allocation methods currently include the following categories: (1) Power allocation methods based on signal processing, such as variational mode decomposition, which distinguish the power of different frequency components by signal decomposition; (2) Power allocation methods based on hierarchical control, which coordinate the power response of batteries and electrolyzers through a multi-layer control architecture; (3) Power allocation methods based on optimization algorithms, such as model predictive control, which determine the power allocation strategy by optimizing the objective function; (4) Power allocation methods based on filters, which utilize the frequency separation characteristics of filters to allocate power fluctuations of different frequencies to different energy storage components. Among these methods, filters have great application potential in hybrid energy storage systems composed of energy storage units with similar or identical performance due to their advantages such as low computational complexity, strong real-time performance, and the ability to achieve efficient power frequency division when designed properly.

[0004] However, for electric-hydrogen hybrid energy storage systems, the electrolyzer has a slow dynamic response speed, which is significantly different from the fast response characteristics of batteries. The existing power distribution methods often result in energy waste during application, or the electrolyzer cannot keep up with the input power, resulting in unstable operation.

[0005] Therefore, it is necessary to propose a new solution to achieve reasonable power allocation in order to solve the above key problems. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the deficiencies in the prior art and provide a fuzzy adaptive electric-hydrogen hybrid energy storage power distribution method.

[0007] To solve the technical problem, the solution of the present invention is:

[0008] A fuzzy adaptive electric-hydrogen hybrid energy storage power distribution method is provided. In a microgrid system including an electric-hydrogen hybrid energy storage system, a Butterworth low-pass filter with adaptively adjusted parameters is used to filter the power to be absorbed, and the low-frequency power and the filtered power are respectively distributed to the hydrogen production electrolyzer and the energy storage battery to realize the joint absorption of the electric-hydrogen hybrid energy storage. During the filtering process, the electrolyzer temperature and the battery SOC are used as input variables and the cutoff frequency of the filter is obtained based on fuzzy inference rules. The parameters of the filter are adaptively adjusted to maintain the electrolyzer absorption power and the battery SOC within a preset range.

[0009] As a preferred solution of the present invention, obtaining the cutoff frequency of the filter based on the fuzzy inference rule specifically includes the following operations:

[0010] (1) Define the input variables as the electrolyzer temperature and battery SOC, and define the output variable as the cutoff frequency of the filter ;

[0011] (2) Define membership functions, including the electrolytic cell temperature membership function, the SOC membership function, and the filter cutoff frequency membership function;

[0012] (3) Define the evaluation rules in the rule base;

[0013] (4) Merge the output fuzzy sets of the rules and use the centroid method to defuzzify the merged membership function to obtain the final filter cutoff frequency .

[0014] As a preferred embodiment of the present invention, the range of the electrolytic cell temperature is arrive , the battery SOC ranges from 0 to 1;

[0015] The electrolytic cell temperature membership function includes a low temperature membership function, a medium temperature membership function and a high temperature membership function, wherein:

[0016] Low temperature membership function : °C to °C, triangular function, peak at °C;

[0017]

[0018] Medium temperature membership function : °C to °C, triangular function, peak at °C;

[0019]

[0020] High temperature membership function : °C to °C, triangular function, peak at °C;

[0021]

[0022] The SOC membership function includes a low SOC membership function, a moderate SOC membership function and a high SOC membership function, wherein:

[0023] Low SOC membership function : 0% to 70%, triangular function, peak at 0%;

[0024]

[0025] Moderate SOC membership function : 40% to 60%, triangular function, peak at 50%;

[0026]

[0027] High SOC membership function : 30% to 100%, triangular function, peak at 100%;

[0028]

[0029] The cutoff frequency membership function of the filter includes a low cutoff frequency membership function, a medium cutoff frequency membership function and a high cutoff frequency membership function, wherein:

[0030] Low cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0031]

[0032] Medium cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0033]

[0034] High cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0035]

[0036] The rule base includes the following evaluation rules:

[0037] Rule 1: If the electrolyzer temperature is high and the SOC is moderate, then the filter cutoff frequency is high; the output high cutoff frequency membership generated by this rule is: ;

[0038] Rule 2: If the electrolyzer temperature is low and the SOC is moderate, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: ;

[0039] Rule 3: If the cell temperature is high and the SOC is low or high, then the filter cutoff frequency is moderate; the output moderate cutoff frequency membership generated by this rule is: ;

