A fire-fighting equipment remote control optimization method and system

Abstract

The present invention relates to the field of fire warning, and specifically to a remote control optimization method and system for fire fighting equipment. A remote control optimization system for fire fighting equipment includes: a regional interference analysis module, a fire condition analysis module and a remote visualization operation module. The present invention independently evaluates the interference degree of smoke, temperature and gas, accurately quantifies the environmental interference factors of each monitoring area, and provides a more reliable data basis for fire condition analysis; this quantitative evaluation allows the fire condition analysis model to focus more on useful information related to the fire condition when processing data, while reducing the interference effect of non-fire factors, so that the fire condition analysis model can more intelligently distinguish between real fire conditions and environmental interference, thereby significantly improving the accuracy and reliability of fire warnings, and providing technical guarantees for accurate fire monitoring and rapid response.

Description

Technical Field

[0001] The present invention relates to the field of fire warning, and in particular to a fire-fighting equipment remote control optimization method and system. Background Art

[0002] With the acceleration of urbanization and the increase in building density, fire safety issues have become increasingly prominent; traditional fire protection systems have certain limitations in fire warning and handling, especially in remote monitoring and linkage control; in the field of fire monitoring, smoke, temperature and gas sensors are commonly used fire detection methods; however, these sensors may be interfered by environmental factors, resulting in false alarms or missed alarms, especially in more complex industrial scenarios; in addition, existing fire protection equipment often lacks sufficient intelligence in data processing and analysis, making it difficult to accurately judge the true situation of the fire.

[0003] In order to improve the accuracy and response speed of fire warning, a fire protection system is needed that can comprehensively consider multiple factors such as smoke, temperature, gas, etc., and perform remote control optimization; In response to the above problems, the present invention proposes a fire protection equipment remote control optimization method and system, which aims to improve the accuracy of fire warning and realize remote control optimization by comprehensively analyzing multiple interference factors such as smoke, temperature, gas, etc. Summary of the invention

[0004] The present invention independently evaluates the interference degree of smoke, temperature and gas, accurately quantifies the environmental interference factors of each monitoring area, and provides a more reliable data basis for fire analysis. This quantitative evaluation allows the fire analysis model to focus more on useful information related to the fire when processing data, while reducing the interference effect of non-fire factors, so that the fire analysis model can more intelligently distinguish between real fire and environmental interference, thereby significantly improving the accuracy and reliability of fire warnings, and providing technical support for accurate fire monitoring and rapid response.

[0005] A fire-fighting equipment remote control optimization method, comprising:

[0006] For any local monitoring area among several local monitoring areas, at the beginning of the current preset time period, the smoke interference degree, temperature interference degree and gas interference degree of the local monitoring area are calculated and obtained; based on the obtained smoke interference degree, temperature interference degree and gas interference degree, the smoke confidence degree, temperature confidence degree and gas confidence degree of the local monitoring area are calculated and obtained, and several analysis time points are evenly set for the local monitoring area within the current preset time period;

[0007] For any local monitoring area, the smoke data, temperature data, gas data and flame data of the local monitoring area are collected at the current monitoring time point. If the flame data collected at the current monitoring time point indicates that a flame exists, a fire alarm for the local monitoring area is issued; if the flame data collected at the current monitoring time point indicates that a flame does not exist, no operation is performed; at the current analysis time point, the smoke data, temperature data and gas data collected at the most recent Q monitoring time points are respectively organized into a smoke time series data set, a temperature time series data set and a gas time series data set in chronological order; a fire analysis model is constructed, and the smoke time series data set, temperature time series data set and gas time series data set obtained at the current analysis time point as well as the smoke confidence, temperature confidence and gas confidence are used as the input of the fire analysis model, and the fire analysis result is output; if the fire analysis result indicates that a fire exists, a fire alarm for the local monitoring area is issued; if the fire analysis result indicates that a fire does not exist, no operation is performed;

[0008] When the central control device receives a fire alarm in any local monitoring area, remote visualization technology is used to remotely operate several buttons of the central control device to control the linked fire-fighting equipment distributed in each local monitoring area. The linked fire-fighting equipment includes a smoke exhaust device, a gas fire extinguishing device, and an audible and visual alarm device.

