Coal mine safety monitoring alarm studying and judging method and system based on data fusion

Abstract

The invention provides a coal mine safety monitoring alarm studying and judging method and system based on data fusion. The method comprises the following steps: establishing an identification framework for representing alarm causes; determining a first support probability of various alarm causes in the recognition framework according to data of an alarm sensor; determining a second support probability of various alarm causes in the recognition framework according to the same type of sensor data located at the upstream and downstream of the alarm sensor; determining a third support probability of various alarm causes in the recognition framework according to different types of sensor data located in the same area as the alarm sensor; and fusing the first support probability, the second support probability and the third support probability of each type of alarm cause based on an evidence theory to obtain an occurrence probability representing each type of alarm cause, and studying and judging the cause of the alarm event according to the occurrence probability. According to the technical scheme provided by the invention, the accuracy and reliability of alarm reason identification are remarkably improved while the research and judgment efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of coal mine safety technology, and in particular to a coal mine safety monitoring and alarm analysis method and system based on data fusion. Background Technology

[0002] Coal mine safety is of paramount importance in mine production. Modern coal mines generally deploy complex safety monitoring systems, using various sensors (such as methane, carbon monoxide, wind speed, and temperature sensors) distributed throughout the mine to monitor hazardous gas concentrations and environmental parameters in real time, generating massive amounts of monitoring data. These systems have become a key technological barrier in preventing major accidents such as gas explosions, fires, and asphyxiation.

[0003] In actual operation, existing coal mine safety monitoring systems face the following prominent technical problems, severely restricting their early warning effectiveness and automation level: 1. Frequent false alarms and severe interference: The underground production environment is extremely complex, with various unfavorable factors such as high humidity, high dust, electromagnetic interference, and equipment vibration. Sensors themselves may generate abnormal signals due to aging, malfunction, or instantaneous strong interference. In addition, the periodic calibration (calibration) operation of sensors according to safety regulations will also generate data that meets the alarm threshold in the system, forming a large number of non-real "process alarms". These "false alarms" triggered by environmental interference, equipment failure, and calibration operations are mixed with "real alarms" generated by real danger signs, forming a massive flow of alarm information that is difficult to distinguish between true and false. 2. Judgment is highly dependent on manual labor, resulting in low efficiency and real-time performance: Currently, the identification, cause analysis, and verification of alarm information mainly rely on the manual experience of the monitoring center staff. Faced with hundreds of thousands to millions of alarm data points that may be generated every day, the workload of manual screening is enormous, and the judgment efficiency is extremely low. This model struggles to meet the stringent requirements of "real-time response" in coal mine safety monitoring, easily leading to delays in responding to real emergencies or causing on-duty personnel to overlook important alarms due to "alarm fatigue," posing significant safety hazards. 3. Lack of intelligent multi-source information collaborative analysis capabilities: While existing systems collect data from multiple types of sensors, alarm judgments are typically based on threshold exceedances of a single sensor, constituting a "single-point judgment." When sensor data is abnormal, the system lacks an effective mechanism to automatically call upon and comprehensively analyze data from upstream and downstream sensors of the same type to verify trends. It also fails to correlate and analyze data from different types of sensors with physical causal logic within the same area (such as methane sensors and air volume sensors, carbon monoxide sensors and temperature / smoke sensors) for collaborative verification. This data silo phenomenon prevents the system from cross-validating and comprehensively judging alarm events from multiple angles and levels, resulting in low accuracy in alarm cause identification and an inability to provide reliable automated support for emergency decision-making. Therefore, there is an urgent need for a solution that can automatically, quickly, and accurately analyze alarm information from coal mine safety monitoring systems to overcome the technical shortcomings of relying on manual labor, low efficiency, high false alarm rate, and lack of multi-source collaborative analysis, thereby truly improving the intelligence level and early warning reliability of coal mine safety monitoring. Summary of the Invention

[0004] This application provides a coal mine safety monitoring alarm analysis method and system based on data fusion, which at least solves the technical problems of existing technologies such as reliance on manual labor, low efficiency, high false alarm rate and lack of multi-source collaborative analysis.

[0005] The first aspect of this application proposes a method for analyzing coal mine safety monitoring alarms based on data fusion, the method comprising: Establish an identification framework for characterizing the causes of alarms; Acquire the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor; The first support probability of various alarm causes in the recognition framework is determined based on the alarm sensor's own data. The second support probability of various alarm causes in the identification framework is determined based on data from similar sensors located upstream and downstream of the alarm sensor. The third support probability of various alarm causes in the identification framework is determined based on data from different types of sensors located in the same area as the alarm sensor. Based on evidence theory, the first support probability, second support probability, and third support probability of the various alarm causes are fused to obtain the occurrence probability of each alarm cause, and the cause of the alarm event is judged based on the occurrence probability.