[0040] Rule 4: If the cell temperature is low and the SOC is low or high, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: ;

[0041] Rule 5: If the cell temperature is moderate and the SOC is low or high, then the filter cutoff frequency is low; this rule produces an output moderate cutoff frequency membership of: ;

[0042] Rule 6: If the electrolyzer temperature is moderate and the SOC is moderate, then the filter cutoff frequency is moderate; the output moderate cutoff frequency membership generated by this rule is: ;

[0043] The membership function after merging the output fuzzy sets is:

[0044]

[0045] The final filter cutoff frequency obtained by defuzzification using the centroid method As shown below:

[0046] ;

[0047] In the above formulas, T is the electrolytic cell temperature; S is the battery SOC; f is the filter cutoff frequency; and are the filter cutoff frequencies corresponding to the lower and upper operating temperature limits of the electrolytic cell respectively; It refers to the rule intersection operation; It refers to the rule merging operation; and are the lower and upper limits of the cutoff frequency range, respectively.

[0048] As a preferred solution of the present invention, the membership function after merging the output fuzzy sets is is a piecewise function, which can be further The segmented interval is split and the cutoff frequency of the segmented interval is calculated;

[0049] Assumptions exist There are different expressions on , then:

[0050]

[0051]

[0052] for In the interval The expression on The total number of different expressions.

[0053] As a preferred solution of the present invention, a frequency response test of a rated voltage step input is performed on the electrolyzer in the electric-hydrogen hybrid energy storage system. and upper operating temperature limit Carry out step response experiments respectively; perform data point identification and frequency response analysis on the experimental results to obtain the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolyzer and .

[0054] As a preferred embodiment of the present invention, the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolytic cell is obtained. and The methods include:

[0055] (1) Multiply the voltage and current output of the electrolyzer to obtain the power curve:

[0056]

[0057] in, is the voltage value of the voltage data point, is the current value of the current data point, For time point;

[0058] (2) Use image processing software to read the step response results in the power curve to obtain step response data;

[0059] (3) The step response data is used as the input of the MATLAB software, and the tfest function is used for fitting. The number of zeros is set to 1, and the number of poles is set to 2 and 1 respectively. The best fitting effect is selected, that is, R 2 The number of poles closest to 1 is used to obtain the fitted transfer function;

[0060] If the number of poles is 1 and the number of zeros is 1, then the transfer function is:

[0061] (1)

[0062] in, is the zero position, is the extreme position, is the transfer function gain factor;

[0063] If the number of poles is 2 and the number of zeros is 1, the general formula of its transfer function is:

[0064] (2)

[0065] in, is the zero position, and is the extreme position, is the transfer function gain factor;

[0066] According to the above fitting process, the parameter values in the general formula of the transfer function are determined;

[0067] (4) Based on the fitted transfer function, the Bode function is used to draw the Bode diagram. According to the logarithmic amplitude-frequency characteristics of the Bode diagram, the frequency corresponding to the gain attenuation of -3dB is found, that is, the cutoff frequencies corresponding to the lower and upper operating temperatures of the electrolytic cell are obtained. and .

[0068] As a preferred embodiment of the present invention, during the operation of the electric-hydrogen hybrid energy storage system, the temperature of the electrolyzer and the SOC of the battery are measured regularly according to the set time interval; the latest measurement data are used to calculate the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolyzer. and , and then further recalculate the cutoff frequency of the filter according to the fuzzy inference rule, thereby realizing adaptive adjustment of the filter parameters.

[0069] As a preferred solution of the present invention, the cutoff frequency obtained based on the fuzzy inference rule is suitable for analog Butterworth filter, which needs to be further converted into a digital filter by bilinear transformation; specifically,

[0070] The transfer function of the analog Butterworth filter is:

[0071]

[0072] ,

[0073] Where, is the angular frequency of the transfer function; is the filter cutoff frequency obtained based on the fuzzy inference rule; s is the complex frequency variable used for Laplace transform;

[0074] make , substituting into the above transfer function, the transfer function of the digital filter is obtained as follows:

[0075]

[0076] Where M is the sampling period; , ; z is the complex variable used for z-transformation.

[0077] As a preferred embodiment of the present invention, the hydrogen production electrolyzer is an alkaline solution electrolyzer; the energy storage battery is a lithium battery cell, or a battery pack composed of multiple lithium battery cells.

