Method for predicting remaining life of solid oxide fuel cell power generation system

Abstract

The application discloses a kind of solid oxide fuel cell power generation system residual life prediction methods, suitable for with methane as fuel, belongs to fuel cell life prediction technical field.The prediction method of the application includes the following steps: continuously operating test is carried out to SOFC, obtains the change data of voltage or power with time and calculates attenuation rate, and obtains the conversion rate of reforming curve of change with time, establishes system overall performance attenuation prediction model and reforming reactor performance attenuation prediction model, then according to the data of having served stage, the future attenuation trend of the stack is predicted, and the residual life of the stack is predicted according to the defined failure threshold;The prediction method of the application can realize the residual life on-line prediction of solid oxide fuel cell power generation system, and the prediction method is simple, reliable, practical, suitable for the rapid prediction of the residual life of field power generation system, and has important significance for the health management of the solid oxide fuel cell power generation system in operation.

Description

Technical Field

[0001] The present invention relates to the technical field of solid oxide fuel cells, and in particular to a method for predicting the remaining life of a solid oxide fuel cell power generation system. Background Art

[0002] Solid oxide fuel cells (SOFCs) are electrochemical devices that convert the chemical energy in fuel directly into electrical energy. As a promising candidate for large-scale, distributed, clean power stations, SOFCs play a crucial role in future energy transformation. The core component of a SOFC power generation system is the stack. Peripheral components include the reformer, heat exchanger, evaporator, mixer, and desulfurization tank. These components ensure stable stack power generation and enable efficient, cascaded heat recovery. Under prolonged, high-temperature operation, SOFC stack electrochemical performance gradually degrades due to complex degradation factors such as electrolyte element diffusion, cathode Cr poisoning, anode carbon deposition, and sulfur poisoning, resulting in a gradual decrease in output power. Furthermore, peripheral components can also be damaged and degraded at high temperatures, such as reformer carbon deposition and reduced heat exchanger efficiency, accelerating stack degradation. Reliable methods for predicting the remaining useful life of SOFC power generation systems are crucial for effectively assessing their health. This approach will facilitate the transition from planned to predictive maintenance, effectively reducing maintenance costs and avoiding the catastrophic consequences of major equipment failures. Therefore, it is a crucial basis for developing SOFC maintenance strategies.

[0003] Current research on remaining life prediction methods mainly includes model-based, data-driven, and hybrid methods. Model-based methods describe degradation phenomena by analyzing the internal mechanisms of fuel cells and establishing corresponding models, generally including parameter identification, electrochemical impedance spectroscopy, and polarization curves. Data-based methods use data mining or statistical methods, without considering the internal degradation mechanisms of fuel cells, but instead predict the remaining service life of the battery through data analysis. Hybrid methods are a combination of the first two types. This method is based on physical equations and estimates the changes in parameters over time by learning from large amounts of data to summarize the laws, thereby achieving the purpose of life prediction. However, the operation of SOFC power generation systems involves complex physical, chemical, and structural changes. The above methods only predict the SOFC cells themselves, without considering the impact of auxiliary components on the degradation of the stack, and cannot accurately predict the remaining life of the SOFC power generation system. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides an online prediction method for the remaining life of a kilowatt-class solid oxide fuel cell stack.

[0005] The technical solution adopted in the present invention is:

[0006] The present invention provides a method for predicting the remaining life of a solid oxide fuel cell power generation system, which is applicable to using methane as fuel, comprising the steps of:

[0007] (1) Analyze the operation process of the solid oxide fuel cell power generation system and determine the health factor to measure its power generation stability, including output voltage and / or output power and reforming conversion rate;

[0008] (2) obtaining the output current, output voltage, and output power data of the solid oxide fuel cell power generation system from the initial operation to the current time in the service stage, and drawing the output voltage or output power attenuation curve, while collecting the inlet gas and outlet gas of the reformer, calculating the reforming conversion rate, and using the obtained output voltage, output power, and reforming conversion rate as historical operation data;

