Cabin battery fire prediction method and system based on big data

Abstract

The invention provides a big data-based cabin battery fire prediction method and system, and the method comprises the steps: obtaining environment data in a cabin and dynamic data of the cabin, and screening environment sub-data and dynamic sub-data related to a battery stored in the cabin from the environment data and the dynamic data; the screened environment sub-data and dynamic sub-data are ensured to have a direct key influence effect on the battery; based on the environment sub-data and the dynamic sub-data, a big data set for predicting the abnormal condition of the battery stored in the cabin is constructed, so that the battery stored in the cabin is predicted, and the event attribute information of the abnormal condition of the battery is obtained; orientation identification is carried out on the characteristics of fire caused by the battery itself from two dimensions of environment and motion; and on the basis of the event attribute information, the characteristic information that the battery reaches a fire state is obtained, so that the battery is subjected to preventive fireproof operation, the possible fire danger of the battery is accurately predicted, and the fireproof control reliability of the battery in the transportation process is improved.

Description

Technical Field

[0001] The present invention relates to the field of cabin fire control, and in particular to a cabin battery fire prediction method and system based on big data. Background Art

[0002] During export transportation, electric vehicle batteries spend extended periods of time in the relatively closed interior of a ship's hold. Due to poor air flow, this environment can lead to localized overheating or excessive pressure. Furthermore, cargo ships experience rocking and pitching during long transoceanic voyages, causing the electric vehicle batteries inside to move. Although these batteries are well-sealed and soft-packed, prolonged and drastic movement can inevitably lead to ruptures in the soft-packaging and leakage from the battery casing. Due to their internal chemical and physical structures, electric vehicle batteries are highly sensitive to changes in the external environment and their own motion. When these factors combine, they can trigger safety incidents such as explosions and fires. Existing technologies lack methods for accurately predicting the potential fire hazards of electric vehicle batteries stored in a cargo hold during cargo ship transportation, influenced by various environmental and motion factors. This makes it difficult to ensure the safe storage of electric vehicle batteries inside a ship's hold and reduces the reliability of fire prevention controls for electric vehicle batteries. Summary of the Invention

[0003] The purpose of the present invention is to provide a cabin battery fire prediction method and system based on big data, which obtains environmental data inside the cabin and dynamic data of the cabin, and filters environmental sub-data and dynamic sub-data associated with the batteries stored inside the cabin, ensuring that the filtered environmental sub-data and dynamic sub-data have a direct and critical impact on the batteries, and providing a reliable and comprehensive data basis for subsequent prediction of battery fire conditions; based on the environmental sub-data and dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored inside the cabin, so as to predict the batteries stored inside the cabin, obtain event attribute information of abnormal conditions of the batteries, and perform position identification of the characteristics of the battery itself causing fire from two dimensions of environment and motion; also based on the event attribute information, characteristic information of the battery reaching a fire state is obtained, so that preventive fire prevention operations are performed on the battery, and possible fire hazards of the battery are accurately predicted, ensuring the storage safety of the battery inside the cabin, and improving the reliability of fire prevention control of the battery during transportation.

[0004] The present invention is achieved through the following technical solutions:

[0005] The big data-based method for predicting battery fires in ship cabins includes:

[0006] Acquire environmental data of the interior of the cabin, analyze the environmental data, and filter out environmental sub-data associated with the batteries stored in the cabin; acquire dynamic data of the cabin, analyze the dynamic data, and filter out dynamic sub-data associated with the batteries stored in the cabin;

[0007] Based on the environmental sub-data and the dynamic sub-data, a big data set is constructed for predicting abnormal conditions of the batteries stored in the cabin; based on the big data set, the batteries stored in the cabin are predicted to obtain event attribute information of abnormal conditions of the batteries;

[0008] Based on the event attribute information, characteristic information indicating that the battery has reached a fire state is obtained; and based on the characteristic information, a preventive fire prevention operation is performed on the battery.

[0009] Optionally, obtaining environmental data of the interior of the cabin, analyzing the environmental data, and screening to obtain environmental sub-data associated with the batteries stored in the cabin; obtaining dynamic data of the cabin, analyzing the dynamic data, and screening to obtain dynamic sub-data associated with the batteries stored in the cabin, including:

[0010] Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use the data as the ambient sub-data;

[0011] The motion action data of the cabin in different directions are obtained, and based on the motion action data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion action sub-data associated with the battery are filtered from the motion action data corresponding to all directions, and used as the dynamic sub-data.