[0078] Description of the invention principle:

[0079] After in-depth research, the applicant's inventor team discovered that the dynamic response characteristics of an alkaline electrolyzer are closely related to its operating temperature. Generally, an electrolyzer responds faster at higher temperatures and slower at lower temperatures. Therefore, during power distribution, the filter cutoff frequency needs to be adaptively adjusted based on the electrolyzer's operating temperature to match the cell's dynamic characteristics and optimize power distribution. Designing a suitable filter for an electric-hydrogen hybrid energy storage system containing an alkaline electrolyzer requires in-depth research into the cell's dynamic characteristics and determining the filter's frequency boundaries based on its dynamic response capabilities. Furthermore, the battery state of charge (SOC) in a hybrid energy storage system is crucial to the system's stable operation. Overcharging or over-discharging can affect battery life and reduce the system's long-term benefits. Therefore, during filter design, the battery SOC needs to be considered, and the filter cutoff frequency needs to be dynamically adjusted based on the SOC level.

[0080] Based on these findings, the applicant proposed designing a filter with a variable cutoff frequency that can adjust the power allocation strategy in real time based on changes in electrolyzer temperature and battery SOC, thereby optimizing the operation of the hybrid energy storage system. Compared to traditional fixed-parameter filters, this approach can dynamically adapt to changes in system status and improve the rationality of power allocation.

[0081] Compared with the prior art, the present invention has the following beneficial effects:

[0082] 1. The present invention adopts fuzzy reasoning to adaptively adjust the filter cutoff frequency according to the dynamic response capability of the electrolytic cell at different temperatures, so that the load of the electrolytic cell is more in line with its dynamic characteristics, reducing energy waste caused by response hysteresis and improving system stability.

[0083] 2. The present invention takes into account the battery SOC to adjust the filter cutoff frequency, which can reduce the risk of battery overcharge and over-discharge, and can extend the battery life compared to traditional power distribution methods.

[0084] 3. The variable parameter filter proposed in the present invention can adjust parameters according to the system status, enhance the system's adaptability, enable the battery and electrolyzer to perform their respective functions, reduce the energy conversion loss of the energy storage system, enable it to achieve better power distribution under different operating conditions, and improve overall operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 The microgrid system structure is an electric-hydrogen hybrid energy storage system containing an electrolyzer.

[0086] Figure 2 This is the voltage-current response diagram of a 10kW electrolyzer at 30 degrees Celsius and 92V voltage step.

[0087] Figure 3 This is the voltage-current response diagram of a 10kW electrolyzer at 75 degrees Celsius and 92V voltage step.

[0088] Figure 4 This is the power response diagram of a 10kW electrolyzer at 30 degrees Celsius and 92V voltage step.

[0089] Figure 5 This is the power response diagram of a 10kW electrolyzer at 75 degrees Celsius and 92V voltage step.

[0090] Figure 6 Bode plot for a 10kW electrolyzer at 30 degrees Celsius.

[0091] Figure 7 Bode plot for a 10kW electrolyzer at 75 degrees Celsius.

[0092] Figure 8 The photovoltaic power variation diagram is a specific example.

[0093] Figure 9 The figure shows the electrolytic cell power results for a specific example.

[0094] Figure 10 The figure shows the battery power results for a specific example.

[0095] Figure 11The figure shows the battery SOC change results for a specific example.

[0096] Figure 12 This is the SOC change result diagram of the traditional fixed-frequency filter example. DETAILED DESCRIPTION

[0097] Part I: Implementation of the Invention

[0098] Based on the dynamic response characteristics of the electrolyzer and the battery state of charge (SOC), this paper proposes a fuzzy adaptive power allocation method for electric-hydrogen hybrid energy storage. Specifically, for a hybrid energy storage system comprising a hydrogen production electrolyzer and an energy storage battery, a Butterworth low-pass filter with adaptive parameter adjustment is used to filter the power to be absorbed, sending low-frequency power to the electrolyzer and filtered power to the battery, thereby achieving hybrid energy storage and combined absorption. For the Butterworth low-pass filter, a filter cutoff frequency is obtained based on the electrolyzer frequency response characteristics, taking into account the electrolyzer temperature and battery SOC, and using fuzzy inference rules. By adaptively adjusting the filter parameters, the electrolyzer can absorb appropriate power while maintaining the battery SOC at a reasonable level.

