Simulink-based airplane inerting system performance model and analysis method thereof

Abstract

The invention belongs to the technical field of aviation fuel inerting system design, and discloses a Simulink-based airplane inerting system performance model and an analysis method thereof. The model comprises a pipeline element mathematical model, an isolating valve element mathematical model, a resistance element mathematical model, a heat exchanger element mathematical model, a double-flow control valve mathematical model and a hollow fiber membrane neural network assembly. According to the method, parameterized models of all the parts of the aircraft inerting system are established, and the models are simplified by combining engineering experience based on the technical characteristics of all the parts, so that the model analysis speed is increased while the precision is ensured; parameterized model matching of all parts of the system is established based on Simulink, the influence of parameters of all the parts on the performance of the system can be dynamically simulated with high precision, early design index distribution is facilitated, and the design iteration period is shortened; an interface model among the components is established, and the input and output cross-linking relationship of the components is identified, so that the integrity of technical index requirements of the system and the equipment is ensured; and 4, performance prediction is carried out based on existing test data by using a neural network method, and the problems of difficult performance analysis and low accuracy of the heat exchanger and the hollow fiber membrane are solved.

Description

Technical Field

[0001] The invention belongs to the technical field of aviation fuel inerting system design, and relates to an aircraft inerting system performance model and an analysis method, and specifically relates to an aircraft inerting system performance model and an analysis method based on Simulink. Background Art

[0002] The civil aircraft fuel inerting system (FITS) usually consists of three subsystems: air handling system (CSAS), inert gas generation system (IGGS) and nitrogen-enriched gas distribution system (NEADS). CSAS includes inlet isolation valve, temperature isolation valve, temperature control valve, heat exchanger, etc., which are used to adjust air parameters so that the inert gas reaches the ideal pressure and temperature to improve the efficiency and life of ASM, and help protect the safety of downstream components and fuel tanks. IGGS usually consists of filters, air separation membranes (ASM), flow control valves and other equipment. The filter filters the air treated by CSAS to remove particles and volatile organic compounds (VOCs) in the airflow, and then passes it into one or more ASMs to separate and produce the required nitrogen-enriched inerting gas (NEA). The excess oxygen-enriched gas (OEA) is usually discharged outside the aircraft. NEADS usually consists of upstream check valves, flame suppressors, distribution check valves and other components. The purpose is to evenly distribute NEA from the onboard nitrogen generation system to the fuel tank to ensure that the oxygen concentration in the fuel tank meets the airworthiness requirements.

[0003] At present, there is a long iteration chain in the design of inerting systems and components in China. The early system design has low accuracy, and the indicators allocated to the equipment are somewhat blind and unreasonable. The application of performance analysis based on Simulink engineering model in early design can minimize the design problems caused by unreasonable demand allocation. Summary of the invention

[0004] In order to solve the above problems, the present invention provides an aircraft inerting system performance model and analysis method based on Simulink. The model can avoid the problems of low system design accuracy and poor equipment index allocation accuracy in the early stage, and effectively reduce the frequency of major problems in the later stage caused by unclear identification of early design requirements.

[0005] The technical solution of the present invention is as follows:

[0006] A Simulink-based aircraft inerting system performance model includes a pipeline component mathematical model, which is constructed by the following method:

[0007] Dynamic modeling of the piping components of the inerting system is performed, taking into account the viscous friction loss between the piping and the gas and the convective heat transfer between the gas and the pipe wall, including:

[0008] The mass conservation equation:

[0009]

[0010] Where: t is time, and are the mass flow rates of the two ports, Tl is the temperature of the internal gas node, Pl is the pressure of the internal gas node, ρI is the density of the internal gas node, and V is the volume of gas in the pipeline;

[0011] Energy conservation equation:

[0012]

[0013] in: and are the heat flow rates at the two ports respectively; is the heat flow rate of heat exchange between the pipeline and the outside world; U is the internal energy of the pipeline gas; P is the pipeline gas pressure;

[0014] Momentum conservation equation:

[0015]

[0016] Where: PA, PB, PI are the pressures of the two ports and the gas node respectively; S is the cross-sectional area of ​​the pipeline; ρA, ρB, ρI are the densities of the two ports and the gas node respectively; ΔPAI, ΔPBI are the pressure losses caused by the viscous friction of the pipeline respectively;

[0017] In laminar flow, the pressure loss caused by viscous friction at the pipe port is:

[0018]

[0019] Where: D h is the hydraulic diameter of the pipe; u I is the dynamic viscosity from the pipe to the internal gas node; L is the pipe length; L eqv is the total equivalent length of the local resistance; f shape is the laminar viscous friction coefficient;

[0020] In turbulent flow, the pressure loss caused by viscous friction at the pipe port is:

[0021]

[0022] Where: f Darcy is the Darcy drag friction factor;

[0023] The convection heat transfer equation between the tube wall and the gas inside the tube is:

[0024]

[0025] Where: Q conv S is the heat change rate of convective heat transfer; surf is the pipe surface area; I is the thermal conductivity of the gas in the pipeline; T H is the gas temperature outside the tube.

[0026] Furthermore, it also includes a mathematical model of the isolation valve component, and the mathematical model of the isolation valve component is constructed by the following method:

[0027] The isolation valve assembly is equipped with A and B inflow and outflow ports, and the concentration interface x O2 and an opening control interface x for receiving a physical signal;

[0028] The mass conservation equation:

[0029]

[0030] in: and are the mass flow rates at the two ports respectively;

[0031] Energy conservation equation:

[0032] φ A +φ B =0

[0033] in: and are the heat flow rates at the two ports respectively;

[0034] Momentum conservation equation:

[0035]

[0036] Where: ΔP is the pressure drop; f Darcy is the Darcy drag friction factor; ρ is the fluid density; v is the fluid velocity;

[0037] The pressure drop equation between flow rate and resistance is:

[0038]

[0039] in: is the mass flow rate; σ is the density correction factor; Z 1 and Z 2 Both are engineering recovery coefficients.

[0040] Furthermore, a resistance element mathematical model is also included, and the resistance element mathematical model is constructed by the following method:

[0041] The mathematical model of the resistance element is only described by the relationship between pressure and flow.

[0042] The mass conservation equation:

[0043]

[0044] in: and are the mass flow rates at the two ports respectively;

[0045] Energy conservation equation:

[0046] φ A +φ B =0

[0047] in: and are the heat flow rates at the two ports respectively;

[0048] Momentum conservation equation:

[0049]

[0050] in: is the mass flow rate; σ is the density correction factor; Z 1 and Z 2 All are engineering recovery coefficients.

[0051] Furthermore, a mathematical model of a heat exchanger element is also included, and the mathematical model of the heat exchanger element is constructed by the following method:

[0052] The mass conservation equation:

[0053]

[0054] in: and are the mass flow rates at the two ports respectively;

[0055] Energy conservation equation:

[0056] φ A +φ B +φ H =0

[0057] in: and are the heat fluxes at ports A and B respectively; is the rate of change of heat transfer of hot and cold fluids;

[0058] Momentum conservation equation:

[0059]

[0060] in: is the mass flow rate; σ is the density correction factor Z1 and Z 2 Both are engineering recovery coefficients.