[0006] Preferably, the various alarm causes in the identification framework include: False alarms include actual alarms, false alarms caused by calibration, and false alarms caused by sensor interference or malfunction.

[0007] Furthermore, determining the first support probability of various alarm causes in the recognition framework based on the alarm sensor's own data includes: Based on the alarm sensor's own data at each moment in the historical period, and based on the own data, it is determined whether the data change process of the alarm sensor is normal; When the data change process of the alarm sensor is abnormal, it is determined that the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. When the data change process of the alarm sensor is normal, the preliminary alarm cause of the alarm sensor is determined according to the preset standard gas concentration used for sensor calibration, and the first support probability of each alarm cause in the identification framework is determined according to the preliminary alarm cause of the alarm sensor.

[0008] Furthermore, the step of determining whether the data change process of the alarm sensor is normal based on its own data includes: The maximum and minimum values ​​of the alarm sensor's own data during the historical working period are determined based on the alarm sensor's own data at each moment within the historical period. Based on the self-data at each moment within the historical period, the rising period and the falling period of self-data are found. In the rising period of self-data, the first self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the first self-data is determined. In the falling period of self-data, the second self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the second self-data is determined. The time period consisting of the time corresponding to the first self-data and the time corresponding to the second self-data is taken as the time period in which the abnormal data exists; Determine whether, within the time period in which the abnormal data exists, the absolute value of the difference between any two adjacent moments is greater than or equal to 90% (X). max -x max If so, the data change process of the alarm sensor is determined to be abnormal, and the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. Otherwise, the data change process of the alarm sensor is determined to be normal. max x is the maximum value of the alarm sensor's own data during the time period in which the abnormal data exists. max This represents the maximum value of the alarm sensor's own data during its historical operating period.

[0009] Furthermore, the step of determining the preliminary alarm cause of the alarm sensor based on the preset standard gas concentration used for sensor calibration, and determining the first support probability of various alarm causes in the identification framework based on the preliminary alarm cause of the alarm sensor, includes: Determine whether ρ*(1-10%)≤X is satisfied. max ≤ρ*(1+10%) and t q -t p If the condition is satisfied (≥60s), the initial alarm cause of the alarm sensor is determined to be a false alarm caused by the calibration process and / or a real alarm. The first support probability corresponding to the real alarm is 0.2, the first support probability corresponding to the false alarm caused by the calibration process is 0.8, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. If the condition is not satisfied, the initial alarm cause of the alarm sensor is determined to be a real alarm. The first support probability corresponding to the real alarm is 1, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. Where ρ is the preset standard gas concentration used for sensor calibration, and t qFor the first time during a period of decline in its own data that the value is less than X. max *95% of the corresponding time, t p For the first time during the rise of its own data within a historical period, the value is greater than or equal to X. max *95% corresponds to the time.

[0010] Furthermore, determining the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor includes: Analyze the data of the same type of sensor upstream and downstream of the alarm sensor at each moment within a first preset time period to determine whether calibration data exists in the data of the same type of sensor at each moment within the first preset time period. If it exists, the second support probability corresponding to the real alarm is 0, the second support probability corresponding to the false alarm caused by the calibration process is 1, and the second support probability corresponding to the false alarm caused by sensor interference or failure is 0. If it does not exist, the second support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the second support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the second support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0011] Furthermore, determining the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor includes: When the alarm sensor is a methane sensor, the airflow sensor data of the duct is judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure. When the alarm sensor is a carbon monoxide sensor, the temperature sensor data and / or smoke sensor data are judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0012] Furthermore, the first support probability, second support probability, and third support probability of the various alarm causes are fused based on evidence theory to obtain the probability of occurrence of each alarm cause, including: Using formula Determine the probability of a real alarm occurring; Using formula Determine the probability of false alarms caused by the calibration process; Using formula To determine the probability of false alarms caused by sensor interference or malfunction; in, This represents the actual probability of an alarm occurring. The first support probability corresponding to a real alarm. The second support probability corresponding to a real alarm. The third support probability corresponding to a real alarm. This represents the probability of false alarms occurring during the calibration process. The first support probability corresponds to the false alarm caused by the calibration process. The second support probability corresponds to the false alarms caused by the calibration process. The third support probability corresponds to false alarms caused by the calibration process. This represents the probability of false alarms caused by sensor interference or malfunction. The first support probability corresponds to false alarms caused by sensor interference or malfunction. This represents the second support probability corresponding to false alarms caused by sensor interference or malfunction. The third support probability corresponds to false alarms caused by sensor interference or malfunction. As the first parameter, For the second parameter, , .

