Intelligent fire-fighting fire detection and analysis method based on artificial intelligence

Abstract

The invention relates to the technical field of fire-fighting fire detection, and particularly discloses an intelligent fire-fighting fire detection analysis method based on artificial intelligence, and the method comprises the steps: obtaining multi-source data from an internal multi-mode sensor network in real time, inputting the data into a risk assessment model, and dynamically calculating a fire risk index, the real-time monitoring and evaluation mechanism can identify potential fire risks in time, especially risks possibly caused by a lithium battery, so that an early warning mechanism is triggered when a risk index exceeds a set threshold value, and alarm information is quickly sent to a safety management system. According to the method, fire accidents can be effectively prevented, safety of personnel and goods is protected, man-made misjudgment and delayed reaction can be reduced through an automatic monitoring and alarming system, and safety and reliability of ship operation are improved. Precise monitoring and quick response to each new energy automobile are ensured, and the probability of occurrence of fire risks is greatly reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of fire detection and relates to an intelligent fire detection and analysis method based on artificial intelligence. Background Art

[0002] With the popularity of new energy vehicles around the world, fire risk management of new energy vehicles on ro-ro ships is particularly important. As the core power source of new energy vehicles, lithium batteries have the characteristics of high energy density, but they also bring potential fire risks. Once a lithium battery has thermal runaway, it may cause a rapidly spreading fire, threatening the safety of ships and personnel. Therefore, it is of great practical significance and urgency to strengthen the prevention measures of lithium battery fires on ro-ro ships and improve the monitoring and emergency response capabilities.

[0003] However, current technologies still have many defects and drawbacks in monitoring lithium battery fire prevention for new energy vehicles carried on ro-ro ships. First, existing monitoring systems mostly rely on a single sensor type and cannot fully capture the multi-dimensional data of the lithium battery status. The limitations of this single data source can easily lead to misjudgment or omission of fire risks. Secondly, the accuracy and sensitivity of real-time monitoring are insufficient, making it difficult to capture subtle changes inside the lithium battery in a timely manner, especially abnormal signals in the early stages. Third, the early warning mechanisms of existing systems are mostly passive, lacking intelligent risk prediction capabilities, and unable to dynamically adjust risk assessment strategies based on real-time data. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention provides an intelligent fire detection and analysis method based on artificial intelligence to solve the above technical problems.

[0005] In order to achieve the above purpose and other purposes, the technical solution adopted by the present invention is as follows: On the one hand, the present invention provides a smart fire detection and analysis method based on artificial intelligence, which comprises the following steps: Step 1: Obtain the total number of new energy vehicles carried by the ro-ro ship, and number each new energy vehicle in turn, and simultaneously obtain multi-source data in real time from the multimodal sensor network inside the ro-ro ship, the multi-source data including environmental condition data and vehicle status data, and use edge computing technology to pre-process the multi-source data to remove noise and outliers; Step 2: Input multi-source data into the risk assessment model to dynamically calculate the fire risk index of each new energy vehicle carried by the ro-ro ship; Step 3: If the fire risk index of a new energy vehicle is greater than the set fire risk threshold, the early warning mechanism is triggered and an alarm message is sent to the ro-ro ship safety management system.

[0006] Exemplarily, the environmental condition data in the multi-source data are divided into the corrected temperature, corrected humidity and hull swing amplitude of the ro-ro ship at each prediction time point; The vehicle status data in the multi-source data is divided into the lithium battery temperature of each new energy vehicle corresponding to each prediction time point; the vehicle status data in the multi-source data also includes the average voltage and current of each new energy vehicle.

[0007] Exemplarily, the specific acquisition logic for acquiring environmental condition data from multi-source data is: S3-1, obtaining the shipping route of the ro-ro ship, and then locating the navigation coordinates of the ro-ro ship at each predicted time point; Based on the forecast temperature of the ro-ro ship at each forecast time point And forecast humidity , j is the number of each prediction time point, ranging from 1 to J, and J is the total number of prediction time points in the prediction time period; Get the corrected temperature of the ro-ro ship at each predicted time point , is the temperature correction term corresponding to the wind speed of the RoRo ship at the jth prediction time point, is the temperature correction term corresponding to the ocean temperature of the ro-ro ship at the jth prediction time point; in , is the reference coefficient of the influence of unit wind speed on temperature, is the wind speed at the jth prediction time point, is the angle between the RoRo ship’s heading and the wind direction at the jth prediction time point; , are the reference coefficients of the influence of unit current velocity and unit sea surface temperature on temperature, They represent the ocean current speed and sea surface temperature corresponding to the navigation coordinates of the RoRo ship at the jth prediction time point; S3-2. Using the calculation formula , calculate the corrected humidity of the ro-ro ship at each predicted time point ,in, The reference coefficient that represents the effect of unit wind speed on humidity. is the reference coefficient of the effect of unit sea surface temperature difference on humidity; S3-3. Through calculation model , the analysis results show that the hull swing amplitude of the ro-ro ship at each predicted time point ;in are the wind influence coefficient and the ocean current influence coefficient, is the angle between the RoRo ship heading and the ocean current direction at the jth prediction time point, f(j) is the correction function based on the navigation coordinates at the jth prediction time point, is the correction value of the hull swing amplitude corresponding to the j-th prediction time point.