[0099] The hydrogen production electrolyzer described in the present invention may be an alkaline solution electrolyzer; the energy storage battery may be a lithium battery cell, or a battery pack composed of multiple lithium battery cells.

[0100] Based on the above innovative ideas, the implementation scheme of the present invention is as follows:

[0101] 1. Calculate the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolytic cell and .

[0102] The frequency response test of the rated voltage step input is carried out on the hydrogen production electrolyzer. and upper operating temperature limit Carry out step response experiments respectively; perform data point identification and frequency response analysis on the experimental results to obtain the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolyzer and .

[0103] Specifically include:

[0104] (1) Multiply the voltage and current output of the electrolyzer to obtain the power curve:

[0105]

[0106] in, is the voltage value of the voltage data point, is the current value of the current data point, For time point;

[0107] (2) Use image processing software to read the step response results in the power curve to obtain step response data;

[0108] (3) The step response data is used as the input of the MATLAB software, and the tfest function is used for fitting. The number of zeros is set to 1, and the number of poles is set to 2 and 1 respectively. The best fitting effect is selected, that is, R 2 The number of poles closest to 1 is used to obtain the fitted transfer function;

[0109] If the number of poles is 1 and the number of zeros is 1, then the transfer function is:

[0110] (1)

[0111] in, is the zero position, is the extreme position, is the transfer function gain factor; s is the complex frequency variable used for Laplace transform;

[0112] If the number of poles is 2 and the number of zeros is 1, the general formula of its transfer function is:

[0113] (2)

[0114] in, is the zero position, and is the extreme position, is the transfer function gain factor;

[0115] According to the above fitting process, the parameter values in the general formula of the transfer function are determined;

[0116] (4) Based on the fitted transfer function, the Bode function is used to draw the Bode diagram. According to the logarithmic amplitude-frequency characteristics of the Bode diagram, the frequency corresponding to the gain attenuation of -3dB is found, that is, the cutoff frequencies corresponding to the lower and upper operating temperatures of the electrolytic cell are obtained. and .

[0117] 2. Define the input variables as the electrolyzer temperature and battery SOC, and define the output variable as the cutoff frequency of the filter The range of electrolytic cell temperature is arrive , the battery SOC ranges from 0 to 1.

[0118] 3. Define membership functions, including: electrolytic cell temperature membership function, SOC membership function and filter cutoff frequency membership function.

[0119] (1) The temperature membership function of the electrolytic cell includes low temperature membership function, medium temperature membership function and high temperature membership function, where:

[0120] Low temperature membership function : °C to °C, triangular function, peak at °C;

[0121]

[0122] Medium temperature membership function : °C to °C, triangular function, peak at °C;

[0123]

[0124] High temperature membership function : °C to °C, triangular function, peak at °C;

[0125]

[0126] (2) The SOC membership function includes low SOC membership function, moderate SOC membership function and high SOC membership function, where:

[0127] Low SOC membership function : 0% to 70%, triangular function, peak at 0%;

[0128]

[0129] Moderate SOC membership function : 40% to 60%, triangular function, peak at 50%;

[0130]

[0131] High SOC membership function : 30% to 100%, triangular function, peak at 100%;

[0132]

[0133] (3) The cutoff frequency membership function of the filter includes a low cutoff frequency membership function, a medium cutoff frequency membership function and a high cutoff frequency membership function, where:

[0134] Low cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0135]

[0136] Medium cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0137]

[0138] High cutoff frequency membership function : Hz to Hz, triangular function, peak at Hz;

[0139]

[0140] 4. Define the evaluation rules in the rule base.

[0141] The rule base includes the following evaluation rules:

[0142] Rule 1: If the electrolyzer temperature is high and the SOC is moderate, then the filter cutoff frequency is high; the output high cutoff frequency membership generated by this rule is: ;

[0143] Rule 2: If the electrolyzer temperature is low and the SOC is moderate, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: ;

[0144] Rule 3: If the cell temperature is high and the SOC is low or high, then the filter cutoff frequency is moderate; the output moderate cutoff frequency membership generated by this rule is: ;

[0145] Rule 4: If the cell temperature is low and the SOC is low or high, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: ;

[0146] Rule 5: If the cell temperature is moderate and the SOC is low or high, then the filter cutoff frequency is low; this rule produces an output moderate cutoff frequency membership of: ;

[0147] Rule 6: If the electrolyzer temperature is moderate and the SOC is moderate, then the filter cutoff frequency is moderate; the output moderate cutoff frequency membership generated by this rule is: ;

[0148] 5. Merge the output fuzzy sets of the rules, use the centroid method to defuzzify the merged membership function, and obtain the final filter cutoff frequency .