[0061] Total heat transfer area and heat transfer coefficient of hot and cold side gases:

[0062]

[0063] Where: K1 and K2 are the heat transfer coefficients of the cold and hot sides respectively; F1 and F2 are the heat transfer areas of the cold and hot sides respectively; are the mass flow rates of the hot and cold sides, respectively;

[0064] η-NTU heat transfer module assembly establishment:

[0065] The η-NTU heat transfer module simulates the heat exchange of cold and hot fluids, and is connected to the cold side gas component and hot side gas component modules to simulate the pressure drop of the heat exchanger and the change of the inlet and outlet temperatures. The calculation method for the total heat transfer coefficient of the gas-to-gas heat exchanger is as follows:

[0066]

[0067] Where: Rw is the wall thermal resistance; K is the total heat transfer coefficient; F is the total heat transfer area;

[0068] The η-NTU heat transfer unit method model defines the heat exchanger efficiency η, the number of heat transfer units NTU, and the heat capacity ratio C* between the cold and hot fluids; the heat exchanger efficiency represents the ratio of the actual heat transfer Q of the heat exchanger to the theoretical maximum possible heat transfer Qmax.

[0069]

[0070] t1' and t2' refer to the cold and hot side inlet temperatures, respectively;

[0071] t1" and t2" refer to the outlet temperatures of the hot and cold sides, respectively;

[0072] Qmax in the heat exchanger efficiency definition is the maximum cooling that the hot fluid can achieve when the heat transfer area is assumed to be infinite;

[0073] NTU refers to the number of heat transfer units, which is defined as

[0074]

[0075] Where: KF represents the heat transferred when the average temperature difference of the heat exchanger is 1°C; Wmin is the smaller heat capacity of the two fluids;

[0076] The heat capacity ratio C* is the ratio of the smaller heat capacity Wmin to the larger heat capacity Wmax;

[0077]

[0078] The heat exchanger efficiency η value equation is:

[0079]

[0080] Therefore, the heat transfer rate of the hot and cold sides of the heat exchanger is:

[0081] φ H =ηW min (t 1 '-t 2 ').

[0082] Furthermore, a mathematical model of a heat exchanger element is also included. The mathematical model of a heat exchanger element can also be constructed by the following method:

[0083] The heat exchanger efficiency is calculated by the neural network method. In the calculation of the heat transfer performance of the heat exchanger, in the case of the existing heat exchanger experimental data, the heat exchange efficiency η is obtained by fitting the inlet temperature and flow rate of the cold end and hot end of the heat exchanger through the neural network; its expression is as follows:

[0084]

[0085] in: is the cold side gas mass flow rate; is the hot side gas mass flow rate; T1 is the cold side gas inlet temperature; T2 is the hot side gas inlet temperature.

[0086] Furthermore, a dual flow control valve mathematical model is also included, and the dual flow control valve mathematical model is constructed by the following method:

[0087] The gas mass accumulation inside the valve is not considered in the mathematical modeling process, and the dual flow control valve is regarded as a local restricted orifice with variable area;

[0088] Mass equation:

[0089]

[0090] in: and are the mass flow rates at the two ports respectively;

[0091] Energy equation:

[0092] φ A +φ B =0

[0093] in: and are the mass flow rates at the two ports respectively;

[0094] The gas inside the dual flow control valve is adiabatic, and the total enthalpy does not change at the inlet A, outlet B and orifice restriction:

[0095]

[0096] Where: h subscripts A, B, and R are the enthalpy values ​​at ports A, B, and R respectively;

[0097]

[0098] Where: S is the cross-sectional area of ​​ports A and B; S R is the cross-sectional area at the restriction; ρ subscripts A, B, and R are the densities at ports A, B, and R, respectively; wA, wB, and wR are the flow velocities at ports A, B, and R, respectively;

[0099] The theoretical mass flow rate without non-ideal effects is:

[0100]

[0101] Where: Cd is the flow coefficient, that is, the ratio of the actual mass flow rate through the restriction position to the theoretical mass flow rate;

[0102] Momentum equation:

[0103] Based on laminar flow:

[0104]

[0105] Δp transition =p avg (1-B lam )

[0106]

[0107] Based on the turbulence state:

[0108]

[0109] Among them, △p lam is the pressure difference between ports A and B in the laminar flow state; ρR is the density at position R, in kg / m3; △p transition is the pressure loss correction factor; r is the throttle aperture of the dual flow cut-off valve; B lam is the pressure loss correction factor at port B; △p AB is the pressure difference between ports A and B; p avg is the average pressure between ports A and B; p A and p B is the pressure at ports A and B.

[0110] Furthermore, it also includes a hollow fiber membrane neural network component, and the hollow fiber membrane neural network component is constructed by the following method:

[0111] The fuel inerting system established requires the inlet pressure, inlet temperature, outlet NEA flow rate and OEA pressure input of the hollow fiber membrane, and the membrane outlet NEA oxygen concentration and inlet air flow rate values ​​are obtained by looking up the table, which are expressed as:

[0112] x O2 =f(p in ,T in ,p OEA ,m NEA )

[0113] m air =f(p in ,T in ,p OEA ,m NEA )

[0114] x O2 is the NEA oxygen concentration at the membrane outlet, m air is the inlet air flow rate value, p in is the inlet pressure, T in is the inlet temperature, p OEA is the OEA pressure input, m NEA It is the egress NEA traffic.

[0115] A method for analyzing an aircraft inerting system performance model based on Simulink, used for analyzing the above-mentioned aircraft inerting system performance model based on Simulink, comprises the following steps:

[0116] S1: Set flight parameters:

[0117] Input the flight altitude, Mach number, and ambient oxygen concentration change over time data into the "Flight Envelope Initial Conditions" module;

[0118] S2: Set system parameters:

[0119] The input includes the bleed air module input parameters, pipeline parameters, inlet isolation valve parameters, ozone converter parameters, heat exchanger parameters, temperature control bypass valve parameters, filter parameters, temperature isolation valve parameters, dual flow control valve parameters, upstream check valve parameters, flame arrester parameters, distribution check valve parameters, aircraft fuel tank parameters, and the hollow fiber membrane neural network is established;

[0120] S3: Perform parameter monitoring:

[0121] View results in the Simulation tab in the Simulink toolstrip, open the Simulation Data Inspector and plot signals, compare individual signals or simulation runs, and compare results using relative, absolute, and time tolerance analysis.

[0122] Furthermore, in S2, the method for forming the hollow fiber membrane neural network is:

[0123] S21: Import the experimental data required for the hollow fiber membrane neural network;

[0124] S22: selecting data of each experimental point outside the range of the hollow fiber membrane neural network learning data to verify the accuracy of the neural network fitting;

[0125] S23: Select "Numerical Matrix" in MATLAB to import data; and also import the data of each experimental point outside the hollow fiber membrane neural network learning data range into the workspace;

[0126] S24: Select the neural network fitting module in MATLAB, import the neural network learning data from the workspace, and set the prediction variables;

[0127] S25: Divide the data into training set, validation set and test set;

[0128] S26: Start training.