[0008] Exemplarily, the specific acquisition logic for acquiring vehicle status data from multi-source data is: S4-1, summing up and averaging the voltage and current values ​​of each new energy vehicle at each current time point in the current monitoring period, and using the calculation results as the voltage and current averages of each new energy vehicle; S4-2. Obtain the temperature of the lithium battery of each new energy vehicle at the last current time point in the current monitoring period, and record it as the initial temperature of each new energy vehicle , h is the serial number of each new energy vehicle; The time step between the last current time point and the first predicted time point in the current monitoring period is taken as the first time step, and the lithium battery temperature of each new energy vehicle in the first time step is calculated. ,in is the number of seconds corresponding to the first time step of the h-th new energy vehicle, is the battery thermal capacity of the hth new energy vehicle, They represent the internal heat and environmental heat dissipation of the h-th new energy vehicle respectively. The specific calculation formula is as follows: ; ; In the above formula They represent the mean battery current and battery internal resistance of the h-th new energy vehicle, They represent the convective heat transfer coefficient and battery surface area of ​​the hth new energy vehicle, is the corrected temperature of the ro-ro ship at the first prediction time point; The time step between the first prediction time point and the second prediction time point is recorded as the second time step, and the lithium battery temperature of each new energy vehicle in the first time step is then used as the initial temperature of each new energy vehicle, and the lithium battery temperature of each new energy vehicle in the second time step is calculated in the same way as the calculation method of the lithium battery temperature of each new energy vehicle in the first time step; Until the lithium battery temperature of each new energy vehicle in the last time step is calculated; The lithium battery temperature of each new energy vehicle in the first time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the first prediction time point, the lithium battery temperature of each new energy vehicle in the second time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the second prediction time point, ... the lithium battery temperature of each new energy vehicle in the last time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the last prediction time point; In this way, the lithium battery temperature of each new energy vehicle corresponding to each predicted time point is obtained.

[0009] For example, the fire risk index of each new energy vehicle carried by a ro-ro ship is dynamically calculated, and the specific calculation process is as follows: Through risk assessment model , dynamically calculate the fire risk index of each new energy vehicle carried by ro-ro ships , τ1, τ2, τ3 and τ4 represent the calculation weight factors corresponding to the temperature risk coefficient, electrical state risk coefficient, thermal runaway risk coefficient and external stress risk coefficient, respectively, and satisfy τ1+τ2+τ3+τ4=1; In the above risk assessment model They respectively represent the temperature risk coefficient, electrical state risk coefficient and thermal runaway risk coefficient of the lithium battery corresponding to the h-th new energy vehicle, and φ is the external stress risk coefficient of the roll-on / roll-off ship.

[0010] Exemplarily, the temperature risk coefficient calculation process of the lithium battery corresponding to each new energy vehicle is as follows: The corrected temperature of the ro-ro ship at each predicted time point is weighted averaged to obtain the corrected temperature mean of the ro-ro ship during the predicted time period. Similarly, the weighted average temperature of the lithium battery of each new energy vehicle corresponding to each predicted time point is calculated to obtain the average temperature of the lithium battery of each new energy vehicle corresponding to the predicted time period. ; The temperature risk coefficient of lithium batteries corresponding to each new energy vehicle is calculated from this , β1 and β2 represent the set adjustment coefficients, is the critical temperature of the lithium battery corresponding to the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, and ζ2 is the ambient temperature change rate: , ; h is the serial number of each new energy vehicle, h=1,2...K, K is the total number of new energy vehicles, They represent the lithium battery temperature of the h-th new energy vehicle at the j+1-th and j-th prediction time points, respectively. Respectively represent the time of the j+1th and jth prediction time points, is the corrected temperature at the j+1th prediction time point.

[0011] For example, the calculation process of the external stress risk factor of a ro-ro ship is as follows: The hull swing amplitude of the ro-ro ship at each predicted time point Perform weighted average calculation to obtain the mean hull swing amplitude of the ro-ro ship during the predicted time period ; Determine the external stress value of the ro-ro ship at each predicted time point , and perform weighted average calculation on it in the same way to obtain the mean external stress of the ro-ro ship in the prediction time period ; The external stress risk coefficient of the ro-ro ship is calculated , where e is a natural constant.