[0009] (3) using the output voltage or output power data obtained in step (2) to calculate the decay rate every △t hours, and obtain the decay rate of the output voltage or output power at different operating time points as the overall performance decay rate of the system, and analyze the decay trend of the fuel cell stack according to the calculated output voltage or output power decay rate, determine the decay trend change node and divide the output voltage or output power decay curve into stages according to the node, with the decay rate η>2% / kh as the first stage, the decay rate 0.5<η≤2% / kh as the second stage, and the decay rate η≤0.5% / kh as the third stage;

[0010] (4) plotting a reforming conversion rate versus time curve based on the reforming conversion rate data obtained in step (2), and establishing a reforming reactor performance decay model regarding the reforming catalyst activity decay rate based on the reforming conversion rate curve;

[0011] (5) defining the failure thresholds of the system overall performance decay rate and the reforming catalyst activity decay rate as the end-of-life values;

[0012] (6) Determine whether the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and / or third stages. If the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and third stages, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rate of the third stage as the remaining life; if the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first and second stages but does not include the third stage, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rates at two different moments of the first and second stages combined with the fuel utilization rate, and the life of the reforming reactor when the failure threshold of the reforming catalyst activity attenuation rate is calculated, and the minimum value of the overall system performance life and the reforming reactor life is taken as the remaining life.

[0013] Furthermore, the reforming conversion rate ξ in step (2) is calculated from the molar flow rate of methane consumed in the reaction and the molar flow rate of methane in the initial reactants:

[0014]

[0015] Among them, ξ n represents the reforming conversion rate of the reformer at the nth moment; L in,n represents the molar flow rate of methane at the reformer inlet (mol / s); L out,n It represents the molar flow rate of methane at the outlet of the reformer (mol / s).

[0016] Furthermore, the calculation formula for the system overall performance attenuation rate in step (3) is:

[0017]

[0018] Among them, η n is the overall system performance attenuation rate per thousand hours at the nth moment, n is the serial number, indicating the nth moment, P0 is the output power when the system starts to operate stably, P n Represents the output power at time n, P n-△t It represents the output power at the time n-△t, where △t is the interval time.

[0019] Furthermore, the performance attenuation model of the reforming reactor in step (4) is:

[0020]

[0021] Among them, α is the activity decay rate of the reforming catalyst, which is expressed as the rate of decrease of the reforming conversion rate.

[0022]

[0023] Among them, ξ n is the reforming conversion rate of the reformer at the nth moment, ξ n-△t is the reforming conversion rate of the reformer at the n-△t moment, δ is the rate of decrease of the reforming conversion rate; φ is the system fuel utilization rate, which is set by the user in the early stage of system operation, w is a parameter related to the operating temperature of the reforming reactor; C is the reforming reaction parameter; t represents the service time.

[0024] Furthermore, in step (5), the failure threshold of the system overall performance attenuation rate is set to 10% to 40%, and the failure threshold of the reforming catalyst activity attenuation rate is set to 0.4 to 0.6 times the failure threshold of the system overall performance attenuation rate.

[0025] Furthermore, in step (6), when the output power attenuation curve of the fuel cell stack to be predicted in the service stage includes the first, second and third stages, the remaining life when the output power reaches the failure threshold of the overall system performance attenuation rate is predicted using the attenuation rate of the third stage, and the calculation formula is:

[0026]

[0027] Among them, t RUL is the service time from the current moment to the failure threshold of the system's overall performance attenuation rate, that is, the remaining life, kh; P0 is the output power when the normal operating voltage is reached, P s is the current output power of the stack, P m The output power when the failure threshold of the overall system performance attenuation rate is reached, P m =(1-m)P0, m is the failure threshold of the system's overall performance attenuation rate, a is the attenuation rate of the third stage, which is 0.5% / kh.

