Hydrogen sealing leakage rate prediction method and detection device

Abstract

The invention discloses a hydrogen sealing leakage rate prediction method and detection device, the device comprises an extrusion mechanism, a gas supply loop, a cone and a cone seat, the extrusion mechanism comprises a motor, a screw, a first piston, a spring, a spring seat, a tension and pressure sensor and a second piston, the motor drives the screw to rotate, one end of the screw abuts against one end of the first piston, and the other end of the screw abuts against the other end of the second piston. The spring seat is connected with one end of a second piston through a pull pressure sensor; the other end of the second piston is propped against a cone; a sealing rubber ring is mounted on one side of the cone; the cone extrudes a cone seat through the sealing rubber ring; and the air supply loop is connected with the cone. According to the method, the numerical relationship between the hydrogen leakage amount and different gas pressures and screw extrusion forces is accurately obtained, an optimal solution is obtained, a basis is provided for optimizing a hydrogen sealing system, the leakage amount is effectively reduced, and fatigue deformation of a cone is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen sealing, and in particular to a method and a detection device for hydrogen sealing leakage. Background Art

[0002] Hydrogen, along with solar energy and nuclear energy, is considered one of the three major new energy sources due to its high combustion efficiency and pollution-free products. As a new energy source, hydrogen is widely used in aviation, power generation, and locomotive fuel cells. However, hydrogen molecules are very small and prone to leakage during storage and use. Hydrogen is unbreathable, colorless, and odorless, making it undetectable to the human nose. Its ignition point is only 585°C, and its concentration in air ranges from 4% to 75%. It explodes when exposed to open flames, making sealing hydrogen crucial. Even small hydrogen leaks waste resources and pose potential safety risks. Therefore, leak monitoring is essential during hydrogen use.

[0003] During hydrogen transmission, sealing is achieved by forming an extruded contact surface between the cone and the valve seat. This component is widely used in hydrogen sealing systems, such as the one-way valve in the hydrogen circuit, where the seal is formed by the extrusion of the cone and the valve seat. The reliability of the sealing performance directly determines the safety of the hydrogen circuit system. However, the long-term opening and closing of the sealing interface causes frequent reciprocating impacts between the cone and the valve seat, which can easily cause seal fatigue failure. Therefore, selecting an appropriate extrusion force between the cone and the valve seat based on the pressure of the hydrogen system is an effective means of achieving a long-term seal between the cone and the valve seat.

[0004] Selecting the extrusion force between the cone and the cone seat requires exploring the relationship between leakage and extrusion force to determine the optimal extrusion force corresponding to the critical leakage. Currently, there are limited testing methods and mathematical model verification methods for the hydrogen sealing performance between the cone and the valve seat. Existing testing devices cannot accurately measure hydrogen leakage or clearly reveal the relationship between gas pressure and leakage. Therefore, a testing device is needed that can accurately and effectively detect hydrogen leakage between the cone and the valve seat and help understand the relationship between gas pressure and leakage, thereby providing support for the appropriate selection of extrusion force. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a hydrogen seal leakage prediction method and detection device. By combining experiments with mathematical model verification, the numerical relationship between hydrogen leakage and different gas pressures and screw extrusion forces can be accurately obtained to obtain the optimal solution, providing a basis for optimizing the hydrogen sealing system, effectively reducing leakage and avoiding cone fatigue deformation.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A hydrogen seal leakage detection device includes an extrusion mechanism, an air supply circuit, a cone, and a cone seat. The extrusion mechanism includes a motor, a screw, a first piston, a spring, a spring seat, a tension and pressure sensor, and a second piston. The motor drives the screw to rotate, one end of the screw abuts one end of the first piston, the other end of the first piston abuts the spring, the spring is located in the spring seat, the spring seat is connected to one end of the second piston via the tension and pressure sensor, the other end of the second piston abuts the cone, a sealing rubber ring is installed on one side of the cone, the cone squeezes the cone seat through the sealing rubber ring, the other side of the cone seat is connected to a hydrogen concentration detection sensor, the air supply circuit is connected to the location of the cone, and a gas pressure sensor is installed in the channel where the cone is located.

[0008] The gas supply circuit includes a compressor, a hydrogen storage tank, a filter, a pressure relay, a dryer, and a shut-off valve. The compressor compresses the hydrogen input from the outside and transports it to the hydrogen storage tank. The hydrogen storage tank is connected to the dryer through a filter. The other side of the dryer is connected to the location of the cone through a shut-off valve. The start and stop of the compressor are controlled by the pressure relay.

[0009] The air supply circuit includes a cooler and an automatic drainer. The cooler is connected between the compressor and the hydrogen storage tank, and the automatic drainer is connected to the bottom of the hydrogen storage tank.

[0010] A method for predicting hydrogen seal leakage includes a leakage calculation model and a fatigue life calculation model. The leakage calculation model calculates leakage by constructing a mathematical model and inputting the screw extrusion force and the gas pressure in the cavity measured by the sensor, thereby verifying theory and experiment. The fatigue life calculation model constructs a fatigue failure model of the rubber seal ring to derive the fatigue life of the sealing cone under alternating load conditions. Finally, based on the dual constraints of leakage and fatigue life, the optimal design range of the extrusion force between the sealing cone and the valve seat is obtained.