[0012] Exemplarily, the calculation process of the electrical state risk coefficient of the lithium battery corresponding to each new energy vehicle is as follows: The weighted average calculation of the corrected humidity of the ro-ro ship at each prediction time point is performed to obtain the corrected humidity mean value of the ro-ro ship during the prediction period. ; Then, by analyzing the formula , get the electrical state risk coefficient of the lithium battery corresponding to each new energy vehicle ,in They represent the battery current mean, voltage homogeneity and battery internal resistance of the h-th new energy vehicle, is the stowage spacing of lithium batteries corresponding to the hth new energy vehicle, β5 and β6 represent the set adjustment coefficients, which are used to adjust the influence of capacity and internal resistance. Represents the total battery capacity of the h-th new energy vehicle.

[0013] Exemplarily, the thermal runaway risk coefficient calculation process of the lithium battery corresponding to each new energy vehicle is as follows: , where exp(·) represents an exponential function with the natural constant e as the base, β7 and β8 represent the set adjustment coefficients, which are used to adjust the temperature change rate and the influence of heat release; They represent the heat released by the lithium battery corresponding to the h-th new energy vehicle and the total heat capacity of the lithium battery; Set a safe temperature for the lithium battery of the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, It is the average temperature of lithium batteries of each new energy vehicle in the corresponding prediction time period.

[0014] Another aspect of the present invention provides an intelligent fire detection and analysis device based on artificial intelligence, comprising a processor, a memory and a communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it is executed to implement an intelligent fire detection and analysis method based on artificial intelligence as described in any one of the present inventions.

[0015] As described above, the present invention provides an intelligent fire detection and analysis method based on artificial intelligence, which has at least the following beneficial effects: The present invention provides a kind of intelligent fire detection and analysis method based on artificial intelligence, which obtains the total number of new energy vehicles carried by roll-on / roll-off ships, numbers each vehicle in turn, and then obtains multi-source data from the internal multimodal sensor network in real time, and inputs these data into the risk assessment model to dynamically calculate the fire risk index. This method can significantly improve the efficiency and accuracy of ship safety management. Specifically, this real-time monitoring and evaluation mechanism can timely identify potential fire risks, especially the risks that may be caused by lithium batteries, so that when the risk index exceeds the set threshold, the early warning mechanism is triggered and the alarm information is quickly sent to the safety management system. This approach can not only effectively prevent the occurrence of fire accidents and protect the safety of personnel and cargo, but also reduce human misjudgment and delayed response through automated monitoring and alarm systems, and improve the safety and reliability of ship operations. By solving specific technical problems, this system ensures accurate monitoring and rapid response to each new energy vehicle, greatly reducing the probability of fire risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 It is a schematic diagram of the connection of each step of the method of the present invention. DETAILED DESCRIPTION

[0018] The above contents in combination with the implementation of the present invention are merely examples and explanations of the concept of the present invention. The technical personnel in the relevant technical field may make various modifications or supplements to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they shall all fall within the protection scope of the present invention.

[0019] Example 1 See also Figure 1 As shown, a smart fire detection and analysis method based on artificial intelligence includes the following steps: Step 1: Obtain the total number of new energy vehicles carried by the ro-ro ship, and number each new energy vehicle in turn, and simultaneously obtain multi-source data in real time from the multimodal sensor network inside the ro-ro ship, the multi-source data including environmental condition data and vehicle status data, and use edge computing technology to pre-process the multi-source data to remove noise and outliers; In the preferred technical solution of the present application, the environmental condition data in the multi-source data are divided into the corrected temperature, corrected humidity and hull swing amplitude of the ro-ro ship at each prediction time point; The vehicle status data in the multi-source data is divided into the lithium battery temperature of each new energy vehicle corresponding to each prediction time point; the vehicle status data in the multi-source data also includes the average voltage and current of each new energy vehicle.

[0020] In the preferred technical solution of the present application, the specific acquisition logic for obtaining environmental condition data from multi-source data is: S3-1, obtaining the shipping route of the ro-ro ship, and then locating the navigation coordinates of the ro-ro ship at each predicted time point; Based on the forecast temperature of the ro-ro ship at each forecast time point And forecast humidity , j is the number of each prediction time point, ranging from 1 to J, and J is the total number of prediction time points in the prediction time period; Get the corrected temperature of the ro-ro ship at each predicted time point , is the temperature correction term corresponding to the wind speed of the RoRo ship at the jth prediction time point, is the temperature correction term corresponding to the ocean temperature of the ro-ro ship at the jth prediction time point; in , is the reference coefficient of the influence of unit wind speed on temperature, is the wind speed at the jth prediction time point, is the angle between the RoRo ship’s heading and the wind direction at the jth prediction time point; , are the reference coefficients of the influence of unit current velocity and unit sea surface temperature on temperature, They represent the ocean current speed and sea surface temperature corresponding to the navigation coordinates of the RoRo ship at the jth prediction time point; It should be added that as well as All of these can be determined through historical data analysis, using regression analysis or machine learning methods. It reflects the sensitivity of wind speed to temperature changes. They respectively reflect the influence of ocean current speed on temperature change and the influence of sea surface temperature on air temperature; All of them can be obtained through marine monitoring equipment.