[0012] Optionally, based on the environmental sub-data and the dynamic sub-data, a big data set is constructed for predicting abnormal conditions of the batteries stored in the cabin; based on the big data set, predictions are made on the batteries stored in the cabin to obtain event attribute information of abnormal conditions of the batteries, including:

[0013] Based on the generation times of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin;

[0014] Based on the big data set, after training a neural network model matching the batteries stored inside the cabin, the time information and location information of the event of abnormal conditions occurring in the batteries are predicted and used as the event attribute information; wherein, the abnormal conditions include leakage, expansion or local excessive temperature of the batteries.

[0015] Optionally, obtaining characteristic information indicating that the battery has reached a fire state based on the event attribute information; and performing a preventive fire prevention operation on the battery based on the characteristic information includes:

[0016] Based on the event occurrence time information and event occurrence location information of the battery abnormality, obtaining the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, and using this as characteristic information of the fire state;

[0017] Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

[0018] Optionally, based on the event occurrence time information and event occurrence location information of the battery abnormality, obtaining the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state includes:

[0019] Step S1: Assume that there are n locations corresponding to the abnormal situation of the battery, and the original temperature value of the battery at the i-th location is T i0 , where i is an integer greater than or equal to 1 and less than or equal to n, and the battery temperature value corresponding to time t is T it , the internal pressure of the battery corresponding to time t is P it , where time t is the time stamp of travel, P ito is the external air pressure value, then the danger index S of the battery at position i it for:

[0020]

[0021] In the above formula (1), α i is the linear thermal expansion coefficient of the battery material at the i-th position, Ei is the material elastic modulus value of the battery at the i-th position;

[0022] Step S2: Determine the battery fire risk assessment value R at the i-th position based on the calculation result of step S1. i (t),

[0023]

[0024] In the above formula (2), a and c are preset adjustment coefficients, and their value range is greater than 0 and less than or equal to 1, t is is the timestamp of the first activation of the battery at the i-th position after installation, and e is a natural constant;

[0025] Step S3: Determine the fire occurrence time M of the battery at the i-th position based on the calculation result of step S2. i ,

[0026]

[0027] In the above formula (3), k is the critical risk value of fire, and tip is the timestamp corresponding to the start time of the abnormal situation at the i-th position. Indicates R i (t) Find all the sets of t values ​​when the integral value is greater than k, It means finding the minimum value of t in the set.

[0028] The big data-based cabin battery fire prediction system includes:

[0029] An environmental data acquisition and screening module, configured to acquire environmental data inside the cabin, analyze the environmental data, and screen environmental sub-data associated with the batteries stored inside the cabin;

[0030] A dynamic data acquisition and screening module, configured to acquire dynamic data of the cabin, analyze the dynamic data, and screen dynamic sub-data associated with batteries stored in the cabin;

[0031] a big data set construction module, configured to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data;

[0032] A battery abnormality prediction module, configured to predict the batteries stored in the cabin based on the big data set, and obtain event attribute information of abnormal conditions occurring in the batteries;

[0033] a battery fire feature determination module, configured to obtain feature information indicating that the battery has reached a fire state based on the event attribute information;

[0034] A preventive fire prevention operation execution module is used to perform a preventive fire prevention operation on the battery based on the characteristic information.

[0035] Optionally, the environmental data acquisition and screening module is used to acquire environmental data inside the cabin, analyze the environmental data, and screen to obtain environmental sub-data associated with the batteries stored inside the cabin, including:

[0036] Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use the data as the ambient sub-data;

[0037] The dynamic data acquisition and screening module is used to acquire dynamic data of the cabin, analyze the dynamic data, and screen to obtain dynamic sub-data associated with the batteries stored in the cabin, including:

[0038] The motion action data of the cabin in different directions are obtained, and based on the motion action data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion action sub-data associated with the battery are filtered from the motion action data corresponding to all directions, and used as the dynamic sub-data.