[0129] The beneficial effects of the present invention are as follows:

[0130] 1. A parametric model of each component of the system was established, and the model was simplified based on the technical characteristics of each component and engineering experience, which improved the speed of model analysis while ensuring accuracy;

[0131] 2. Based on Simulink, the parametric model matching of each component of the system was established, which can dynamically simulate the impact of each component parameter on the performance of the system with high accuracy, which is helpful for early design index allocation and shortening the design iteration cycle;

[0132] 3. An interface model between components was established, and the input and output cross-linking relationship of each component was identified, which is conducive to ensuring the integrity of the system and equipment technical indicators;

[0133] 4. Using the neural network method, performance prediction is performed based on existing test data, solving the problem of difficult and low accuracy performance analysis of heat exchangers and hollow fiber membranes;

[0134] 5. Through the model parameter control module, the comprehensive performance of the system (heat exchanger outlet temperature, ASM inlet temperature, NEA outlet oxygen concentration, NEA mass flow rate and ASM inlet pressure, etc.) is evaluated to determine whether the performance requirements of the whole machine for the system can be met based on the current architecture and system parameters, which is conducive to the system solution demonstration and system indicator determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0135] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0136] Figure 1 The present invention is a schematic diagram of a performance model of a civil aircraft inerting system based on Simulink.

[0137] Figure 2 It is the pipeline parameter model.

[0138] Figure 3 It is the parameter model of the isolation valve.

[0139] Figure 4 It is the parameter model of the resistance element.

[0140] Figure 5 It is the parameter model of the heat exchanger element.

[0141] Figure 6 Schematic diagram of the hollow fiber membrane neural network process.

[0142] Figure 7 It is the parameter model of the neural network component of the hollow fiber membrane.

[0143] Figure 8 It is the parameter model of dual flow cut-off valve.

[0144] Fig. 9 Schematic diagram for flight parameter settings.

[0145] Fig.10 Schematic diagram for monitoring parameter configuration.

[0146] Fig.11 Schematic diagram of monitoring parameter results. DETAILED DESCRIPTION

[0147] This section is an embodiment of the present invention, which is used to explain and illustrate the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0148] Embodiment 1:

[0149] Modeling process:

[0150] A performance model and analysis method of civil aircraft inerting system based on Simulink, the mathematical modeling mainly includes the following aspects:

[0151] (I) Mathematical model of pipeline components

[0152] For the dynamic modeling of the piping components of the inerting system, it is necessary to consider the viscous friction loss between the pipeline and the gas and the convective heat transfer between the gas and the pipe wall.

[0153] The following assumptions are made for pipeline modeling:

[0154] 1) The volume of gas in the pipeline is constant and the gas in the pipeline is an ideal gas;

[0155] 2) The pressure and temperature vary according to the compressibility and heat capacity of the gas volume;

[0156] 3) Considering the impact of blockage on the pipeline, when the outlet velocity reaches the speed of sound, a blocked flow is generated;

[0157] (1) Conservation of mass

[0158] Relating mass flow rate to kinetics, the equation is as follows:

[0159]

[0160] The partial derivative equation of pipeline gas mass M with respect to pressure P and temperature T:

[0161]

[0162] Therefore, we can get:

[0163]

[0164] in:

[0165] t is time, in seconds;

[0166] and are the mass flow rates of the two ports, in kg / s;

[0167] TI is the temperature of the internal gas node, in K;

[0168] PI is the pressure of the internal gas node, in Pa;

[0169] ρI is the density of the internal gas node, in kg / m 3 ,

[0170] V is the volume of gas in the pipeline, in m 3 .

[0171] (2) Conservation of Energy

[0172] The conservation of energy relates the energy and heat flow rates to the dynamics of the pressure and temperature of the internal nodes representing the gas volume, and its equation is:

[0173]

[0174] The partial derivative equation of pipeline gas internal energy U with respect to pressure and temperature:

[0175]

[0176] in:

[0177] and are the heat flow rates at the two ports, in W;

[0178] It is the heat flow rate of heat exchange between the pipeline and the outside world, in W;

[0179] hI is the specific enthalpy of the internal gas node, in J / kg;

[0180] cpI is the constant pressure specific heat of the gas, in J / kg·℃;

[0181] Z is the compression factor;

[0182] R is the gas constant.

[0183] (3) Conservation of momentum

[0184] The momentum conservation equation captures the pressure drop in each half of the pipe due to momentum flux and viscous friction:

[0185]

[0186] The heat through port H is added to the internal gas node through energy conservation. For half of the pipe ports and the node connection between the internal gas is an adiabatic process, the relationship is:

[0187]

[0188] in:

[0189] PA, PR, and PI are the pressures of the two ports and the gas node, respectively, in Pa;

[0190] S is the cross-sectional area of ​​the pipeline, in m 2 ;

[0191] BA, ρB, ρI are the densities of the two ports and the gas node, respectively, in kg / m3.

[0192] ΔPAI and ΔPFI are the pressure losses caused by pipeline viscous friction, in Pa.

[0193] Along-the-line drag plays a role in flows in pipes or channels with inner surfaces. When a fluid flows through a pipe, there is friction between the fluid and the inner wall of the pipe, which slows down the movement of the fluid and causes energy loss. The resistance caused by this friction is called along-the-line drag. The magnitude of the along-the-line drag depends on several factors, including the geometry of the pipe, the roughness of the inner wall of the pipe, the properties of the fluid, and the flow rate. Generally, the along-the-line drag can be calculated using Darcy's formula. In laminar flow, the pressure loss caused by viscous friction at the pipe port is:

[0194]

[0195] in:

[0196] D h is the hydraulic diameter of the pipe in m;

[0197] u I is the dynamic viscosity from the tube to the internal gas node, in N·s / ㎡;

[0198] L is the pipeline length, in meters;

[0199] L eqv The total equivalent length of the local resistance, in meters;

[0200] f shape is the laminar viscous friction coefficient.

[0201] Friction coefficient f shape It is a key parameter that is related to the smoothness of the inner wall of the pipe and the viscosity of the fluid. The friction coefficient will also vary depending on the different types of pipe wall materials and fluid properties. Engineers need to select the appropriate friction coefficient based on the specific application scenario and fluid characteristics.

[0202] Darcy's formula is applicable to both laminar flow and turbulent flow, except that the friction coefficient λ is different in the two cases. In turbulent flow, the pressure loss caused by viscous friction at the pipe port is:

[0203]

[0204] Where: f Darcy is the Darcy drag friction factor.

[0205] The Moody diagram provides friction coefficient data for various fluids at different relative roughness, making it easier to obtain the value of the friction coefficient when applying the Darcy-Weisbach formula.

[0206] The Darcy friction factor is calculated based on the Haaland correlation:

[0207]

[0208] Where: rough It is the absolute roughness parameter of the inner surface of the pipeline.

[0209] (4) Convective heat exchange between pipeline and outside

[0210] The convection heat transfer equation between the tube wall and the gas inside the tube is:

[0211]

[0212] in:

[0213] Q conv is the rate of change of heat transfer due to convection, in W;

[0214] S surf is the surface area of ​​the pipeline, in m 2 ;

[0215] λ I is the thermal conductivity of the gas in the pipeline, in W / (m·K);

[0216] T H is the gas temperature outside the tube, in K.

[0217] Assuming that the temperature along the pipe is exponentially distributed, the convective heat transfer is

[0218]

[0219] in:

[0220] T in is the inlet temperature in the flow direction, in K;

[0221] c pavg is the specific heat of the gas at the average temperature in the pipeline, in J / kg·℃;

[0222] is the average mass flow rate of gas in the pipeline, in kg / s;

[0223] h coeff is the convection heat transfer coefficient between gas and tube wall, unit is W / (m 2 ·K).