[0021] S3-2. Using the calculation formula , calculate the corrected humidity of the ro-ro ship at each predicted time point ,in, The reference coefficient that represents the effect of unit wind speed on humidity. is the reference coefficient of the effect of unit sea surface temperature difference on humidity; and All of these can be determined through historical data analysis, regression analysis or machine learning methods. It reflects the influence of sea surface temperature on humidity changes; The above formulas for correcting temperature and humidity provide fine-tuning of numerical weather forecast data by comprehensively considering multiple factors such as wind speed, wind direction, and ocean conditions; it can more accurately reflect the changes in the real-time environment during the ship's voyage, thereby improving the accuracy and reliability of the forecast. The wind speed and wind direction corrections reflect the impact of wind on heat and humidity transmission, while the ocean condition corrections integrate the effects of ocean currents and sea surface temperature. The introduction of these corrections is necessary because they can capture subtle changes in the local environment, especially in a dynamic and complex environment such as the sea, which helps the safe navigation of ships and effective risk management.

[0022] S3-3. Through calculation model , the analysis results show that the hull swing amplitude of the ro-ro ship at each predicted time point ;in are the wind influence coefficient and the ocean current influence coefficient, is the correction value of the hull swing amplitude corresponding to the j-th prediction time point, specifically: , c2 is the influence of each meter per second on the swing amplitude, the unit is degree / (meter / second); c1 is the influence of each meter on the swing amplitude, the unit is degree / meter; ZX represents the center of gravity height of the ro-ro ship, is the speed of the ro-ro ship corresponding to the j-th prediction time point; is the angle between the RoRo ship heading and the ocean current direction at the jth prediction time point, and f(j) is the correction function based on the navigation coordinates at the jth prediction time point. The specific calculation process is as follows: The navigation coordinates (longitude and latitude) of the RoRo ship at the jth prediction time point are used to extract local environmental conditions, such as wind field, wave field, etc., from the environmental data model; Use spatial interpolation techniques (such as bilinear interpolation and Kriging interpolation) to estimate the environmental conditions of the navigation coordinate points of the RoRo ship at the jth prediction time point; f(j)=α1×wind speed correction+α2×wave correction+α3×current correction; Wind speed correction: Adjust according to the local wind speed data extracted from the coordinates; Wave correction: Adjustment based on local wave data extracted from coordinates; Current correction: Adjust according to the local current data extracted from the coordinates; α1, α2 and α3 are weight coefficients used to adjust the impact of various factors, which are calibrated according to historical data and ship characteristics.

[0023] In the preferred technical solution of the present application, the specific acquisition logic for acquiring vehicle status data from multi-source data is: S4-1, summing up and averaging the voltage and current values ​​of each new energy vehicle at each current time point in the current monitoring period, and using the calculation results as the voltage and current averages of each new energy vehicle; S4-2. Obtain the temperature of the lithium battery of each new energy vehicle at the last current time point in the current monitoring period, and record it as the initial temperature of each new energy vehicle , h is the serial number of each new energy vehicle; The time step between the last current time point and the first predicted time point in the current monitoring period is taken as the first time step, and the lithium battery temperature of each new energy vehicle in the first time step is calculated. ,in is the number of seconds corresponding to the first time step of the h-th new energy vehicle, is the battery thermal capacity of the hth new energy vehicle, They represent the internal heat and environmental heat dissipation of the h-th new energy vehicle respectively. The specific calculation formula is as follows: ; ; In the above formula They represent the mean battery current and battery internal resistance of the h-th new energy vehicle, They represent the convective heat transfer coefficient and battery surface area of ​​the hth new energy vehicle, is the corrected temperature of the ro-ro ship at the first prediction time point; It needs to be explained that the calculation formula of the lithium battery temperature of each new energy vehicle in the first time step is based on the principle of thermal balance, which accurately describes the change in the lithium battery temperature and the internal heat in the first time step. and external heat dissipation By multiplying the temperature change by the time step, the formula combines the net heat input (i.e., internal heat minus external heat dissipation) with the battery's thermal capacity to give the temperature increment. This relationship conforms to the basic principle of thermodynamics, that changes in heat directly affect the temperature of an object; Secondly, the rationality of the formula is reflected in its ability to effectively simulate the temperature changes of the battery during actual use. During transportation, factors such as the battery's charge and discharge state, ambient temperature, and convection heat transfer will affect its temperature. By gradually updating the temperature, the formula can adapt to these dynamic changes, making the model more realistic and practical; Finally, the use of this formula simplifies the complex heat transfer process, making temperature prediction more intuitive and easy to implement. Through discretization, the temperature change of the battery can be quickly calculated within each time step, which is convenient for real-time monitoring and management.

