Unmanned cabin fire intelligent monitoring method and device based on multi-source information fusion

Abstract

The invention relates to an unmanned cabin fire intelligent monitoring method and device based on multi-source information fusion. The method comprises the following steps: acquiring a multi-source fire sensor real-time signal through a signal acquisition module, and judging the validity of the acquired sensor data to acquire original sensor data x (t); constructing a filter based on a fixed-point moving average method to eliminate environmental battery interference influence, and obtaining effective multi-source sensor data y (t); obtaining a feature value H after multi-source sensor data fusion based on a multi-source information fusion algorithm, comparing the feature value H with a feature value HO in a normal state, and judging a fire alarm state of the unmanned cabin; starting a fire extinguishing device to dispose fire based on grading alarm strategy sequence combination; the problem of low fire alarm accuracy of the unmanned cabin is solved, and the fire is intelligently and efficiently disposed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fire control systems and methods, and in particular to an intelligent fire monitoring method and device for unmanned cabins based on multi-source information fusion. Background Art

[0002] Intelligent unmanned ships are one of the fastest-growing and most promising directions in the field of ship transportation. In recent years, the global market size of the surface unmanned ship industry has been growing rapidly year by year. Against the background of the rapid development of global high-precision positioning technology and high labor costs, the development prospect of the intelligent unmanned ship industry is very broad.

[0003] The operating environment of unmanned ships is very complex and is often affected by harsh weather and other environments, which puts higher requirements on the intelligence and automation of electromechanical systems and equipment. If a fire breaks out on an unmanned ship and cannot be disposed of in time, it will lead to the occurrence of a ship out-of-control accident, and even endanger the safety of waterways and docks. Therefore, it is of great significance to monitor the fire situation in the cabins of unmanned ships, improve the stability and reliability of fire monitoring and disposal devices, and implement fire control measures in a timely manner at the initial stage of a fire to prevent further expansion of damage.

[0004] The publication number is CN218685836U, which discloses a fire monitoring and extinguishing integrated device, including a fixing frame, an impact component and a limiting component; the fixing frame: is of a U-shaped structure, and cross beams are symmetrically arranged at the lower ends between the two vertical plates of the fixing frame, and support rods are evenly arranged between the two cross beams, and fire extinguishing balls are placed on the upper ends of adjacent support rods; the impact component: is arranged on the lower surface of the horizontal plate of the fixing frame, and the impact component is arranged in cooperation with the fire extinguishing balls. The impact component includes a sleeve, a support plate, a spring and a firing pin. The sleeves and the springs are both arranged in an array on the lower surface of the horizontal plate of the fixing frame, and firing pins are slidably connected inside the sleeves; the limiting component: is arranged between the two vertical plates of the fixing frame, and the limiting component is arranged in cooperation with the impact component. This fire monitoring and extinguishing integrated device uses two methods, namely a flame detector and fire extinguishing balls, to monitor fires, avoiding the situation where monitoring cannot be carried out due to a circuit interruption, and has high reliability.

[0005] It uses the traditional flame detector method to monitor fires, and the source of fire monitoring information is single, with a certain false alarm rate. The traditional fire monitoring method requires manual judgment and disposal of the fire situation, and the reaction time will cause the fire to spread further; the fire alarm method based on a single information source such as smoke concentration and temperature has low accuracy. Improving the fire alarm accuracy of unmanned cabins is a problem to be solved by those skilled in the art. Summary of the Invention

[0006] In view of the shortcomings of the above-mentioned existing production technology, the applicant provides an unmanned cabin fire intelligent monitoring method and device based on multi-source information fusion, thereby greatly improving the working reliability and ensuring the accuracy.

[0007] The technical solution adopted by the present invention is as follows:

[0008] An intelligent fire monitoring method for unmanned cabins based on multi-source information fusion comprises the following steps:

[0009] S1: The real-time signals of multi-source fire sensors in the unmanned cabin are obtained through the signal acquisition module. In order to remove the data whose amplitude exceeds the preset amplitude range due to sensor failure, the validity of the collected sensor data is judged to obtain the original fire sensor data; the multi-source fire sensors are: temperature sensors, smoke sensors and flame sensors;

[0010] S2: Construct a filter based on the fixed-point sliding average method to eliminate the impact of spikes in sensor data caused by environmental electromagnetic interference;

[0011] S3: Based on the information fusion algorithm, the multi-source fire sensor information is integrated, and the characteristic value H obtained after integration is compared and analyzed with the preset threshold to determine the cabin fire alarm status;

[0012] S4: According to the graded alarm strategy, the fire extinguishing devices of the fire extinguishing units are opened in sequence and combined to deal with the fire in time.