[0149] The membership function after merging the output fuzzy sets is:

[0150]

[0151] The final filter cutoff frequency obtained by defuzzification using the centroid method As shown below:

[0152] ;

[0153] In the above formulas, T is the electrolytic cell temperature; S is the battery SOC; f is the filter cutoff frequency; and are the filter cutoff frequencies corresponding to the lower and upper operating temperature limits of the electrolytic cell respectively; It refers to the rule intersection operation; It refers to the rule merging operation; and are the lower and upper limits of the cutoff frequency range, respectively.

[0154] The membership function after merging the output fuzzy sets is a piecewise function, which can be further The segmented interval is split and the cutoff frequency of the segmented interval is calculated;

[0155] Assumptions exist There are different expressions on , then:

[0156]

[0157]

[0158] in, for In the interval The expression on The total number of different expressions.

[0159] The cutoff frequency obtained above based on fuzzy inference rules is suitable for analog Butterworth filter, which needs to be further converted into digital filter using bilinear transformation;

[0160] First, the second-order Butterworth low-pass filter has a flat passband response. Based on the above-obtained cutoff frequency, the second-order Butterworth low-pass filter is designed. For the analog Butterworth filter, the transfer function is:

[0161]

[0162] ,

[0163] Where, is the angular frequency of the transfer function; is the filter cutoff frequency obtained based on the fuzzy inference rule; s is the complex frequency variable used for Laplace transform;

[0164] Then, the above analog filter is converted into a digital filter using bilinear transformation:

[0165] make , substituting into the above transfer function, the transfer function of the digital filter is obtained as follows:

[0166]

[0167] Where M is the sampling period; , ; z is the complex variable used for z-transformation.

[0168] During the operation of the electric-hydrogen hybrid energy storage system, it is necessary to regularly measure the temperature of the electrolyzer and the SOC of the battery according to the set time interval; the latest measurement data is used to calculate the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolyzer. and , and then further calculate the cutoff frequency of the filter according to the fuzzy inference rule, thereby realizing adaptive adjustment of the filter parameters.

[0169] As is common sense that can be understood by those skilled in the art, in order to implement the power distribution method of the electric-hydrogen hybrid energy storage described in the present invention, it is necessary to apply a computing device, which includes: a memory configured to store instructions; and a processor configured to call the instructions from the memory and implement the aforementioned power distribution method of the electric-hydrogen hybrid energy storage when executing the instructions. At the same time, it is also necessary to apply a computer-readable storage medium, which stores instructions for causing the computer to execute the aforementioned power distribution method of the electric-hydrogen hybrid energy storage. The computing device can be a personal computer, a server, or a network device, etc. The computer storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0170] In practice, the calculation and control process can be implemented using software embedded in the microgrid system's control equipment. Alternatively, the measurement data can be uploaded to a local computer or cloud server via a network connection. After the calculations are complete, control signals are fed back to the hydrogen production electrolyzer and energy storage battery to achieve power distribution and maintain the electrolyzer's power consumption and battery SOC within a preset range.

[0171] Part II Application Examples and Verification

[0172] 1. Figure 1 The following is an example of a microgrid system structure with an electric-hydrogen hybrid energy storage system. It primarily consists of a photovoltaic array, energy storage batteries (such as lithium batteries), a hydrogen electrolyzer, loads, and a DC-DC converter. The energy storage system includes both a battery-based electric energy storage system and an electrolyzer-based hydrogen energy storage system. In actual microgrid systems, fuel cells are optional, but this invention only uses the hydrogen electrolyzer and energy storage battery as the coupling elements for electric-hydrogen hybrid energy storage.