[0011] The calculation process of the leakage calculation model is as follows: According to the measured screw extrusion force F, the gas pressure p in the cavity, and the surface roughness power spectrum density C(q), the contact area ratio A(ξ c ) / A0, and the critical magnification ξ is obtained. c , overall average height The height of the critical contraction part is u1(ξ), and the leakage rate Q0 of each square is calculated by u1(ξ), and the total leakage rate Q=Ly / LxQ0 is obtained.

[0012] The entire nominal contact area between the cone and the cone seat is a rectangle with an area of ​​L x ×L y , divide it into L y / L x The side length is L x、The area is A0=L x 2 When the magnification reaches the critical magnification ξ c When , a seepage channel will appear throughout the entire contact area. Assuming that all leakage occurs in the seepage channel and the pressure drop Δp occurs in the critical contraction part, the leakage rate Q0 of each square can be calculated by formula (1):

[0013]

[0014] Where μ is the dynamic viscosity of the fluid, α is a correction factor, α is equal to 1, along the circumferential direction, the circumference is Ly, then the number of squares on the contact surface is Ly / Lx, and the total leakage rate can be calculated by Q = Ly / LxQ0.

[0015] When the magnification is ξ, the contact area ratio A(ξ) / A0 can be obtained by formula (2):

[0016]

[0017] Where p0 is the applied pressure, erf is the error function, and G represents the effective stiffness of the surface roughness, which can be expressed as follows when the magnification is ξ:

[0018]

[0019] Where q is the wave vector, q0 is the reference wave vector, and E * is the equivalent elastic modulus, which can be obtained by formula (4), C(q) is the power spectrum density PSD of the isotropic surface roughness, which can be obtained by the one-dimensional PSD function C 1-D (q) is obtained, as shown in formula (5),

[0020] E * =E / (1-ν 2 )=1 / (1-ν 2 / E1+1-ν 2 / E2) (4)

[0021] Where ν is Poisson’s ratio, E1 and E2 are the elastic moduli of the contacting objects, respectively;

[0022]

[0023] The height of the critical constriction, u1(ξ), is considered to be the distance between the two surfaces when the first channel penetrates the contact area. u1(ξ) can be calculated as follows:

[0024]

[0025] in, is the overall average height of the two surfaces at magnification ξ, u'(ξ) and A'(ξ) are the derivatives of u(ξ) and the contact area A(ξ) with respect to ξ, respectively. It can be obtained from formula (7) to formula (9),

[0026]

[0027] Among them, q0 is the reference wave vector, q1 is the maximum wave vector, w(q) is a weight function related to the surface roughness power spectrum C(q), p(ξ) is the nominal pressure at the current magnification, w(q,ξ) is a function related to the local contact stiffness, s(q,ξ) is a function related to the local pressure and elastic deformation, and s(q,ξ)=w(q,ξ) / E * , P(q,p,ξ) is an error function, which represents the influence of local contact pressure p on elastic contact probability under specific magnification ξ and wave vector q. Since the elastic energy stored in the contact area is less than the elastic energy in the full contact area, a correction factor γ0 is introduced, and γ0 is taken as 0.4.

[0028] The calculation process of the fatigue life calculation model: Based on the parameters of the rubber material and the length of the initial crack c0, the elastic strain energy U is obtained, the strain energy release rate G and the crack growth rate dc / dN are calculated, and then the number of cycles N and fatigue life T are obtained.

[0029] According to the energy method, the released elastic strain energy U drives the crack to expand.

[0030]

[0031] σ1, σ2, and σ3 are the principal stresses in the x, y, and z directions, respectively; ε1, ε2, and ε3 are the strains in the three directions, respectively;

[0032] The strain energy release rate G is introduced to represent the energy released when the initial crack increases per unit area. G is defined as the negative partial derivative of the strain energy U with respect to the crack area C, that is, the energy released when the initial crack increases per unit area. It can be expressed by formula (11):

[0033]

[0034] During the crack propagation process, the release mechanism of strain energy U is divided into two parts: one part is due to the redistribution of elastic strain energy generated by crack propagation inside the material, and the elastic strain energy originally stored around the crack is released along with the newly added crack surface; the other part is related to the work done by the external force W. Under the continuous action of the external force, the structural deformation mode changes when the crack propagates, and the work done by the external force affects the change of strain energy. The calculation method of the strain energy release rate is shown in formula (12):

[0035]

[0036] Where: G represents the strain energy release rate; W represents the work done by the external force; U represents the strain energy; B represents the material constant, which is related to factors such as the material's geometry and size; k(λ) represents the proportional coefficient related to the stretch ratio λ; c represents the crack length; w0 represents the strain energy density, which is the strain energy stored in a unit volume of material and is related to the material's mechanical properties such as elastic modulus and Poisson's ratio as well as the stress state.