[0009] Preferably, the specific operations for calculating the smoke interference degree, the temperature interference degree and the gas interference degree are as follows:

[0010] Calculation of smoke interference level: based on the data collected within the latest preset time period Smoke data , =1, 2, …, ; Using the formula Calculate the smoke standard deviation ,in, for Smoke data The average value of smoke standard deviation is set to 0 to , and set the baseline smoke level threshold to record Smoke data The number of smoke levels above the baseline threshold ; Using the formula Calculate the degree of smoke interference ;

[0011] Calculation of temperature interference degree: based on the data collected within the latest preset time period Temperature data , using the formula Calculate the average temperature change rate; set the temperature change rate limit range to arrive , and set the reference temperature threshold to record Temperature data The number of ; Using the formula Calculate the degree of temperature interference ;

[0012] Calculation of gas interference level: based on the target detection gas data collected within the most recent preset time period , using the formula Calculate the gas standard deviation ,in, for Target detection gas data The gas standard deviation is set to a value between 0 and , and set the baseline gas level threshold to record Target detection gas data The number of gas levels above the baseline threshold ; Using the formula Calculate the gas interference level .

[0013] Preferably, the specific operations for calculating the smoke confidence, temperature confidence and gas confidence are as follows:

[0014] Using the formula Calculate the smoke confidence; use the formula Calculate the temperature confidence; use the formula Calculate the gas confidence.

[0015] Preferably, the specific operations for setting the analysis time point for each local monitoring area are as follows:

[0016] Set the minimum analysis interval and the highest analysis interval , for any local monitoring area, based on the currently acquired smoke interference level of the local monitoring area , Temperature interference degree and gas interference level , using the formula Calculate the comprehensive interference level of the local monitoring area ; Then use the formula Calculate the time interval between every two analysis time points applied to the local monitoring area within the current preset time period .

[0017] Preferably, the fire situation analysis model is established based on the LSTM model, including an input layer, a feature extraction layer, a feature fusion layer, a fully connected layer and an output layer;

[0018] The input layer is used to receive smoke time series data set, temperature time series data set and gas time series data set as well as smoke confidence, temperature confidence and gas confidence;

[0019] The feature extraction layer includes three parallel first LSTM layers, second LSTM layers, and third LSTM layers. The first LSTM layer is used to extract the temporal features of the smoke time series data set and obtain the smoke time series feature vector; the second LSTM layer is used to extract the temporal features of the temperature time series data set and obtain the temperature time series feature vector; the third LSTM layer is used to extract the temporal features of the gas time series data set and obtain the gas time series feature vector;

[0020] The feature fusion layer is used to use the smoke confidence, temperature confidence and gas confidence as the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector respectively, and perform weighted fusion on the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector to obtain a fused time series feature vector;

[0021] The fully connected layer is used to further extract features from the fused time series feature vector;

[0022] The output layer is used to output the fire situation analysis results, which include abnormal or normal.

[0023] Preferably, the specific operations for training the fire situation analysis model are as follows:

[0024] Obtain several fire analysis training samples with labeled fire analysis results, each fire analysis training sample contains a set of historical smoke time series data sets, historical temperature time series data sets and historical gas time series data sets as well as smoke confidence, temperature confidence and gas confidence; divide all fire analysis training samples into training set and verification set; use the training set to train the fire analysis model with initialized parameters, and then use the verification set to verify the fire analysis model to obtain the verification result; set the training condition, and judge whether the verification result meets the training condition. If so, output the trained fire analysis model; if not, continue to train the fire analysis model with the training set.

[0025] Preferably, when the central control device receives a fire alarm in any local monitoring area, the remote visualization technology is used to perform remote key operations on several keys of the central control device. The specific operations are as follows:

[0026] Use the camera to obtain a high-definition image of the fire control panel of the central control device, apply computer vision technology to identify the key area, and determine the center coordinates of each key; then record the function and coordinate information corresponding to each key and store it in the database; set the reference position of the robotic arm, and apply visual calibration technology to ensure the accurate correspondence between the robotic arm and the key coordinates;

[0027] View the real-time image of the fire control panel transmitted by the camera through the remote client; the client interface includes each button and the corresponding function label; click the button to be operated, and according to the selected button function, retrieve the corresponding button coordinates from the database, generate the robot arm movement command, and apply the robot arm movement command to control the robot arm to perform button operations.