[0013] Furthermore, the step of analyzing the cause of the alarm event based on the probability of occurrence includes: The probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are sorted from left to right from largest to smallest to form a first sequence, and the cause corresponding to the first value on the left in the first sequence is taken as the cause of the alarm event. When the probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are equal, the probability of a real alarm occurring is updated to be equal to the first support probability corresponding to the real alarm, the probability of a false alarm occurring due to the calibration process is equal to the first support probability corresponding to the false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction is equal to the first support probability corresponding to the false alarm occurring due to sensor interference or malfunction.

[0014] The second aspect of this application proposes a coal mine safety monitoring and alarm analysis system based on data fusion, comprising: Establish a module to build an identification framework for characterizing the causes of alarms; The acquisition module is used to acquire the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor. The first determining module is used to determine the first support probability of various alarm causes in the identification framework based on the alarm sensor's own data. The second determining module is used to determine the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor. The third determining module is used to determine the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor. The analysis module is used to fuse the first support probability, second support probability, and third support probability of the various alarm causes based on evidence theory to obtain the occurrence probability of each alarm cause, and to analyze the cause of the alarm event based on the occurrence probability.

[0015] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects: This application proposes a data fusion-based method and system for analyzing alarms in coal mine safety monitoring. The method includes: establishing an identification framework to characterize alarm causes; acquiring alarm sensor data corresponding to the alarm event, data from similar sensors located upstream and downstream of the alarm sensor, and data from different types of sensors in the same area as the alarm sensor; determining a first support probability for each alarm cause in the identification framework based on the alarm sensor data; determining a second support probability for each alarm cause in the identification framework based on the data from similar sensors located upstream and downstream of the alarm sensor; determining a third support probability for each alarm cause in the identification framework based on the data from different types of sensors in the same area as the alarm sensor; fusing the first, second, and third support probabilities of each alarm cause based on evidence theory to obtain the occurrence probability characterizing each alarm cause, and analyzing the cause of the alarm event based on the occurrence probability. The technical solution proposed in this application improves the efficiency of analysis while significantly enhancing the accuracy and reliability of alarm cause identification.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a coal mine safety monitoring and alarm analysis method based on data fusion, according to an embodiment of this application. Figure 2 A detailed flowchart of a coal mine safety monitoring alarm judgment method based on data fusion according to an embodiment of this application; Figure 3 This is a structural diagram of a coal mine safety monitoring and alarm analysis system based on data fusion, according to an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0019] This application proposes a data fusion-based method and system for analyzing alarms in coal mine safety monitoring. The method includes: establishing an identification framework to characterize alarm causes; acquiring alarm sensor data corresponding to the alarm event, data from similar sensors located upstream and downstream of the alarm sensor, and data from different types of sensors in the same area as the alarm sensor; determining a first support probability for each alarm cause in the identification framework based on the alarm sensor data; determining a second support probability for each alarm cause in the identification framework based on the data from similar sensors located upstream and downstream of the alarm sensor; determining a third support probability for each alarm cause in the identification framework based on the data from different types of sensors in the same area as the alarm sensor; fusing the first, second, and third support probabilities of each alarm cause based on evidence theory to obtain the occurrence probability characterizing each alarm cause, and analyzing the cause of the alarm event based on the occurrence probability. The technical solution proposed in this application improves the efficiency of analysis while significantly enhancing the accuracy and reliability of alarm cause identification.

[0020] The following description, with reference to the accompanying drawings, illustrates a coal mine safety monitoring and alarm analysis method and system based on data fusion, according to an embodiment of this application.

[0021] Example 1 Figure 1 This is a flowchart illustrating a coal mine safety monitoring and alarm analysis method based on data fusion, according to an embodiment of this application. Figure 1 As shown, the method includes: Step 1: Establish an identification framework to characterize the causes of alarms; It should be noted that the various alarm causes in the aforementioned identification framework include: False alarms include actual alarms, false alarms caused by calibration, and false alarms caused by sensor interference or malfunction.

[0022] It should be noted that the alarm types of the sensors (mainly for methane and carbon monoxide sensors) are divided into three categories: real alarms, false alarms caused by the calibration process, and false alarms caused by sensor interference or malfunction. Based on the alarm type, the identification framework Θ={A1,A2,A3} is determined, where A1 represents a real alarm, A2 represents a false alarm caused by the calibration process, and A3 represents a false alarm caused by sensor interference or malfunction.

