Intelligent fire door control method, system, device and medium based on artificial intelligence

Abstract

The present invention relates to an artificial intelligence-based intelligent fire door control method, system, device, and medium. The method comprises: acquiring device data and sensor data of the intelligent fire door in real time, extracting features from the sensor data to obtain a feature data set; evaluating and processing environmental data based on the feature data set to obtain corresponding environmental assessment information; performing state analysis on the device data to obtain corresponding fire door status data; transmitting the environmental assessment information and fire door status data to a preset control model; performing command analysis on the fire door status data using the control model to obtain corresponding control commands; generating corresponding command tasks based on the environmental assessment information and the control commands; controlling the intelligent fire door according to the command tasks, obtaining feedback information from the intelligent fire door, and optimizing model parameters of the control model based on the feedback information to obtain an optimized control model. The present invention can accurately identify subtle signs of an early fire.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fire doors, and in particular to an artificial intelligence-based intelligent fire door control method, system, device and medium. Background Art

[0002] In modern buildings, intelligent fire doors, as a crucial firefighting feature, are widely used in various locations to improve fire safety. Traditional fire door control systems often rely on simple sensor signals to trigger switches, lacking the ability to intelligently analyze environmental changes and device status. Furthermore, existing intelligent fire doors have limited capabilities for collecting and analyzing environmental data, making them unable to accurately identify subtle signs of a fire in its early stages. Summary of the Invention

[0003] The main purpose of the present invention is to provide an artificial intelligence-based intelligent fire door control method, system, device and medium, which can improve the analysis ability of environmental data and accurately identify subtle signs of early fire.

[0004] To achieve the above objectives, the present invention provides an intelligent fire door control method based on artificial intelligence, comprising:

[0005] Acquire device data and sensor data of the intelligent fire door in real time, perform feature extraction on the sensor data, and obtain a feature data set;

[0006] Evaluate and process the environmental data according to the feature data set to obtain corresponding environmental evaluation information, and perform status analysis on the equipment data to obtain corresponding fire door status data;

[0007] Transmitting the environmental assessment information and the fire door status data to a preset control model, performing instruction analysis on the fire door status data through the control model to obtain corresponding control instructions, and generating corresponding instruction tasks based on the environmental assessment information and the control instructions;

[0008] The intelligent fire door is controlled according to the instruction task, and feedback information of the intelligent fire door is obtained. The model parameters of the control model are optimized according to the feedback information to obtain the optimized control model.

[0009] Furthermore, the real-time acquisition of device data and sensor data of the intelligent fire door, and feature extraction of the sensor data to obtain a feature data set include:

[0010] Acquire external raw data collected by external sensors deployed around the intelligent fire door in real time, and pre-process the external raw data to obtain the corresponding sensor data;

[0011] Obtaining original internal data of the intelligent fire door, and preprocessing the original internal data to obtain the device data;

[0012] Classifying the sensor data according to a preset classification rule to obtain corresponding environmental data, and performing initial feature calculation on the sensor data to obtain the feature data set;

[0013] The environmental data is calculated into environmental types to obtain smoke concentration data, temperature data, flame data and airflow number.

[0014] Furthermore, the evaluation process is performed based on the feature data set to obtain corresponding environmental evaluation information, including:

[0015] Reading the characteristic data set to obtain a first smoke concentration change rate, a first temperature rise rate, and a first thermal radiation change amount;

[0016] Performing a deviation analysis on the first smoke concentration change rate based on the smoke concentration data and the airflow data to obtain corresponding smoke deviation data, performing a deviation analysis on the first temperature rise rate based on the temperature data and the airflow data to obtain corresponding temperature deviation data, and performing a deviation analysis on the first thermal radiation change amount based on the temperature data, the flame data, and the airflow data to obtain corresponding thermal radiation deviation data;

[0017] Determine whether the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data meet the preset deviation threshold requirements; when any one of the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data does not meet the deviation threshold requirements, calculate the risk characteristics based on the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data in combination with the environmental data to obtain a corrected characteristic data set;

[0018] An environmental assessment is performed on the intelligent fire door based on the environmental data and the corrected feature data set to obtain the environmental assessment information.

[0019] Furthermore, the state analysis of the device data to obtain corresponding fire door state data includes:

[0020] Read the device data by type to obtain the working status of the fire door and the motor status data;

[0021] Perform status analysis on the working status of the fire door to obtain the open / close status and lock status of the fire door;

[0022] Perform health assessment on motor status data to obtain motor health status information;

[0023] The switch status, lock status and motor health status information are integrated to obtain the fire door status data.