[0028] A fire-fighting equipment remote control optimization system, comprising:

[0029] The regional interference analysis module includes an interference degree calculation unit, a confidence calculation unit and an analysis frequency adjustment unit; the interference degree calculation unit is used to calculate and obtain the smoke interference degree, temperature interference degree and gas interference degree of each local monitoring area at the beginning of any preset time period; the confidence calculation unit is used to calculate and obtain the smoke confidence, temperature confidence and gas confidence of each local monitoring area by using the smoke interference degree, temperature interference degree and gas interference degree obtained for each local monitoring area; the analysis frequency adjustment unit is used to evenly set a number of analysis time points for each local monitoring area within the current preset time period;

[0030] The fire analysis module includes a data acquisition unit, a first judgment unit, a fire analysis unit and a second judgment unit; the data acquisition unit is used to collect smoke data, temperature data, gas data and flame data of any local monitoring area at the current monitoring time point; the first judgment unit is used to judge whether the flame data collected at the current monitoring time point indicates the existence of flame, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed; the fire analysis unit is used to form a smoke time series data set, a temperature time series data set and a gas time series data set in chronological order at the current analysis time point; the smoke time series data set, the temperature time series data set and the gas time series data set obtained at the current analysis time point as well as the smoke confidence, the temperature confidence and the gas confidence are used as inputs of the fire analysis model, and the fire analysis result is output; the second judgment unit is used to judge whether the fire analysis result indicates the existence of fire, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed;

[0031] The remote visualization operation module is used to use remote visualization technology to operate several buttons of the central control device when the central control device receives a fire alarm from any local monitoring area, thereby controlling the linked fire-fighting equipment distributed in each local monitoring area.

[0032] The present invention has the following advantages:

[0033] The present invention independently evaluates the interference degree of smoke, temperature and gas, accurately quantifies the environmental interference factors of each monitoring area, and provides a more reliable data basis for fire analysis. This quantitative evaluation allows the fire analysis model to focus more on useful information related to the fire when processing data, while reducing the interference effect of non-fire factors, so that the fire analysis model can more intelligently distinguish between real fire and environmental interference, thereby significantly improving the accuracy and reliability of fire warnings, and providing technical support for accurate fire monitoring and rapid response.

[0034] The present invention provides a remote visualization operation module, which allows operators to monitor and operate the fire control panel in real time through a remote client; by utilizing computer vision technology and precise control of a robotic arm, the system can quickly execute preset fire equipment operations when a fire alarm is triggered; this remote control method not only improves the response speed, but also reduces the risk of on-site operations, making the management of fire equipment more flexible and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the structure of the fire-fighting equipment remote control optimization system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0037] Embodiment 1, a fire-fighting equipment remote control optimization method, comprising:

[0038] For any local monitoring area among several local monitoring areas, at the beginning of the current preset time period, the smoke interference degree, temperature interference degree and gas interference degree of the local monitoring area are calculated and obtained. The smoke interference degree, temperature interference degree and gas interference degree are used to evaluate the fluctuation or abnormality of the smoke sensor data, temperature sensor data and gas sensor data in the specific local monitoring area respectively; by quantifying these interference degrees, the system can identify and distinguish between normal environmental changes and possible signs of fire; based on the obtained smoke interference degree, temperature interference degree and gas interference degree, the smoke confidence, temperature confidence and gas confidence of the local monitoring area are calculated and obtained. The confidence provides a quantitative indicator to help the system judge the reliability of sensor data, so as to give appropriate attention in fire analysis; several analysis time points are evenly set for the local monitoring area within the current preset time period;

[0039] In industrial scenarios, smoke interference may be caused by dust, steam or smoke generated in the daily operation of the factory, which may cause the fire alarm system to misjudge. Temperature interference may come from the heat dissipation of mechanical equipment, heat sources in the production process or changes in external climate conditions. These factors may cause fluctuations in temperature sensors, thereby misleading the fire detection system. Gas interference may be caused by chemicals used in industrial production, leaked gases or the operation of ventilation systems. These gases may be detected by gas sensors and cause false alarms. The existence of these interference factors may cause the fire monitoring system to mistakenly judge it as a fire when there is no actual fire, thereby increasing the risk of false alarms. Therefore, the present invention provides more representative data for fire analysis by quantifying these interference factors.

