CN105137362A - Nondestructive online detection and fault diagnosis method for stack - Google Patents

Abstract

The invention discloses a non-destructive on-line detection and fault diagnosis method for an electric stack, which belongs to the technical field of solid oxide fuel cells and electrolytic cell stacks. Use EIS method to obtain impedance spectrum data; use ADIS to analyze impedance spectrum data, and preliminarily determine the number of characteristic peaks and corresponding frequency range; use DRT to analyze impedance spectrum data, select regularization factor, and obtain relaxation time distribution curve The number and frequency range of characteristic peaks; establish an alternative equivalent circuit; select the correct equivalent circuit and determine the initial value of the fitting, and obtain quantitative analysis results; obtain the DRT analysis results under different operating conditions, and compare the influencing factors of each characteristic peak to find The attribution electrochemical process of the DRT characteristic peak is obtained, and the pile fault diagnosis comparison table is obtained; the fault cause analysis is carried out for the repeating unit with poor performance in the test or rapid performance degradation during operation. The experimental efficiency of the stack can be greatly improved and the operating cost can be reduced.

Description

A non-destructive on-line detection and fault diagnosis method for electric stacks

technical field

The invention belongs to the technical field of solid oxide fuel cells and electrolytic cell stacks, in particular to a method for non-destructive on-line detection and fault diagnosis of electric stacks.

Background technique

Solid Oxide Fuel Cells (Solid Oxide Fuel Cells, SOFC) convert the chemical energy of fuel into electrical energy through electrochemical reactions. The solid oxide electrolytic cell (SolidOxideElectrolyticCells, SOEC) is theoretically the reverse operation of SOFC, that is, the raw material water is electrolyzed into hydrogen and oxygen by using external electric energy. In order to increase the power generation capacity of SOFC or the hydrogen production rate of SOEC, multiple repeating units of SOFC/SOEC can be stacked to form a stack.

The operating conditions of the cell stack are harsh (high temperature, high humidity), involving many components and materials. During the experimental test and operation, the performance of each cell is often inconsistent. The performance of one or several cells is weak or missing and affects the performance of the entire electrolytic cell stack. Condition. In addition, operational accidents are unavoidable in actual operation. For example, failure of the preheating system and power outage may cause the electrolytic cell stack to cool down rapidly, and a failure of the constant current and constant voltage source may cause the electrolytic cell stack to operate at an excessively high voltage. At this time, the commonly used on-line testing methods of stacks such as constant current/constant voltage curve method and I-V curve method cannot give more electrochemical information except the apparent electrolytic performance, and it is difficult to diagnose and distinguish the cause and effect of the fault. At present, the fault diagnosis method for the electrolytic cell stack is generally to carry out destructive disassembly, and analyze and find the problem after the experiment. However, this destructive post-disassembly analysis is off-line and non-in-situ. The influence on the coordination state and parameters of each component in the stack is irreversible before and after disassembly, and it is often difficult to find out the cause of the failure even after disassembly. . Therefore, it is necessary to develop a non-destructive and efficient online stack detection method for performance evaluation and fault diagnosis.

Electrochemical impedance spectroscopy (EIS) is a method that uses alternating current signals to disturb the electrochemical system to be measured, measures the feedback voltage signal, and understands the impedance behavior of the electrochemical system to be measured by analyzing the response relationship of the alternating current signal. The EIS method is widely used in the research of SOFC and SOEC single cells and related materials. Since the total impedance can be further distinguished and separated, it is considered to be a powerful tool for battery and material performance evaluation and attenuation analysis. However, such studies mainly focus on the composition, structure, interfacial properties, and transport processes of materials. It is important to understand and optimize the electrolysis performance from the mechanism, but to realize the commercialization of SOFC and SOEC technology, research on battery stacks and modular battery stacks is still needed. Considering that the battery stack not only involves a variety of materials and components, but also has a relatively complex structure. At the same time, due to the increase in the area of the cell, the uneven distribution of the reaction gas and current causes the diffusion of the relaxation phase, which also greatly reduces the reliability of the measured EIS and increases the difficulty of analysis. Therefore, there are very few reports on the application of EIS measurement and analysis methods to battery stacks.