[0039] Optionally, the big data set construction module is used to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data, including:

[0040] Based on the generation times of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin;

[0041] The battery abnormality prediction module is used to predict the batteries stored in the cabin based on the big data set, and obtain event attribute information of the battery abnormality, including:

[0042] Based on the big data set, after training a neural network model matching the batteries stored inside the cabin, the time information and location information of the event of abnormal conditions occurring in the batteries are predicted and used as the event attribute information; wherein, the abnormal conditions include leakage, expansion or local excessive temperature of the batteries.

[0043] Optionally, the battery fire feature determination module is configured to obtain feature information indicating that the battery has reached a fire state based on the event attribute information, including:

[0044] Based on the event occurrence time information and event occurrence location information of the battery abnormality, obtaining the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, and using this as characteristic information of the fire state;

[0045] The preventive fire prevention operation execution module is configured to perform a preventive fire prevention operation on the battery based on the characteristic information, including:

[0046] Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The big data-based cabin battery fire prediction method and system provided in the present application obtains the environmental data inside the cabin and the dynamic data of the cabin, and filters the environmental sub-data and dynamic sub-data associated with the batteries stored inside the cabin, ensuring that the filtered environmental sub-data and dynamic sub-data have a direct and critical impact on the batteries, and providing a reliable and comprehensive data basis for subsequent prediction of battery fire conditions; based on the environmental sub-data and dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored inside the cabin, so as to predict the batteries stored inside the cabin, obtain event attribute information of abnormal conditions of the batteries, and perform positional identification of the characteristics of the battery itself that causes fire from two dimensions: environment and motion; based on the event attribute information, characteristic information of the battery reaching a fire state is obtained, so that preventive fire prevention operations are performed on the battery, and possible fire hazards of the battery are accurately predicted, so as to ensure the storage safety of the battery inside the cabin and improve the reliability of fire prevention control of the battery during transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. Among them:

[0050] Figure 1 This is a flow chart of the big data-based cabin battery fire prediction method provided by the present invention.

[0051] Figure 2 This is a schematic diagram of the structure of the big data-based cabin battery fire prediction system provided by the present invention. DETAILED DESCRIPTION

[0052] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It will be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some, rather than all, structures related to the present application are shown in the accompanying drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0053] As used herein, the terms "comprise," "comprising," and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0054] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] See also Figure 1 As shown, an embodiment of the present application provides a method for predicting battery fires in a ship cabin based on big data. The method for predicting battery fires in a ship cabin based on big data includes:

[0056] Acquire environmental data of the cabin, analyze the environmental data, and filter out environmental sub-data associated with the batteries stored in the cabin; acquire dynamic data of the cabin, analyze the dynamic data, and filter out dynamic sub-data associated with the batteries stored in the cabin;

[0057] Based on the environmental sub-data and the dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored in the cabin; based on the big data set, the batteries stored in the cabin are predicted to obtain event attribute information of abnormal conditions of the batteries;

[0058] Based on the event attribute information, characteristic information indicating that the battery has reached a fire state is obtained; based on the characteristic information, preventive fire prevention operations are performed on the battery.

[0059] The beneficial effects of the above embodiments are as follows: the cabin battery fire prediction method based on big data obtains the environmental data inside the cabin and the dynamic data of the cabin, and filters the environmental sub-data and dynamic sub-data associated with the batteries stored inside the cabin, ensuring that the filtered environmental sub-data and dynamic sub-data have a direct and critical impact on the batteries, and providing a reliable and comprehensive data basis for subsequent prediction of battery fire conditions; based on the environmental sub-data and dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored inside the cabin, thereby predicting the batteries stored inside the cabin, obtaining event attribute information of abnormal conditions of the batteries, and locating the characteristics of the battery itself causing fire from two dimensions: environment and motion; based on the event attribute information, characteristic information of the battery reaching a fire state is obtained, and preventive fire prevention operations are performed on the battery, and possible fire hazards of the battery are accurately predicted, ensuring the storage safety of the battery inside the cabin, and improving the reliability of fire prevention control of the battery during transportation.

[0060] In another embodiment, environmental data of the interior of the cabin is obtained, the environmental data is analyzed, and environmental sub-data associated with batteries stored in the cabin is obtained by screening; dynamic data of the cabin is obtained, the dynamic data is analyzed, and dynamic sub-data associated with batteries stored in the cabin is obtained by screening, including:

[0061] Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use as the ambient sub-data;

[0062] The motion data of the cabin in different directions are obtained, and based on the motion data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion sub-data associated with the battery are filtered from the motion data corresponding to all directions, and used as the dynamic sub-data.