[0023] Step 2: Obtain the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor; Step 3: Determine the first support probability of each alarm cause in the recognition framework based on the alarm sensor's own data; In this embodiment of the disclosure, step 3 specifically includes: Based on the alarm sensor's own data at each moment in the historical period, and based on the own data, it is determined whether the data change process of the alarm sensor is normal; When the data change process of the alarm sensor is abnormal, it is determined that the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. When the data change process of the alarm sensor is normal, the preliminary alarm cause of the alarm sensor is determined according to the preset standard gas concentration used for sensor calibration, and the first support probability of each alarm cause in the identification framework is determined according to the preliminary alarm cause of the alarm sensor.

[0024] It should be noted that the process of determining whether the data change process of the alarm sensor is normal based on its own data includes: The maximum and minimum values ​​of the alarm sensor's own data during the historical working period are determined based on the alarm sensor's own data at each moment within the historical period. Based on the self-data at each moment within the historical period, the rising period and the falling period of self-data are found. In the rising period of self-data, the first self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the first self-data is determined. In the falling period of self-data, the second self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the second self-data is determined. The time period consisting of the time corresponding to the first self-data and the time corresponding to the second self-data is taken as the time period in which the abnormal data exists; Determine whether, within the time period in which the abnormal data exists, the absolute value of the difference between any two adjacent moments is greater than or equal to 90% (X). max -x max If so, the data change process of the alarm sensor is determined to be abnormal, and the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. Otherwise, the data change process of the alarm sensor is determined to be normal. maxx is the maximum value of the alarm sensor during the time period in which the abnormal data exists. max This represents the maximum value of the alarm sensor's own data during its historical operating period.

[0025] Furthermore, the step of determining the preliminary alarm cause of the alarm sensor based on the preset standard gas concentration used for sensor calibration, and determining the first support probability of various alarm causes in the identification framework based on the preliminary alarm cause of the alarm sensor, includes: Determine whether ρ*(1-10%)≤X is satisfied. max ≤ρ*(1+10%) and t q -t p If the condition is satisfied (≥60s), the initial alarm cause of the alarm sensor is determined to be a false alarm caused by the calibration process and / or a real alarm. The first support probability corresponding to the real alarm is 0.2, the first support probability corresponding to the false alarm caused by the calibration process is 0.8, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. If the condition is not satisfied, the initial alarm cause of the alarm sensor is determined to be a real alarm. The first support probability corresponding to the real alarm is 1, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. Where ρ is the preset standard gas concentration used for sensor calibration, and t q For the first time during a period of decline in its own data that the value is less than X. max *95% of the corresponding time, t p For the first time during the rise of its own data within a historical period, the value is greater than or equal to X. max *95% corresponds to the time.

[0026] It should be noted that after the coal mine safety monitoring system uploads alarm data, it retrieves one month's historical data from the alarm sensor and calculates the maximum value x of that sensor. max Minimum value x min The maximum value of the alarm sensor's own data within the historical alarm period is X. max The corresponding time is T. max Find data to reach X max Previously min(x) m -x max The corresponding time t) m And the data descent process satisfies min(x) m+n -x max The corresponding time t) m+n , then t m and t m+n This refers to the time period during which the abnormal data exists.

[0027] In (t) m , t m+n Does abs(x) exist within the time period? m+k+1 -x m+k )≥90% (X max -x max If it exists, then the data change process is abnormal, and the preliminary judgment is that the alarm is a false alarm caused by sensor interference or malfunction. In this case, m1(A1)=0, m1(A2)=0, m1(A3)=1; if abs(x m+k+1 -x m+k <90% (X) max -x max If so, then the sensor data change process is normal.

[0028] Let the concentration of the standard gas used for sensor calibration be ρ. Determine whether ρ*(1-10%)≤X is satisfied. max ≤ρ*(1+10%) and find the first time X is reached during the data's ascent. max *95% corresponds to time t p And the first time the data falls below X during the decline process max *95% corresponds to time t q , t q -t p If the above conditions are met (≥60s), the alarm is most likely a false alarm caused by the calibration process, but the possibility of a real alarm cannot be ruled out. In this case, m1(A1) = 0.2, m1(A2) = 0.8, and m1(A3) = 0. If the above conditions are not met, the alarm is most likely a real alarm. In this case, m1(A1) = 1, m1(A2) = 0, and m1(A3) = 0.