[0040] For any local monitoring area, the smoke data, temperature data, gas data and flame data of the local monitoring area are collected at the current monitoring time point. Among them, the flame data provides the most direct fire indication, allowing the system to quickly respond to possible fire situations. If the flame data collected at the current monitoring time point shows that there is a flame, a fire alarm for the local monitoring area is issued; if the flame data collected at the current monitoring time point shows that there is no flame, no operation is performed; at the current analysis time point, the smoke data, temperature data and gas data collected at the most recent Q monitoring time points are respectively organized into a smoke time series data set, a temperature time series data set and a gas time series data set in chronological order; a fire analysis model is constructed, and the smoke time series data set, temperature time series data set and gas time series data set obtained at the current analysis time point as well as the smoke confidence, temperature confidence and gas confidence are used as the fire The fire analysis model is used as the input, and the fire analysis result is output; if the fire analysis result is that there is a fire, a fire alarm is issued for the local monitoring area; if the fire analysis result is that there is no fire, no operation is performed; the purpose of this paragraph is to establish an efficient fire judgment mechanism: flame data, as a direct indicator, can instantly reflect the core characteristics of a fire, so it can be quickly used to trigger a fire alarm and achieve a rapid response; smoke, temperature and gas data are easily affected by non-fire factors, so they need to be deeply analyzed by building a fire analysis model to distinguish between real fires and environmental interference to ensure the accuracy of judgment; at the same time, considering the direct judgment of flame data and the analysis of multi-parameter models, it not only ensures the response speed, but also improves the accuracy of judgment, thereby achieving a balance between timeliness and accuracy, and significantly improving the effectiveness of the fire monitoring system;

[0041] When the central control device receives a fire alarm in any local monitoring area, the remote visualization technology is used to remotely operate several buttons of the central control device, thereby controlling the linked fire-fighting equipment distributed in each local monitoring area, and the linked fire-fighting equipment includes a smoke exhaust device, a gas fire extinguishing device and an audible and visual alarm device; as described above, the present invention provides an efficient, flexible and responsive fire response mechanism; when the central control device receives a fire alarm, the remote visualization technology enables the operator to view the fire control panel in real time through high-definition images, and use computer vision technology to accurately identify and operate related buttons; this technology allows the operator to remotely control the linked fire-fighting equipment such as the smoke exhaust device, the gas fire extinguishing device and the audible and visual alarm device, so as to take quick action when a fire occurs, effectively control and extinguish the fire, and reduce the losses caused by the fire; in addition, remote operation also reduces the safety risks of on-site personnel and improves the efficiency and accuracy of fire response.

[0042] The specific operations for calculating the smoke interference degree, temperature interference degree and gas interference degree are as follows:

[0043] Calculation of smoke interference level: based on the data collected within the latest preset time period Smoke data , =1, 2, …, ; Using the formula Calculate the smoke standard deviation ,in, for Smoke data The average value of smoke standard deviation is set to 0 to , and set the baseline smoke level threshold to record Smoke data The number of smoke levels above the baseline threshold ; Using the formula Calculate the degree of smoke interference ;

[0044] Calculation of temperature interference degree: based on the data collected within the latest preset time period Temperature data , using the formula Calculate the average temperature change rate; set the temperature change rate limit range to arrive , and set the reference temperature threshold to record Temperature data The number of ; Using the formula Calculate the degree of temperature interference ;

[0045] Calculation of gas interference level: based on the target detection gas data collected within the most recent preset time period , using the formula Calculate the gas standard deviation ,in, for Target detection gas data The gas standard deviation is set to a value between 0 and , and set the baseline gas level threshold to record Target detection gas data The number of gas levels above the baseline threshold ; Using the formula Calculate the gas interference level .

[0046] The specific operations for calculating smoke confidence, temperature confidence, and gas confidence are as follows:

[0047] Using the formula Calculate the smoke confidence; use the formula Calculate the temperature confidence; use the formula Calculate the gas confidence.