[0063] The beneficial effect of the above embodiment is that the interior of the cargo ship's cabin is a relatively closed space environment during ocean transportation, and the air flow inside the cabin is poor, which makes it easy for local temperatures to be too high or local air pressure to be too high inside the cabin. The batteries for electric vehicles are more sensitive to changes in the temperature or air pressure of the external environment. Small changes in temperature or air pressure may cause adverse disturbances to the batteries for electric vehicles and then trigger drastic changes inside the batteries for electric vehicles, thereby causing safety problems such as battery explosion and fire. First, obtain the ambient temperature and pressure data for all sub-areas within the cabin. Since each sub-area's relative positional relationship with the battery (e.g., relative distance and / or relative orientation) is different, the heat from each sub-area's own heat exerts different thermal radiation and pressure effects on the battery. Consequently, each sub-area exerts different temperature and pressure disturbances on the battery. Therefore, based on the relative positional relationship between each sub-area and the battery stored within the cabin, determine the characteristic information of each sub-area's thermal radiation effect (e.g., thermal radiation rate per unit area) and pressure effect (e.g., pressure per unit area) on the battery. This determines whether each sub-area exerts an effective temperature and pressure disturbance on the battery. If so, the ambient temperature and pressure data for the corresponding sub-area are determined as the ambient temperature and pressure sub-data associated with the battery. Furthermore, cargo ships can experience rocking and pitching during navigation, which can impact the batteries stored within the cabin, thereby disturbing the batteries and making them more susceptible to explosion and fire. To this end, the motion action data of the cabin in different directions (such as at least one of the motion speed, motion acceleration and running distance in the X, Y and Z axis directions in three-dimensional space) are obtained to determine the probability information of battery damage in each direction. When the probability of battery damage due to movement in a certain direction exceeds the preset probability threshold, the corresponding transportation action data in this direction is used as the motion action sub-data associated with the battery.

[0064] In another embodiment, a big data set is constructed based on the environmental sub-data and the dynamic sub-data to predict abnormal conditions of batteries stored in the cabin. Based on the big data set, the batteries stored in the cabin are predicted to obtain event attribute information of abnormal conditions of the batteries, including:

[0065] Based on the generation times of the ambient temperature sub-data and ambient pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin;

[0066] Based on this big data set, a neural network model matching the batteries stored inside the cabin is trained to predict the time and location of abnormal battery conditions as attribute information of the event. The abnormal conditions include battery leakage, expansion, or local overtemperature.

[0067] The beneficial effect of the above embodiment is that the screened ambient temperature sub-data, ambient air pressure sub-data and motion sub-data have a direct and high correlation with whether the outside and inside of the battery are degraded and damaged, and the degree of influence of ambient temperature, ambient air pressure and motion on battery degradation and damage continues to deepen over time. For this reason, based on the generation time of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the ambient sub-data and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data and the motion sub-data are time-domain sorted and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of batteries stored inside the cabin, so that the big data set can comprehensively reflect the degree of disturbance influence of ambient temperature, ambient air pressure and motion on the battery. Based on this big data set, a neural network model matching the batteries stored inside the cabin is trained to predict the time and location of abnormal battery conditions. This allows for identification of abnormal conditions that could easily cause fires, such as battery leakage, expansion, or local overheating, at the time and space domain levels.

[0068] In another embodiment, characteristic information indicating that the battery has reached a fire state is obtained based on the event attribute information; and preventive fire prevention operations are performed on the battery based on the characteristic information, including:

[0069] Based on the time information and location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained as characteristic information of the fire state;

[0070] Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching the fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

[0071] The beneficial effects of the above embodiments are as follows: based on the time information and location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained. This allows for accurate time and location prediction of battery fires caused by disturbances in ambient temperature, ambient air pressure, and / or motion, facilitating subsequent timely and targeted fire prevention operations for the battery. In addition, based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the battery area and its surrounding adjacent areas (such as pre-spraying fire extinguishing foam on the battery area and its surrounding adjacent areas). This ensures the storage safety of the battery inside the cabin and improves the reliability of fire prevention control of the battery during transportation.