[0029] Step 4: Determine the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor; In this embodiment of the disclosure, step 4 specifically includes: Analyze the data of the same type of sensor upstream and downstream of the alarm sensor at each moment within a first preset time period to determine whether calibration data exists in the data of the same type of sensor at each moment within the first preset time period. If it exists, the second support probability corresponding to the real alarm is 0, the second support probability corresponding to the false alarm caused by the calibration process is 1, and the second support probability corresponding to the false alarm caused by sensor interference or failure is 0. If it does not exist, the second support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the second support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the second support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0030] It should be noted that data from upstream and downstream sensors of the same type, taken two hours before and after the alarm time period, are retrieved and analyzed. If calibration data is found, the probability of a false alarm caused by the calibration process is relatively high, and m2(A1)=0, m2(A2)=1, m2(A3)=0. If no calibration data is found in the upstream and downstream sensor data, the cause of the alarm is determined based on the preliminary judgment result, and m2(A1)=m1(A1), m2(A2)=m1(A2), m2(A3)=m1(A3).

[0031] Step 5: Determine the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor; In this embodiment of the disclosure, step 5 specifically includes: When the alarm sensor is a methane sensor, the airflow sensor data of the duct is judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure. When the alarm sensor is a carbon monoxide sensor, the temperature sensor data and / or smoke sensor data are judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0032] It should be noted that, for the methane sensor, the airflow sensor in the same area should also be analyzed for abnormality: if the airflow sensor data is abnormal, it can further confirm that the alarm of the methane sensor is a real alarm caused by the abnormality of the ventilation system, then m3(A1)=1, m1(A2)=0, m1(A3)=0; otherwise, the cause of the alarm should be determined according to the preliminary judgment result, m3(A1)=m1(A1), m3(A2)=m1(A2), m3(A3)=m1(A3); For carbon monoxide sensors, if the temperature and smoke sensors show abnormal data, it can be further inferred that the alarm cause of the carbon monoxide sensor is a real alarm caused by a fire, then m3(A1)=1, m1(A2)=0, m1(A3)=0; if the temperature and smoke sensors are normal, the alarm cause can be determined based on the preliminary judgment results, then m3(A1)=m1(A1), m3(A2)=m1(A2), m3(A3)=m1(A3).

[0033] Step 6: Based on the DS evidence theory, the first support probability, the second support probability, and the third support probability of the various alarm causes are fused to obtain the occurrence probability of each alarm cause, and the cause of the alarm event is judged according to the occurrence probability.

[0034] In this embodiment of the disclosure, step 6 specifically includes: Using formula Determine the probability of a real alarm occurring; Using formula Determine the probability of false alarms caused by the calibration process; Using formula To determine the probability of false alarms caused by sensor interference or malfunction; in, This represents the actual probability of an alarm occurring. The first support probability corresponding to a real alarm. The second support probability corresponding to a real alarm. The third support probability corresponding to a real alarm. This represents the probability of false alarms occurring during the calibration process. The first support probability corresponds to the false alarm caused by the calibration process. The second support probability corresponds to the false alarms caused by the calibration process. The third support probability corresponds to false alarms caused by the calibration process. This represents the probability of false alarms caused by sensor interference or malfunction. The first support probability corresponds to false alarms caused by sensor interference or malfunction. This represents the second support probability corresponding to false alarms caused by sensor interference or malfunction. The third support probability corresponds to false alarms caused by sensor interference or malfunction. As the first parameter, For the second parameter, , .

[0035] The probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are sorted from left to right from largest to smallest to form a first sequence, and the cause corresponding to the first value on the left in the first sequence is taken as the cause of the alarm event. When the probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are equal, the probability of a real alarm occurring is updated to be equal to the first support probability corresponding to the real alarm, the probability of a false alarm occurring due to the calibration process is equal to the first support probability corresponding to the false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction is equal to the first support probability corresponding to the false alarm occurring due to sensor interference or malfunction.

[0036] It should be noted that the detailed process of the coal mine safety monitoring and alarm analysis method based on data fusion proposed in this embodiment is as follows: Figure 2 As shown, it will not be elaborated further here.

[0037] This embodiment constructs a sensor alarm model covering three typical scenarios: real alarms, calibration false alarms, and false alarms caused by sensor interference or malfunction (focusing on methane and carbon monoxide sensors), enabling preliminary judgment of alarm causes. Furthermore, by fusing monitoring data from upstream and downstream sensors of the alarm sensor, as well as from different types of sensors within the same area, a multi-source information fusion algorithm based on DS evidence theory is used to collaboratively analyze and make decisions about alarm events. This significantly improves the accuracy and reliability of alarm cause identification while enhancing analysis efficiency.

[0038] In summary, the data fusion-based alarm analysis method for coal mine safety monitoring proposed in this embodiment improves analysis efficiency while significantly enhancing the accuracy and reliability of alarm cause identification.