[0037]

[0038] ε represents engineering strain;

[0039] Substituting equation (11) into equation (12) yields the expression for the number of cycles N required for the initial crack length to expand from c0 to c:

[0040]

[0041] Where: β represents the crack growth parameter, c0 can be determined by experimental method;

[0042] The expression for the number of cycles required for the initial crack length to expand from c0 to the final crack length c is shown in formula (15):

[0043]

[0044] The fatigue life T is obtained by multiplying the number of cycles by the cycle time per cycle t:

[0045] T=Nt (16)

[0046] The beneficial effects of the present invention are:

[0047] The leakage calculation model calculates leakage by constructing a mathematical model that uses sensor-measured screw extrusion force and in-chamber gas pressure as input, thus validating theory and experiment. The fatigue life calculation model constructs a rubber seal fatigue failure model to derive the fatigue life of the sealing cone under alternating load conditions. Finally, based on the dual constraints of leakage and fatigue life, the optimal design range for the extrusion force between the sealing cone and the valve seat is determined. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a cross-sectional view of the hydrogen seal leakage detection device.

[0049] Figure 2 This is the pneumatic circuit diagram of the hydrogen seal leakage detection device.

[0050] Figure 3 Flow chart for leakage calculation.

[0051] Figure 4 Flowchart for fatigue life calculation.

[0052] Figure 5 This is the relationship diagram of maximum cyclic stress-fatigue life.

[0053] Figure 6 This is the relationship diagram between screw extrusion force and leakage.

[0054] Figure 7 This is the relationship diagram between the gas pressure in the sealing chamber and the leakage rate.

[0055] In the figure: motor 101, screw 102, sealing ring 103, first piston 104, hydrogen concentration alarm sensor 105, spring 106, spring seat 107, tension pressure sensor 108, second piston 109, tension pressure sensor display 110, cone 201, air supply circuit 202, gas pressure sensor 203, sealing rubber ring 204, cone seat 205, hydrogen concentration detection sensor 206, compressor 301, cooler 302, hydrogen storage tank 303, overflow valve 304, automatic drainer 305, filter 306, pressure relay 307, dryer 308, stop valve 309. DETAILED DESCRIPTION

[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0057] like Figure 1-Figure 2 As shown, this embodiment discloses a hydrogen seal leakage detection device, including an extrusion mechanism, an air supply circuit 202, a cone 201, and a cone seat 205. The extrusion mechanism includes a motor 101, a screw 102, a first piston 104, a spring 106, a spring seat 107, a tension and pressure sensor 108, and a second piston 109. The motor 101 drives the screw 102 to rotate, thereby squeezing the spring 106 through the first piston 104. A sealing ring 103 is installed on the outer wall of the first piston 104. One end of the screw 102 abuts against one end of the first piston 104, and the other end of the first piston 104 abuts against the spring 106. The spring 106 is located in the spring seat 107. The spring seat 107 is connected to one end of the second piston 109 through the tension and pressure sensor 108, and the other end of the second piston 109 abuts against the cone 201.

[0058] According to Hooke's law, F = kx (where F is the force applied to the spring, k is the spring constant, and x is the spring deformation), spring 106, when squeezed by screw 102, generates a force proportional to the deformation. This force is transmitted via spring seat 107 to tension and pressure sensor 108. This highly accurate sensor converts the received force signal into an electrical signal, which is then displayed on tension and pressure sensor display 110 as the compressive force exerted by second piston 109 on cone 201. The force applied by spring 106 is calculated by measuring the deformation x of spring 106 (which can be calculated by, for example, the number of revolutions of screw 102). This is then combined with the measured value from tension and pressure sensor 108 for calibration and verification, ensuring accurate measurement of the compressive force and minimizing measurement errors. Spring seat 107, through tension and pressure sensor 108, drives second piston 109 to squeeze cone 201, generating a high compressive force between cone 201 and cone seat 205, thereby achieving a good seal and preventing hydrogen leakage from the cavity. A sealing rubber ring 204 is installed on one side of the cone 201, and the sealing rubber ring 204 squeezes the cone seat 205 to ensure sealing. The other side of the cone seat 205 is connected to a hydrogen concentration detection sensor 206. This sensor has high sensitivity and can accurately measure the hydrogen content in a small volume, thereby providing accurate data for analyzing the sealing performance. A gas pressure sensor 203 is installed in the channel where the cone 201 is located. The gas pressure sensor 203 can detect and record the gas pressure in the sealed cavity in real time, explore the changes in the system pressure, and combine with the hydrogen concentration measurement to obtain the corresponding relationship between the change in gas leakage and the pressure in the cavity. A hydrogen concentration alarm sensor 105 is installed in the channel where the first piston 104 is located. This sensor has high sensitivity and can send an alarm signal at the first time of hydrogen leakage, prompting the operator to take measures in time.