[0024] The time step between the first prediction time point and the second prediction time point is recorded as the second time step, and the lithium battery temperature of each new energy vehicle in the first time step is then used as the initial temperature of each new energy vehicle, and the lithium battery temperature of each new energy vehicle in the second time step is calculated in the same way as the calculation method of the lithium battery temperature of each new energy vehicle in the first time step; Until the lithium battery temperature of each new energy vehicle in the last time step is calculated; The lithium battery temperature of each new energy vehicle in the first time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the first prediction time point, the lithium battery temperature of each new energy vehicle in the second time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the second prediction time point, ... the lithium battery temperature of each new energy vehicle in the last time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the last prediction time point; In this way, the lithium battery temperature of each new energy vehicle corresponding to each predicted time point is obtained.

[0025] Step 2: Input multi-source data into the risk assessment model to dynamically calculate the fire risk index of each new energy vehicle carried by the ro-ro ship; In the preferred technical solution of this application, the fire risk index of each new energy vehicle carried by the ro-ro ship is dynamically calculated, and the specific calculation process is: Through risk assessment model , dynamically calculate the fire risk index of each new energy vehicle carried by ro-ro ships , τ1, τ2, τ3 and τ4 represent the calculation weight factors corresponding to the temperature risk coefficient, electrical state risk coefficient, thermal runaway risk coefficient and external stress risk coefficient, respectively, and satisfy τ1+τ2+τ3+τ4=1; In the above risk assessment model They respectively represent the temperature risk coefficient, electrical state risk coefficient and thermal runaway risk coefficient of the lithium battery corresponding to the h-th new energy vehicle, and φ is the external stress risk coefficient of the roll-on / roll-off ship.

[0026] In the preferred technical solution of this application, the temperature risk coefficient calculation process of the lithium battery corresponding to each new energy vehicle is as follows: The corrected temperature of the ro-ro ship at each predicted time point is weighted averaged to obtain the corrected temperature mean of the ro-ro ship during the predicted time period. Similarly, the weighted average temperature of the lithium battery of each new energy vehicle corresponding to each predicted time point is calculated to obtain the average temperature of the lithium battery of each new energy vehicle corresponding to the predicted time period. ; The temperature risk coefficient of lithium batteries corresponding to each new energy vehicle is calculated from this , β1 and β2 represent the set adjustment coefficients, is the critical temperature of the lithium battery corresponding to the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, and ζ2 is the ambient temperature change rate: , ; h is the serial number of each new energy vehicle, h=1,2...K, K is the total number of new energy vehicles, They represent the lithium battery temperature of the h-th new energy vehicle at the j+1-th and j-th prediction time points, respectively. Respectively represent the time of the j+1th and jth prediction time points, is the corrected temperature at the j+1th prediction time point.

[0027] It should be added that the above formula provides a more comprehensive risk assessment method by considering the dynamic changes of lithium battery and ambient temperature, as well as the rate of temperature change. Each term in the formula is standardized and dimensionless (such as using the (tanh) function) to ensure unit consistency. Through time-weighted averaging, the formula can capture the trend of temperature changes over time, which is crucial for thermal management and safety assessment of lithium batteries. The adjustment coefficients β1 and β2 allow optimization based on actual data to reflect the impact of the rate of temperature change on risk. This method not only improves the accuracy of risk assessment, but also enhances the sensitivity to rapid temperature changes, and is suitable for risk monitoring in dynamic environments.

[0028] In the preferred technical solution of this application, the calculation process of the external stress risk coefficient of the ro-ro ship is as follows: The hull swing amplitude of the ro-ro ship at each predicted time point Perform weighted average calculation to obtain the mean hull swing amplitude of the ro-ro ship during the predicted time period ; Determine the external stress value of the ro-ro ship at each predicted time point ; ω1, ω2 and ω3 represent the set weight coefficients respectively, and satisfy ω1+ω2+ω3=1; is the wind force at the jth prediction time point, , ρ is the air density, A1 is the windward area of ​​the ro-ro ship, is the wind speed at the jth prediction time point; is the sea wave at the j-th prediction time point, , d is the empirical coefficient, ZQ is the wave period, which indicates the time required for the wave to reach the next peak from one crest. The wave period is an important parameter to measure the characteristics of waves and affects the force of waves on ships; is the wave height at the j-th prediction time point; is the water flow at the jth prediction time point, , A2 is the water receiving area of ​​the ro-ro ship’s hull, is the water density, is the water flow velocity at the jth prediction time point; The weighted average calculation is performed in the same way to obtain the mean external stress of the ro-ro ship in the predicted time period. ; The external stress risk coefficient of the ro-ro ship is calculated , where e is a natural constant.