The invention proposes a method for non-destructive on-line detection and fault diagnosis of electric stacks, using electrochemical impedance spectroscopy (EIS) as a testing method; impedance spectrum difference analysis method (ADIS), relaxation time distribution method (DRT), equivalent circuit The organic combination of ECM (ECM) is used as an analysis method to realize the qualitative and quantitative analysis of each electrochemical process in the stack; through the established relationship between the electrochemical process and the DRT characteristic peak, the online non-destructive diagnosis of the performance difference and the cause of the fault can be realized; Heap experimental efficiency, reduce operating costs.

Contents of the invention

The object of the present invention is to propose a method for non-destructive on-line detection and fault diagnosis of electric stacks, which is characterized in that it includes the following steps:

1) Connect the impedance spectrum testing instrument to the stack, and use the electrochemical impedance spectroscopy method (EIS) as a test method to obtain the impedance spectrum data of each unit in the stack;

2) Analyze the impedance spectrum data obtained in step 1) by using the impedance spectrum difference analysis method (ADIS), and initially determine the number of characteristic peaks and the corresponding frequency range;

3) Using the relaxation time distribution method (DRT) to analyze the impedance spectrum data, compare the number of characteristic peaks and the corresponding frequency range to select the regularization factor L, and obtain the number of characteristic peaks and frequency range of the relaxation time distribution curve;

4) Establish an alternative equivalent circuit according to the number of characteristic peaks and frequency range after DRT analysis; compare the characteristic frequency of the equivalent circuit unit components with the characteristic peak-to-peak frequency after DRT analysis, select the correct equivalent circuit and determine the initial fitting value to obtain quantitative analysis results;

5) Repeat step 1) to step 4) to obtain the DRT analysis results under different operating conditions; compare the influencing factors of each characteristic peak to find out the attribution electrochemical process of the DRT characteristic peak, and obtain the pile fault diagnosis comparison table;

6) Using the stack fault diagnosis comparison table obtained in step 5), analyze the fault cause of the repeating unit with poor performance in the test or rapid performance degradation during operation, and find out the reason for the lack of performance and the faulty component.

The key parameters required for the DRT analysis are obtained based on the ADIS analysis results.

The establishment, screening, and selection of the initial fitting value of the equivalent circuit are based on the corresponding relationship between the time constant of the equivalent circuit unit and the characteristic peak in the relaxation time distribution curve.

The pile fault diagnosis comparison table is obtained according to the corresponding relationship between operating conditions and DRT analysis results.

The beneficial effect of the present invention is to propose a non-destructive on-line detection and fault detection method for electrolytic cell stacks, aiming at the problem that the current electrolytic cell stack fault diagnosis method needs to be destructively disassembled, and the problem of off-line and non-in-situ analysis is performed after dismantling. Diagnosis method: Electrochemical impedance spectroscopy (EIS) is used as the test method; Impedance difference analysis method (ADIS), relaxation time distribution method (DRT), and equivalent circuit method (ECM) are organically combined as analysis methods to realize the Qualitative and quantitative analysis of each electrochemical process; through the establishment of the corresponding relationship between the electrochemical process and DRT characteristic peaks, the online non-destructive diagnosis of performance differences and fault causes can be greatly improved. The experimental efficiency of the stack can be greatly improved and the operating cost can be reduced.

Description of drawings

Fig. 1 is a flow chart of a method for non-destructive on-line detection and fault diagnosis of an electric stack.

Figure 2 is a graph of ADIS analysis results.

Figure 3 is a diagram of the DRT analysis results.

Fig. 4 is a schematic diagram of equivalent circuit EC-1 screened by DRT analysis results.

Fig. 5 is a schematic diagram of equivalent circuit EC-2 screened by DRT analysis results.

Figure 6 is the I-V curve of four cells in the stack.

Fig. 7 is a comparison chart of ohmic impedance and polarization impedance of Cell-2 and Cell-3.

Fig. 8 is a schematic diagram of Cell-2 and Cell-3 DRT results in SOFC mode.

Fig. 9 is a graph showing the relationship between the impedance and the temperature of Cell-2 and Cell-3 in SOFC mode.

Figure 10 is a schematic diagram of Nyquist and DRT before and after the accident.