[0024] Furthermore, the fire door status data is subjected to instruction analysis by the control model to obtain corresponding control instructions, and corresponding instruction tasks are generated based on the environmental assessment information and the control instructions, including:

[0025] Performing format conversion on the input environmental assessment information and the fire door status data through the input layer of the control model to obtain an environmental assessment matrix, the switch status, the locking status, and the motor health status information;

[0026] Comprehensively analyzing the switch state, the lock state, and the motor health status information through the double-layer convolutional layer and the recurrent neural network of the control model to obtain corresponding instruction features, performing instruction matching based on the instruction features to obtain corresponding control instructions, and performing task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task;

[0027] The instruction task is outputted through the output layer of the control model.

[0028] Furthermore, the control model performs a comprehensive analysis of the switch state, the lock state, and the motor health status information through the double-layer convolutional layer and the recurrent neural network to obtain corresponding instruction features, performs instruction matching based on the instruction features to obtain corresponding control instructions, and performs task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task, including:

[0029] Inputting the switch state and the lock state into a first convolutional layer, performing feature recognition on the switch state and the lock state through the first convolutional layer to obtain corresponding fire door control features;

[0030] Inputting the motor health status information into the second convolutional layer, performing feature recognition on the motor health status information through the second convolutional layer to obtain corresponding fire door motor features;

[0031] Inputting the environmental assessment matrix into a recurrent neural network, performing risk assessment on the environmental assessment matrix through the recurrent neural network, and obtaining a corresponding environmental risk level;

[0032] The fire door control features and the fire door motor features are input into the activation function layer, and the fire door control features and the fire door motor features are nonlinearly processed by the activation function layer to obtain the corresponding nonlinear fire door control features and nonlinear fire door motor features;

[0033] The nonlinear fire door control features and the nonlinear fire door motor features are input into the pooling layer, and the nonlinear fire door control features and the nonlinear fire door motor features are globally pooled by the pooling layer to obtain the comprehensive instruction pooling features;

[0034] Perform instruction matching on the comprehensive instruction pooling features to obtain the corresponding control instructions, and associate the control instructions with the environmental risk level;

[0035] Performing task analysis on the control instructions based on the environmental risk level to obtain the instruction tasks.

[0036] Furthermore, controlling the intelligent fire door according to the instruction task, obtaining feedback information of the intelligent fire door, and optimizing the model parameters of the control model according to the feedback information to obtain the optimized control model include:

[0037] Controlling the intelligent fire door to execute the instruction according to the instruction task, and obtaining the feedback information sent by the intelligent fire door after executing the instruction task;

[0038] Analyze the feedback information to obtain the current fire door status, fire door execution result, fire door execution time and fire door abnormality;

[0039] Performing an expected analysis on the command task to obtain a corresponding expected execution threshold, performing an error analysis on the current fire door status, fire door execution result, and fire door execution time based on the expected execution threshold to obtain corresponding status error, execution error, and time error, performing an abnormality analysis on the fire door abnormality to obtain abnormal information data;

[0040] Based on the abnormal information data, the model parameters of the state error, execution error and time error are analyzed to obtain the corresponding optimization parameters;

[0041] Parameter optimization is performed on the control model based on the optimization parameters to obtain the optimized control model.

[0042] The present invention further provides an artificial intelligence-based intelligent fire door control system, which is applied to any of the above-mentioned artificial intelligence-based intelligent fire door control methods, comprising:

[0043] An acquisition module is used to acquire device data and sensor data of the intelligent fire door in real time, perform feature extraction on the sensor data, and obtain a feature data set;

[0044] An analysis module, configured to evaluate and process the environmental data based on the feature data set to obtain corresponding environmental evaluation information, and to perform status analysis on the device data to obtain corresponding fire door status data;

[0045] a processing module, the processing module being configured to transmit the environmental assessment information and the fire door status data to a preset control model, perform instruction analysis on the fire door status data through the control model to obtain corresponding control instructions, and generate corresponding instruction tasks based on the environmental assessment information and the control instructions;

[0046] A control module is used to control the intelligent fire door according to the instruction task, obtain feedback information of the intelligent fire door, optimize the model parameters of the control model according to the feedback information, and obtain the optimized control model.

[0047] The present invention also provides an artificial intelligence-based intelligent fire door control device, comprising:

[0048] Memory, used to store programs;

[0049] A processor is used to execute the program to implement each step of the artificial intelligence-based intelligent fire door control method described above.

[0050] The present invention also provides a storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute any of the above methods.

