Intelligent fire detection method based on high-temperature cabin area of high-speed aircraft

Abstract

The application belongs to the field of aircraft fire protection system design, and particularly relates to a fire intelligent detection method based on high-temperature cabin area of high-speed aircraft. The detector adopts a split design, the high-temperature-resistant detection part is arranged in the fire area to collect cabin fire information, and the information is transmitted to the controller through the transmission part. The controller is arranged in a relatively low-temperature area, and the transmission part (cable or optical cable) can be isolated from the high temperature through a high-temperature-resistant coating to reduce the influence of high temperature on signal transmission. According to the structured segmented fire area environment, combined with the temperature field, flow field, pressure field distribution and fire development characteristics, the fire in each cabin area is detected through a combined detection method (collecting temperature, flame light, combustion gas composition, image and other information streams), and the fire data collected by the detectors in each cabin area is centrally processed through the controller. The state of the fire is simulated and reconstructed through an algorithm, the development trend is predicted, and the fire is accurately judged.

Description

Technical Field

[0001] This application belongs to the field of aircraft fire protection system design, and specifically relates to an intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft. Background Technology

[0002] Aircraft fires are among the most serious safety incidents during aircraft use and maintenance. Fires in areas such as the engine compartment, equipment compartment, and landing gear compartment are significant contributing factors. These compartments are confined spaces, and leaks in fuel lines can easily ignite upon contact with heat sources, spreading rapidly and causing damage to onboard systems with severe consequences. As new-generation aircraft evolve towards high speed and stealth capabilities, the power of electrical equipment, hydraulic drives, and aerodynamic heat have increased dramatically. Different compartments within an aircraft exhibit varying temperatures, pressures, and ventilation volumes, with some areas (such as the engine compartment) reaching temperatures as high as 600°C. Fires in different compartments develop and progress in different ways (smoke, flames, temperature changes, etc.), and these factors change in real time after a fire begins. Therefore, fire prevention systems must quickly and accurately detect and precisely control the fire in each area under these complex and variable conditions.

[0003] Traditional aircraft fire detection systems only target temperature or flame light, using aerodynamic or integrated flame detectors for fire monitoring. These two detection methods cannot meet the fire detection and control requirements of high-speed aircraft in wide temperature range cabins. The specific reasons are as follows:

[0004] 1) Areas of high-speed aircraft prone to fire (such as engine compartments) have high temperatures, reaching up to 600°C. Pneumatic or integrated flame detectors are used to detect fires, but these two types of detectors have poor performance under high-temperature conditions and cannot meet the fire detection requirements of wide temperature range compartments of high-speed aircraft.

[0005] 2) New-generation high-speed aircraft have more high-temperature compartments, increasing the areas prone to fire and the volume of the fire zone. According to current traditional fire suppression system design methods, the extinguishing agent dosage is designed based on maximum fire suppression requirements. In the event of a fire, all the extinguishing agent is released to bring the entire fire zone to the required concentration; the amount of extinguishing agent carried by the aircraft can even be 6-10 times the required concentration. To control the amount of extinguishing agent carried, the system should have the ability to target the fire zone with a minimum extinguishing agent spray volume. Therefore, the detection system needs to be able to accurately identify the temperature distribution and development trend of the fire zone.

[0006] Therefore, fire detection systems need to be able to accurately identify the temperature field distribution and flame development trend in the fire zone, but existing fire detection methods cannot achieve this goal.

[0007] Therefore, it is essential to conduct research on fire detection methods based on the high-temperature compartments of high-speed aircraft. Summary of the Invention

[0008] The purpose of this application is to provide an intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft, so as to solve the problem of difficulty in accurately identifying the temperature field distribution and flame development trend of the fire zone.

[0009] The technical solution of this application is: a method for intelligent fire detection in the high-temperature compartment of a high-speed aircraft, comprising:

[0010] Acquire the layout and geometric data of each high-temperature compartment of the aircraft, construct a general fire model of the high-temperature compartment of the aircraft, perform CFD simulation calculations, and obtain simulation results;

[0011] Based on the simulation results, the changes in each parameter domain during the fire development process are obtained. Different types of detectors are deployed in the high-temperature zone of the aircraft to collect the changes in each parameter domain during the fire development process. The data is then transmitted to the controller located in the non-high-temperature zone, and the detection algorithm of the controller is designed. Then, the data stream of the detectors is processed according to the changes in each parameter domain to determine the fire data in each high-temperature zone.