[0013] Its further technical solution is:

[0014] In S1, the method for determining the acquisition of original multi-source sensor data is: by acquiring the time domain characteristic value of the sensor data and presetting the limit value, removing the data exceeding the limit value, and the time domain characteristic value of the sensor is the peak-to-peak value.

[0015] In S2, the specific process of obtaining valid sensor data is as follows:

[0016] S2.1: Use the rectangular window function f(x) to segment the original sensor signal and obtain a multi-segment sensor data stream sequence x(t);

[0017] S2.2: Construct filter based on fixed-point sliding average method;

[0018] S2.3: Obtain valid sensor data y(t) based on the sensor data x(t).

[0019] The expression of rectangular window function f(x) is:

[0020]

[0021] Where x is a variable;

[0022] l is the length of the data intercepted by the rectangular window function.

[0023] The effective sensor data sequence y(t) after filter processing is:

[0024]

[0025] Where, t is a natural number and 1≤t≤l-(k-1);

[0026] k is a natural number and 3≤k≤10;

[0027] l is the length of data intercepted by the rectangular window function;

[0028] x(t) is the sensor data sequence.

[0029] The standardized calculation method is:

[0030]

[0031] The calculation method of multi-source sensor data fusion is:

[0032]

[0033] In S4, the fire alarm status determination method is:

[0034]

[0035] Where:

[0036] D(H) is the cabin fire alarm status;

[0037] H O Calculated threshold for normal sensor data.

[0038] In S3, the comparative analysis and alarm process is as follows:

[0039] When the eigenvalue H obtained after multi-source sensor data fusion is equal to the eigenvalue H calculated in the normal state O A comparison is made to determine the fire status of the cabin. When it exceeds the preset threshold, it is determined that a fire has occurred, the corresponding fire extinguishing device is activated, and an alarm is sent to the shore-based monitoring center.

[0040] A device for an intelligent fire monitoring method for unmanned cabins based on multi-source information fusion, characterized in that it includes a fire detection unit, a fire alarm unit, a fire extinguishing control unit and a fire extinguishing unit, the fire detection unit is connected to the fire monitoring part of the unmanned cabin and collects signals for monitoring the fire situation, the fire extinguishing control unit is connected to the fire detection unit and can process the signals of the fire detection unit, the fire alarm unit is connected to the fire extinguishing control unit and can issue an alarm when there is a fire situation, and the fire extinguishing unit is connected to the fire extinguishing control unit and can receive fire extinguishing instructions to extinguish the fire in the designated cabin.

[0041] The beneficial effects of the present invention are as follows:

[0042] The present invention has a compact and reasonable structure and is easy to operate. Through the cooperation between the fire detection unit, the fire alarm unit, the fire extinguishing control unit and the fire extinguishing unit, the fire alarm work of the unmanned cabin can be easily completed, and the accuracy is high and the working reliability is good.

[0043] At the same time, the present invention also has the following advantages:

[0044] (1) The present invention is based on the actual operating environment of unmanned cabins. The current fire monitoring technology may have the problem of false alarms. A filter is constructed by a fixed-point sliding average method to eliminate the influence of signal spikes caused by electromagnetic interference. Information fusion is performed at the information level. The cabin fire alarm status is determined based on the characteristic value obtained based on the fusion result, thereby avoiding the occurrence of false alarms and improving the accuracy of fire alarms.