[0224] (II) Mathematical model of isolation valve components

[0225] The isolation valve is a switch valve, which is mainly composed of a valve body, a valve cover, a valve disc, a sealing plate, a valve core, a valve seat, etc. It can only be in two states: open or closed, and has requirements for leakage levels. The isolation valve assembly is provided with an orifice with a variable opening area, which is controlled by a sliding ball mechanism to provide the FTIS with a deflation opening and closing function. The opening area changes with the relative position of the ball. The larger the opening area, the greater the valve lift, and the greater the flow rate of gas flowing through the valve assembly. The isolation valve assembly is provided with A and B inflow and outflow ports, and the concentration interface x O2 And an opening control interface x that receives a physical signal.

[0226] Since the local flow area of ​​the gas in the valve becomes narrower after flowing through the valve, when the orifice position reaches the speed of sound, a blocking state will occur. The local resistance limit node state is adiabatic and does not exchange heat with the external environment. When flowing through the resistance position, the flow area suddenly contracts and then expands. The gas accelerates during the contraction process, resulting in a pressure drop, and then suddenly expands. The gas separates from the wall in the process, and the pressure is only partially restored to produce a pressure drop.

[0227] (1) Conservation of mass

[0228] For a valve, the volume and mass of the fluid inside the valve are very small and can be ignored, so not much gas can accumulate in the valve. According to the principle of conservation of mass, the mass flow rate at the valve inlet is equal to the mass flow rate at the outlet, with inflow being positive and outflow being negative:

[0229]

[0230] in: and are the mass flow rates of the two ports respectively, in kg / s.

[0231] (2) Conservation of Energy

[0232] The valve element can be regarded as an adiabatic element, and no heat exchange occurs between the fluid and the wall around it. Energy flows from port A to port B in an advection manner.

[0233] φ A +φ B =0

[0234] in: and are the heat flow rates at the two ports, in W;

[0235] (3) Conservation of momentum

[0236] For valves, the momentum equation is usually used to calculate the pressure drop across the valve, usually using the DarcyWeisbach equation:

[0237]

[0238] in:

[0239] ΔP is the pressure drop, in Pa;

[0240] fDarcy is the Darcy drag friction factor;

[0241] ρ is the fluid density in kg / m 3 ;

[0242] v is the fluid velocity in m / s.

[0243] However, in order to simplify the calculation, the relationship between flow and resistance is usually simplified by an empirical formula. This method is used by foreign inertial system manufacturers. The simplified pressure drop equation is:

[0244]

[0245] in:

[0246] is the mass flow rate, in g / s;

[0247] σ is the density correction factor,

[0248] Z1 and Z2 are engineering recovery coefficients (each unit is selected according to the equipment)

[0249] The calculation formula is as follows:

[0250]

[0251] T ref =15°C(288.15K)

[0252] P ref =101325Pa(14.696psia)

[0253] Currently, only two states, open and closed, are considered for isolation valve assemblies. When the isolation valve assembly is opened, the system runs, and when the isolation valve assembly is closed, the system stops.

[0254] (III) Mathematical model of resistance element

[0255] At present, due to the initial gas state, the presence of ozone, water vapor and other impurities is not considered. Currently, only the ozone converter, filter and flame arrester are regarded as resistance elements. For resistance elements, mass accumulation is not considered, and they are only described by the relationship between pressure and flow.

[0256] (1) Conservation of mass

[0257] For the resistance element, not much gas can accumulate in the element. Without considering mass accumulation, the pressure difference between its inlet and outlet is calculated only by flow. According to the principle of mass conservation, the mass flow rate flowing through the inlet of the resistance element is equal to the mass flow rate at the outlet, with inflow being positive and outflow being negative:

[0258]

[0259] in: and are the mass flow rates of the two ports respectively, in kg / s.

[0260] (2) Conservation of Energy

[0261] The resistance element can be regarded as an insulating element. The gas flowing into and out of the port will not cause heat change, and there is no heat exchange between the fluid and the surrounding wall. Energy flows from port A to port B. The rate of change of heat flowing through the inlet is equal to the rate of change of heat flowing through the outlet:

[0262] φ A +φ B =0

[0263] in: and are the heat flow rates at the two ports, in W;

[0264] (3) Conservation of momentum

[0265] The external force of the fluid in the resistance element mainly lies in the viscous friction between the port and the gas and the component wall. At present, the empirical formula between pressure drop and flow rate is used abroad:

[0266]

[0267] in:

[0268] is the mass flow rate, in g / s;

[0269] σ is the density correction factor.

[0270] This equation is a linear combination of the mass flow rate and its square, an empirical relationship derived from experimental data obtained from component testing performed by the FTIS equipment supplier. This equation was introduced to simplify the complex FTIS model using the Darcy-Weisbach standard equation described above.

[0271] (III) Mathematical model of heat exchanger elements

[0272] According to foreign architecture references, the cross-flow plate heat exchanger is very suitable for aircraft fuel tank inerting systems due to its high efficiency, compact structure, light weight, pressure resistance, easy maintenance and cleaning, and strong adaptability.

[0273] The modeling process assumes that:

[0274] The fluid is single-phase; since the current system does not consider the existence of water vapor, the latent heat of vaporization is not considered; there is no chemical reaction inside the heat exchanger; there are enough ribs, and the temperature change of the fluid between each baffle can be ignored compared with the total change; the heat exchange plate is thin, the axial and radial heat conduction on the plate can be ignored, and the lateral temperature is uniform; the inlet effect of the fluid is ignored; it is assumed that the temperature is mixed evenly on the cross section of the flow channel; only the heat exchange between the cold and hot fluids is considered, and the heat exchange between the heat exchanger and the external environment is not considered.

[0275] (1) Conservation of mass

[0276] Assuming that the fluid is not compressible in the heat exchanger, its mass conservation is expressed as the sum of the mass flow rate at port A and the mass flow rate at port B is zero.

[0277]

[0278] in: and are the mass flow rates of the two ports respectively, in kg / s.

[0279] (2) Conservation of Energy

[0280] Energy can be transferred via heat flow between ports A and B and via the hot and cold side heat exchange from port H. Inflow is positive and outflow is negative.

[0281] φ A +φ B +φ H =0

[0282] in:

[0283] and are the heat fluxes at ports A and B respectively;

[0284] is the rate of change of heat transfer of hot and cold fluids, in W;

[0285] (3) Conservation of momentum

[0286] The pressure drop of gas flowing through a heat exchanger is a function of mass flow rate given by an empirical formula and has a certain flow range limitation.

[0287]

[0288] in:

[0289] is the mass flow rate, in g / s;

[0290] σ is the density correction factor.

[0291] The total heat transfer area and heat transfer coefficient for the hot and cold side gases can be expressed in the form of an exponential mass flow rate.

[0292]

[0293] in:

[0294] K1 and K2 are the heat transfer coefficients of the hot and cold sides, respectively, in W / m 2 K;

[0295] F1 and F2 are the heat transfer areas of the hot and cold sides, respectively, in m 2 ;

[0296] are the mass flow rates of the hot and cold sides respectively, in kg / s.

[0297] In the formula, a and b are empirical coefficients that can be obtained through experiments.