[0059] Because the first piston 104 and spring seat 107 are both equipped with rubber rings with good sealing effects, an effective sealing effect can be produced. Therefore, hydrogen leakage to the left can be ignored. The other channel joints and sensor joint positions all have good sealing rings, so the entire hydrogen leakage path is along the metal dual interface between the cone 201 and the cone seat 205. To accurately detect the amount of hydrogen leakage between the cone 201 and the cone seat 205, a hydrogen concentration detection sensor 206 is set at the right outlet position. The leakage amount is calculated by detecting the outlet hydrogen concentration and combining the outlet cavity volume. At the same time, to prevent hydrogen leakage caused by failure or installation problems of the sealing rings of the left first piston 104 and spring seat 107, a hydrogen concentration alarm sensor 105 is added to the corresponding position of the groove opened by the first piston 10. If hydrogen is detected at this position, an alarm is issued, indicating that hydrogen is leaking from the left side, and the entire test needs to be stopped, thereby ensuring the accuracy of the hydrogen leakage amount between the right cone 201 and the cone seat 205.

[0060] In order to simulate the fatigue load that the cone seat 205 is subjected to when it is repeatedly opened during actual working, the force applied to the cone seat 205 is changed by adjusting the forward and reverse rotation of the motor 101. The number of turns that the motor 101 drives the screw 102 to rotate in the same direction determines the forward displacement of the screw 102 and the amount of extrusion between the cone 201 and the cone seat 205. The frequency of the forward and reverse rotation of the motor 101 determines the frequency of the alternating load on the cone 201, thereby completing the alternating loading of the load between the cone and the cone seat, providing experimental support for the performance and life prediction of the seal under the action of alternating loads.

[0061] The gas supply circuit 202 is connected to the cone 201 and includes a compressor 301, a hydrogen storage tank 303, a relief valve 304, a dryer 308, and a shut-off valve 309. The compressor 301 compresses hydrogen gas from the outside. If the compressed hydrogen contains water, it is cooled by the cooler 302 to separate the condensed water. The condensed water accumulates at the bottom of the hydrogen storage tank 303 and is discharged by the automatic drain 305. The compressed hydrogen from the hydrogen storage tank 303 passes through the filter 306 and then enters the dryer 308 for water removal. It then enters the sealed chamber through the shut-off valve 309. The pressure of the gas supply circuit 202 is controlled by a pressure relay 307, which controls the start and stop of the compressor 301, maintaining the pressure in the sealed chamber and the hydrogen storage tank 303 at the desired value. This allows the investigation of hydrogen leakage while maintaining constant gas pressure. When the gas pressure in the sealed chamber reaches a certain value, the shut-off valve 309 is closed, forming a closed space, allowing the relationship between leakage and changes in gas pressure within the chamber to be investigated. After the test is completed, the stop valve 309 is opened to release the hydrogen in the sealed chamber.

[0062] The detection device is used as follows: motor 101 drives screw 102 to compress spring 106, sealing cone 201 and cone seat 205 while recording the sealing pressure. Hydrogen is introduced into the cavity through gas supply circuit 202 to reach a certain pressure, and then shut-off valve 309 is closed to seal the cavity. Leakage is detected by hydrogen concentration sensor 206, and abnormal leakage is monitored by hydrogen concentration alarm sensor 105. After the test is completed, hydrogen is discharged from the cavity. The device uses gas pressure sensor 203 to explore changes in system pressure. Combined with leaked hydrogen content measurement, this helps to understand the corresponding relationship between gas leakage changes and cavity pressure, providing support for the research and optimization of the hydrogen sealing performance of cone valves. Furthermore, motor 101 drives screw 102 to reciprocate, simulating the repeated opening and closing of cone seat 205 in actual working conditions, thereby placing the sealing rubber ring 204 in an alternating working condition. The forward and reverse rotation frequency of motor 101 determines the frequency of the alternating load, and the displacement range of screw 102 determines the change in the extrusion force between cone 201 and cone seat 205. The squeeze force is monitored in real time by the pull pressure sensor 108, the gas pressure sensor 203 records changes in the gas pressure within the sealed chamber, and the hydrogen concentration sensor 206 detects leakage. During the test, the number of alternating load cycles is recorded. The test is terminated when the hydrogen leakage exceeds a set threshold, thereby assessing the fatigue life of the sealing system.

[0063] The test process of the detection device is divided into two parts:

[0064] Initial pressure constant test: Hydrogen is introduced into the sealed chamber through the gas supply circuit 202. The hydrogen pressure in the chamber is monitored in real time by the gas pressure sensor 203. When the hydrogen pressure reaches the set value, the shut-off valve 309 is closed, forming a sealed space in the sealed chamber. The hydrogen concentration sensor 206 monitors the amount of hydrogen leakage in real time. Simultaneously, the gas pressure sensor 203 monitors the change in gas pressure in the chamber in real time, establishing a corresponding relationship between compressed hydrogen leakage and pressure decay. This relationship between compressed hydrogen leakage and pressure change in the sealed container is obtained, so that in subsequent tests, the amount of leakage can be predicted by monitoring the pressure change in the sealed container without the need for a hydrogen concentration sensor.