[0012] Different fire control strategies were set according to the different fire data of each high-temperature zone, and fire extinguishing tests were carried out. The detection algorithm of the controller was adjusted according to the results of the fire extinguishing tests until the difference between the fire data of the high-temperature zone detected by the detector algorithm and the fire data obtained from the fire extinguishing test was within the set range.

[0013] Preferably, the detectors include detectors for temperature, flame light, combustion gas concentration, and flame images; after different types of data are detected by different types of detectors, they are transmitted to the controller in the form of data streams; the controller simulates and reconstructs the fire distribution in each fire zone and predicts the trend of fire development.

[0014] Preferably, the specific method for reconstructing the fire distribution in each fire zone is as follows:

[0015] The controller is configured with an optimization algorithm to obtain the current ventilation data of the aircraft. Then, the changes in each parameter domain during the fire development process collected by the sensors are input into the generalized fire model. The generalized fire model is then called and the fire situation in each high-temperature zone is determined through the optimization algorithm.

[0016] Preferably, during the fire extinguishing test, the risk level and fuel type are first set. Then, based on the current distribution of the extinguishing agent and the extinguishing agent nozzles, the fuel is ignited to generate a fire of the set risk level, and the fire is extinguished using the extinguishing agent. The changes in each parameter domain collected by detectors of different detection types are input into the CFD model. The fire data of the high-temperature zone output by the CFD model is compared with the actual fire data of the high-temperature zone to determine the difference in fire data and adjust the optimization algorithm. The fire extinguishing test is repeated until the difference in fire data is within the set range.

[0017] Preferably, when performing CFD simulation calculations, a typical mission profile of the current aircraft is obtained, along with the flight altitude, speed, and ambient temperature under the current high-temperature compartment layout, and CFD simulation calculations are carried out based on the geometric structure data.

[0018] Preferably, the simulation results include the development process of fires in different compartments and locations from occurrence to combustion over a certain period of time, and obtain the changes of various parameter domains in the compartments over time after the fire occurs, including the changes of temperature field, light intensity and combustion gas concentration with respect to the parameter domains, as characteristic parameters for judging the fire and predicting the fire trend.

[0019] 1. Preferably, a generalized fire model for high-temperature compartments of aircraft is constructed based on the mass conservation equation, energy conservation equation, and heat conservation equation, specifically as follows:

[0020] The mass conservation equation can be described as follows: Within a unit time, the net mass flowing into a fluid element is equal to the increase in the element's mass; the expression is:

[0021] ;

[0022] In the formula, ρ represents fluid density; u, v, and w represent fluid velocities in different directions; t represents time.

[0023] The energy conservation equation can be described as follows: the rate of increase of energy within a fluid element is equal to the sum of the net heat flow into the element and the work done on the element by the surrounding environment; the expression is:

[0024] ;

[0025] In the formula, The average velocity of the fluid; It is a volume force; P x P y P z These are surface forces in the x, y, and z directions;

[0026] The heat conservation equation is:

[0027] ;

[0028] In the formula, C p is the specific heat capacity at constant pressure of the fluid; k is the heat transfer coefficient of the fluid; 𝑆 𝑇 The heat source within the fluid and the increase in heat due to friction caused by viscosity; T is the temperature.

[0029] Preferably, the CFD simulation calculation uses the RNG k-ε two-equation model, and the k-equation and ε-equation in the RNG k-ε model are as follows:

[0030] ;

[0031] ;

[0032] In the formula, , , , , , , , , , , .

[0033] Preferably, a multi-parameter joint detection model of the confined space of the aircraft is established by using a Bayesian network by inputting specific attributes of the measuring points in the fire scene, and the fire control algorithm of the controller is designed.