[0048] The specific operations for setting the analysis time point for each local monitoring area are as follows:

[0049] Set the minimum analysis interval and the highest analysis interval , for any local monitoring area, based on the currently acquired smoke interference level of the local monitoring area , Temperature interference degree and gas interference level , using the formula

[0050] Calculate the comprehensive interference level of the local monitoring area ; Then use the formula Calculate the time interval between every two analysis time points applied to the local monitoring area within the current preset time period .

[0051] The fire analysis model is built based on the LSTM model, including input layer, feature extraction layer, feature fusion layer, fully connected layer and output layer;

[0052] The input layer is used to receive smoke time series data set, temperature time series data set and gas time series data set as well as smoke confidence, temperature confidence and gas confidence;

[0053] The feature extraction layer includes three parallel first LSTM layers, second LSTM layers, and third LSTM layers. The first LSTM layer is used to extract the temporal features of the smoke time series data set and obtain the smoke time series feature vector; the second LSTM layer is used to extract the temporal features of the temperature time series data set and obtain the temperature time series feature vector; the third LSTM layer is used to extract the temporal features of the gas time series data set and obtain the gas time series feature vector;

[0054] The feature fusion layer is used to use the smoke confidence, temperature confidence and gas confidence as the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector respectively, and perform weighted fusion on the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector to obtain a fused time series feature vector;

[0055] The fully connected layer is used to further extract features from the fused time series feature vector;

[0056] The output layer is used to output the fire situation analysis results, which include abnormal or normal.

[0057] The specific operations for training the fire situation analysis model are as follows:

[0058] Obtain several fire analysis training samples with labeled fire analysis results, each fire analysis training sample contains a set of historical smoke time series data sets, historical temperature time series data sets and historical gas time series data sets as well as smoke confidence, temperature confidence and gas confidence; divide all fire analysis training samples into training set and verification set; use the training set to train the fire analysis model with initialized parameters, and then use the verification set to verify the fire analysis model to obtain the verification result; set the training condition, and judge whether the verification result meets the training condition. If so, output the trained fire analysis model; if not, continue to train the fire analysis model with the training set.

[0059] When the central control device receives a fire alarm in any local monitoring area, the remote visualization technology is used to remotely operate several buttons of the central control device. The specific operations are as follows:

[0060] Use the camera to obtain a high-definition image of the fire control panel of the central control device, apply computer vision technology to identify the key area, and determine the center coordinates of each key; then record the function and coordinate information corresponding to each key and store it in the database; set the reference position of the robotic arm, and apply visual calibration technology to ensure the accurate correspondence between the robotic arm and the key coordinates;

[0061] View the real-time image of the fire control panel transmitted by the camera through the remote client; the client interface includes each button and the corresponding function label; click the button to be operated, and according to the selected button function, retrieve the corresponding button coordinates from the database, generate the robot arm movement command, and apply the robot arm movement command to control the robot arm to perform button operations.

[0062] Embodiment 2, a fire equipment remote control optimization system, such as Figure 1 As shown, including:

[0063] The regional interference analysis module includes an interference degree calculation unit, a confidence calculation unit and an analysis frequency adjustment unit; the interference degree calculation unit is used to calculate and obtain the smoke interference degree, temperature interference degree and gas interference degree of each local monitoring area at the beginning of any preset time period; the confidence calculation unit is used to calculate and obtain the smoke confidence, temperature confidence and gas confidence of each local monitoring area by using the smoke interference degree, temperature interference degree and gas interference degree obtained for each local monitoring area; the analysis frequency adjustment unit is used to evenly set a number of analysis time points for each local monitoring area within the current preset time period;

[0064] The fire analysis module includes a data acquisition unit, a first judgment unit, a fire analysis unit and a second judgment unit; the data acquisition unit is used to collect smoke data, temperature data, gas data and flame data of any local monitoring area at the current monitoring time point; the first judgment unit is used to judge whether the flame data collected at the current monitoring time point indicates the existence of flame, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed; the fire analysis unit is used to form a smoke time series data set, a temperature time series data set and a gas time series data set in chronological order at the current analysis time point; the smoke time series data set, the temperature time series data set and the gas time series data set obtained at the current analysis time point as well as the smoke confidence, the temperature confidence and the gas confidence are used as inputs of the fire analysis model, and the fire analysis result is output; the second judgment unit is used to judge whether the fire analysis result indicates the existence of fire, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed;

[0065] The remote visualization operation module is used to use remote visualization technology to operate several buttons of the central control device when the central control device receives a fire alarm from any local monitoring area, thereby controlling the linked fire-fighting equipment distributed in each local monitoring area.