[0065] Constant pressure test during the whole process: hydrogen is continuously introduced into the cavity through the gas supply circuit 202, the stop valve 309 is always in the open state, and the pressure of the system is controlled by the relief valve 304. If the set pressure is reached, the pressure relay 307 is disconnected and the compressor 301 stops working. If it is detected that the pressure is lower than the set pressure, the pressure relay 307 sends a signal to the compressor 301 to continue working. Through this method, even if there is a leak, the hydrogen pressure in the cavity can be maintained at a constant value, thereby obtaining the hydrogen leakage amount under constant gas pressure. By moving the first piston 104 through the screw 102, different extrusion pressures between the cone 201 and the cone seat 205 can be obtained, thereby obtaining the relationship between the hydrogen leakage amount and the extrusion pressure between the cone 201 and the cone seat 205 under constant pressure; similarly, by keeping the position of the screw 102 constant (that is, the extrusion pressure between the cone 201 and the cone seat 205 remains unchanged), changing the pressure setting value of the pressure relay 307 of the air supply circuit 202, different hydrogen pressures can be obtained, thereby obtaining the relationship between the hydrogen leakage amount and the gas pressure in the cavity under constant extrusion pressure.

[0066] During the experiment, a mathematical model for predicting hydrogen leakage was used. The extrusion force exerted by screw 102 on cone 201 and the pressure within the sealing chamber were substituted into the model formula algorithm. Parameters such as contact area, leakage, and critical contraction thickness at different magnifications were calculated based on the model formula algorithm. The change in gas pressure and hydrogen leakage under the current extrusion force of screw 102 was predicted, thereby calculating the leakage under these conditions. The model accuracy was corrected using experimentally measured leakage data to obtain a leakage correction factor. By measuring leakage under different extrusion force and gas pressure conditions, an accurate cone-valve seat seal hydrogen leakage model was ultimately obtained, enabling accurate prediction of hydrogen leakage and providing support for setting the extrusion force between the cone and cone seat in hydrogen sealing systems.

[0067] This test provides measured data support for model modification and optimization, continuously improving the fatigue life prediction model to make it more suitable for actual working conditions. It can effectively detect hydrogen leakage between cone 201 and cone seat 205 and obtain the fatigue life of the sealing rubber under alternating working conditions. Combining measurement with mathematical prediction models, it can accurately obtain the relationship between leakage, extrusion pressure, and gas pressure fatigue life, and help to understand the relationship between them. Therefore, the appropriate extrusion pressure and extrusion frequency between cone 201 and cone seat 205 can be selected according to the hydrogen system pressure. While ensuring that the leakage meets the safety permit, the cone valve seal can operate stably and effectively for a long time, thus ensuring the safety of the hydrogen sealing system. It has good application prospects.

[0068] The hydrogen seal leakage prediction method includes a leakage calculation model and a fatigue life calculation model. The leakage calculation model calculates leakage by constructing a mathematical model that uses the screw extrusion force and in-chamber gas pressure measured by sensors as input, thus validating theory and experiment. The fatigue life calculation model constructs a fatigue failure model for the rubber seal ring to derive the fatigue life of the seal cone under alternating load conditions. Finally, based on the dual constraints of leakage and fatigue life, the optimal design range for the extrusion force between the seal cone and the valve seat is determined.

[0069] The theoretical basis of the leakage calculation model is to establish a quantitative relationship between hydrogen leakage Q, screw extrusion force F, and sealing chamber gas pressure p through the combination of experiments and mathematical models, and to reveal the influence of sealing interface morphology and material parameters on leakage.

[0070] Based on the theory of magnification contact model, the entire nominal contact area between the cone and the cone seat is a rectangle with an area of ​​L x ×L y , which can be divided into L y / L x The side length is L x 、The area is A0=L x 2 The magnification factor ξ is a key parameter in the theory. At the minimum magnification factor (ξ=1), the two surfaces are completely in contact with each other, that is, A(1)=A0. As the magnification factor increases, the details of the surface roughness become apparent, some areas become non-contact areas, and the actual contact area A(ξ) decreases. When the magnification factor reaches the critical magnification factor ξ c When , a percolation path will appear throughout the entire contact area, and the thickness (spacing) of the contraction part is defined as u1(ξ c ).

[0071] Leakage calculation: Assuming that all leakage occurs in the seepage channel and the pressure drop Δp occurs in the critical contraction part, then each square (L x ×L x ) can be calculated by formula (1).

[0072]

[0073] Where μ is the dynamic viscosity of the fluid and α is a correction factor to account for the shape of the contraction and the effect of smaller channels at higher magnifications. Since the critical contraction is simplified as a rectangular hole and the actual contraction shape cannot be obtained, α is expected to be equal to 1. Along the circumference, the circumference is L y , then the number of squares on the contact surface is L y / L x The total leakage rate can be calculated by Q = Ly / L x Q0 is calculated.

[0074] Relationship between contact area and magnification: non-contact area ratio is P c-n =1-A(ξ) / A0. As the magnification increases, the contact area decreases. Although the actual contact area is difficult to define and obtain, the theory based on the magnification combines the contact area with the percolation theory to obtain information about the percolation channel. According to the percolation theory, for an isotropic surface, when the non-contact area ratio (P c-n ) reaches about 0.6, a channel connecting the two ends of the contact area will appear. At this time, the corresponding contact area ratio A(ξ c ) / A0 is 0.4, which is used as the standard for determining the critical magnification.