[0298] (4) Establishment of η-NTU heat transfer module components

[0299] In heat exchanger analysis, if the inlet and outlet temperatures of the fluid are known or can be calculated using energy balance, the logarithmic mean temperature difference can be used for analysis, but if there is no outlet temperature and efficiency, the NTU method can be used for analysis. The η-NTU method is usually used in the design calculation of plate heat exchangers. The heat exchanger component is a composite subsystem, including the cold side gas component, the hot side gas component and the η-NTU heat transfer module component.

[0300] The η-NTU heat transfer module component simulates the heat exchange between cold and hot fluids, and is connected to the cold side gas component and hot side gas component modules to simulate the pressure drop of the heat exchanger and the changes in the outlet and inlet temperatures.

[0301] Calculation method for the total heat transfer coefficient of the gas-to-gas heat exchanger:

[0302]

[0303] in:

[0304] Rw is the wall thermal resistance;

[0305] K is the overall heat transfer coefficient;

[0306] K1 and K2 are the heat transfer coefficients of the hot and cold sides, respectively, in W / m 2 K;

[0307] F is the total heat transfer area, in m 2 ;

[0308] F1 and F2 are the heat transfer areas of the hot and cold sides, respectively, in m 2 .

[0309] The η-NTU heat transfer unit method model defines the heat exchanger efficiency η between the cold and hot fluids, the number of heat transfer units NTU, and the heat capacity ratio C*. The heat exchanger efficiency represents the ratio of the actual heat transfer Q of the heat exchanger to the theoretical maximum possible heat transfer Qmax.

[0310]

[0311] Qmax in the heat exchanger efficiency definition is the maximum cooling that the hot fluid can achieve when the heat transfer area is assumed to be infinite.

[0312] t1' and t2' refer to the cold and hot side inlet temperatures, respectively;

[0313] t1" and t2" refer to the outlet temperatures of the hot and cold sides, respectively;

[0314] NTU refers to the number of heat transfer units, which is defined as

[0315]

[0316] in:

[0317] KF represents the heat transferred when the average temperature difference of the heat exchanger is 1°C;

[0318] Wmin is the smaller heat capacity of the two fluids.

[0319] Therefore, NTU is an important dimensionless quantity that represents the heat transfer capacity of a heat exchanger. The heat capacity ratio C* is the ratio of the smaller heat capacity Wmin to the larger heat capacity Wmax.

[0320]

[0321] Under certain flow conditions, the size of NTU and C* values ​​directly affects the size of the heat exchanger efficiency η, that is, η is a function of NTU and C*. The plate heat exchanger used in this project belongs to a single-flow cross-flow flow mode in which the two fluids are not mixed, because the cold and hot fluids pass through separate flow channels and do not mix. The efficiency approximate relationship formula proposed by Drake is currently commonly used in engineering calculations as follows:

[0322]

[0323] This formula is simple to calculate and has high accuracy.

[0324] Therefore, the heat transfer rate of the hot and cold sides of the heat exchanger can be calculated by the following formula:

[0325] φ H =ηW min (t 1 '-t 2 ')

[0326] (V) Neural Network Method for Calculating Heat Exchanger Efficiency [Alternative to (IV)]

[0327] In addition to using the NTU method to calculate the heat exchanger efficiency, the heat exchanger efficiency can also be calculated using a neural network method. Neural networks are good at fitting functions. By training neural networks, nonlinear relationships between data can be learned, which can better adapt to complex relationships in actual data, and adaptively adjust parameters to adapt to different data. It has better generalization ability, and can perform parallel calculations on multiple processors at the same time, with fast calculation speed.

[0328] The calculation of heat transfer performance of heat exchangers involves the specific heat capacity, density and thermal conductivity of the heat exchanger tube wall, the specific heat capacity, thermal conductivity, viscosity and density of the fluid, and the saturated vapor pressure and latent heat of vaporization of the fluid when there is a phase change. These physical properties change to varying degrees with temperature, and the calculation iteration process is relatively complicated. In the actual process, the heat exchange area of ​​the heat exchanger with complex structure is difficult to measure accurately. Therefore, in the process of heat exchanger mechanism modeling and design calculation, only the change characteristics of certain specific positions or specific parameters of the heat exchanger are concerned.

[0329] Whether it is heat exchanger design calculation or verification calculation, heat exchanger efficiency is always the key parameter for calculating thermodynamic performance. In the case of existing heat exchanger experimental data, the heat exchange efficiency η that is more accurate and fits the actual situation can be obtained through neural network fitting through the inlet temperature and flow rate of the cold end and hot end of the heat exchanger. Its expression is as follows:

[0330]

[0331] in:

[0332] is the cold side gas mass flow rate;

[0333] is the hot side gas mass flow rate, in kg / s;

[0334] T1 is the cold side gas inlet temperature, in K;

[0335] T2 is the hot side gas inlet temperature, in K.

[0336] (VI) Mathematical model of dual flow control valve

[0337] The dual flow control valve operates in different flow modes depending on the aircraft's flight phase. It also serves as an auxiliary means to prevent fuel / vapor from entering the FTIS from the fuel tank.

[0338] The gas mass accumulation inside the valve is not considered in the mathematical modeling process, and the dual flow control valve is regarded as a local restricted orifice with variable area. When the speed of sound is reached at the orifice restriction, a choked flow will occur. The choked position is adiabatic and no heat exchange occurs with the environment.

[0339] (1) Mass equation

[0340]

[0341] (2) Energy equation

[0342] φ A +φ B =0

[0343] The gas inside the element is adiabatic, and the total enthalpy does not change at the inlet A, outlet B, and orifice restriction:

[0344]

[0345] in:

[0346] hA, hB, hR are the enthalpy values ​​at ports A, B, and R.

[0347] WA, WB, WR are the flow velocities at A, B, and R, in m / s;

[0348]

[0349] in:

[0350] S is the cross-sectional area of ​​ports A and B;

[0351] SR is the cross-sectional area at the restriction, in m2;

[0352] ρ is the density at different positions, in kg / m3.

[0353] The theoretical mass flow rate without non-ideal effects is:

[0354]

[0355] Where: C d is the flow coefficient, which is the ratio of the actual mass flow rate through the restriction to the theoretical mass flow rate.

[0356] (3) Momentum equation

[0357] The pressure difference between ports A and B is the momentum balance based on the contraction of the inlet and restriction areas plus the momentum balance based on the expansion of the outlet and restriction areas.

[0358] Based on laminar flow:

[0359]

[0360] Δp transition =p avg (1-B lam )

[0361]

[0362] Based on the turbulence state:

[0363]

[0364] in:

[0365] △p lam is the pressure difference between ports A and B in the laminar flow state;

[0366] ρR is the density at position R, in kg / m 3 ;

[0367] △p transition is the pressure loss correction factor;

[0368] r is the throttle aperture of the dual flow cut-off valve;

[0369] B lam is the pressure loss correction factor at port B;

[0370] △p AB is the pressure difference between ports A and B;

[0371] p avg is the average pressure between ports A and B;

[0372] p A and p B is the pressure at ports A and B;

[0373] When the flow through the restriction becomes a blocking condition, further flow calculations depend on upstream conditions and are not affected by downstream conditions. For dual flow control valves, the outlets are generally blocked:

[0374]

[0375] Where a is the speed of sound in m / s.

[0376]

[0377] (VII) Method for establishing hollow fiber membrane neural network components

[0378] Hollow fiber membrane is a key component of fuel inerting system, because it generates NEA through engine bleed air or compressed cabin air to inertify the fuel tank. The basic principle of oxygen and nitrogen separation by hollow fiber membrane is that the partial pressure difference of oxygen and nitrogen inside and outside the membrane is different, and the mass transfer rate through the membrane is different, thus achieving the purpose of separation.