[0066] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A fire equipment remote control optimization method, characterized in that: include: For any local monitoring area among several local monitoring areas, at the beginning of the current preset time period, the smoke interference degree, temperature interference degree and gas interference degree of the local monitoring area are calculated and obtained; based on the obtained smoke interference degree, temperature interference degree and gas interference degree, the smoke confidence degree, temperature confidence degree and gas confidence degree of the local monitoring area are calculated and obtained, and several analysis time points are evenly set for the local monitoring area within the current preset time period; For any local monitoring area, the smoke data, temperature data, gas data and flame data of the local monitoring area are collected at the current monitoring time point. If the flame data collected at the current monitoring time point indicates that there is a flame, a fire alarm for the local monitoring area is issued; If the flame data collected at the current monitoring time point shows that there is no flame, no operation is performed; at the current analysis time point, the smoke data, temperature data, and gas data collected at the most recent Q monitoring time points are respectively combined into a smoke time series data set, a temperature time series data set, and a gas time series data set in chronological order; Construct a fire situation analysis model, use the smoke time series data set, temperature time series data set and gas time series data set obtained at the current analysis time point as well as the smoke confidence, temperature confidence and gas confidence as the input of the fire situation analysis model, and output the fire situation analysis results; If the fire situation analysis result indicates that a fire exists, a fire alarm is issued for the local monitoring area; If the fire analysis result shows that there is no fire, no operation is performed; When the central control device receives a fire alarm in any local monitoring area, remote visualization technology is used to remotely operate several buttons of the central control device to control the linked fire-fighting equipment distributed in each local monitoring area. The linked fire-fighting equipment includes a smoke exhaust device, a gas fire extinguishing device, and an audible and visual alarm device.

2. A fire equipment remote control optimization method according to claim 1, characterized in that: The specific operations for calculating the smoke interference degree, temperature interference degree and gas interference degree are as follows: Calculation of smoke interference level: based on the data collected within the latest preset time period Smoke data , =1, 2, …, ; Using the formula Calculate the smoke standard deviation ,in, for Smoke data The average value of smoke standard deviation is set to 0 to , and set the baseline smoke level threshold to record Smoke data The number of smoke levels above the baseline threshold ; Using the formula Calculate the degree of smoke interference ; Calculation of temperature interference degree: based on the data collected within the latest preset time period Temperature data , using the formula Calculate the average temperature change rate; set the temperature change rate limit range to arrive , and set the reference temperature threshold to record Temperature data The number of ; Using the formula Calculate the degree of temperature interference ; Calculation of gas interference level: based on the target detection gas data collected within the most recent preset time period , using the formula Calculate the gas standard deviation ,in, for Target detection gas data The gas standard deviation is set to a value between 0 and , and set the baseline gas level threshold to record Target detection gas data The number of gas levels above the baseline threshold ; Using the formula Calculate the gas interference level .

3. A fire-fighting equipment remote control optimization method according to claim 2, characterized in that: The specific operations for calculating smoke confidence, temperature confidence, and gas confidence are as follows: Using the formula Calculate the smoke confidence; use the formula Calculate the temperature confidence; use the formula Calculate the gas confidence.

4. A fire-fighting equipment remote control optimization method according to claim 3, characterized in that: The specific operations for setting the analysis time point for each local monitoring area are as follows: Set the minimum analysis interval and the highest analysis interval , for any local monitoring area, based on the currently acquired smoke interference level of the local monitoring area , Temperature interference degree and gas interference level , using the formula Calculate the comprehensive interference level of the local monitoring area ; Then use the formula Calculate the time interval between every two analysis time points applied to the local monitoring area within the current preset time period .