[0072] In another embodiment, based on the event occurrence time information and event occurrence location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained, including:

[0073] Step S1: Assume that there are n locations corresponding to the abnormal situation of the battery, and the original temperature value of the battery at the i-th location is T i0 , where i is an integer greater than or equal to 1 and less than or equal to n, and the battery temperature value corresponding to time t is T it , the internal pressure of the battery corresponding to time t is P it , where time t is the time stamp of travel, P ito is the external air pressure value, then the danger index S of the battery at position i it for:

[0074]

[0075] In the above formula (1), α i is the linear thermal expansion coefficient of the battery material at the i-th position, E i is the material elastic modulus value of the battery at the i-th position; the above formula (1) comprehensively evaluates the battery expansion caused by temperature and pressure changes due to abnormal battery conditions, which is of great reference significance for evaluating battery performance and health status, and also helps to evaluate whether the battery is still working within the specified performance parameters;

[0076] Step S2: Determine the battery fire risk assessment value R at the i-th position based on the calculation result of step S1. i (t),

[0077]

[0078] In the above formula (2), a and c are preset adjustment coefficients, and their value range is greater than 0 and less than or equal to 1, t is is the timestamp of the first activation of the battery at the i-th position after installation, e is a natural constant with a value of 2.71;

[0079] Step S3: Determine the fire occurrence time M of the battery at the i-th position based on the calculation result of step S2. i ,

[0080]

[0081] In the above formula (3), k is the critical risk value of fire, and tip is the timestamp corresponding to the start time of the abnormal situation at the i-th position. Indicates R i (t) Find all the sets of t values ​​when the integral value is greater than k, represents finding the minimum value of t within the set. By accurately identifying the potential fire hazard based on battery usage time, the time of abnormality, and changes in internal and external pressure and temperature, operators are given sufficient time to respond and take action, improving the reliability and safety of the battery system.

[0082] The beneficial effects of the above embodiments are as follows: by comprehensively evaluating the battery expansion caused by temperature and air pressure changes due to abnormal battery conditions, and based on the battery usage time and the time when the abnormal condition occurs, as well as the pressure changes and temperature changes inside and outside the battery, the time point when the battery may catch fire can be accurately identified in advance, thereby providing accurate information for subsequent preventive physical fire isolation operations, thereby avoiding the situation where the subsequent preventive physical fire isolation operations are delayed due to inaccurate fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, causing major safety accidents and serious economic losses.

[0083] See also Figure 2 As shown, an embodiment of the present application provides a cabin battery fire prediction system based on big data. The cabin battery fire prediction system based on big data includes:

[0084] An environmental data acquisition and screening module is used to acquire environmental data inside the cabin, analyze the environmental data, and screen to obtain environmental sub-data associated with the batteries stored inside the cabin;

[0085] A dynamic data acquisition and screening module is used to acquire dynamic data of the cabin, analyze the dynamic data, and screen the dynamic sub-data associated with the batteries stored in the cabin;

[0086] a big data set construction module, configured to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data;

[0087] A battery abnormality prediction module is used to predict the batteries stored in the cabin based on the big data set and obtain event attribute information of abnormal conditions of the batteries;

[0088] a battery fire feature determination module, configured to obtain feature information indicating that the battery has reached a fire state based on the event attribute information;

[0089] The preventive fire prevention operation execution module is used to perform a preventive fire prevention operation on the battery based on the characteristic information.

[0090] The beneficial effects of the above embodiments are as follows: the cabin battery fire prediction system based on big data obtains the environmental data inside the cabin and the dynamic data of the cabin, and filters the environmental sub-data and dynamic sub-data associated with the batteries stored inside the cabin, ensuring that the filtered environmental sub-data and dynamic sub-data have a direct and critical impact on the batteries, and providing a reliable and comprehensive data basis for subsequent prediction of battery fire conditions; based on the environmental sub-data and dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored inside the cabin, thereby predicting the batteries stored inside the cabin, obtaining event attribute information of abnormal conditions of the batteries, and locating the characteristics of the battery itself causing fire from two dimensions: environment and motion; based on the event attribute information, characteristic information of the battery reaching a fire state is obtained, and preventive fire prevention operations are performed on the battery, and possible fire hazards of the battery are accurately predicted, ensuring the storage safety of the battery inside the cabin, and improving the reliability of fire prevention control of the battery during transportation.