[0039] Example 2 Figure 3 This is a structural diagram of a coal mine safety monitoring and alarm analysis system based on data fusion, according to an embodiment of this application. Figure 3 As shown, the system includes: Establish module 100 to build an identification framework for characterizing alarm causes; It should be noted that the various alarm causes in the aforementioned identification framework include: False alarms include actual alarms, false alarms caused by calibration, and false alarms caused by sensor interference or malfunction.

[0040] The acquisition module 200 is used to acquire the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor. The first determining module 300 is used to determine the first support probability of various alarm causes in the identification framework based on the alarm sensor's own data. The second determining module 400 is used to determine the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor. The third determination module 500 is used to determine the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor. The analysis module 600 is used to fuse the first support probability, the second support probability, and the third support probability of the various alarm causes based on evidence theory to obtain the occurrence probability of each alarm cause, and to analyze the cause of the alarm event based on the occurrence probability.

[0041] In this embodiment of the disclosure, the first determining module 300 is further configured to: Based on the alarm sensor's own data at each moment in the historical period, and based on the own data, it is determined whether the data change process of the alarm sensor is normal; When the data change process of the alarm sensor is abnormal, it is determined that the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. When the data change process of the alarm sensor is normal, the preliminary alarm cause of the alarm sensor is determined according to the preset standard gas concentration used for sensor calibration, and the first support probability of each alarm cause in the identification framework is determined according to the preliminary alarm cause of the alarm sensor.

[0042] Furthermore, the first determining module 300 is also used for: The maximum and minimum values ​​of the alarm sensor's own data during the historical working period are determined based on the alarm sensor's own data at each moment within the historical period. Based on the self-data at each moment within the historical period, the rising period and the falling period of self-data are found. In the rising period of self-data, the first self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the first self-data is determined. In the falling period of self-data, the second self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the second self-data is determined. The time period consisting of the time corresponding to the first self-data and the time corresponding to the second self-data is taken as the time period in which the abnormal data exists; Determine whether, within the time period in which the abnormal data exists, the absolute value of the difference between any two adjacent moments is greater than or equal to 90% (X). max -x max If so, the data change process of the alarm sensor is determined to be abnormal, and the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. Otherwise, the data change process of the alarm sensor is determined to be normal. max x is the maximum value of the alarm sensor's own data during the time period in which the abnormal data exists. max This represents the maximum value of the alarm sensor's own data during its historical operating period.

[0043] Furthermore, the first determining module 300 is also used for: Determine whether ρ*(1-10%)≤X is satisfied. max ≤ρ*(1+10%) and t q -t pIf the condition is satisfied (≥60s), the initial alarm cause of the alarm sensor is determined to be a false alarm caused by the calibration process and / or a real alarm. The first support probability corresponding to the real alarm is 0.2, the first support probability corresponding to the false alarm caused by the calibration process is 0.8, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. If the condition is not satisfied, the initial alarm cause of the alarm sensor is determined to be a real alarm. The first support probability corresponding to the real alarm is 1, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. Where ρ is the preset standard gas concentration used for sensor calibration, and t q For the first time during a period of decline in its own data that the value is less than X. max *95% of the corresponding time, t p For the first time during the rise of its own data within a historical period, the value is greater than or equal to X. max *95% corresponds to the time.

[0044] In this embodiment of the disclosure, the second determining module 400 is further configured to: Analyze the data of the same type of sensor upstream and downstream of the alarm sensor at each moment within a first preset time period to determine whether calibration data exists in the data of the same type of sensor at each moment within the first preset time period. If it exists, the second support probability corresponding to the real alarm is 0, the second support probability corresponding to the false alarm caused by the calibration process is 1, and the second support probability corresponding to the false alarm caused by sensor interference or failure is 0. If it does not exist, the second support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the second support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the second support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0045] In this embodiment of the disclosure, the third determining module 500 is further configured to: When the alarm sensor is a methane sensor, the airflow sensor data of the duct is judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure. When the alarm sensor is a carbon monoxide sensor, the temperature sensor data and / or smoke sensor data are judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

[0046] In this embodiment of the disclosure, the analysis module 600 is further configured to: Using formula Determine the probability of a real alarm occurring; Using formula Determine the probability of false alarms caused by the calibration process; Using formula To determine the probability of false alarms caused by sensor interference or malfunction; in, This represents the actual probability of an alarm occurring. The first support probability corresponding to a real alarm. The second support probability corresponding to a real alarm. The third support probability corresponding to a real alarm. This represents the probability of false alarms occurring during the calibration process. The first support probability corresponds to the false alarm caused by the calibration process. The second support probability corresponds to the false alarms caused by the calibration process. The third support probability corresponds to false alarms caused by the calibration process. This represents the probability of false alarms caused by sensor interference or malfunction. The first support probability corresponds to false alarms caused by sensor interference or malfunction. This represents the second support probability corresponding to false alarms caused by sensor interference or malfunction. The third support probability corresponds to false alarms caused by sensor interference or malfunction. As the first parameter, For the second parameter, , .