[0028] Furthermore, in step (6), if the output power attenuation curve of the fuel cell stack to be predicted in the service stage includes the first and second stages but does not include the third stage, the minimum value of the remaining life of the overall system performance and the remaining life of the reforming reactor is used as the remaining life, that is:

[0029] t RUL ={t 整体 , t 重整} min (6)

[0030] t 整体 According to formula (7), we can get:

[0031]

[0032] t 重整 According to formula (8), we can get:

[0033]

[0034] Among them, t 整体 is the remaining life obtained by predicting the overall performance degradation of the system, t 重整 is the remaining life of the reforming reactor; P0 is the output power when the normal operating voltage is reached, P m is the output power when the failure threshold is reached, that is, the end of life, P m =(1-m)P0, m is the failure threshold of the system's overall performance attenuation rate, A and B are attenuation rate coefficients, α 失效is the activity decay rate of the reforming catalyst when it reaches the failure threshold, β is the multiple of the failure threshold of the overall performance decay rate of the system corresponding to the failure threshold of the reforming reactor, φ is the system fuel utilization rate, w is a parameter related to the operating temperature of the reforming reactor; C is the reforming reaction parameter.

[0035] Furthermore, the step (2) further includes filtering and smoothing the collected historical operation data to obtain stable historical operation data. The calculation formula for filtering and smoothing is:

[0036]

[0037] Where n is the sequence number, indicating the nth moment, h(n) is the impulse response at the nth moment, τ n is the analysis time point at moment n.

[0038] The beneficial effects of the present invention are:

[0039] The present invention provides a method for predicting the remaining life of a solid oxide fuel cell power generation system. Based on the historical operating data of the power generation system, an output voltage or power attenuation curve and a reforming conversion rate change curve of a solid oxide fuel cell stack are constructed. A system overall performance attenuation prediction model is established according to the output voltage or output power attenuation curve. A reforming reactor performance attenuation prediction model is established according to the reforming conversion rate change curve. The future attenuation trend of the stack can be predicted based on the data of the stack in the service stage, and the remaining life of the stack can be predicted based on the defined failure threshold. The prediction method of the present invention can realize online prediction of the remaining life of a solid oxide fuel cell power generation system, and the prediction method is simple, reliable, and highly practical. It is suitable for rapid prediction of the remaining life of an on-site power generation system, and is of great significance to the health management of a running solid oxide fuel cell power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 is a flow chart of the life prediction method of the present invention;

[0042] Figure 2 Output voltage historical operating data obtained by the embodiment of the present invention;

[0043] Figure 3 The historical operating data of the reforming conversion rate obtained in the embodiment of the present invention;

[0044] Figure 4 It is the historical operating data of output voltage after filtering;

[0045] Figure 5 is the historical operating data of the reforming conversion rate after filtering;

[0046] Figure 6 It is the historical operating data of output voltage after being divided into stages. DETAILED DESCRIPTION

[0047] The present invention provides a method for predicting the remaining life of a solid oxide fuel cell power generation system. To clarify the objectives, technical solutions, and effects of the present invention, the present invention is described in further detail below. It should be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention.

[0048] The present invention will be described in detail below with reference to the accompanying drawings.

[0049] Reference Figure 1 This embodiment provides a method for predicting the remaining life of a solid oxide fuel cell power generation system, which is applicable to using methane as fuel, and includes the following steps:

[0050] (1) Analyze the operation process of the solid oxide fuel cell power generation system and determine the health factor to measure its power generation stability, including output voltage and / or output power and reforming conversion rate;

[0051] (2) Obtain the output current, output voltage, and output power data of the solid oxide fuel cell power generation system from the initial operation to the current time in the service stage, draw the output voltage or output power attenuation curve, and at the same time collect the inlet gas and outlet gas of the reformer, calculate the reforming conversion rate, and use the obtained output voltage, output power, and reforming conversion rate as historical operation data; filter and smooth the historical operation data to obtain stable historical operation data, and the calculation formula for the above filtering and smoothing is:

[0052]

[0053] Where n is the sequence number, indicating the nth moment, h(n) is the impulse response at the nth moment, t n is the analysis time point at moment n;

[0054] The above reforming conversion rate ξ is calculated from the molar flow rate of methane consumed in the reaction and the molar flow rate of methane in the initial reactants:

[0055]

[0056] Among them, ξ nrepresents the reforming conversion rate of the reformer at the nth moment; L in,n represents the molar flow rate of methane at the reformer inlet (mol / s); L out,n It represents the molar flow rate of methane at the outlet of the reformer (mol / s).