[0029] In the preferred technical solution of this application, the calculation process of the electrical state risk coefficient of the lithium battery corresponding to each new energy vehicle is as follows: The weighted average calculation of the corrected humidity of the ro-ro ship at each prediction time point is performed to obtain the corrected humidity mean value of the ro-ro ship during the prediction period. ; Then, by analyzing the formula , get the electrical state risk coefficient of the lithium battery corresponding to each new energy vehicle ,in They represent the battery current mean, voltage homogeneity and battery internal resistance of the h-th new energy vehicle, is the stowage spacing of lithium batteries corresponding to the hth new energy vehicle, β5 and β6 represent the set adjustment coefficients, which are used to adjust the influence of capacity and internal resistance. Represents the total battery capacity of the h-th new energy vehicle.

[0030] It should be added that all parameters have been converted into dimensionless form before calculating the electrical status risk coefficient of each new energy vehicle. The first item in the above calculation formula describes the influence of voltage and current under humidity and loading spacing conditions. The second item in the formula takes into account the influence of battery capacity on risk. Small capacity may be more prone to overload. The third item in the formula takes into account the influence of battery internal resistance on risk. Large internal resistance may lead to higher heat generation.

[0031] The above formula provides a more comprehensive and dynamic method for fire risk assessment of new energy vehicles carried by ro-ro ships by comprehensively considering multiple factors such as voltage, current, humidity, stowage spacing, battery capacity and internal resistance. By introducing the influence of battery capacity and internal resistance, the formula can more accurately reflect the true state of the battery under different conditions. Traditional risk assessment methods may only focus on voltage and current, while ignoring the impact of changes in battery capacity and internal resistance on heat generation and risks. When the battery capacity is small, overload is more likely to occur, increasing the risk of thermal runaway; and larger internal resistance may lead to more heat accumulation. These factors are reflected in the formula through exponential and logarithmic functions, making risk assessment more targeted; Secondly, the consideration of humidity and stowage spacing in the formula reflects the complex impact of environmental conditions on electrical status. Changes in humidity may affect the heat dissipation of the battery, while stowage spacing directly affects the conduction and diffusion of heat. By incorporating these factors into the calculation, it is possible to better predict the changes in risks under different environmental conditions. This approach can help ship operators take appropriate precautions in different navigation environments, thereby improving safety; Furthermore, the nonlinear structure of the formula enables it to capture the complex interactions between various factors. The traditional weighted summation method often assumes that the factors are linearly independent, but in reality, there may be complex interactions between the factors. For example, changes in voltage and current may be affected by humidity and stowage spacing at the same time, and the weights of these effects may be different under different conditions. By using nonlinear functions such as exponentials and logarithms, the formula can be more flexible to adapt to different risk scenarios; In addition, by acquiring data on battery status and environmental conditions in real time, the formula can instantly calculate the current fire risk index. This is particularly important for risk management in a dynamic environment such as a ro-ro vessel. Real-time risk assessment can help crew members identify potential risks in a timely manner and take appropriate measures, such as adjusting stowage spacing, changing sailing routes or adjusting the ship's ventilation system, thereby effectively preventing fire accidents.

[0032] In the preferred technical solution of this application, the thermal runaway risk coefficient calculation process of the lithium battery corresponding to each new energy vehicle is as follows: , where exp(·) represents an exponential function with the natural constant e as the base, β7 and β8 represent the set adjustment coefficients, which are used to adjust the temperature change rate and the influence of heat release; They represent the heat released by the lithium battery corresponding to the h-th new energy vehicle and the total heat capacity of the lithium battery; Set a safe temperature for the lithium battery of the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, It is the average temperature of lithium batteries of each new energy vehicle in the corresponding prediction time period.

[0033] It should be added that all parameters have been converted into dimensionless form before calculating the thermal runaway risk coefficient of each new energy vehicle. The first item in the above calculation formula describes the increased risk when the temperature approaches the critical point. The second item in the formula takes into account the impact of the temperature change rate on the risk. Rapid temperature rise may indicate thermal runaway. The third item in the formula takes into account the relationship between heat release and battery thermal capacity. Excessive heat release may lead to thermal runaway.

[0034] The above thermal runaway risk formula provides a method for dynamically evaluating the risk of thermal runaway of batteries by comprehensively considering battery temperature, temperature change rate, heat release and battery thermal capacity. Compared with the existing technology, the advantage of this formula is that it can reflect the changes in battery status in real time, especially in the case of rapid temperature rise or heat accumulation, and provide early warning of potential risks. The necessity of this method lies in that it can more accurately capture the key factors that lead to thermal runaway and reduce the problem of missed judgments in traditional methods due to single parameter monitoring. Through this multi-parameter integrated evaluation, the defects of the existing technology in insufficient adaptability to complex dynamic environments can be effectively solved, and the ability to predict and manage the safety of lithium batteries can be improved.