[0091] In another embodiment, the environmental data acquisition and screening module is used to acquire environmental data inside the cabin, analyze the environmental data, and screen to obtain environmental sub-data associated with the batteries stored inside the cabin, including:

[0092] Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use as the ambient sub-data;

[0093] The dynamic data acquisition and screening module is used to acquire dynamic data of the cabin, analyze the dynamic data, and screen to obtain dynamic sub-data associated with the batteries stored in the cabin, including:

[0094] The motion data of the cabin in different directions are obtained, and based on the motion data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion sub-data associated with the battery are filtered from the motion data corresponding to all directions, and used as the dynamic sub-data.

[0095] The beneficial effect of the above embodiment is that the interior of the cargo ship's cabin is a relatively closed space environment during ocean transportation, and the air flow inside the cabin is poor, which makes it easy for local temperatures to be too high or local air pressure to be too high inside the cabin. The batteries for electric vehicles are more sensitive to changes in the temperature or air pressure of the external environment. Small changes in temperature or air pressure may cause adverse disturbances to the batteries for electric vehicles and then trigger drastic changes inside the batteries for electric vehicles, thereby causing safety problems such as battery explosion and fire. First, obtain the ambient temperature and pressure data for all sub-areas within the cabin. Since each sub-area's relative positional relationship with the battery (e.g., relative distance and / or relative orientation) is different, the heat from each sub-area's own heat exerts different thermal radiation and pressure effects on the battery. Consequently, each sub-area exerts different temperature and pressure disturbances on the battery. Therefore, based on the relative positional relationship between each sub-area and the battery stored within the cabin, determine the characteristic information of each sub-area's thermal radiation effect (e.g., thermal radiation rate per unit area) and pressure effect (e.g., pressure per unit area) on the battery. This determines whether each sub-area exerts an effective temperature and pressure disturbance on the battery. If so, the ambient temperature and pressure data for the corresponding sub-area are determined as the ambient temperature and pressure sub-data associated with the battery. Furthermore, cargo ships can experience rocking and pitching during navigation, which can impact the batteries stored within the cabin, thereby disturbing the batteries and making them more susceptible to explosion and fire. To this end, the motion action data of the cabin in different directions (such as at least one of the motion speed, motion acceleration and running distance in the X, Y and Z axis directions in three-dimensional space) are obtained to determine the probability information of battery damage in each direction. When the probability of battery damage due to movement in a certain direction exceeds the preset probability threshold, the corresponding transportation action data in this direction is used as the motion action sub-data associated with the battery.

[0096] In another embodiment, the big data set construction module is used to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data, including:

[0097] Based on the generation times of the ambient temperature sub-data and ambient pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin;

[0098] The battery abnormality prediction module is used to predict the batteries stored in the cabin based on the big data set, and obtain event attribute information of the battery abnormality, including:

[0099] Based on this big data set, a neural network model matching the batteries stored inside the cabin is trained to predict the time and location of abnormal battery conditions as attribute information of the event. The abnormal conditions include battery leakage, expansion, or local overtemperature.

[0100] The beneficial effect of the above embodiment is that the screened ambient temperature sub-data, ambient air pressure sub-data and motion sub-data have a direct and high correlation with whether the outside and inside of the battery are degraded and damaged, and the degree of influence of ambient temperature, ambient air pressure and motion on battery degradation and damage continues to deepen over time. For this reason, based on the generation time of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the ambient sub-data and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data and the motion sub-data are time-domain sorted and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of batteries stored inside the cabin, so that the big data set can comprehensively reflect the degree of disturbance influence of ambient temperature, ambient air pressure and motion on the battery. Based on this big data set, a neural network model matching the batteries stored inside the cabin is trained to predict the time and location of abnormal battery conditions. This allows for identification of abnormal conditions that could easily cause fires, such as battery leakage, expansion, or local overheating, at the time and space domain levels.

[0101] In another embodiment, the battery fire feature determination module is configured to obtain feature information indicating that the battery has reached a fire state based on the event attribute information, including:

[0102] Based on the time information and location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained as characteristic information of the fire state;

[0103] The preventive fire prevention operation execution module is used to perform a preventive fire prevention operation on the battery based on the characteristic information, including:

[0104] Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching the fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

[0105] The beneficial effects of the above embodiments are as follows: based on the time information and location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained. This allows for accurate time and location prediction of battery fires caused by disturbances in ambient temperature, ambient air pressure, and / or motion, facilitating subsequent timely and targeted fire prevention operations for the battery. In addition, based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the battery area and its surrounding adjacent areas (such as pre-spraying fire extinguishing foam on the battery area and its surrounding adjacent areas). This ensures the storage safety of the battery inside the cabin and improves the reliability of fire prevention control of the battery during transportation.