Detailed ways

The present invention proposes a non-destructive on-line detection and fault diagnosis method for electric stacks. The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

Figure 1 shows a flow chart of a non-destructive on-line detection and fault diagnosis method for an electric stack, including the following steps:

1) Connect the impedance spectrum testing instrument to the stack, and use the electrochemical impedance spectroscopy method (EIS) as a test method to obtain the impedance spectrum data of each unit in the stack;

2) Analyze the impedance spectrum data obtained in step 1) by using the impedance spectrum difference analysis method (ADIS), and initially determine the number of characteristic peaks and the corresponding frequency range;

3) Using the relaxation time distribution method (DRT) to analyze the impedance spectrum data, compare the number of characteristic peaks and the corresponding frequency range to select the regularization factor L, and obtain the number of characteristic peaks and frequency range of the relaxation time distribution curve;

4) Establish an alternative equivalent circuit according to the number of characteristic peaks and frequency range after DRT analysis; compare the characteristic frequency of the equivalent circuit unit components with the characteristic peak-to-peak frequency after DRT analysis, select the correct equivalent circuit and determine the initial fitting value to obtain quantitative analysis results;

5) Repeat step 1) to step 4) to obtain the DRT analysis results under different operating conditions; compare the influencing factors of each characteristic peak to find out the attribution electrochemical process of the DRT characteristic peak, and obtain the pile fault diagnosis comparison table;

6) Using the stack fault diagnosis comparison table obtained in step 5), analyze the fault cause of the repeating unit with poor performance in the test or rapid performance degradation during operation, and find out the reason for the lack of performance and the faulty component.

Wherein, the key parameters required for the DRT analysis are obtained based on the ADIS analysis results.

Wherein, the establishment, screening, and selection of the initial fitting value of the equivalent circuit are based on the corresponding relationship between the time constant of the equivalent circuit unit and the characteristic peak in the relaxation time distribution curve.

Wherein, the pile fault diagnosis comparison table is obtained according to the corresponding relationship between operating conditions and DRT analysis results.

At 700°C, SOFC mode, when the hydrogen electrode side gas composition is 100% H ₂ , test the EIS of the stack under different oxygen electrode atmospheres; implement ADIS analysis, the results are shown in Figure 2; compare the impedance in Figure 2 The regularization factor used for DRT analysis is selected for differential change, and its value is 350. The obtained DRT analysis results are shown in Figure 3. In Figure 3, at least four electrochemical processes are distinguished: from low frequency to high frequency, they are called P1, P2 , P3 and P4; use DRT analysis results to screen equivalent circuits EC-1 (as shown in Figure 4) and EC-2 (as shown in Figure 5) and select appropriate fitting initial values to complete the quantitative analysis;

DRT analysis gives the relaxation time distribution corresponding to the impedance spectrum. The extreme value of each characteristic peak corresponds to the characteristic time of an electrochemical process, which is called the relaxation time τ. The calculation formula is:

$\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}\mathrm{<span\; class="notranslate">\; f</span>}}$

where f is the relaxation frequency.

ECM simulates the impedance behavior of the electrochemical system with equivalent circuit elements by establishing an equivalent circuit, where each group/equivalent element corresponds to an electrochemical process. Therefore, the characteristic times of the equivalent elements in a reasonable ECM fitting result should coincide with the relaxation times in the DRT.

For RQ components, the time constant T _RQ can be expressed as:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}\mathrm{<span\; class="notranslate">\; Q</span>}}<span\; class="notranslate">\; =</span>\sqrt[\mathrm{<span\; class="notranslate">\; Q</span>}_{\mathrm{<span\; class="notranslate">\; no</span>}}]{\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; Q</span>}}$

Among them, R is resistance; Q is constant phase angle component value; n _Q is dimensionless parameter.

The time constant T _w of the Warburg element is shown in the following formula:

${\mathrm{<span\; class="notranslate">\; Z</span>}}_{\mathrm{<span\; class="notranslate">\; W</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; w</span>}}\frac{\mathrm{<span\; class="notranslate">\; tanh</span>}\sqrt{{\mathrm{<span\; class="notranslate">\; wxya</span>}}_{\mathrm{<span\; class="notranslate">\; w</span>}}}}{\sqrt{{\mathrm{<span\; class="notranslate">\; wxya</span>}}_{\mathrm{<span\; class="notranslate">\; w</span>}}}}$

Among them, Z _W is the impedance of the Warburg element; R _W is the resistance of the Warburg element; i is the imaginary number symbol; ω is the angular frequency.

The time constant T _G of the Gerischer element satisfies:

${\mathrm{<span\; class="notranslate">\; Z</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}}{\sqrt{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; wxya</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}}}$

Among them, Z _G is the impedance of the Gerischer element; i is the imaginary number symbol; ω is the angular frequency; R _G is the resistance of the Gerischer element.