[0035] Step 3: If the fire risk index of a new energy vehicle is greater than the set fire risk threshold, the early warning mechanism is triggered and an alarm message is sent to the ro-ro ship safety management system.

[0036] Example 2 An intelligent fire detection and analysis device based on artificial intelligence, comprising a processor, a memory and a communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it is executed to implement an intelligent fire detection and analysis method based on artificial intelligence as described in any one of the present inventions.

[0037] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0038] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0039] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0040] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A fire detection and analysis method based on artificial intelligence, characterized in that: include: Obtain the total number of new energy vehicles carried by the ro-ro ship, number each new energy vehicle in turn, and simultaneously obtain multi-source data in real time from the multimodal sensor network inside the ro-ro ship. The multi-source data includes environmental condition data and vehicle status data, and use edge computing technology to pre-process the multi-source data to remove noise and outliers; Input multi-source data into the risk assessment model to dynamically calculate the fire risk index of each new energy vehicle carried by a ro-ro ship; If the fire risk index of a new energy vehicle is greater than the set fire risk threshold, the early warning mechanism will be triggered and an alarm message will be sent to the roll-on / roll-off ship safety management system.

2. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 1 is characterized in that: The environmental condition data in the multi-source data are divided into the corrected temperature, corrected humidity and hull swing amplitude of the ro-ro ship at each prediction time point; The vehicle status data in the multi-source data is divided into the lithium battery temperature of each new energy vehicle corresponding to each prediction time point; the vehicle status data in the multi-source data also includes the average voltage and current of each new energy vehicle.

3. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 2 is characterized in that: The specific acquisition logic for obtaining environmental condition data from multi-source data is: S3-1, obtaining the shipping route of the ro-ro ship, and then locating the navigation coordinates of the ro-ro ship at each predicted time point; Based on the forecast temperature of the ro-ro ship at each forecast time point And forecast humidity , j is the number of each prediction time point, ranging from 1 to J, and J is the total number of prediction time points in the prediction time period; Get the corrected temperature of the ro-ro ship at each predicted time point , is the temperature correction term corresponding to the wind speed of the RoRo ship at the jth prediction time point, is the temperature correction term corresponding to the ocean temperature of the ro-ro ship at the jth prediction time point; in , is the reference coefficient of the influence of unit wind speed on temperature, is the wind speed at the jth prediction time point, is the angle between the RoRo ship’s heading and the wind direction at the jth prediction time point; , are the reference coefficients of the influence of unit current velocity and unit sea surface temperature on temperature, They represent the ocean current speed and sea surface temperature corresponding to the navigation coordinates of the RoRo ship at the jth prediction time point; S3-2. Using the calculation formula , calculate the corrected humidity of the ro-ro ship at each predicted time point ,in, The reference coefficient that represents the effect of unit wind speed on humidity. is the reference coefficient of the effect of unit sea surface temperature difference on humidity; S3-3. Through calculation model , the analysis results show that the hull swing amplitude of the ro-ro ship at each predicted time point ;in are the wind influence coefficient and the ocean current influence coefficient, is the angle between the RoRo ship heading and the ocean current direction at the jth prediction time point, f(j) is the correction function based on the navigation coordinates at the jth prediction time point, is the correction value of the hull swing amplitude corresponding to the j-th prediction time point.

4. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 1 is characterized in that: The specific acquisition logic for obtaining vehicle status data from multi-source data is: S4-1, summing up and averaging the voltage and current values ​​of each new energy vehicle at each current time point in the current monitoring period, and using the calculation results as the voltage and current averages of each new energy vehicle; S4-2. Obtain the temperature of the lithium battery of each new energy vehicle at the last current time point in the current monitoring period, and record it as the initial temperature of each new energy vehicle , h is the serial number of each new energy vehicle; The time step between the last current time point and the first predicted time point in the current monitoring period is taken as the first time step, and the lithium battery temperature of each new energy vehicle in the first time step is calculated. ,in is the number of seconds corresponding to the first time step of the h-th new energy vehicle, is the battery thermal capacity of the hth new energy vehicle, They represent the internal heat and environmental heat dissipation of the h-th new energy vehicle respectively. The specific calculation formula is as follows: ； ； In the above formula They represent the mean battery current and battery internal resistance of the h-th new energy vehicle, They represent the convective heat transfer coefficient and battery surface area of ​​the hth new energy vehicle, is the corrected temperature of the ro-ro ship at the first prediction time point; The time step between the first prediction time point and the second prediction time point is recorded as the second time step, and the lithium battery temperature of each new energy vehicle in the first time step is then used as the initial temperature of each new energy vehicle, and the lithium battery temperature of each new energy vehicle in the second time step is calculated in the same way as the calculation method of the lithium battery temperature of each new energy vehicle in the first time step; Until the lithium battery temperature of each new energy vehicle in the last time step is calculated; The lithium battery temperature of each new energy vehicle in the first time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the first prediction time point, the lithium battery temperature of each new energy vehicle in the second time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the second prediction time point, ... the lithium battery temperature of each new energy vehicle in the last time step is recorded as the lithium battery temperature of each new energy vehicle corresponding to the last prediction time point; In this way, the lithium battery temperature of each new energy vehicle corresponding to each predicted time point is obtained.

5. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 1 is characterized in that: Dynamically calculate the fire risk index of each new energy vehicle carried by a ro-ro ship. The specific calculation process is as follows: Through risk assessment model , dynamically calculate the fire risk index of each new energy vehicle carried by ro-ro ships , τ1, τ2, τ3 and τ4 represent the calculation weight factors corresponding to the temperature risk coefficient, electrical state risk coefficient, thermal runaway risk coefficient and external stress risk coefficient, respectively, and satisfy τ1+τ2+τ3+τ4=1; In the above risk assessment model They respectively represent the temperature risk coefficient, electrical state risk coefficient and thermal runaway risk coefficient of the lithium battery corresponding to the h-th new energy vehicle, and φ is the external stress risk coefficient of the roll-on / roll-off ship.

6. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 5 is characterized in that: The calculation process of the temperature risk coefficient of lithium batteries corresponding to each new energy vehicle is as follows: The corrected temperature of the ro-ro ship at each predicted time point is weighted averaged to obtain the corrected temperature mean of the ro-ro ship during the predicted time period. Similarly, the weighted average temperature of the lithium battery of each new energy vehicle corresponding to each predicted time point is calculated to obtain the average temperature of the lithium battery of each new energy vehicle corresponding to the predicted time period. ; The temperature risk coefficient of lithium batteries corresponding to each new energy vehicle is calculated from this , β1 and β2 represent the set adjustment coefficients, is the critical temperature of the lithium battery corresponding to the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, and ζ2 is the ambient temperature change rate: ， ； h is the serial number of each new energy vehicle, h=1,2...K, K is the total number of new energy vehicles, They represent the lithium battery temperature of the h-th new energy vehicle at the j+1-th and j-th prediction time points, respectively. Respectively represent the time of the j+1th and jth prediction time points, is the corrected temperature at the j+1th prediction time point.

7. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 5 is characterized in that: The calculation process of the external stress risk factor of ro-ro ships is as follows: The hull swing amplitude of the ro-ro ship at each predicted time point Perform weighted average calculation to obtain the mean hull swing amplitude of the ro-ro ship during the predicted time period ; Determine the external stress value of the ro-ro ship at each predicted time point , and perform weighted average calculation on it in the same way to obtain the mean external stress of the ro-ro ship in the prediction time period ; The external stress risk coefficient of the ro-ro ship is calculated , where e is a natural constant.

8. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 5 is characterized in that: The calculation process of the electrical status risk factor of lithium batteries corresponding to each new energy vehicle is as follows: The weighted average calculation of the corrected humidity of the ro-ro ship at each prediction time point is performed to obtain the corrected humidity mean value of the ro-ro ship during the prediction period. ; Then, by analyzing the formula , get the electrical state risk coefficient of the lithium battery corresponding to each new energy vehicle ,in They represent the battery current mean, voltage homogeneity and battery internal resistance of the h-th new energy vehicle, is the stowage spacing of lithium batteries corresponding to the hth new energy vehicle, β5 and β6 represent the set adjustment coefficients, which are used to adjust the influence of capacity and internal resistance. Represents the total battery capacity of the h-th new energy vehicle.

9. The method for intelligent fire detection and analysis based on artificial intelligence according to claim 5 is characterized in that: The calculation process of the thermal runaway risk coefficient of lithium batteries corresponding to each new energy vehicle is as follows: , where exp(·) represents an exponential function with the natural constant e as the base, β7 and β8 represent the set adjustment coefficients, which are used to adjust the temperature change rate and the influence of heat release; They represent the heat released by the lithium battery corresponding to the h-th new energy vehicle and the total heat capacity of the lithium battery; Set a safe temperature for the lithium battery of the hth new energy vehicle, is the temperature change rate of the lithium battery of the hth new energy vehicle, It is the average temperature of lithium batteries of each new energy vehicle in the corresponding prediction time period.

10. An intelligent fire detection and analysis device based on artificial intelligence, characterized in that: It is implemented based on an artificial intelligence-based smart fire detection and analysis method according to any one of claims 1 to 9, comprising a processor, a memory and a communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it implements the artificial intelligence-based smart fire detection and analysis method as described in any one of claims 1-9.

Images