[0034] Preferably, the Bayesian network is specifically designed as follows:

[0035] The child node and the parent node are linked by a conditional probability value, namely W. i-j The joint probability value for each node can be expressed by the following formula:

[0036] ;

[0037] In the formula: P(C) represents the joint probability, p j (x i ) is x i From the parent node of node x, we can know that node x i The edge density probability is shown in the following formula:

[0038] ;

[0039] In the formula: x i =k represents the state of the node (0 or 1), i.e., whether a fire has occurred. C = {x1, x2, x3, ..., xk} j} represents j nodes in a Bayesian network; at this time, if x jIf there are m parent nodes, then the corresponding conditional probability table contains 2m conditional probability values;

[0040] The Bayesian rule is defined as follows:

[0041] ;

[0042] In the formula: P(h) represents the prior probability of hypothesis h, P(D|h) represents the conditional probability of D given that hypothesis h is true, P(D) represents the probability of D when the hypothesis is not known to be true, and P(h|D) represents the posterior probability of h.

[0043] The intelligent fire detection method based on the high-temperature cabin area of ​​a high-speed aircraft proposed in this application adopts a split design for the detector. The high-temperature resistant detection part is placed in the fire area to collect fire information inside the cabin. The information is transmitted to the controller through the transmission part. The controller is placed in a relatively low-temperature area. The transmission part (cable or optical cable) can be isolated from the influence of high temperature on signal transmission through a high-temperature resistant coating.

[0044] Based on the structured segmentation of the fire zone environment, combined with the distribution of temperature field, flow field, pressure field and fire development characteristics, the fire situation in each compartment is detected by a combined detection method (collecting information streams such as temperature, flame light, combustion gas composition, and images). The fire data collected by the detectors in each compartment is centrally processed by the controller, and the state of the fire is reconstructed by algorithm simulation and the development trend is predicted to accurately determine the fire situation. Attached Figure Description

[0045] To more clearly illustrate the technical solutions provided in this application, the accompanying drawings will be briefly described below. Obviously, the drawings described below are merely some embodiments of this application.

[0046] Figure 1 This is a schematic diagram of the overall process of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] A smart fire detection method based on the high-temperature compartment of a high-speed aircraft, such as Figure 1 As shown, it includes the following steps:

[0049] Step S100: Obtain the layout and geometric structure data of each high-temperature compartment of the aircraft, construct a general fire model of the high-temperature compartment of the aircraft, perform CFD simulation calculations, and obtain simulation results.

[0050] The generalized fire model includes temperature field, flow field, pressure-turbulence model, and combustion model.

[0051] Preferably, when performing CFD simulation calculations, a typical mission profile of the current aircraft is obtained, and the changes in multiple parameters such as flight altitude, speed and ambient temperature under the current high-temperature compartment layout are obtained, and CFD simulation calculations are carried out based on the geometric structure data.

[0052] The simulation results include the development process of fires in different compartments and locations from occurrence to combustion over a certain period of time, and obtain the changes of various parameter domains in the compartments over time after the fire occurs, including the changes of temperature field, light intensity and combustion gas concentration, etc., as characteristic parameters for judging the fire and predicting the fire trend.

[0053] The causes and processes of fires in the high-temperature compartments of aircraft were analyzed. Based on the above CFD simulation results, the fire development process was simulated and calculated.

[0054] Preferably, a generalized fire model for high-temperature compartments of aircraft is constructed based on the mass conservation equation, energy conservation equation, and heat conservation equation, specifically as follows:

[0055] The law of conservation of mass can be described as follows: Within a unit time, the net mass flowing into a fluid element is equal to the increase in the element's mass. The expression is:

[0056] (1)

[0057] In the formula, denoted as fluid density; u, v, and w represent fluid velocities in different directions; and t represents time.

[0058] The energy conservation equation can be described as follows: the rate of increase of energy within a fluid element is equal to the sum of the net heat flow into the element and the work done on the element by the surrounding environment. The expression is:

[0059] (2)

[0060] In the formula, The average velocity of the fluid; It is a volume force; P x P y P z These are the surface forces in the x, y, and z directions.