[0045] (2) The present invention provides a graded alarm based on the cabin fire characteristic value H, H≥2H O If it is judged as an initial fire, you can choose to open the gas fire extinguishing device / fine water mist fire extinguishing device / carbon dioxide fire extinguishing device to deal with the initial fire; H O ≤H<2H O If it is determined to be a serious fire, the first step is to issue a gas fire extinguishing device fire extinguishing command to extinguish the open flames, and the second step is to issue a water mist fire extinguishing device fire extinguishing command to continuously lower the temperature of the fire scene and further suppress the spread of the fire. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 The figure is a flow chart of the fire monitoring method of the present invention.

[0047] Figure 2 This is a schematic diagram of the composition of the intelligent monitoring and disposal device of the present invention.

[0048] Figure 3 This is a data processing flow chart of the multi-source information fusion algorithm of the present invention. DETAILED DESCRIPTION

[0049] The specific implementation of the present invention will be described below in conjunction with the accompanying drawings.

[0050] like Figure 1 As shown, the present invention provides an intelligent fire monitoring method for unmanned cabins based on multi-source information fusion, and the specific method is:

[0051] S1, obtains real-time signals from multi-source fire sensors through the signal acquisition module, and judges the validity of the collected sensor data to obtain the original fire sensor data, in order to remove data whose amplitude exceeds the preset amplitude range due to sensor failure and improve the alarm accuracy.

[0052] The method for determining and obtaining the original fire sensor data is: by obtaining the time domain characteristic value of the normal sensor signal and presetting the over-limit value, the sensor over-limit value data is removed; judging whether the sensor is faulty, and removing the data when the sensor is faulty. The time domain characteristic value of the sensor signal is the peak-to-peak value, and the multi-source sensor types are: temperature sensor, smoke sensor, and flame sensor.

[0053] S2, a filter is constructed based on the fixed-point sliding average method to eliminate the impact of spikes in sensor data caused by environmental electromagnetic interference.

[0054] like Figure 3 As shown in Figure 2, the specific process of constructing filters and obtaining effective multi-source sensor data is as follows:

[0055] S2.1, use the rectangular window function f(x) to segment the original sensor signal and obtain a multi-segment sensor data stream sequence x(t);

[0056] The rectangular window function f(x) is expressed as:

[0057]

[0058] Where x is a variable;

[0059] l is the length of the data intercepted by the rectangular window function.

[0060] S2.2, construct a filter based on the fixed-point sliding average method to obtain the effective sensor data y(t);

[0061]

[0062] Where, t is a natural number and 1≤t≤l-(k-1);

[0063] k is a natural number and 3≤k≤10;

[0064] l is the length of data intercepted by the rectangular window function;

[0065] x(t) is the sensor data sequence.

[0066] The specific process of constructing the filter is:

[0067] Input sensor data sequence x(t). In this embodiment, the amplitudes corresponding to the first three points of the input sensor data sequence are averaged as the first value of the effective sensor data sequence y(t) after filter processing. The amplitudes corresponding to the third point from the second point to the fourth point of the input sensor data sequence x(t) are averaged as the second value of the sensor data sequence y(t) after filter processing, and so on. The last two points of the sensor data sequence y(t) after filter processing are meaningless, so the subsequent t is 1≤t≤l-2.

[0068] The effective sensor data sequence y(t) after filter processing is:

[0069]

[0070] Where t is a natural number and 1≤t≤l-2, and is the point in the sensor data at equal time intervals;

[0071] l is the length of data intercepted by the rectangular window function;

[0072] S3, based on the information fusion algorithm, fuses the multi-source fire sensor information, compares the fused index H with the preset threshold, and determines the cabin fire alarm status.

[0073] S3.1, construct the multi-source sensor normalization matrix z based on the valid sensor data sequence y(t) ij ;

[0074] Assume that there are j types of fire sensors and 1≤j≤n, the length of each sensor data point is m and 1≤i≤m, then the multi-source sensor data matrix y ij for:

[0075]

[0076] Standardize the data:

[0077]

[0078] Furthermore, the multi-source sensor normalization matrix z ij for:

[0079]

[0080] Among them, the matrix column elements are the time domain signals of sensors of the same type, and the matrix row elements are the data points of sensors of different types at the same time.

[0081] S3.2, Determine the weights w of different sensor data based on entropy method j .