[0173] 2. Take a 10kW alkali electrolyzer as an example. Its operating temperature range is 30 degrees Celsius to 70 degrees Celsius. At 30 degrees Celsius and 70 degrees Celsius, a 92V voltage step signal (slightly lower than the rated voltage 100V for safety reasons) is input to obtain the voltage and current curves at the output end, as shown below: Figure 2 and Figure 3 The upper line is the voltage, the lower line is the current, the voltage is 40V per grid, the current is 10A per grid, the horizontal axis is the time, each grid is 1s, and the power response curve is obtained by reading the voltage and current curve data and multiplying them, as shown in the figure. Figure 4 and Figure 5According to the power response curve, the transfer function and Bode plot corresponding to the two curves can be obtained by using the tfest and bode functions in Matlab. After comparison, the transfer function with 2 poles and 1 zero is the most suitable, as shown in Figure 6 and Figure 7 As shown, according to the Bode diagram, we can know the highest frequency that the alkali electrolyzer can respond to at these two temperatures, that is, the cutoff frequency corresponding to the lower temperature limit and the upper temperature limit. and , 0.132Hz and 0.22Hz respectively.

[0174] 3. Taking the light intensity data of a certain area in Zhejiang Province as an example, the maximum power of the photovoltaic panel is 20kW. The change of its photovoltaic power within about 75 minutes is as follows: Figure 8 As shown in the figure, the battery capacity is set to 0.05 kWh. Due to instrument limitations, the sampling period is set to 4 Hz. The method proposed in this invention uses a Butterworth filter to decompose the photovoltaic power. Low-frequency power is input to the electrolyzer and high-frequency power is input to the battery. The input variables of the fuzzy inference model (electrolyzer temperature and battery SOC) are re-measured every 5 minutes. The filter cutoff frequency is adaptively adjusted. According to the fuzzy inference method proposed in this invention, the corresponding filter cutoff frequency is inferred based on the electrolyzer temperature and SOC at the sampling time point. The inference results are shown below.

[0175] Based on the input temperature of 50.00 °C and SOC of 0.50, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0176] Based on the input temperature of 51.08 °C and SOC of 0.55, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0177] Based on the input temperature of 52.19 °C and SOC of 0.53, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0178] Based on the input temperature of 53.35 °C and SOC of 0.51, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0179] Based on the input temperature of 54.57 °C and SOC of 0.48, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0180] Based on the input temperature of 55.82 °C and SOC of 0.58, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0181] Based on the input temperature of 57.08 °C and SOC of 0.59, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0182] Based on the input temperature of 58.42 °C and SOC of 0.62, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0183] Based on the input temperature of 59.75 °C and SOC of 0.37, the filter cutoff frequency is adjusted to: 0.17 Hz;

[0184] Based on the input temperature of 61.19 °C and SOC 0.07, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0185] Based on the input temperature of 62.73 °C and SOC of 0.13, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0186] Based on the input temperature of 64.32 °C and SOC of 0.22, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0187] Based on the input temperature of 65.93 °C and SOC of 0.23, the filter cutoff frequency is adjusted to: 0.18 Hz;

[0188] Based on the input temperature of 67.55 °C and SOC of 0.30, the filter cutoff frequency is adjusted to: 0.18 Hz.

[0189] 4. In the power distribution process, the Butterworth filter is used and the filter parameters are adaptively adjusted, which can filter in real time and achieve reasonable and effective power distribution between the electrolyzer and the battery.

[0190] Based on the regulation of the method of the present invention, the obtained electrolytic cell power consumption curve and battery power consumption curve are Figure 9 and Figure 10 As shown, Figure 11 The SOC curve of the battery shows that the SOC is maintained within a suitable range. When it approaches the lower limit, it can be adjusted by adjusting the filter cutoff frequency.

[0191] 5. Use a fixed-frequency Butterworth filter to distribute and control power for the same electric-hydrogen hybrid energy storage system. The experimental results show that the fixed-frequency filter cannot adjust the frequency according to the battery SOC and temperature. When the set frequency is too low, assuming it is 0.14Hz, the battery SOC result is as follows: Figure 12 As shown, the battery SOC will be lower than the lower limit, which will have an adverse impact on the battery life and the safety and stability of the system; when the set frequency is too high, assuming it is 0.2Hz, when the electrolyzer temperature is low, it will not be able to follow and cause energy waste, and may also affect its operating stability.

[0192] In contrast, the adaptive Butterworth filter proposed in the present invention can ensure that the SOC is within a safe range and the cutoff frequency ensures that the electrolyzer follows its power input as closely as possible.