[0106] In general, the big data-based cabin battery fire prediction method and system obtains the environmental data inside the cabin and the dynamic data of the cabin, and filters the environmental sub-data and dynamic sub-data associated with the batteries stored in the cabin, ensuring that the filtered environmental sub-data and dynamic sub-data have a direct and critical impact on the batteries, and providing a reliable and comprehensive data basis for subsequent prediction of battery fire conditions; based on the environmental sub-data and dynamic sub-data, a big data set is constructed for predicting abnormal conditions of batteries stored in the cabin, so as to predict the batteries stored in the cabin, obtain event attribute information of abnormal conditions of the batteries, and perform position identification of the characteristics of the battery itself that causes fire from two dimensions: environment and motion; based on the event attribute information, characteristic information of the battery reaching a fire state is obtained, so that preventive fire prevention operations are performed on the battery, and possible fire hazards of the battery are accurately predicted, ensuring the storage safety of the battery in the cabin, and improving the reliability of fire prevention control of the battery during transportation.

[0107] The above is only a specific embodiment of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting cabin battery fire based on big data, characterized in that: include: Acquiring environmental data inside the cabin, analyzing the environmental data, and filtering to obtain environmental sub-data associated with the batteries stored inside the cabin; Acquiring dynamic data of the cabin, analyzing the dynamic data, and filtering to obtain dynamic sub-data associated with batteries stored in the cabin; Based on the environmental sub-data and the dynamic sub-data, a big data set is constructed for predicting abnormal conditions of the batteries stored inside the cabin; based on the big data set, the batteries stored inside the cabin are predicted to obtain event attribute information of abnormal conditions of the batteries; based on the event attribute information, characteristic information of the batteries reaching a fire state is obtained; based on the characteristic information, preventive fire prevention operations are performed on the batteries.

2. The method for predicting battery fire in a ship cabin based on big data according to claim 1, characterized in that: Acquiring environmental data of the interior of the cabin, analyzing the environmental data, and filtering to obtain environmental sub-data associated with the batteries stored in the cabin; acquiring dynamic data of the cabin, analyzing the dynamic data, and filtering to obtain dynamic sub-data associated with the batteries stored in the cabin, including: Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use the data as the ambient sub-data; The motion action data of the cabin in different directions are obtained, and based on the motion action data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion action sub-data associated with the battery are filtered from the motion action data corresponding to all directions, and used as the dynamic sub-data.

3. The method for predicting battery fire in a ship cabin based on big data according to claim 1, characterized in that: constructing a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data; Based on the big data set, the batteries stored in the cabin are predicted to obtain event attribute information of abnormal conditions of the batteries, including: Based on the generation times of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin; Based on the big data set, after training a neural network model matching the batteries stored inside the cabin, the time information and location information of the event of abnormal conditions occurring in the batteries are predicted and used as the event attribute information; wherein, the abnormal conditions include leakage, expansion or local excessive temperature of the batteries.

4. The method for predicting battery fire in a ship cabin based on big data according to claim 1, characterized in that: Based on the event attribute information, obtaining characteristic information that the battery has reached a fire state; Based on the characteristic information, preventive fire prevention operations are performed on the battery, including: Based on the event occurrence time information and event occurrence location information of the battery abnormality, obtaining the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, and using this as characteristic information of the fire state; Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