[0082] Calculate each z ij Proportion:

[0083]

[0084] Calculate the entropy value e of each indicator j :

[0085]

[0086] Calculate the weight w of each sensor indicator j :

[0087]

[0088] The multi-source sensor data index H after fusion is:

[0089]

[0090] Where:

[0091] y j is the data of j different fire sensors at the same time point;

[0092] w j are the weights of different sensor indicators.

[0093] S3.3, the method for determining the cabin fire alarm status is:

[0094]

[0095] Where:

[0096] D(H) is the cabin fire alarm status;

[0097] H O Calculated threshold for normal sensor data.

[0098] The process of comparative analysis and alarm is as follows: when the index H after multi-source sensor data fusion exceeds the preset fire alarm threshold H O When a fire occurs in the area, the fire extinguishing device of the corresponding fire extinguishing unit should be automatically turned on. Furthermore, the mobile communication technology is used to send an alarm to the shore-based monitoring center, and the time and location of the fire are sent when the alarm is sent, so that subsequent management personnel can take reasonable and effective disposal measures.

[0099] S4, according to the hierarchical alarm strategy, the fire extinguishing unit fire extinguishing device is opened in sequence and combined to deal with the fire in time. The hierarchical alarm strategy is:

[0100]

[0101] S4.1, when the characteristic value H representing the fire state of the compartment O ≤H<2H O At this time, it is determined that an initial fire has occurred in the cabin. At this time, the fire extinguishing control unit sends a fire extinguishing command to the fire extinguishing unit, and selects to open the gas fire extinguishing device / fine water mist fire extinguishing device / carbon dioxide fire extinguishing device based on the characteristics of the fire hazard source.

[0102] S4.2: When the characteristic value H ≥ 2H representing the fire state of the cabin O At this time, it is determined that a serious fire has occurred in the cabin. At this time, the fire extinguishing control unit sends two instructions to the fire extinguishing unit in sequence. First, it sends a gas fire extinguishing device activation instruction to extinguish the open fire, and then sends a water mist fire extinguishing device activation instruction to continuously reduce the fire scene temperature and further prevent the fire from spreading.

[0103] like Figure 2 As shown, the device used for the above-mentioned unmanned cabin fire intelligent monitoring method based on multi-source information fusion has the following specific structure and functions:

[0104] It mainly includes: fire detection unit, fire alarm unit, fire extinguishing control unit and fire extinguishing unit.

[0105] Among them, the fire detection unit is connected to the fire monitoring part of the unmanned cabin and collects signals for monitoring the fire situation. The fire extinguishing control unit is connected to the fire detection unit and can process the signals of the fire detection unit. The fire alarm unit is connected to the fire extinguishing control unit and can issue an alarm when there is a fire. The fire extinguishing unit is connected to the fire extinguishing control unit and can receive fire extinguishing instructions to extinguish the fire in the designated cabin.

[0106] Among them, the fire detection unit includes: temperature-sensitive fire sensors, smoke-sensitive fire sensors, flame-sensitive fire sensors, and further fire sensors are optional: combustible gas fire sensors, suction-type smoke fire sensors and other special fire sensors.

[0107] Among them, the fire alarm unit includes: sound and light alarm.

[0108] Among them, the fire extinguishing control unit includes: information acquisition module, central processing unit, power module, LCD display control panel, fire extinguishing control module, communication module, and linkage control module.

[0109] Among them, the information acquisition module is mainly used to collect information such as multi-source fire sensor data of the fire detection unit and transmit the information to the central processing unit.

[0110] Among them, the central processing unit is mainly used to process multi-source information from the information acquisition module, send fire extinguishing signals to the fire extinguishing unit, and send control instructions to the linkage device through the output module.

[0111] Among them, the power module is mainly used to supply power to the information acquisition module, central processing unit, LCD display control panel, fire extinguishing control module and other equipment in the system.

[0112] Among them, the LCD display control panel displays the fire alarm information from the central processing unit module and queries historical data by pressing buttons.

[0113] Among them, the fire extinguishing control module is mainly used to automatically open and close fire extinguishing devices such as gas fire extinguishing devices, fine water mist fire extinguishing devices, and foam fire extinguishing devices.

[0114] Among them, the communication module is mainly used to upload the central processing unit module information to the remote fire monitoring center and receive the fire extinguishing instructions issued by the remote fire monitoring center.