[0057] (3) Calculate the attenuation rate every △t hours using the filtered output power data of step (2), and obtain the output power attenuation rate at different operating time points as the overall performance attenuation rate of the system, and analyze the attenuation trend of the fuel cell stack based on the calculated output power attenuation rate, determine the attenuation trend change node and divide it into stages according to the node, and take the output power attenuation rate η>2% / kh as the first stage, that is, the rapid attenuation stage of the fuel cell stack in the early stage, and take the output power attenuation rate 0.5<η≤2% / kh as the second stage, that is, the slow attenuation stage of the fuel cell stack in the early stage, and take the output power attenuation rate η≤0.5% / kh as the third stage, that is, the stable operation stage of the fuel cell stack in the late stage;

[0058] The calculation formula for the overall performance attenuation rate of the above system is:

[0059]

[0060] Among them, η n is the overall system performance attenuation rate per thousand hours at the nth moment, n is the serial number, indicating the nth moment, P0 is the output power when the system starts to operate stably, P n Represents the output power at time n, P n-△t It represents the output power at the time n-△t, △t is the interval time;

[0061] (4) plotting a reforming conversion rate versus time curve based on the filtered reforming conversion rate data obtained in step (2), and establishing a reforming reactor performance attenuation model for the reforming catalyst activity attenuation rate based on the reforming conversion rate curve;

[0062] The above reforming reactor performance attenuation model is:

[0063]

[0064] Wherein, α is the activity decay rate of the reforming catalyst, which is expressed as the decreasing rate δ of the reforming conversion rate;

[0065]

[0066] Among them, ξ n is the reforming conversion rate of the reformer at the nth moment, ξ n-△t is the reforming conversion rate of the reformer at the n-△t moment; φ is the system fuel utilization rate, w is a parameter related to the operating temperature of the reforming reactor; C is the reforming reaction parameter; t represents the service time;

[0067] (5) defining the failure thresholds of the system overall performance attenuation rate and the reforming catalyst activity attenuation rate as the end-of-life values; that is, setting the failure threshold of the system overall performance attenuation rate to 10% to 40%, and the failure threshold of the reforming catalyst activity attenuation rate to 0.4 to 0.6 times the failure threshold of the system overall performance attenuation rate;

[0068] (6) Determine whether the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and / or third stages. If the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and third stages, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rate of the third stage as the remaining life; if the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first and second stages but does not include the third stage, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rates at two different moments of the first and second stages combined with the fuel utilization rate, and the life of the reforming reactor when the failure threshold of the reforming catalyst activity attenuation rate is calculated, and the minimum value of the overall system performance life and the reforming reactor life is used as the remaining life; specifically:

[0069] When the output power attenuation curve of the fuel cell stack to be predicted in the service stage includes the first, second and third stages, the remaining life when the output power reaches the failure threshold of the system overall performance attenuation rate is predicted using the attenuation rate of the third stage. The calculation formula is:

[0070]

[0071] Among them, t RUL is the service time from the current moment to the failure threshold of the system's overall performance attenuation rate, that is, the remaining life, kh; P0 is the output power when the normal operating voltage is reached, P s is the current output power of the stack, P m The output power when the failure threshold of the overall system performance attenuation rate is reached, P m =(1-m)P0, where m is the failure threshold of the system's overall performance attenuation rate, and a is the attenuation rate of the third stage, which is 0.5% / kh;

[0072] When the output power attenuation curve of the fuel cell stack in service to be predicted includes the first and second stages but does not include the third stage, the minimum value of the remaining life of the overall system performance and the remaining life of the reforming reactor is taken as the remaining life, that is:

[0073] t RUL =[t 整体 , t 重整} min (6)

[0074] t 整体 According to formula (7), we can get:

[0075]

[0076] A and B are obtained by the following formula:

[0077]

[0078] t 重整 According to formula (8), we can get:

[0079] Among them, t 整体 is the remaining life obtained by predicting the overall performance degradation of the system, t 重整 is the remaining life of the reforming reactor; P0 is the output power when the normal operating voltage is reached, P m is the output power when the failure threshold is reached, that is, the end of life, P m =(1-m)P0, m is the failure threshold of the system's overall performance attenuation rate, A and B are attenuation rate coefficients, C N is the relative attenuation of a certain period of time in the Nth stage, ΔP is the power attenuation of a certain period of time in the Nth stage, N is 1 and 2, t N is the selected relative attenuation C N The corresponding running time, kh, α 失效 is the activity decay rate of the reforming catalyst when it reaches the failure threshold, β is the multiple of the failure threshold of the overall performance decay rate of the system corresponding to the failure threshold of the reforming reactor, φ is the system fuel utilization rate, w is a parameter related to the operating temperature of the reforming reactor; C is the reforming reaction parameter.

[0080] In addition, during the actual operation of the power generation system, the first stage is relatively short. When the power generation system is predicted, the first stage is usually completed. Therefore, the present invention mainly studies the situation where the service stage includes the first and second stages or the first, second and third stages. If, when predicting the remaining life of a system, it is calculated that the voltage or power attenuation curve of the system in the service stage only includes the first stage, it is not accurate to use only the data of the first stage for prediction. In this case, the system can continue to operate for a period of time so that the system can run to the second stage or even the third stage.

[0081] In addition, it should be noted that the output power in the above calculation process is obtained based on the output voltage * current, and since the current is kept constant during system operation, the output power and output voltage have the same attenuation trend.

[0082] Example

[0083] The life prediction method of the present invention is illustrated using a 2kW SOFC power generation system as an example. The operating temperature is 750°C, the discharge current is 30A, the fuel gas is methane (anode gas) with a flow rate of 10L / min, and the cathode gas is air with a flow rate of 30L / min. After activation and stabilization, a long-term stability test is conducted. Based on the first 2500 hours of operating data, the subsequent voltage decay trend is predicted and the remaining service life is obtained. The specific steps are as follows:

[0084] (1) Analyze the operation process of the solid oxide fuel cell power generation system and determine the health factor to measure its power generation stability, including output voltage and / or output power and reforming conversion rate;

[0085] (2) Conduct a continuous operation test on the solid oxide fuel cell power generation system, allowing the system to operate continuously and stably for about 2000 hours, and record the output voltage and output power of the system during the continuous and stable operation according to the set interval time. At the same time, collect the inlet gas and outlet gas of the reformer, calculate the reforming conversion rate, and use the obtained output voltage, output power, and reforming conversion rate as historical operation data; specifically:

[0086] (21) An activation test was conducted on the stack, i.e. nitrogen was introduced into the anode of the stack and air was introduced into the cathode. The stack was heated from room temperature to 800°C at a rate of 1°C / min, hydrogen was introduced into the anode for reduction for 4 h, and then the temperature was lowered to the operating temperature of 750°C.

[0087] (22) Carry out a loading test on the stack, i.e., keep the operating temperature stable and change the current in a step-by-step manner, starting from 0A and increasing the current by 5A each time, and keep it for 1 minute until it is adjusted to the rated current of 30A output by the system;

[0088] (23) The stack was subjected to a continuous operation test, i.e., the output rated current remained unchanged, and the system operated stably for 2542 h. The output voltage and output power were extracted every 1 h. At the same time, a mass flow meter was used to record the fuel flow entering the SOFC power generation system and the gas flow leaving the system every 24 h. The inlet gas and outlet gas of the reformer were collected through a gas sampling bag. The volume fraction of methane in the inlet gas and the volume fraction of methane in the outlet gas were analyzed using a gas chromatograph, and the reforming conversion rate was calculated. The output voltage and reforming conversion rate change curves with time were obtained as shown in Fig. Figure 2 and 3 As shown;

[0089] (24) The collected historical operation data is filtered and smoothed to obtain stable historical operation data, such as Figure 4 and 5 As shown;

[0090] (3) First, remove the data from the early activation stage, and then use the above formula (2) to calculate the decay rate every 100 hours for the output power data after filtering in step (3), and obtain the decay rate of the output power at different operating time points as the overall performance decay rate of the system, as shown in Table 1;