The relaxation time obtained by the DRT method and the time constant of each equivalent element in the ECM fitting result are shown in Table 1;

Table 1 Comparison table of relaxation time of DRT and time constant of ECM

In Table 1, the relaxation times t ₁ : t ₄ correspond to P1-P4 in the DRT diagram. It can be seen that when EC-1 is used for fitting, the time constants of RQ and Warburg elements in the low frequency band are inconsistent with the relaxation times _t1 and _t2 , and the fitting error is 6.3%. After replacing the Warburg elements with RQ elements, the t ₂ situation is improved, but the gap in t ₁ is still large, and the fitting error is 5.9%. Then, the new initial value was used to re-fit, and the ideal fitting result was obtained, the fitting error was only 1.0%, and each time constant corresponded to the relaxation time one by one. The obtained fitting results corresponding to EC-2 are shown in Table 2;

Table 2 Equivalent circuit fitting result parameters

The corresponding relationship between the electrochemical process and the characteristic peaks of DRT determined through the experiment of changing the operating parameters is shown in Table 3; the above analysis results are used for stack diagnosis.

Table 3 DRT characteristic peak comparison table

(a) Cause Analysis of Questionable Sheets

Figure 6 shows the IV curves of four cells in the stack (700°C, SOFC mode, hydrogen electrode gas composition of 80% H ₂ , 20% H ₂ O), from the IV curves in Figure 6 it can be seen that Cell-2 is four The worst performance problem cell in the cell stack, now try to use the above method to find the reason for the performance difference.

The ohms of Cell-2 and Cell-3 at 700°C, hydrogen electrode atmosphere is 100% hydrogen; 700°C, hydrogen electrode water vapor content is 20%; 750°C, hydrogen electrode water vapor content is 20% The impedance and polarization impedance are compared, as shown in Figure 7; it can be seen that under the three conditions, there is no significant difference in the polarization impedance of Cell-2 and Cell-3, and the difference in resistance is mainly due to the ohmic impedance.

To further prove this conclusion, compare the DRT results of Cell-2 and Cell-3 in SOFC mode when the hydrogen electrode atmosphere changes, the oxygen electrode atmosphere changes, and the temperature changes, as shown in Figure 8. Cell- DRT results of 2 and Cell-3 under (a) hydrogen electrode side atmosphere change, (b) oxygen electrode side atmosphere change, (c) temperature change; because the polarization impedance is related to the peak area of the DRT characteristic peak. It can be seen intuitively that the magnitude and change relationship of the impedances of Cell-2 and Cell-3 under various operating conditions are consistent. It is thus determined that the difference between Cell-2 and Cell-3 does not lie in the polarization impedance.

The ohmic impedance is the main source of the performance difference between Cell-2 and Cell-3, and the ohmic impedance in the battery stack mainly includes the ohmic resistance of the battery sheet and the contact resistance in the stack. Under the condition of good contact, the resistance value of the contact resistance varies with temperature, which is similar to the relationship between the contact material itself and the temperature. Since the same material, composition and manufacturing process are used, when the contact is good, the relationship between the ohmic impedance and the temperature of Cell-2 and Cell-3 should be consistent. It can be seen from Figure 9 that the relationship parameters between the ohmic impedance and temperature of Cell-2 and Cell-3 are basically the same, so it can be judged that the performance difference between the two cells is mainly due to the influence of contact resistance.

(b) Accident diagnosis

Due to operational errors during the experiment, the battery stack was operated in SOEC mode for a period of time under the condition that the water content of the hydrogen electrode was 0%, which resulted in a rapid decline in the performance of the electrolytic cell stack. The diagnostic method of the present invention is used to analyze the faulty cell Cell-2 and the normal cell-3 to find out the main reason for the decline in electrical performance caused by the accident.

Table 4 shows the changes in ohmic impedance and polarization impedance of the problematic slice Cell-2 and the normal slice Cell-3 before and after the accident.

Table 4 Impedance comparison of Cell-2 and Cell-3 before and after the accident

Figure 10 is (a) Nyquist diagram and (b) DRT diagram (700°C, SOFC mode, hydrogen electrode is 100% H2, oxygen electrode is air) before and after the accident.