[0061] The heat conservation equation is:

[0062] (3)

[0063] In the formula, C p is the specific heat capacity at constant pressure of the fluid; k is the heat transfer coefficient of the fluid; 𝑆 𝑇 The heat source within the fluid and the increase in heat due to friction caused by viscosity; T is the temperature.

[0064] Preferably, the CFD simulation uses the RNG k-ε two-equation model, and the standard wall function is selected. Currently, the two-equation model is the most widely used in engineering, and the most basic two-equation model is the standard k-ε model. The RNG k-ε model is derived from the Standard k-ε model. The RNG k-ε model incorporates the effects of small scales into large-scale motion and the modified viscous terms, thus removing small-scale motion systems from the governing equations. The k (turbulent kinetic energy) equation and ε (dissipation rate) equation in the RNG k-ε model are as follows:

[0065] (4)

[0066] (5)

[0067] In the formula, , , , , , , , , , , .

[0068] The RNG k-ε model, by correcting for turbulent viscosity, can simulate rotating and swirling flows in time-averaged flow; the dissipation rate equation considers the mainstream time-averaged strain rate E. ij Therefore, the generation terms of the RNG k-ε model are not only related to the flow but also become functions of the flow field coordinates. For near-wall flows and flows with low Reynolds numbers, wall functions are used for processing.

[0069] The causes and processes of fires in high-temperature compartments of aircraft were analyzed. Based on the aforementioned CFD simulation results, the fire development process was simulated. The simulation calculated the development process of fires in different compartments and locations from occurrence to combustion over a certain period of time, and obtained the changes of various parameter domains within the compartments over time after the fire occurred, such as the changes in temperature field, light intensity, and combustion gas concentration. These parameters serve as characteristic parameters for judging the fire situation and predicting fire trends.

[0070] Step S200: Based on the simulation results, obtain the changes in each parameter domain during the fire development process, deploy detectors of different detection types in the high-temperature zone of the aircraft, collect the changes in each parameter domain during the fire development process, transmit the data to the controller located in the non-high-temperature zone, and carry out the detection algorithm design of the controller; then, based on the changes in each parameter domain, process the data stream of the detectors to determine the fire data in each high-temperature zone.

[0071] Preferably, the detectors include detectors for temperature, flame light, combustion gas concentration, and flame images; after different types of data are detected by different types of detectors, they are transmitted to the controller in the form of data streams; the controller simulates and reconstructs the fire distribution in each fire zone and predicts the trend of fire development.

[0072] Preferably, the specific method for reconstructing the fire distribution in each fire zone is as follows:

[0073] The controller is configured with an optimization algorithm to obtain the current ventilation data of the aircraft. Then, the changes in each parameter domain during the fire development process collected by the sensors are input into the generalized fire model. The generalized fire model is then called and the fire situation in each high-temperature zone is determined through the optimization algorithm.

[0074] To meet the usage requirements of high-temperature chambers, the detector adopts a split design, with the high-temperature resistant detection part placed in the fire zone to collect fire information inside the chamber. This information is then transmitted to the controller via the transmission part, which is located in a relatively low-temperature area. The transmission part (cable or optical cable) can be isolated from the influence of high temperature on signal transmission through a high-temperature resistant coating.

[0075] Traditional neural network algorithms used for fire detection cannot accurately describe the correlations between various parameters or the causal relationships between model parameters. Bayesian networks, on the other hand, are, firstly, a probabilistic knowledge representation and reasoning model that visualizes multivariate knowledge graphs; secondly, they can effectively represent and fuse multi-source information, incorporating various information related to fault diagnosis and maintenance decisions into the network structure, processing them uniformly at the node level, and connecting the nodes.

[0076] Therefore, by inputting specific attributes of measuring points in the fire scene, a multi-parameter joint detection model of the confined space of the aircraft can be established using a Bayesian network, and the fire control algorithm of the controller can be designed to achieve accurate prediction of the fire situation in each high-temperature compartment of the aircraft.