[0091] Table 1

[0092] Time / h 100 200 300 400 500 600 700 800 900 1000 1100 Attenuation rate / % 7.68 6.56 5.44 3.27 3.07 2.42 2.41 1.95 1.69 0.91 0.71 Time / h 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 Attenuation rate / % 0.64 0.62 0.61 0.58 0.52 0.48 0.44 0.44 0.44 0.44 0.43

[0093] According to the output power attenuation rate calculated in Table 1, the attenuation trend of the fuel cell stack is analyzed, the attenuation trend change nodes are determined and divided into stages according to the nodes. The output power attenuation rate η>2% / kh is regarded as the first stage, that is, the rapid attenuation stage in the early stage of the fuel cell stack; the output power attenuation rate 0.5<η≤2% / kh is regarded as the second stage, that is, the slow attenuation stage in the early stage of the fuel cell stack; the output power attenuation rate η≤0.5% / kh is regarded as the third stage, that is, the stable operation stage in the late stage of the fuel cell stack. Figure 6 As shown;

[0094] (4) Based on the reforming conversion rate data filtered in step (2), a reforming conversion rate change curve is drawn. Based on the reforming conversion rate curve, a reforming reactor performance decay model regarding the reforming catalyst activity decay rate is established; and the reforming reactor performance decay model is obtained by fitting as follows:

[0095]

[0096] Wherein, T is the operating temperature 750℃;

[0097] (5) The failure threshold of the system overall performance attenuation rate is defined as 40%, and the failure threshold of the reforming catalyst activity attenuation rate is defined as 0.4 times the failure threshold of the system overall performance attenuation rate;

[0098] (6) When the output power attenuation curve of the above power generation system in service stage includes the first, second and third stages, the remaining life when the output power reaches the failure threshold of the system overall performance attenuation rate is predicted to be 71040h based on the attenuation rate of the third stage, and the corresponding output power is 1125W.

[0099] In addition, the above method can also predict the overall performance degradation trend of the system through the following formula, namely:

[0100] P f =P s -(P0×a3×t f )

[0101] Among them, P f The predicted fuel cell stack will continue to serve based on the current service time. fP0 is the output power when the system begins to run steadily, a3 is the attenuation rate of the third stage, taking 0.5% / kh, t f kh is the time for future service, P s P is the current output power of the stack to be predicted.

[0102] According to the formula, the output power after any time of future service can be calculated, such as the output power after 100,000 hours of future service is 899W.

[0103] It should be noted that the parts not mentioned in the present application can be realized by using or referring to the existing technology.

[0104] Of course, the above description is not a limitation of the present application, and the present application is not limited to the above examples, and the changes, modifications, additions or replacements made by the person skilled in the art within the essential scope of the present application should also belong to the protection scope of the present application.

Claims

1. A method for predicting the remaining life of a solid oxide fuel cell power generation system, characterized in that: Including steps: (1) Analyze the operation process of the solid oxide fuel cell power generation system and determine the health factors that measure its power generation stability. The health factors include output voltage and / or output power and reforming conversion rate; (2) Obtain the output current, output voltage, and output power data of the solid oxide fuel cell power generation system from the initial operation to the current time in the service stage, and draw the output voltage or output power attenuation curve. At the same time, collect the inlet gas and outlet gas of the reformer, calculate the reforming conversion rate, and use the obtained output voltage, output power, and reforming conversion rate as historical operation data; (3) Using the output voltage or output power data obtained in step (2), calculate the attenuation rate every △t hours, and obtain the output voltage or output power attenuation rate at different operating time points as the overall performance attenuation rate of the system. Analyze the attenuation trend of the battery stack based on the calculated output voltage or output power attenuation rate, determine the attenuation trend change node, and divide the output voltage or output power attenuation curve into stages according to the node, and calculate the attenuation rate. As a first stage, the decay rate As a second stage, the decay rate As the third stage; (4) plotting a reforming conversion rate versus time curve based on the reforming conversion rate data obtained in step (2), and establishing a reforming reactor performance decay model regarding the reforming catalyst activity decay rate based on the reforming conversion rate curve; (5) Define the failure thresholds of the system overall performance decay rate and the reforming catalyst activity decay rate as the end-of-life values; (6) Determine whether the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and / or third stages. If the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first, second and third stages, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rate of the third stage as the remaining life; if the output voltage or output power attenuation curve of the fuel cell stack in the service stage includes the first and second stages but does not include the third stage, the remaining life of the overall system performance when the failure threshold is reached is calculated using the attenuation rates at two different moments of the first and second stages combined with the fuel utilization rate, and the life of the reforming reactor when the failure threshold of the reforming catalyst activity attenuation rate is calculated, and the minimum value of the overall system performance life and the reforming reactor life is taken as the remaining life; The performance attenuation model of the reforming reactor in step (4) is: （3）； in, is the activity decay rate of the reforming catalyst, is the system fuel utilization rate, is a parameter related to the operating temperature of the reforming reactor, is the reforming reaction parameter, Indicates the time of service.

2. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, characterized in that: The reforming conversion rate in step (2) is calculated from the molar flow rate of methane consumed in the reaction and the molar flow rate of methane in the initial reactants: （1）； in, Indicates the Reforming conversion rate of the reformer at any moment; represents the molar flow rate of methane at the reformer inlet, mol / s; It represents the molar flow rate of methane at the outlet of the reformer, mol / s.

3. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, characterized in that: The calculation formula for the overall system performance attenuation rate in step (3) is: （2）； in, For the The overall system performance degradation rate every thousand hours, Is a serial number, indicating the time, The output power at which the system starts to operate stably. Indicates The output power at the moment, Indicates The output power at the moment, For the interval time.

4. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, wherein: The activity decay rate of the reforming catalyst in step (4) is the rate of decrease of the reforming conversion rate. express: （4）； in, For the The reforming conversion rate of the reformer at that moment, For the The reforming conversion rate of the reformer at that moment.

5. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, characterized in that: In the step (5), the failure threshold of the system overall performance attenuation rate is set to 10% to 40%, and the failure threshold of the reforming catalyst activity attenuation rate is set to 0.4 to 0.6 times the failure threshold of the system overall performance attenuation rate.

6. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, characterized in that: In step (6), when the output power attenuation curve of the fuel cell stack in service includes the first, second, and third stages, the remaining life when the output power reaches the failure threshold of the overall system performance attenuation rate is predicted using the attenuation rate of the third stage. The calculation formula is: （5）； in, The service time from the current moment to the failure threshold of the system's overall performance attenuation rate, that is, the remaining life, kh; In order to achieve the output power at normal working voltage, is the current output power of the stack to be predicted, The output power when the failure threshold of the overall system performance attenuation rate is reached is: , is the failure threshold of the system’s overall performance attenuation rate, is the attenuation rate of the third stage, which is 0.5% / kh.

7. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, characterized in that: In step (6), if the output power attenuation curve of the fuel cell stack in service includes the first and second stages but does not include the third stage, the minimum value of the remaining life of the overall system performance and the remaining life of the reforming reactor is used as the remaining life, that is: （6）； According to formula (7), we can get: （7）； According to formula (8), we can get: （8）； in, is the remaining life obtained by predicting the overall performance degradation of the system, is the remaining life of the reforming reactor; In order to achieve the output power at normal working voltage, is the output power when the failure threshold is reached, i.e. the end of life, , is the failure threshold of the system’s overall performance attenuation rate, 、 is the attenuation coefficient, is the multiple of the failure threshold of the overall system performance attenuation rate corresponding to when the reforming reactor reaches the failure threshold, is the system fuel utilization rate, is a parameter related to the operating temperature of the reforming reactor; are the reforming reaction parameters.

8. The method for predicting the remaining life of a solid oxide fuel cell power generation system according to claim 1, wherein: The step (2) further includes filtering and smoothing the collected historical operation data to obtain stable historical operation data. The calculation formula for filtering and smoothing is: （9）； in, Is a serial number, indicating the time, For the The impulse response of the moment, For the The analysis time point of the moment.

Images