5. The method for predicting battery fire in a ship cabin based on big data according to claim 4, characterized in that: Based on the event occurrence time information and event occurrence location information of the battery abnormality, the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state are obtained, including: Step S1: Assume that there are n locations corresponding to the abnormal situation of the battery, and the original temperature value of the battery at the i-th location is T i0 , where i is an integer greater than or equal to 1 and less than or equal to n, and the battery temperature value corresponding to time t is T it , the internal pressure of the battery corresponding to time t is P it , where time t is the time stamp of travel, P ito is the external air pressure value, then the danger index S of the battery at position i it for: In the above formula (1), α i is the linear thermal expansion coefficient of the battery material at the i-th position, E i is the material elastic modulus value of the battery at the i-th position; Step S2: Determine the battery fire risk assessment value R at the i-th position based on the calculation result of step S1. i (t), In the above formula (2), a and c are preset adjustment coefficients, and their value range is greater than 0 and less than or equal to 1, t is is the timestamp of the first activation of the battery at the i-th position after installation, and e is a natural constant; Step S3: Determine the fire occurrence time M of the battery at the i-th position based on the calculation result of step S2. i , In the above formula (3), k is the critical risk value of fire, and tip is the timestamp corresponding to the start time of the abnormal situation at the i-th position. Indicates R i (t) Find all the sets of t values ​​when the integral value is greater than k, It means finding the minimum value of t in the set.

6. The cabin battery fire prediction system based on big data is characterized by: include: An environmental data acquisition and screening module is used to acquire environmental data of the cabin, analyze the environmental data, and screen to obtain environmental sub-data associated with the batteries stored in the cabin; a dynamic data acquisition and screening module is used to acquire dynamic data of the cabin, analyze the dynamic data, and screen to obtain dynamic sub-data associated with the batteries stored in the cabin; a big data set construction module, configured to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data; A battery abnormality prediction module, configured to predict the batteries stored in the cabin based on the big data set and obtain event attribute information of abnormal conditions occurring in the batteries; a battery fire feature determination module, configured to obtain feature information indicating that the battery has reached a fire state based on the event attribute information; A preventive fire prevention operation execution module is used to perform a preventive fire prevention operation on the battery based on the characteristic information.

7. The big data-based cabin battery fire prediction system according to claim 6, characterized in that: The environmental data acquisition and screening module is used to acquire environmental data inside the cabin, analyze the environmental data, and screen to obtain environmental sub-data associated with the batteries stored inside the cabin, including: Acquiring ambient temperature data and ambient air pressure data for all sub-areas within the cabin, determining characteristic information of thermal radiation effects and air pressure effects of each sub-area on the battery based on the relative positional relationship between each sub-area and the battery stored within the cabin, and filtering ambient temperature sub-data and ambient air pressure sub-data associated with the battery from the ambient temperature data and ambient air pressure data of all sub-areas to use the data as the ambient sub-data; The dynamic data acquisition and screening module is used to acquire dynamic data of the cabin, analyze the dynamic data, and screen to obtain dynamic sub-data associated with the batteries stored in the cabin, including: The motion action data of the cabin in different directions are obtained, and based on the motion action data corresponding to each direction, the probability information of the battery being damaged in each direction is determined; based on the probability information, the motion action sub-data associated with the battery are filtered from the motion action data corresponding to all directions, and used as the dynamic sub-data.

8. The big data-based cabin battery fire prediction system according to claim 6, characterized in that: The big data set construction module is used to construct a big data set for predicting abnormal conditions of batteries stored in the cabin based on the environmental sub-data and the dynamic sub-data, including: Based on the generation times of the ambient temperature sub-data and ambient air pressure sub-data associated with the battery contained in the environmental sub-data, and the motion sub-data associated with the battery contained in the dynamic sub-data, the ambient temperature sub-data, the ambient air pressure sub-data, and the motion sub-data are sorted in the time domain and erroneous sub-data are eliminated to construct a big data set for predicting abnormal conditions of the batteries stored in the cabin; The battery abnormality prediction module is used to predict the batteries stored in the cabin based on the big data set, and obtain event attribute information of the battery abnormality, including: Based on the big data set, after training a neural network model matching the batteries stored inside the cabin, the time information and location information of the event of abnormal conditions occurring in the batteries are predicted and used as the event attribute information; wherein, the abnormal conditions include leakage, expansion or local excessive temperature of the batteries.

9. The big data-based cabin battery fire prediction system according to claim 6, characterized in that: The battery fire feature determination module is configured to obtain feature information of the battery reaching a fire state based on the event attribute information, including: Based on the event occurrence time information and event occurrence location information of the battery abnormality, obtaining the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, and using this as characteristic information of the fire state; The preventive fire prevention operation execution module is configured to perform a preventive fire prevention operation on the battery based on the characteristic information, including: Based on the fire occurrence time information and fire source location information corresponding to the battery itself reaching a fire state, preventive physical fire isolation operations are performed on the area where the battery is located and its surrounding adjacent areas.

Images