[0077] Preferably, the Bayesian network is specifically designed as follows:

[0078] A Bayesian network is a probabilistic model that combines probability theory and graph theory. A Bayesian network consists of a directed acyclic graph and a set of conditional probabilities. Child nodes and parent nodes are linked by conditional probability values, i.e., W. i-j The joint probability value for each node can be expressed by the following formula:

[0079] (6)

[0080] In the formula: P(C) represents the joint probability, p j (x i ) is x i From the parent node of node x, we can know that node x i The edge density probability is shown in the following formula:

[0081] (7)

[0082] In the formula: x i =k represents the state of the node (0 or 1), i.e., whether a fire has occurred. C = {x1, x2, x3, ..., xk} j} represents j nodes in a Bayesian network. At this point, if x... j If there are m parent nodes, then the corresponding conditional probability table contains 2m conditional probability values.

[0083] Bayes' rule provides a method for calculating the probability of a hypothesis. It is based on the prior probability of the hypothesis, the probability of observing different data given the hypothesis, and the prior probability of the observation probability itself. The definition of Bayes' rule is as follows:

[0084] (8)

[0085] In the formula: P(h) represents the prior probability of hypothesis h, P(D|h) represents the conditional probability of D given that hypothesis h is true, P(D) represents the probability of D when the hypothesis is not known to be true, and P(h|D) represents the posterior probability of h.

[0086] Step S300: Set different fire control strategies according to different fire data in each high-temperature zone, conduct fire extinguishing tests, and adjust the controller's detection algorithm according to the fire extinguishing test results until the difference between the fire data in the high-temperature zone detected by the detector's detection algorithm and the fire data obtained from the fire extinguishing test is within the set range.

[0087] Preferably, during the fire extinguishing test, the risk level and fuel type are first set. Then, based on the current distribution of the extinguishing agent and the extinguishing agent nozzles, the fuel is ignited to generate a fire of the set risk level, and the fire is extinguished using the extinguishing agent. The changes in each parameter domain collected by detectors of different detection types are input into the CFD model. The fire data of the high-temperature zone output by the CFD model is compared with the actual fire data of the high-temperature zone to determine the difference in fire data and adjust the optimization algorithm. The fire extinguishing test is repeated until the difference in fire data is within the set range.

[0088] In summary, to meet the usage requirements of high-temperature chambers, this application adopts a split design for the detector, with the high-temperature resistant detection part placed in the fire zone to collect fire information inside the chamber. This information is then transmitted to the controller via the transmission part, which is located in a relatively low-temperature area. The transmission part (cable or optical cable) can be isolated from the influence of high temperature on signal transmission through a high-temperature resistant coating.

[0089] Based on the structured segmentation of the fire zone environment, combined with the distribution of temperature field, flow field, pressure field and fire development characteristics, the fire situation in each compartment is detected by a combined detection method (collecting information streams such as temperature, flame light, combustion gas composition, and images). The fire data collected by the detectors in each compartment is centrally processed by the controller, and the state of the fire is reconstructed by algorithm simulation and the development trend is predicted to accurately determine the fire situation.

[0090] Finally, it should be noted that the accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0091] In conclusion, 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 within the protection scope of the present invention.