[0193] After comparison and verification, the advantages and effects of the present invention over the existing technology are mainly reflected in the following aspects: (1) improving the overall efficiency of the energy storage system and reducing unnecessary energy loss; (2) extending the battery life and avoiding performance degradation caused by overcharging or over-discharging; (3) improving the response capability of the electrolyzer, enabling it to better adapt to power fluctuations and improve the operating stability of the system; (4) enhancing the adaptability of the system, enabling it to achieve optimal power distribution under different operating conditions, and improving the economic benefits and reliability of the energy storage system.

[0194] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art may make various variations or modifications within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A fuzzy adaptive electric-hydrogen hybrid energy storage power allocation method, characterized in that: In a microgrid system that includes an electric-hydrogen hybrid energy storage system, a Butterworth low-pass filter with adaptive parameter adjustment is used to filter the power to be absorbed, and the low-frequency power and filtered power are respectively distributed to the hydrogen production electrolyzer and the energy storage battery to achieve joint absorption of the electric-hydrogen hybrid energy storage. During the filtering process, the electrolyzer temperature and the battery SOC are used as input variables and the cutoff frequency of the filter is obtained based on fuzzy inference rules. The filter parameters are adaptively adjusted to maintain the electrolyzer absorption power and the battery SOC within a preset range. The method of obtaining the cutoff frequency of the filter based on the fuzzy inference rule specifically includes the following operations: (1) Define the input variables as the electrolytic cell temperature and battery SOC, and define the output variable as the filter cutoff frequency f cut ; (2) defining membership functions, including the electrolytic cell temperature membership function, the SOC membership function, and the filter cutoff frequency membership function; wherein the electrolytic cell temperature membership function includes a low temperature membership function μ TLow (T), medium temperature membership function μ TMedium (T) and high temperature membership function μ THigh (T); The SOC membership function includes a low SOC membership function μ SLow (S), moderate SOC membership function μ SMedium (S) and high SOC membership function μ SHigh (S); The cutoff frequency membership function of the filter includes a low cutoff frequency membership function μ FLow (f), medium cutoff frequency membership function μ FMedium (f) and high cutoff frequency membership function μ FHigh (f); (3) Define the evaluation rules in the rule base; The rule base includes the following evaluation rules: Rule 1: If the cell temperature is high and the SOC is moderate, then the filter cutoff frequency is high; the output high cutoff frequency membership generated by this rule is: μ1(f) = μ THigh (T)∧μ SMedium (S)∧μf High (f); Rule 2: If the cell temperature is low and the SOC is moderate, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: μ2(f) = μ TLow (T)∧μ SMedium (S)∧μf Low (f); Rule 3: If the cell temperature is high and the SOC is low or high, then the filter cutoff frequency is moderate; this rule produces an output moderate cutoff frequency membership of: μ3(f) = μ THigh (T)∧(μ SLow (S)∨μ SHigh (S))∧μf Medium (f); Rule 4: If the cell temperature is low and the SOC is low or high, then the filter cutoff frequency is low; the output low cutoff frequency membership generated by this rule is: μ4(f) = μ TLow (T)∧(μ SLow (S)∨μ SHigh (S))∧μf Low (f); Rule 5: If the cell temperature is moderate and the SOC is low or high, then the filter cutoff frequency is low; this rule produces an output with a moderate cutoff frequency membership of: μ5(f) = μ TMedium (T)∧(μ SLow (S)∨μ SHigh (S))∧μf Low (f); Rule 6: If the cell temperature is moderate and the SOC is moderate, then the filter cutoff frequency is moderate; this rule produces an output medium cutoff frequency membership of: μ6(f) = μ TMedium (T)∧μ SMedium (S)∧μf Medium (f); (4) Merge the output fuzzy sets of the rules and use the centroid method to defuzzify the merged membership function to obtain the final filter cutoff frequency in, The membership function after merging the output fuzzy sets is: μ(f)=max(μ1(f),μ2(f),μ3(f),μ4(f),μ5(f),μ6(f)) The final filter cutoff frequency obtained by defuzzification using the centroid method As shown below: In the above formulas, T is the electrolytic cell temperature; S is the battery SOC; f is the filter cutoff frequency; and are the filter cutoff frequencies corresponding to the lower and upper operating temperature limits of the electrolytic cell respectively; ∧ refers to the rule intersection operation; ∨ refers to the rule merging operation; f min and f max are the lower and upper limits of the cutoff frequency range, respectively.