[0047] Furthermore, the analysis module 600 is also used for: The probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are sorted from left to right from largest to smallest to form a first sequence, and the cause corresponding to the first value on the left in the first sequence is taken as the cause of the alarm event. When the probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are equal, the probability of a real alarm occurring is updated to be equal to the first support probability corresponding to the real alarm, the probability of a false alarm occurring due to the calibration process is equal to the first support probability corresponding to the false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction is equal to the first support probability corresponding to the false alarm occurring due to sensor interference or malfunction.

[0048] In summary, the coal mine safety monitoring and alarm analysis system based on data fusion proposed in this embodiment improves the analysis efficiency while significantly enhancing the accuracy and reliability of alarm cause identification.

[0049] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0050] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0051] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for analyzing coal mine safety monitoring alarms based on data fusion, characterized in that, The method includes: Establish an identification framework for characterizing the causes of alarms; Acquire the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor; The first support probability of various alarm causes in the recognition framework is determined based on the alarm sensor's own data. The second support probability of various alarm causes in the identification framework is determined based on data from similar sensors located upstream and downstream of the alarm sensor. The third support probability of various alarm causes in the identification framework is determined based on data from different types of sensors located in the same area as the alarm sensor. Based on evidence theory, the first support probability, second support probability, and third support probability of the various alarm causes are fused to obtain the occurrence probability of each alarm cause, and the cause of the alarm event is judged based on the occurrence probability.

2. The method as described in claim 1, characterized in that, The various alarm causes in the identification framework include: False alarms include actual alarms, false alarms caused by calibration, and false alarms caused by sensor interference or malfunction.

3. The method as described in claim 2, characterized in that, The step of determining the first support probability of various alarm causes in the recognition framework based on the alarm sensor's own data includes: Based on the alarm sensor's own data at each moment in the historical period, and based on the own data, it is determined whether the data change process of the alarm sensor is normal; When the data change process of the alarm sensor is abnormal, it is determined that the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. When the data change process of the alarm sensor is normal, the preliminary alarm cause of the alarm sensor is determined according to the preset standard gas concentration used for sensor calibration, and the first support probability of each alarm cause in the identification framework is determined according to the preliminary alarm cause of the alarm sensor.

4. The method as described in claim 3, characterized in that, The step of determining whether the data change process of the alarm sensor is normal based on its own data includes: The maximum and minimum values ​​of the alarm sensor's own data during the historical working period are determined based on the alarm sensor's own data at each moment within the historical period. Based on the self-data at each moment within the historical period, the rising period and the falling period of self-data are found. In the rising period of self-data, the first self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the first self-data is determined. In the falling period of self-data, the second self-data corresponding to the minimum difference between the maximum value of self-data in the historical working period is found, and then the time corresponding to the second self-data is determined. The time period consisting of the time corresponding to the first self-data and the time corresponding to the second self-data is taken as the time period in which the abnormal data exists; Determine whether, within the time period in which the abnormal data exists, the absolute value of the difference between any two adjacent moments of its own data is greater than or equal to 90% (X). max -x max If so, the data change process of the alarm sensor is determined to be abnormal, and the alarm cause of the alarm sensor is a false alarm caused by interference or malfunction of the sensor. The first support probability corresponding to the real alarm is 0, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by interference or malfunction of the sensor is 1. Otherwise, the data change process of the alarm sensor is determined to be normal. max x is the maximum value of the alarm sensor during the time period in which the abnormal data exists. max This represents the maximum value of the alarm sensor's own data during its historical operating period.

5. The method as described in claim 4, characterized in that, The step of determining the preliminary alarm cause of the alarm sensor based on a preset standard gas concentration used for sensor calibration, and determining the first support probability of various alarm causes in the identification framework based on the preliminary alarm cause of the alarm sensor, includes: Determine whether ρ*(1-10%)≤X is satisfied. max ≤ρ*(1+10%) and t q -t p If the condition is satisfied (≥60s), the initial alarm cause of the alarm sensor is determined to be a false alarm caused by the calibration process and / or a real alarm. The first support probability corresponding to the real alarm is 0.2, the first support probability corresponding to the false alarm caused by the calibration process is 0.8, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. If the condition is not satisfied, the initial alarm cause of the alarm sensor is determined to be a real alarm. The first support probability corresponding to the real alarm is 1, the first support probability corresponding to the false alarm caused by the calibration process is 0, and the first support probability corresponding to the false alarm caused by sensor interference or malfunction is 0. Where ρ is the preset standard gas concentration used for sensor calibration, and t q For the first time during a period of decline in its own data that the value is less than X. max *95% of the corresponding time, t p For the first time during the rise of its own data within a historical period, the value is greater than or equal to X. max *95% corresponds to the time.