[0379] The process is roughly divided into three steps: adsorption, dissolution, diffusion, and desorption. The gas molecules on the raw material side contact the dense membrane surface, adsorb and dissolve on the membrane surface; because of the enrichment of gas molecules, a concentration gradient is generated. Driven by the concentration difference, the gas molecules diffuse in the membrane to the permeation surface and finally desorb. Therefore, the permeability of oxygen and nitrogen on polymer membranes is different, depending on the ability of adsorption, dissolution and diffusion.

[0380] The factors affecting the performance of hollow fiber membrane modules are relatively complex. The accuracy of a single mathematical model is currently unknown, so the table lookup method is usually used for calculation at home and abroad. The established FTIS (fuel inerting system) requires the input of inlet pressure, inlet temperature, outlet NEA flow rate and OEA pressure for the ASM (hollow fiber membrane), and the table lookup is used to obtain the membrane outlet NEA oxygen concentration and inlet air flow value. It can be expressed as:

[0381] x O2 =f(p in ,T in ,p OEA ,m NEA )

[0382] m air =f(p in ,T in ,p OEA ,m NEA )

[0383] x O2 is the NEA oxygen concentration at the membrane outlet, m air is the inlet air flow rate value, p in is the inlet pressure, T in is the inlet temperature, p OEA is the OEA pressure input, m NEA is the export NEA traffic

[0384] However, since the four-dimensional function table lookup in Simulink is relatively complex and the search speed is slow, and the change relationship of the data points given by the experiment requires the use of a more complex nonlinear function model, it has a huge impact on the calculation speed of the ASM model.

[0385] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

[0386] Analysis process:

[0387] A performance model and analysis method of civil aircraft inerting system based on Simulink, the application of mathematical model mainly includes the following steps:

[0388] 1. Flight parameter settings

[0389] 1. Click the flight parameter input form in the initial folder, enter the flight altitude, Mach number, and ambient oxygen concentration over time data into the corresponding position, click Save and close the Excel data form. The time is in seconds, the flight altitude is in meters, the Mach number is 1, and the ambient oxygen concentration is 1;

[0390] 2. Double-click the "Flight Envelope Initial Conditions" block to open the Simulink signal group input function interface, and click the title bar "File-Import from File" with the left mouse button;

[0391] 3. In the prompt for entering the file to be imported, click "Browse" in the lower right corner, select the spreadsheet file "flight customer input.xlsx" to be imported and open it.

[0392] 4. In the data selection to be imported, check the options of "Altitude", "Mach number", and "Air oxygen concentration". In the data placement selection, select "Replace existing data set" and click to confirm the selection. Click OK below to successfully import the data.

[0393] The flight parameter settings mainly include the input of flight altitude, Mach number, oxygen concentration in the outside air, and vertical speed in the model, the calculation of ambient temperature and pressure at different flight altitudes, the calculation of recovery temperature and recovery pressure of the boundary layer on the aircraft surface, and the calculation of flight airspeed and vertical speed. The changes of flight altitude, Mach number, oxygen concentration in the outside air, and vertical speed over time can be imported into the flight envelope initial condition module through a custom Excel spreadsheet. The flight parameters support user-defined modification.

[0394] 2. System parameter settings

[0395] The system parameter settings mainly include air bleed module input parameter settings, pipeline parameter settings, inlet isolation valve parameter settings, ozone converter parameter settings, heat exchanger parameter settings, temperature control bypass valve parameter settings, filter parameter settings, temperature isolation valve parameter settings, dual flow control valve parameter settings, upstream check valve parameter settings, flame suppressor parameter settings, distribution check valve settings, aircraft fuel tank settings, and hollow fiber membrane neural network formation.

[0396] 1. Input parameter setting of air bleed module

[0397] Double-click the engine bleed air module with the left mouse button, enter the engine bleed air temperature (K), engine bleed air temperature (Pa), and click OK;

[0398] 2. Pipeline parameter setting

[0399] Double-click the pipeline module with the left mouse button, enter the pipeline length (m), pipeline cross-sectional area (m^2), hydraulic diameter (m) and equivalent length of pipeline local resistance (m), and click OK after the settings are completed. In addition, the initial values ​​of the gas pressure, temperature and density in the pipe can be modified.

[0400] 3. Inlet isolation valve parameter setting

[0401] Double-click the inlet isolation valve module icon with the left mouse button. You need to enter the minimum resistance area (m^2) (also known as the leakage area of ​​the valve) and the maximum resistance area (m^2) of the isolation valve. In addition, you need to obtain the cross-sectional area of ​​the port (m^2). Note that the minimum resistance area needs to be smaller than the maximum resistance area, and both are smaller than the cross-sectional area of ​​the port.

[0402] 4. Ozone converter parameter settings

[0403] Double-click the ozone converter module icon with the left mouse button to enter the cross-sectional area of ​​its port (m^2).

[0404] 5. Heat exchanger parameter setting

[0405] Double-click the heat exchanger module icon with the left mouse button to modify the heat exchanger parameters: heat exchanger wall thermal resistance (K / W); cold end port cross-sectional area (m^2), thermal resistance coefficient and piezoresistance coefficient; heat exchanger hot end port cross-sectional area (m^2), thermal resistance coefficient and piezoresistance coefficient. Click OK after the settings are completed.

[0406] 6. Temperature control bypass valve parameter setting

[0407] Double-click the temperature control valve module with the left button of the mouse, and set the heat exchanger outlet target temperature (K) and flow area (m^2). After the settings are completed, click OK.

[0408] 7. Filter parameter settings

[0409] Double-click the filter module with the left mouse button to enter the cross-sectional area of ​​its port (m^2). Click OK after the setting is completed.

[0410] 8. Temperature isolation valve parameter setting

[0411] Double-click the temperature isolation valve module icon with the left mouse button. You need to enter the minimum resistance area (m^2) (also known as the leakage area of ​​the valve) and the maximum resistance area (m^2) of the isolation valve. In addition, you need to obtain the cross-sectional area of ​​the port (m^2). Note that the minimum resistance area needs to be smaller than the maximum resistance area, and both are smaller than the cross-sectional area of ​​the port.

[0412] 9. Dual flow control valve parameter setting

[0413] Double-click the dual flow control valve module icon with the left mouse button. You need to enter the resistance area of ​​the smaller hole of the valve (m^2) and the resistance area of ​​the larger hole (m^2). In addition, you need to obtain the cross-sectional area of ​​the port (m^2). At this position, the flow is usually blocked because the resistance aperture of the valve is very small. The flow rate depends on the size of the resistance aperture area. You can manually select the flow mode, 1-small flow, 2-medium flow, 3-large flow.

[0414] 10. Upstream check valve parameter setting

[0415] Double-click the upstream check valve module icon with the left mouse button. You need to enter the smaller hole resistance area (m^2) and the larger hole resistance area (m^2) of the isolation valve, and also obtain the cross-sectional area (m^2) of the port. At this position, the flow is usually blocked because the resistance aperture of the valve is very small, and the flow rate depends on the size of the resistance aperture area.