5. A fire-fighting equipment remote control optimization method according to claim 4, characterized in that: The fire analysis model is built based on the LSTM model, including input layer, feature extraction layer, feature fusion layer, fully connected layer and output layer; The input layer is used to receive smoke time series data set, temperature time series data set and gas time series data set as well as smoke confidence, temperature confidence and gas confidence; The feature extraction layer includes three parallel first LSTM layers, second LSTM layers, and third LSTM layers. The first LSTM layer is used to extract the temporal features of the smoke time series data set and obtain the smoke time series feature vector; The second LSTM layer is used to extract the temporal characteristics of the temperature time series data set and obtain the temperature time series feature vector; the third LSTM layer is used to extract the temporal characteristics of the gas time series data set and obtain the gas time series feature vector; The feature fusion layer is used to use the smoke confidence, temperature confidence and gas confidence as the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector respectively, and perform weighted fusion on the weights of the smoke time series feature vector, temperature time series feature vector and gas time series feature vector to obtain a fused time series feature vector; The fully connected layer is used to further extract features from the fused time series feature vector; The output layer is used to output the fire situation analysis results, which include abnormal or normal.

6. A fire-fighting equipment remote control optimization method according to claim 5, characterized in that: The specific operations for training the fire situation analysis model are as follows: Obtain several fire analysis training samples with annotated fire analysis results, each of which contains a set of historical smoke time series data sets, historical temperature time series data sets, and historical gas time series data sets, as well as smoke confidence, temperature confidence, and gas confidence; divide all fire analysis training samples into a training set and a validation set; The fire analysis model with initialized parameters is trained using the training set, and then the fire analysis model is verified using the verification set to obtain the verification results; the training conditions are set to determine whether the verification results meet the training conditions, and if so, the trained fire analysis model is output; if not, the fire analysis model is continued to be trained using the training set.

7. A fire-fighting equipment remote control optimization method according to claim 6, characterized in that: When the central control device receives a fire alarm in any local monitoring area, the remote visualization technology is used to remotely operate several buttons of the central control device. The specific operations are as follows: Use the camera to obtain a high-definition image of the fire control panel of the central control device, apply computer vision technology to identify the key area, and determine the center coordinates of each key; then record the function and coordinate information corresponding to each key and store it in the database; set the reference position of the robotic arm, and apply visual calibration technology to ensure the accurate correspondence between the robotic arm and the key coordinates; View the real-time image of the fire control panel transmitted by the camera through the remote client; the client interface includes each button and the corresponding function label; click the button to be operated, and according to the selected button function, retrieve the corresponding button coordinates from the database, generate the robot arm movement command, and apply the robot arm movement command to control the robot arm to perform button operations.

8. A fire equipment remote control optimization system, characterized in that: The system is applied to a fire-fighting equipment remote control optimization method as described in any one of claims 1 to 7, comprising: The regional interference analysis module includes an interference degree calculation unit, a confidence calculation unit and an analysis frequency adjustment unit; the interference degree calculation unit is used to calculate and obtain the smoke interference degree, temperature interference degree and gas interference degree of each local monitoring area at the beginning of any preset time period; the confidence calculation unit is used to calculate and obtain the smoke confidence, temperature confidence and gas confidence of each local monitoring area by using the smoke interference degree, temperature interference degree and gas interference degree obtained for each local monitoring area; the analysis frequency adjustment unit is used to evenly set a number of analysis time points for each local monitoring area within the current preset time period; The fire analysis module includes a data acquisition unit, a first judgment unit, a fire analysis unit and a second judgment unit; the data acquisition unit is used to collect smoke data, temperature data, gas data and flame data of any local monitoring area at the current monitoring time point; the first judgment unit is used to judge whether the flame data collected at the current monitoring time point indicates the existence of flame, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed; the fire analysis unit is used to form a smoke time series data set, a temperature time series data set and a gas time series data set in chronological order at the current analysis time point; the smoke time series data set, the temperature time series data set and the gas time series data set obtained at the current analysis time point as well as the smoke confidence, the temperature confidence and the gas confidence are used as inputs of the fire analysis model, and the fire analysis result is output; the second judgment unit is used to judge whether the fire analysis result indicates the existence of fire, and if so, a fire alarm for the local monitoring area is issued; if not, no operation is performed; The remote visualization operation module is used to use remote visualization technology to operate several buttons of the central control device when the central control device receives a fire alarm from any local monitoring area, thereby controlling the linked fire-fighting equipment distributed in each local monitoring area.

Images