[0115] Among them, the linkage control module is mainly used to receive control instructions sent by the central processing unit module, and automatically control the shutdown of the unmanned cabin fan, fire damper, etc.

[0116] Among them, the fire extinguishing unit includes: gas fire extinguishing device, fine water mist fire extinguishing device, foam fire extinguishing device. Further, the extinguishing agent of the gas fire extinguishing device can be carbon dioxide, heptafluoropropane, perfluorohexanone, and the fine water mist fire extinguishing device can be a closed fine water mist system or an open fine water mist system.

[0117] The whole processing device has a compact structure and works stably and reliably.

[0118] The above description is an explanation of the present invention, not a limitation of the present invention. The scope of the present invention is defined in the claims. Any form of modification may be made within the scope of protection of the present invention.

Claims

1. An intelligent fire monitoring method for unmanned cabins based on multi-source information fusion, characterized in that: The steps include: S1: Acquire the real-time signals of multi-source fire sensors in the unmanned cabin through the signal acquisition module. In order to remove the data whose amplitude exceeds the preset amplitude range due to sensor failure, the validity of the collected sensor data is judged to obtain the original fire sensor data; Multi-source fire sensors include: temperature sensors, smoke sensors and flame sensors; S2: Construct a filter based on the fixed-point sliding average method to eliminate the impact of spikes in sensor data caused by environmental electromagnetic interference; S3: Based on the information fusion algorithm, the multi-source fire sensor information is integrated, and the characteristic value H obtained after integration is compared and analyzed with the preset threshold to determine the cabin fire alarm status; S4: According to the graded alarm strategy, the fire extinguishing devices of the fire extinguishing units are opened in sequence and combined to deal with the fire in time.

2. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: In S1, the method for determining the acquisition of original multi-source sensor data is: by acquiring the time domain characteristic value of the sensor data and presetting the limit value, removing the data exceeding the limit value, and the time domain characteristic value of the sensor is the peak-to-peak value.

3. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: In S2, the specific process of obtaining valid sensor data is as follows: S2.1: Use the rectangular window function f(x) to segment the original sensor signal and obtain a multi-segment sensor data stream sequence x(t); S2.2: Construct filter based on fixed-point sliding average method; S2.3: Obtain valid sensor data y(t) based on the sensor data x(t).

4. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 3, characterized in that: The expression of rectangular window function f(x) is: Where x is a variable; l is the length of the data intercepted by the rectangular window function.

5. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 3, characterized in that: The effective sensor data sequence y(t) after filter processing is: Where, t is a natural number and 1≤t≤l-(k-1); k is a natural number and 3≤k≤10; l is the length of data intercepted by the rectangular window function; x(t) is the sensor data sequence.

6. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: The standardized calculation method is:

7. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: The calculation method of multi-source sensor data fusion is:

8. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: In S4, the fire alarm status determination method is: Where: D(H) is the cabin fire alarm status; H O Calculated threshold for normal sensor data.

9. The method for intelligent fire monitoring in unmanned cabins based on multi-source information fusion according to claim 1, characterized in that: In S3, the comparative analysis and alarm process is as follows: When the eigenvalue H obtained after multi-source sensor data fusion is equal to the eigenvalue H calculated in the normal state O A comparison is made to determine the fire status of the cabin. When it exceeds the preset threshold, it is determined that a fire has occurred, the corresponding fire extinguishing device is activated, and an alarm is sent to the shore-based monitoring center.

10. A device for the unmanned cabin fire intelligent monitoring method based on multi-source information fusion as claimed in claim 1, characterized in that: It includes a fire detection unit, a fire alarm unit, a fire extinguishing control unit and a fire extinguishing unit. The fire detection unit is connected to the fire monitoring part of the unmanned cabin and collects signals for monitoring the fire situation. The fire extinguishing control unit is connected to the fire detection unit and can process the signals of the fire detection unit. The fire alarm unit is connected to the fire extinguishing control unit and can issue an alarm when there is a fire. The fire extinguishing unit is connected to the fire extinguishing control unit and can receive fire extinguishing instructions to extinguish the fire in the designated cabin.

Images