[0075] When the magnification is ξ, the contact area ratio A(ξ) / A0 can be obtained by formula (2):

[0076]

[0077] Where p0 is the applied pressure, is the contact pressure when the magnification factor ξ is 1, erf is the error function, and G represents the effective stiffness of the surface roughness, which can be expressed as:

[0078]

[0079] Where q is the wave vector, q0 is the reference wave vector, and E * is the equivalent elastic modulus, which can be obtained by formula (4), C(q) is the power spectrum density PSD of the isotropic surface roughness, which can be obtained by the one-dimensional PSD function C 1-D (q) is obtained, as shown in formula (5),

[0080] E * =E / (1-ν 2 )=1 / (1-ν 2 / E1+1-ν 2 / E2) (4)

[0081] Among them, ν is Poisson's ratio, E1 and E2 are the elastic moduli of the contact objects (i.e., the sealing rubber ring and the cone seat), respectively.

[0082]

[0083] Calculation of the critical pinch thickness: The critical pinch height u1(ξ) is considered to be the height of separation between the two surfaces when the first path penetrates the contact area. Therefore, u1(ξ) is defined as the height at which the two surfaces that originally appeared to be in contact will separate when the magnification is reduced by an infinitesimal amount (Δξ). u1(ξ) can be calculated using the following formula:

[0084]

[0085] in, is the overall average height of the two surfaces at magnification ξ, u'(ξ) and A'(ξ) are the derivatives of u(ξ) and the contact area A(ξ) with respect to ξ, respectively. It can be obtained from formula (7) to formula (9),

[0086]

[0087] Among them, q0 is the reference wave vector, q1 is the maximum wave vector, w(q) is a weight function related to the surface roughness power spectrum C(q), p(ξ) is the nominal pressure at the current magnification,

[0088] w(q,ξ) is a function related to the local contact stiffness, s(q,ξ) is a function related to the local pressure and elastic deformation, and s(q,ξ)=w(q,ξ) / E * P(q,p,ξ) is an error function that represents the effect of the local contact pressure p on the elastic contact probability for a specific magnification factor ξ and wave vector q. Because the elastic energy stored in the contact area is less than the elastic energy in the full contact area, a correction factor γ0 is introduced. γ0 is less than 1 and is set to 0.4 in this example.

[0089] The entire leakage calculation process is as follows Figure 3 As shown in the figure, the contact area ratio A(ξ c ) / A0, and the critical magnification ξ is obtained. c , overall average height The height of the critical contraction part is u1(ξ), and the leakage rate Q0 of each square is calculated by u1(ξ), and the total leakage rate Q=Ly / LxQ0 is obtained. The key parameters are the critical magnification ξ0, the average spacing and the thickness u1 of the critical shrinkage portion.

[0090] Calculation of sealing rubber fatigue life: Revealing the relationship between alternating load frequency, amplitude, and life. The load frequency and amplitude are set by motor 101, and the load magnitude can be measured by tension and pressure sensor 108.

[0091] Fracture mechanics methods were used to study the fracture behavior of sealing rubber under cyclic alternating load conditions. Finite element calculations were used to determine the internal stress and strain changes during extrusion, and fracture mechanics theory was used to determine the fatigue life of rubber sealing rings under alternating loads. Fracture mechanics theory assumes that all materials naturally contain micron-scale defects (such as bubbles, impurities, and processing damage). These defects originate from inherent defects in the material preparation process (such as microvoids caused by uneven mixing, residual vulcanization bubbles, or filler agglomeration) or from microcracks that initiate during fatigue / aging processes, and are the starting point for crack propagation.

[0092] The fatigue life calculation process is as follows Figure 4 As shown in the figure, according to the parameters of the rubber material and the length of the initial crack c0, the elastic strain energy U is obtained, the strain energy release rate G and the crack growth rate dc / dN are calculated, and then the number of cycles N and fatigue life T are obtained.

[0093] First, stress simulation was performed using finite element simulation software to determine the strain state of the sealing rubber at the load amplitude after the alternating load was applied. Meshing software was used to create a finite element model of the seal. Based on the alternating load conditions to which the rubber ring was subjected, the finite element analysis software was used to extract the nominal strain results, strain amplitude, and hazardous unit number of the seal in all directions. Based on the load history, the nominal strain values ​​were input into the fatigue analysis file to calculate the equivalent stress.

[0094] According to the energy method, the released elastic strain energy U drives the crack to expand.

[0095]

[0096] σ1, σ2, and σ3 are the principal stresses in the x, y, and z directions, respectively; ε1, ε2, and ε3 are the strains in the three directions, respectively.

[0097] The strain energy release rate G is introduced to represent the energy released when the initial crack increases per unit area. G is defined as the negative partial derivative of the strain energy U with respect to the crack area C, that is, the energy released when the initial crack increases per unit area. It can be expressed by formula (11):

[0098]

[0099] During crack propagation, the release mechanism of strain energy U is divided into two parts: one part comes from the redistribution of elastic strain energy generated by crack propagation within the material, and the elastic strain energy originally stored around the crack is released along with the newly added crack surface; the other part is related to the work done by the external force W. Under the continuous action of the external force, the structural deformation mode changes as the crack propagates, and the work done by the external force affects the change in strain energy.