First, analyze the normal slice Cell-3. Before and after the accident, the ohmic impedance of Cell-3 remained almost unchanged, while the polarization impedance increased sharply by 42%. The corresponding DRT analysis results are shown in Fig. 10(b), and the increase in polarization impedance is the characteristic peak P4 related to the charge transfer at the three-phase interface of the main hydrogen electrode and the ion conduction process in the YSZ framework. Considering that under the experimental conditions, since the hydrogen electrode does not have water vapor for the electrolysis reaction, the electrolysis current may change the microstructure at the three-phase interface, resulting in performance degradation.

Secondly, considering the problematic slice Cell-2, its attenuation is more severe than that of Cell-3, the ohmic part has increased by 65%, and the polarized part has increased by 145%. This is the constant current mode of electrolysis. Since Cell-2 has the worst performance, the cell has the highest partial voltage during the process, so more serious attenuation occurs.

To sum up, the operation accident of constant current electrolysis under the dry hydrogen electrode atmosphere caused the apparent performance attenuation of the stack. Using the fault diagnosis method of the present invention, it can be learned that: on the one hand, the attenuation mainly occurs in the polarization impedance part, specifically , the accident destroyed the electrode microstructure at the three-phase interface of the hydrogen electrode, resulting in a sharp drop in performance; When the electrolysis voltage is high, the problem sheet is more affected.

The invention proposes a method for non-destructive on-line detection and fault diagnosis of electric piles; the electrochemical impedance spectroscopy method (EIS) is used as the testing method; the impedance spectrum difference analysis method (ADIS), the relaxation time distribution method (DRT), the equivalent The electronic circuit method (ECM) is organically combined as an analysis method to realize the qualitative and quantitative analysis of each electrochemical process in the stack; by establishing the correspondence between the electrochemical process and the DRT characteristic peaks, the stack unit can be analyzed without cooling down and without disassembling the stack. Non-destructive diagnosis of the performance difference between the stacks and the cause of the stack failure not only avoids the change and damage to the stack structure and surface caused by operations such as cooling and dismantling, which will affect the diagnosis results, but also helps to optimize the experimental operation by performing online fault cause analysis , reduce the possibility of attenuation and failure of the stack, and even complete the troubleshooting, thereby greatly improving the experimental efficiency of the stack and reducing operating costs.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (4)

1. the lossless audio coding of pile and a method for diagnosing faults, is characterized in that, comprise the steps:

1) by impedance spectrum testing tool access pile, adopt electrochemical impedance spectral method (EIS) as means of testing, obtain the impedance spectrum data of unit in pile;

2) impedance spectrum discriminant analysis (ADIS) is utilized to step 1) the impedance spectrum data analysis that obtains, tentatively determine the quantity of characteristic peak and corresponding frequency range;

3) utilize relaxation time distribution (DRT) to impedance spectrum data analysis, quantity and the corresponding frequency range at contrast characteristic peak select regularization factors L, obtain characteristic peak quantity and the frequency range of relaxation time distribution curve;

4) the characteristic peak quantity after analyzing according to DRT and frequency range set up alternative equivalent electrical circuit; Characteristic peak crest frequency after the characteristic frequency of contrast equivalent electrical circuit unit block and DRT analyze, selects correct equivalent electrical circuit and determines matching initial value, obtaining quantitative analysis results;

5) step 1 is repeated) to step 4), obtain the DRT analysis result under different operating condition; The influence factor contrasting each characteristic peak finds out the ownership electrochemical process of DRT characteristic peak, obtains the pile fault diagnosis table of comparisons;

6) utilize step 5) the pile fault diagnosis table of comparisons that obtains, to poor-performing in test or repetitive that the very fast decay of performance occurs in running carry out failure reason analysis, find out performance deficiencies reason and failure problems assembly.

2. a kind of lossless audio coding of pile and method for diagnosing faults according to claim 1, it is characterized in that, described DRT analyzes required key parameter and obtains based on ADIS analysis result.

3. a kind of lossless audio coding of pile and method for diagnosing faults according to claim 1, it is characterized in that, the selection of the foundation of described equivalent electrical circuit, screening and matching initial value is the corresponding relation based on characteristic peaks in the time constant of equivalent electrical circuit unit and relaxation time distribution curve.

4. a kind of lossless audio coding of pile and method for diagnosing faults according to claim 1, it is characterized in that, the described pile fault diagnosis table of comparisons obtains according to the corresponding relation of operating conditions and DRT analysis result.