2. The method according to claim 1, characterized in that The range of the electrolytic cell temperature is T min to T max , the battery SOC range is 0 to 1; where, Low temperature membership function μ TLow (T): T min ℃ to Triangular function, with a peak at Medium temperature membership function μ TMedium (T): arrive Triangular function, with a peak at High temperature membership function to T max ℃, triangular function, peak at Low SOC membership function 0% to 70%, triangular function, peak at 0%; Moderate SOC membership function 40% to 60%, triangular function, peak at 50%; High SOC membership function 30% to 100%, triangular function, peak at 100%; Low cutoff frequency membership function arrive Triangular function, with a peak at Medium cutoff frequency membership function arrive Triangular function, with a peak at High cutoff frequency membership function arrive Triangular function, with a peak at 3. The method according to claim 1, characterized in that The membership function μ(f) after merging the output fuzzy sets is a piecewise function, which can be further split according to the segmented intervals of μ(f) to calculate the cutoff frequency of the segmented intervals; Assume μ(f) is in [f1,f2],[f2,f3],...,[f n ,f n+1 ] have different expressions, then: Among them, μ i (f) is μ(f) in the interval [f i ,f i+1 ]; n is the total number of different expressions for μ(f).

4. The method according to claim 1, wherein The frequency response test of the rated voltage step input of the electrolyzer in the electric-hydrogen hybrid energy storage system is carried out at the lower limit of the electrolyzer operating temperature T min and upper operating temperature limit T max Carry out step response experiments respectively; perform data point identification and frequency response analysis on the experimental results to obtain the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolyzer and 5. The method according to claim 4, characterized in that Get the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolytic cell and The methods include: (1) Multiply the voltage and current output of the electrolyzer to obtain the power curve: p(t)=u(t)·i(t) Where u(t) is the voltage value of the voltage data point, i(t) is the current value of the current data point, and t is the time point; (2) Using image processing software to read the step response results in the power curve to obtain step response data; (3) The step response data was used as the input of the MATLAB software and the tfest function was used for fitting. The number of zeros was set to 1, and the number of poles was set to 2 and 1 for fitting respectively. The best fitting result was selected, i.e., R 2 The number of poles closest to 1 is used to obtain the fitted transfer function; If the number of poles is 1 and the number of zeros is 1, then the transfer function is: Where z is the zero position, p is the pole position, and K is the transfer function gain factor; If the number of poles is 2 and the number of zeros is 1, the general formula of its transfer function is: Where z is the zero position, p1 and p2 are the pole positions, and K is the transfer function gain factor; According to the above fitting process, the parameter values in the general formula of the transfer function are determined; (4) Based on the fitted transfer function, the Bode function is used to draw the Bode diagram. According to the logarithmic amplitude-frequency characteristics of the Bode diagram, the frequency corresponding to the gain attenuation of -3dB is found, that is, the cutoff frequencies corresponding to the lower and upper operating temperatures of the electrolytic cell are obtained. and 6. The method according to any one of claims 1 to 5, characterized in that During the operation of the electric-hydrogen hybrid energy storage system, the temperature of the electrolyzer and the SOC of the battery are measured regularly according to the set time intervals; The latest measurement data is used to calculate the cutoff frequency corresponding to the lower and upper operating temperature limits of the electrolytic cell. and Then, the cutoff frequency of the filter is further recalculated according to the fuzzy inference rule, thereby realizing adaptive adjustment of the filter parameters.

7. The method according to any one of claims 1 to 5, characterized in that The cutoff frequency obtained based on the fuzzy inference rule is suitable for analog Butterworth filter, which needs to be further converted into a digital filter using bilinear transformation; specifically, The transfer function of the analog Butterworth filter is: Where, ω c is the angular frequency of the transfer function; is the filter cutoff frequency obtained based on fuzzy inference rules; s is the complex frequency variable used for Laplace transform; make Substituting the above transfer function, the transfer function of the digital filter is obtained as follows: Where M is the sampling period; z is the complex variable used for z-transformation.

8. The method according to any one of claims 1 to 5, characterized in that The hydrogen production electrolyzer is an alkaline solution electrolyzer; the energy storage battery is a lithium battery cell, or a battery pack composed of multiple lithium battery cells.

Images