[0100] Therefore, the calculation method of strain energy release rate is shown in formula (12):

[0101]

[0102] Where: G represents the strain energy release rate, which reflects the driving force of crack propagation; W represents the work done by the external force, in joules (J); U represents the strain energy, in joules (J); B represents the material constant, which is related to the geometric shape, size and other factors of the material; k (λ) represents the proportional coefficient related to the stretching ratio λ, which describes the mechanical behavior of the material under different stretching states; c represents the crack length, in meters (m); w0 represents the strain energy density, which is the strain energy stored in a unit volume of material, and is related to the mechanical properties parameters such as the elastic modulus and Poisson's ratio of the material and the stress state, in joules per cubic meter (J / m 3 ).

[0103]

[0104] ε represents the engineering strain.

[0105] Substituting equation (11) into equation (12) yields the expression for the number of cycles N required for the initial crack length to expand from c0 to c:

[0106]

[0107] Where: β represents the crack growth parameter, c0 can be determined by experimental method: using scanning electron microscope (SEM) or laser confocal microscope to directly measure surface microcracks (accuracy can reach 1μm).

[0108] Therefore, the expression for the number of cycles required for the initial crack length to expand from c0 to the final crack length c is shown in Equation (15):

[0109]

[0110] The fatigue life T is obtained by multiplying the number of cycles by the cycle time per cycle t:

[0111] T=Nt (16)

[0112] During the sealing test, the extrusion force on the sealing ring changes back and forth during the reciprocating motion of the screw. Figure 6 The leakage volume at different screw extrusion pressures can provide a data basis for the subsequent fatigue analysis of the maximum cyclic stress, where the maximum cyclic stress is the maximum extrusion force of the screw on the sealing ring during the reciprocating cycle, that is, the peak value.

[0113] Combine Figure 5-Figure 7 , based on the dual constraints of leakage and fatigue life, the optimal design range of the extrusion force between the sealing cone and the valve seat is obtained.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A hydrogen seal leakage detection device, characterized in that: The invention comprises an extrusion mechanism, an air supply circuit (202), a cone (201), and a cone seat (205); the extrusion mechanism comprises a motor (101), a screw (102), a first piston (104), a spring (106), a spring seat (107), a tension pressure sensor (108), and a second piston (109); the motor (101) drives the screw (102) to rotate; one end of the screw (102) abuts against one end of the first piston (104); the other end of the first piston (104) abuts against the spring (106); the spring (106) is located in the spring seat (107); the The spring seat (107) is connected to one end of the second piston (109) through a tension pressure sensor (108), and the other end of the second piston (109) is against the cone (201). A sealing rubber ring (204) is installed on one side of the cone (201). The cone (201) squeezes the cone seat (205) through the sealing rubber ring (204). The other side of the cone seat (205) is connected to a hydrogen concentration detection sensor (206). The gas supply circuit (202) is connected to the location of the cone (201), and a gas pressure sensor (203) is installed in the channel where the cone (201) is located.

2. A hydrogen seal leakage detection device as claimed in claim 1, characterized in that: The gas supply circuit (202) comprises a compressor (301), a hydrogen storage tank (303), a filter (306), a pressure relay (307), a dryer (308), and a stop valve (309). The compressor (301) compresses hydrogen input from the outside and transports it to the hydrogen storage tank (303). The hydrogen storage tank (303) is connected to the dryer (308) via the filter (306). The other side of the dryer (308) is connected to the location of the cone (201) via the stop valve (309). The start and stop of the compressor (301) are controlled by the pressure relay (307).

3. A hydrogen seal leakage detection device as claimed in claim 2, characterized in that: The air supply circuit (202) comprises a cooler (302) and an automatic drainer (305), wherein the cooler (302) is connected between the compressor (301) and the hydrogen storage tank (303), and the automatic drainer (305) is connected to the bottom of the hydrogen storage tank (303).

4. A method for predicting hydrogen seal leakage, using the hydrogen seal leakage detection device according to any one of claims 1 to 3, characterized in that: It includes a leakage calculation model and a fatigue life calculation model. The leakage calculation model calculates the leakage by constructing a mathematical model and inputting the screw extrusion force and the gas pressure in the cavity measured by the sensor, so that the theory and the experiment can be verified with each other; the fatigue life calculation model constructs a rubber sealing ring fatigue failure model to obtain the fatigue life of the sealing cone under alternating load conditions; finally, based on the dual constraints of leakage and fatigue life, the optimal design range of the extrusion force between the sealing cone and the valve seat is obtained.

5. A method for predicting hydrogen seal leakage as claimed in claim 4, characterized in that: The calculation process of the leakage calculation model is as follows: According to the measured screw extrusion force F, the gas pressure p in the cavity, and the surface roughness power spectrum density C(q), the contact area ratio A(ξ c ) / A0, and the critical magnification ξ is obtained. c , overall average height The height of the critical contraction part is u1(ξ), and the leakage rate Q0 of each square is calculated by u1(ξ), and the total leakage rate Q=Ly / LxQ0 is obtained.