[0416] 11. Flame arrester parameter setting

[0417] Double-click the fire damper module with the left mouse button to enter the cross-sectional area of ​​its port (m^2). Click OK after the setting is completed.

[0418] 12. Dispensing check valve setting

[0419] Double-click the check valve module icon with the left mouse button. You need to enter the maximum opening pressure (Pa), cracking pressure (Pa), maximum opening area (m^2), escape area (m^2), and flow coefficient of the check valve. In addition, you also need to obtain the cross-sectional area of ​​the port (m^2) and the time required for the valve to open (s).

[0420] 13. Aircraft fuel tank settings

[0421] Double-click the aircraft fuel tank module icon with the left button of the mouse, and you need to enter the volume (m^3), initial fuel loading rate, and inflow orifice area (m^2) of the left wing fuel tank, center fuel tank, and right wing fuel tank.

[0422] 14. Hollow fiber membrane neural network construction

[0423] a. Open the file in the unzipped folder and import the experimental data required for the hollow fiber membrane neural network.

[0424] b. Open the file in the unzipped folder, select the data of each experimental point outside the range of hollow fiber membrane neural network learning data, and input them into the "fiber_test" excel file to verify the accuracy of the neural network fitting.

[0425] c. Click "Import Data" in the title bar of the MATLAB main interface and choose to open the "fiber_in" file.

[0426] d. Select "Numeric Matrix" in "Import-Output Type", modify the name of the workspace input data matrix, select the table lookup input data column (independent variable), and finally click Import.

[0427] e. Modify the output data matrix name and import the output data column (dependent variable) into the workspace.

[0428] f. Similarly, import the neural network test data of the "fiber_test" excel file into the workspace through the above operations. At this time, the workspace has the following 6 numerical matrices for neural network fitting and testing.

[0429] g. Click "app" in the title bar of the MATLAB main interface, then click all app icons, select the "neural network fitting icon" under the machine learning and deep learning column, and click it to pop up the neural network fitting learning interface.

[0430] h. Import the neural network learning data from the workspace, select the workspace "Membrane input learning parameter fiberin_find" from the drop-down list on the right side of the predictor variable, select the workspace "Membrane output learning parameter fiberin_NEAxo2" from the drop-down list on the right side of the response variable, and select "Row" for the observation value. Click OK.

[0431] i. Split the data into training, validation, and test sets. Keep the default settings. The data is split into: 70% is usually set for training; 15% is used to verify that the network is generalizing and stop training before overfitting; 15% is used to independently test the generalization of the network.

[0432] j. Select Training > Levenberg-Marquardt Training. This is the default training algorithm and has the same effect as click training. This method has the advantage of fast calculation speed.

[0433] k. In the training pane, you can see the training progress and stop training when the verification conditions are met.

[0434] l. You can further analyze the results by generating plots. To plot the linear regression, in the Plot section, click Regression. The regression plot shows the network predictions (outputs) for the response (target) for the training set, validation set, and test set.

[0435] The parameter monitoring module includes an indicator light module and a real-time data module. The indicator light module includes an inlet isolation valve indicator light, a temperature isolation valve indicator light, a temperature control valve indicator light, an upstream check valve indicator light, a distribution check valve indicator light, a flow control valve indicator light, etc., which can display the equipment working status under different system parameter settings. The real-time data module mainly displays the system operating time, flight altitude, engine bleed air inlet pressure, heat exchanger outlet temperature, ASM inlet temperature, NEA outlet oxygen concentration, NEA mass flow rate and ASM inlet pressure.

[0436] 2. Parameter Monitoring

[0437] In the model's Configuration Parameters, select Data Import / Export > Record logged workspace data in Simulation Data Inspector to make the logged state and Simscape data available in the Simulation Data Inspector after simulation. If you want to view the signals without logging data, consider using a Scope block, Floating Scope block, or ScopeViewer.

[0438] Open the simulation Data Inspector from Simulation > Data Inspector or by clicking a signal logging marker. When you open the simulation Data Inspector by clicking a logging marker on a signal, the system automatically plots the signal.

[0439] (1) In the Simulink toolbar, on the Simulation tab, under View Results, click Data Inspector.

[0440] (2) Click the streaming marker on the signal to open the Simulation Data Inspector and plot the signal.

[0441] (3) Compare single signals or simulation runs and compare results using relative, absolute, and time tolerance analysis.

[0442] (4) To import data, select the Import button in the Simulation Data Inspector. You can choose to import data from the workspace or from a file. The table below the options shows the data available for import.

[0443] (5) You can use the cursor to check the correspondence between the horizontal and vertical coordinate signal values.

[0444] The parameter monitoring module includes an indicator light module and a real-time data module. The indicator light module includes an inlet isolation valve indicator light, a temperature isolation valve indicator light, a temperature control valve indicator light, an upstream check valve indicator light, a distribution check valve indicator light, a flow control valve indicator light, etc., which can display the equipment working status under different system parameter settings. The real-time data module mainly displays the system operating time, flight altitude, engine bleed air inlet pressure, heat exchanger outlet temperature, ASM inlet temperature, NEA outlet oxygen concentration, NEA mass flow rate and ASM inlet pressure.

[0445] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the field can easily think of various equivalent modifications or substitutions within the technical scope disclosed by the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A Simulink-based aircraft inerting system performance model, characterized in that: Including the mathematical model of piping components, the mathematical model of piping components is constructed by the following method: Dynamic modeling of the piping components of the inerting system is performed, taking into account the viscous friction loss between the piping and the gas and the convective heat transfer between the gas and the pipe wall, including: The mass conservation equation: Where: t is time, and are the mass flow rates at the two ports, T I is the temperature of the internal gas node, P I is the pressure of the internal gas node, ρ I is the density of the internal gas node, V is the volume of gas in the pipeline; Energy conservation equation: in: and are the heat flow rates at the two ports respectively; is the heat flow rate of heat exchange between the pipeline and the outside world; U is the internal energy of the pipeline gas; P is the pipeline gas pressure; Momentum conservation equation: Where: P A ,P B ,P I are the pressures of the two ports and the gas node respectively; S is the cross-sectional area of ​​the pipeline; ρ A ,ρ B ,ρ I are the densities of the two ports and the gas node respectively; ΔP AI ,ΔP BI are the pressure losses caused by pipeline viscous friction; In laminar flow, the pressure loss caused by viscous friction at the pipe port is: Where: D h is the hydraulic diameter of the pipe; u I is the dynamic viscosity from the pipe to the internal gas node; L is the pipe length; L eqv is the total equivalent length of the local resistance; f shape is the laminar viscous friction coefficient; In turbulent flow, the pressure loss caused by viscous friction at the pipe port is: Where: f Darcy is the Darcy drag friction factor; The convection heat transfer equation between the tube wall and the gas inside the tube is: Where: Q conv S is the heat change rate of convective heat transfer; surf is the pipe surface area; I is the thermal conductivity of the gas in the pipeline; T H is the gas temperature outside the tube.

2. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: It also includes a mathematical model of the isolation valve component, which is constructed by the following method: The isolation valve assembly is equipped with A and B inflow and outflow ports, and the concentration interface x O2 and an opening control interface x for receiving a physical signal; The mass conservation equation: in: and are the mass flow rates at the two ports respectively; Energy conservation equation: f A +φ B =0 in: and are the heat flow rates at the two ports respectively; Momentum conservation equation: Where: ΔP is the pressure drop; f Darcy is the Darcy drag friction factor; ρ is the fluid density; v is the fluid velocity; The pressure drop equation between flow rate and resistance is: in: is the mass flow rate; σ is the density correction factor; Z1 and Z2 are both engineering recovery coefficients.

3. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: It also includes a resistance element mathematical model, which is constructed by the following method: The mathematical model of the resistance element is only described by the relationship between pressure and flow. The mass conservation equation: in: and are the mass flow rates at the two ports respectively; Energy conservation equation: f A +φ B =0 in: and are the heat flow rates at the two ports respectively; Momentum conservation equation: in: is the mass flow rate; σ is the density correction factor; Z1 and Z2 are both engineering recovery coefficients.

4. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: It also includes a mathematical model of the heat exchanger element, which is constructed by the following method: The mass conservation equation: in: and are the mass flow rates at the two ports respectively; Energy conservation equation: f A +φ B +φ H =0 in: and are the heat fluxes at ports A and B respectively; is the rate of change of heat transfer of hot and cold fluids; Momentum conservation equation: in: is the mass flow rate; σ is the density correction factor; Z1 and Z2 are both engineering recovery coefficients. Total heat transfer area and heat transfer coefficient of hot and cold side gases: Where: K1 and K2 are the heat transfer coefficients of the cold and hot sides respectively; F1 and F2 are the heat transfer areas of the cold and hot sides respectively; are the mass flow rates of the hot and cold sides respectively; η-NTU heat transfer module assembly establishment: The η-NTU heat transfer module simulates the heat exchange of cold and hot fluids, and is connected to the cold side gas component and hot side gas component modules to simulate the pressure drop of the heat exchanger and the change of the inlet and outlet temperatures. The calculation method for the total heat transfer coefficient of the gas-to-gas heat exchanger is as follows: Where: Rw is the wall thermal resistance; K is the total heat transfer coefficient; F is the total heat transfer area; The η-NTU heat transfer unit method model defines the heat exchanger efficiency η, the number of heat transfer units NTU, and the heat capacity ratio C* between the cold and hot fluids; the heat exchanger efficiency represents the ratio of the actual heat transfer Q of the heat exchanger to the theoretical maximum possible heat transfer Qmax. t1' and t2' refer to the cold and hot side inlet temperatures, respectively; t1" and t2" refer to the outlet temperatures of the hot and cold sides, respectively; Qmax in the heat exchanger efficiency definition is the maximum cooling that the hot fluid can achieve when the heat transfer area is assumed to be infinite; NTU refers to the number of heat transfer units, which is defined as Where: KF represents the heat transferred when the average temperature difference of the heat exchanger is 1°C; Wmin is the smaller heat capacity of the two fluids; The heat capacity ratio C* is the ratio of the smaller heat capacity Wmin to the larger heat capacity Wmax; The heat exchanger efficiency η value equation is: Therefore, the heat transfer rate of the hot and cold sides of the heat exchanger is: f H =ηW min (t1'-t2').

5. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: It also includes a mathematical model of the heat exchanger element, which is constructed by the following method: The heat exchanger efficiency is calculated by the neural network method. In the calculation of the heat transfer performance of the heat exchanger, in the case of the existing heat exchanger experimental data, the heat exchange efficiency η is obtained by fitting the inlet temperature and flow rate of the cold end and hot end of the heat exchanger through the neural network; Its expression is as follows: in: is the cold side gas mass flow rate; is the hot side gas mass flow rate; T1 is the cold side gas inlet temperature; T2 is the hot side gas inlet temperature.

6. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: It also includes a dual flow control valve mathematical model, which is constructed by the following method: The gas mass accumulation inside the valve is not considered in the mathematical modeling process, and the dual flow control valve is regarded as a local restricted orifice with variable area; Mass equation: in: and are the mass flow rates at the two ports respectively; Energy equation: f A +φ B =0 in: and are the mass flow rates at the two ports respectively; The gas inside the dual flow control valve is adiabatic, and the total enthalpy does not change at the inlet A, outlet B and orifice restriction: Where: h subscripts A, B, and R are the enthalpy values ​​at ports A, B, and R respectively; Where: S is the cross-sectional area of ​​ports A and B; S R is the cross-sectional area at the restriction; ρ subscripts A, B, and R are the densities at ports A, B, and R, respectively; W A , W B , W R are the flow rates at ports A, B, and R, respectively; The theoretical mass flow rate without non-ideal effects is: Where: Cd is the flow coefficient, that is, the ratio of the actual mass flow rate through the restriction position to the theoretical mass flow rate; Momentum equation: Based on laminar flow: Δp transition =p avg (1-B lam ) Based on the turbulence state: Among them, △p lam is the pressure difference between ports A and B in the laminar flow state; ρ R is the R position density; △p transition is the pressure loss correction factor; r is the throttle aperture of the dual flow cut-off valve; B lam is the pressure loss correction factor at port B; △p AB is the pressure difference between ports A and B; p avg is the average pressure between ports A and B; p A and p B is the pressure at ports A and B.

7. The aircraft inerting system performance model based on Simulink according to claim 1, characterized in that: Also included is a hollow fiber membrane neural network component, which is constructed by the following method: The fuel inerting system established requires the inlet pressure, inlet temperature, outlet NEA flow rate and OEA pressure input of the hollow fiber membrane, and the membrane outlet NEA oxygen concentration and inlet air flow rate values ​​are obtained by looking up the table, which are expressed as: x O2 =f(p in ,T in ,p OEA ,m NEA ) m air =f(p in ,T in ,p OEA ,m NEA ) x O2 is the NEA oxygen concentration at the membrane outlet, m air is the inlet air flow rate value, p in is the inlet pressure, T in is the inlet temperature, p OEA is the OEA pressure input, m NEA It is the egress NEA traffic.

8. A method for analyzing an aircraft inerting system performance model based on Simulink, used for analyzing an aircraft inerting system performance model based on Simulink as claimed in any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Set flight parameters: Input the flight altitude, Mach number, and ambient oxygen concentration change over time data into the "Flight Envelope Initial Conditions" module; S2: Set system parameters: The input includes the bleed air module input parameters, pipeline parameters, inlet isolation valve parameters, ozone converter parameters, heat exchanger parameters, temperature control bypass valve parameters, filter parameters, temperature isolation valve parameters, dual flow control valve parameters, upstream check valve parameters, flame arrester parameters, distribution check valve parameters, aircraft fuel tank parameters, and the hollow fiber membrane neural network is established; S3: Perform parameter monitoring: View results in the Simulation tab in the Simulink toolstrip, open the Simulation Data Inspector and plot signals, compare individual signals or simulation runs, and compare results using relative, absolute, and time tolerance analysis.

9. The analysis method of the aircraft inerting system performance model based on Simulink according to claim 8, characterized in that: In S2, the method for forming the hollow fiber membrane neural network is: S21: Import the experimental data required for the hollow fiber membrane neural network; S22: selecting data of each experimental point outside the range of the hollow fiber membrane neural network learning data to verify the accuracy of the neural network fitting; S23: Select "Numerical Matrix" in MATLAB to import data; and also import the data of each experimental point outside the hollow fiber membrane neural network learning data range into the workspace; S24: Select the neural network fitting module in MATLAB, import the neural network learning data from the workspace, and set the prediction variables; S25: Divide the data into training set, validation set and test set; S26: Start training.

Images