Claims

1. A method for intelligent fire detection in the high-temperature compartment of a high-speed aircraft, characterized in that, include: Acquire the layout and geometric data of each high-temperature compartment of the aircraft, construct a general fire model of the high-temperature compartment of the aircraft, perform CFD simulation calculations, and obtain simulation results; Based on the simulation results, the changes in each parameter domain during the fire development process are obtained. Different types of detectors are deployed in the high-temperature zone of the aircraft to collect the changes in each parameter domain during the fire development process. The data is then transmitted to the controller located in the non-high-temperature zone, and the detection algorithm of the controller is designed. Then, the data stream of the detectors is processed according to the changes in each parameter domain to determine the fire data in each high-temperature zone. Different fire control strategies are set according to different fire data in each high-temperature zone, and fire extinguishing tests are carried out. The detection algorithm of the controller is adjusted according to the results of the fire extinguishing tests until the difference between the fire data of the high-temperature zone detected by the detector algorithm and the fire data obtained from the fire extinguishing test is within the set range. A generalized fire model for high-temperature compartments of aircraft is constructed based on the mass conservation equation, energy conservation equation, and heat conservation equation. Specifically: The mass conservation equation can be described as follows: Within a unit time, the net mass flowing into a fluid element is equal to the increase in the element's mass; the expression is: ； In the formula, ρ represents the fluid density; u, v, and w represent the fluid velocities in different directions. t represents time; The energy conservation equation can be described as follows: the rate of increase of energy within a fluid element is equal to the sum of the net heat flow into the element and the work done on the element by the external environment. The expression is: ； In the formula, The average velocity of the fluid; It is a volume force; P x P y P z These are surface forces in the x, y, and z directions; The heat conservation equation is: ； In the formula, C p is the specific heat capacity at constant pressure of the fluid; k is the heat transfer coefficient of the fluid; 𝑆 𝑇 The heat source within the fluid and the increase in heat due to friction caused by viscosity; T is the temperature; The CFD simulation calculation uses the RNG k-ε two-equation model. The k-equation and ε-equation in the RNG k-ε model are as follows: ； ； In the formula, , , , , , , , , , , ; By inputting specific attributes of measuring points in the fire scene, a multi-parameter joint detection model of the confined space of the aircraft is established using a Bayesian network, and the fire control algorithm of the controller is designed. The Bayesian network is specifically designed as follows: The child node and the parent node are linked by a conditional probability value, namely W. i-j The joint probability value for each node can be expressed by the following formula: ； In the formula: P(C) represents the joint probability, p j (x i ) is x i From the parent node of node x, we can know that node x i The edge density probability is shown in the following formula: ； In the formula: x i =k represents the state of the node (0 or 1), i.e., whether a fire has occurred. C = {x1, x2, x3, ..., xk} j } represents j nodes in a Bayesian network; at this time, if x j If there are m parent nodes, then the corresponding conditional probability table contains 2m conditional probability values; The Bayesian rule is defined as follows: ； In the formula: P(h) represents the prior probability of hypothesis h, P(D|h) represents the conditional probability of D given that hypothesis h is true, P(D) represents the probability of D when the hypothesis is not known to be true, and P(h|D) represents the posterior probability of h.

2. The intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft as described in claim 1, characterized in that: The detectors include detectors for temperature, flame light, combustion gas concentration, and flame images. After different types of data are detected by different types of detectors, they are transmitted to the controller in the form of data streams. The controller simulates and reconstructs the fire distribution in each fire zone and predicts the trend of fire development.

3. The intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft as described in claim 2, characterized in that, The specific method for reconstructing the fire distribution in each fire zone is as follows: The controller is configured with an optimization algorithm to obtain the current ventilation data of the aircraft. Then, the changes in each parameter domain during the fire development process collected by the sensors are input into the generalized fire model. The generalized fire model is then called and the fire situation in each high-temperature zone is determined through the optimization algorithm.

4. The intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft as described in claim 3, characterized in that: During the fire extinguishing test, the risk level and fuel type are first set. Then, based on the current distribution of the extinguishing agent and extinguishing agent nozzles, the fuel is ignited to generate a fire of the set risk level, and the fire is extinguished using the extinguishing agent. The changes in each parameter domain collected by detectors of different detection types are input into the CFD model. The fire data of the high-temperature zone output by the CFD model is compared with the actual fire data of the high-temperature zone to determine the difference in fire data and adjust and optimize the algorithm. The fire extinguishing test is repeated until the difference in fire data is within the set range.

5. The intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft as described in claim 1, characterized in that: When performing CFD simulation calculations, a typical mission profile of the current aircraft is obtained, along with the flight altitude, speed, and ambient temperature under the current high-temperature compartment layout. CFD simulation calculations are then carried out based on the geometric data.

6. The intelligent fire detection method based on the high-temperature compartment of a high-speed aircraft as described in claim 1, characterized in that: The simulation results include the development process of fires in different compartments and locations from occurrence to combustion over a certain period of time, and obtain the changes of various parameter domains in the compartments over time after the fire occurs, including the changes of temperature field, light intensity and combustion gas concentration with respect to the parameter domains, as characteristic parameters for judging the fire and predicting the fire trend.

Images