6. A method for predicting hydrogen seal leakage as claimed in claim 5, characterized in that: The entire nominal contact area between the cone (201) and the cone seat (205) is a rectangle with an area of ​​L x ×L y , divide it into L y / L x The side length is L x 、The area is A0=L x 2 When the magnification reaches the critical magnification ξ c When , a seepage channel will appear throughout the entire contact area. Assuming that all leakage occurs in the seepage channel and the pressure drop Δp occurs in the critical contraction part, the leakage rate Q0 of each square can be calculated by formula (1): Where μ is the dynamic viscosity of the fluid, α is a correction factor, α is equal to 1, along the circumferential direction, the circumference is Ly, then the number of squares on the contact surface is Ly / Lx, and the total leakage rate can be calculated by Q = Ly / LxQ0.

7. A method for predicting hydrogen seal leakage as claimed in claim 6, characterized in that: When the magnification is ξ, the contact area ratio A(ξ) / A0 can be obtained by formula (2): Where p0 is the applied pressure, is the contact pressure when the magnification factor ξ is 1, erf is the error function, and G represents the effective stiffness of the surface roughness, which can be expressed as: Where q is the wave vector, q0 is the reference wave vector, and E * is the equivalent elastic modulus, which can be obtained by formula (4), C(q) is the power spectrum density PSD of the isotropic surface roughness, which can be obtained by the one-dimensional PSD function C 1-D (q) is obtained, as shown in formula (5), E * =E / (1-ν 2 )=1 / (1-n 2 / E1+1-n 2 / E2) (4) Where ν is Poisson’s ratio, E1 and E2 are the elastic moduli of the contacting objects, respectively; 8. A method for predicting hydrogen seal leakage as claimed in claim 7, characterized in that: The height of the critical constriction, u1(ξ), is considered to be the distance between the two surfaces when the first channel penetrates the contact area. u1(ξ) can be calculated as follows: in, is the overall average height of the two surfaces at magnification ξ, u'(ξ) and A'(ξ) are the derivatives of u(ξ) and the contact area A(ξ) with respect to ξ, respectively. It can be obtained from formula (7) to formula (9), Among them, q0 is the reference wave vector, q1 is the maximum wave vector, w(q) is a weight function related to the surface roughness power spectrum C(q), p(ξ) is the nominal pressure at the current magnification, w(q,ξ) is a function related to the local contact stiffness, s(q,ξ) is a function related to the local pressure and elastic deformation, and s(q,ξ)=w(q,ξ) / E * , P(q,p,ξ) is an error function, which represents the influence of local contact pressure p on elastic contact probability under specific magnification ξ and wave vector q. Since the elastic energy stored in the contact area is less than the elastic energy in the full contact area, a correction factor γ0 is introduced, and γ0 is taken as 0.

4.

9. A method for predicting hydrogen seal leakage as claimed in claim 4, characterized in that: The calculation process of the fatigue life calculation model: Based on the parameters of the rubber material and the length of the initial crack c0, the elastic strain energy U is obtained, the strain energy release rate G and the crack growth rate dc / dN are calculated, and then the number of cycles N and fatigue life T are obtained.

10. A method for predicting hydrogen seal leakage according to claim 9, characterized in that: According to the energy method, the released elastic strain energy U drives the crack to expand. σ1, σ2, and σ3 are the principal stresses in the x, y, and z directions, respectively; ε1, ε2, and ε3 are the strains in the three directions, respectively; The strain energy release rate G is introduced to represent the energy released when the initial crack increases per unit area. G is defined as the negative partial derivative of the strain energy U with respect to the crack area C, that is, the energy released when the initial crack increases per unit area. It can be expressed by formula (11): During the crack propagation process, the release mechanism of strain energy U is divided into two parts: one part is due to the redistribution of elastic strain energy generated by crack propagation inside the material, and the elastic strain energy originally stored around the crack is released along with the newly added crack surface; the other part is related to the work done by the external force W. Under the continuous action of the external force, the structural deformation mode changes when the crack propagates, and the work done by the external force affects the change of strain energy. The calculation method of the strain energy release rate is shown in formula (12): Where: G represents the strain energy release rate; W represents the work done by the external force; U represents the strain energy; B represents the material constant, which is related to factors such as the material's geometry and size; k(λ) represents the proportional coefficient related to the stretch ratio λ; c represents the crack length; w0 represents the strain energy density, which is the strain energy stored in a unit volume of material and is related to the material's mechanical properties such as elastic modulus and Poisson's ratio as well as the stress state. ε represents engineering strain; Substituting equation (11) into equation (12) yields the expression for the number of cycles N required for the initial crack length to expand from c0 to c: Where: β represents the crack growth parameter, c0 can be determined by experimental method; The expression for the number of cycles required for the initial crack length to expand from c0 to the final crack length c is shown in formula (15): The fatigue life T is obtained by multiplying the number of cycles by the cycle time per cycle t: T=Nt (16)

Images