6. The method as described in claim 5, characterized in that, The step of determining the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor includes: Analyze the data of the same type of sensor upstream and downstream of the alarm sensor at each moment within a first preset time period to determine whether calibration data exists in the data of the same type of sensor at each moment within the first preset time period. If it exists, the second support probability corresponding to the real alarm is 0, the second support probability corresponding to the false alarm caused by the calibration process is 1, and the second support probability corresponding to the false alarm caused by sensor interference or failure is 0. If it does not exist, the second support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the second support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the second support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

7. The method as described in claim 6, characterized in that, The determination of the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor includes: When the alarm sensor is a methane sensor, the airflow sensor data of the duct is judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure. When the alarm sensor is a carbon monoxide sensor, the temperature sensor data and / or smoke sensor data are judged to be abnormal based on the data of different types of sensors in the same area as the alarm sensor. If abnormal, the third support probability corresponding to the real alarm is 1, the third support probability corresponding to the false alarm caused by the calibration process is 0, and the third support probability corresponding to the false alarm caused by sensor interference or failure is 0. Otherwise, the third support probability corresponding to the real alarm is equal to the first support probability corresponding to the real alarm, the third support probability corresponding to the false alarm caused by the calibration process is equal to the first support probability corresponding to the false alarm caused by the calibration process, and the third support probability corresponding to the false alarm caused by sensor interference or failure is equal to the first support probability corresponding to the false alarm caused by sensor interference or failure.

8. The method as described in claim 7, characterized in that, The first, second, and third support probabilities of the various alarm causes are fused based on evidence theory to obtain the probability of occurrence of each alarm cause, including: Using formula Determine the probability of a real alarm occurring; Using formula Determine the probability of false alarms caused by the calibration process; Using formula To determine the probability of false alarms caused by sensor interference or malfunction; in, This represents the actual probability of an alarm occurring. The first support probability corresponding to a real alarm. The second support probability corresponding to a real alarm. The third support probability corresponding to a real alarm. This represents the probability of false alarms occurring during the calibration process. The first support probability corresponds to the false alarm caused by the calibration process. The second support probability corresponds to the false alarms caused by the calibration process. The third support probability corresponds to false alarms caused by the calibration process. This represents the probability of false alarms caused by sensor interference or malfunction. The first support probability corresponds to false alarms caused by sensor interference or malfunction. This represents the second support probability corresponding to false alarms caused by sensor interference or malfunction. The third support probability corresponds to false alarms caused by sensor interference or malfunction. As the first parameter, For the second parameter, , .

9. The method as described in claim 8, characterized in that, The step of analyzing the cause of the alarm event based on the probability of occurrence includes: The probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are sorted from left to right from largest to smallest to form a first sequence, and the cause corresponding to the first value on the left in the first sequence is taken as the cause of the alarm event. When the probability of a real alarm occurring, the probability of a false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction are equal, the probability of a real alarm occurring is updated to be equal to the first support probability corresponding to the real alarm, the probability of a false alarm occurring due to the calibration process is equal to the first support probability corresponding to the false alarm occurring due to the calibration process, and the probability of a false alarm occurring due to sensor interference or malfunction is equal to the first support probability corresponding to the false alarm occurring due to sensor interference or malfunction.

10. A coal mine safety monitoring and alarm analysis system based on data fusion, characterized in that, The system includes: Establish a module to build an identification framework for characterizing the causes of alarms; The acquisition module is used to acquire the alarm sensor's own data corresponding to the alarm event, the data of the same type of sensor located upstream and downstream of the alarm sensor, and the data of different types of sensors located in the same area as the alarm sensor. The first determining module is used to determine the first support probability of various alarm causes in the identification framework based on the alarm sensor's own data. The second determining module is used to determine the second support probability of various alarm causes in the identification framework based on data from similar sensors located upstream and downstream of the alarm sensor. The third determining module is used to determine the third support probability of various alarm causes in the identification framework based on data from different types of sensors located in the same area as the alarm sensor. The analysis module is used to fuse the first support probability, second support probability, and third support probability of the various alarm causes based on evidence theory to obtain the occurrence probability of each alarm cause, and to analyze the cause of the alarm event based on the occurrence probability.

Images