[0051] The artificial intelligence-based intelligent fire door control method, system, device, and medium provided by the present invention have the following beneficial effects:

[0052] In-depth analysis of sensor data enables more accurate identification of early signs of a fire, such as abnormally high temperatures and increased smoke concentrations, enabling early warning and rapid response. Comprehensive analysis of environmental data and feature datasets enables a more precise assessment of the current safety status of the environment, enabling the development of appropriate response strategies and ensuring a prompt and optimal response should a fire occur. Status analysis of equipment data effectively monitors the operating status of fire doors, promptly identifying potential faults and hazards and implementing preventive measures to ensure they remain in optimal working order. By combining environmental assessment information with fire door status data and generating command tasks through an intelligent control model, the system can automatically adjust its operating mode to suit different environments, enhancing its flexibility and adaptability. Continuously optimizing control model parameters based on actual feedback from the fire doors enables the system to continuously learn and improve, continuously enhancing its performance over time and better adapting to complex and changing operating environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of the intelligent fire door control method based on artificial intelligence provided by the present invention;

[0054] Figure 2 This is a structural diagram of the artificial intelligence-based intelligent fire door control system provided by the present invention;

[0055] Figure 3 This is a structural diagram of the artificial intelligence-based intelligent fire door control device provided by the present invention.

[0056] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0059] Reference Figure 1 As shown, the present invention provides an intelligent fire door control method based on artificial intelligence, comprising:

[0060] Step S1: acquiring device data and sensor data of the intelligent fire door in real time, performing feature extraction on the sensor data, and obtaining a feature data set;

[0061] Step S2: evaluating and processing the environmental data according to the feature data set to obtain corresponding environmental evaluation information, and performing status analysis on the equipment data to obtain corresponding fire door status data;

[0062] Step S3: transmitting the environmental assessment information and the fire door status data to a preset control model, performing instruction analysis on the fire door status data through the control model to obtain corresponding control instructions, and generating corresponding instruction tasks based on the environmental assessment information and the control instructions;

[0063] Step S4: controlling the intelligent fire door according to the instruction task, obtaining feedback information of the intelligent fire door, optimizing the model parameters of the control model according to the feedback information, and obtaining the optimized control model.

[0064] Based on the above steps, the detailed steps are as follows:

[0065] Step S1: Collect real-time data from hardware devices such as fire door controllers and motor drive units, such as motor current, voltage, and temperature. Collect data from sensors installed around the fire door, such as smoke sensors, temperature sensors, flame detectors, and airflow sensors.

[0066] Remove invalid, erroneous, or inconsistent data points from the collected data and convert the data into a unified format for subsequent processing. Ensure that the timestamps of all sensor data are synchronized.

[0067] Extract the smoke concentration trend from the smoke sensor data, extract the temperature change trend over time from the temperature sensor data, and identify the presence of flames from the flame detector data.

[0068] Combining the above characteristics, a feature data set is formed to characterize the current environmental risk level.

[0069] Step S2: Based on the extracted risk signature data, the system assesses the risk level of the current environment and determines whether a fire or other emergency situation exists, such as smoke concentration exceeding a safety threshold, a sudden temperature increase, or flame detection. The system then outputs an environmental assessment, including the risk level, the probability of an emergency, and recommended response measures.

[0070] Check whether the hardware components of the fire door are operating normally, such as whether the motor is overheating or worn. Confirm whether the fire door functions properly, such as whether the door can open and close smoothly and whether the sensor is accurate.

[0071] All equipment status information is integrated to form the fire door status data of the fire door.

[0072] Step S3: The environmental assessment information and the fire door status data are transmitted to a preset control model, and the control model performs instruction analysis on the fire door status data. Based on the environmental assessment information, it is determined whether it is necessary to change the status of the fire door (for example, to close it or keep it open) to reduce the risk.

[0073] Based on the results of fire door status command analysis and environmental assessment command analysis, specific control instructions are generated. For example, if rising smoke concentrations are detected, the fire door may need to be closed immediately; if the temperature is normal and there are no other emergency situations, the current state will be maintained.

[0074] Combining control instructions with environmental assessment information generates a series of specific instruction tasks, including adjusting the position of fire doors, activating alarm systems, notifying relevant personnel, etc.

[0075] Step S4:

[0076] Collect feedback information after the fire door executes instructions, including actual door status changes, sensor data changes, etc., compare the expected control effect with the actual feedback information, and evaluate the performance of the control model.

[0077] Based on the evaluation results, parameters in the control model are adjusted to improve its accuracy and responsiveness.

[0078] Use new data and feedback to retrain the control model to better adapt it to the current environment and device characteristics. Apply the optimized control model to subsequent control processes to achieve closed-loop control and continuously iterate and improve the control strategy.

[0079] The artificial intelligence-based intelligent fire door control method provided by the present invention can more accurately identify various signs of the early stages of a fire, such as abnormally high temperature and increased smoke concentration, by conducting in-depth analysis of sensor data, thereby achieving early warning and rapid response. By comprehensively analyzing environmental data and feature data sets, the safety status of the current environment can be more accurately assessed, and then a reasonable response strategy can be formulated to ensure that the most appropriate response can be made quickly when a fire occurs. The equipment data is analyzed for status, the working status of the fire door is effectively monitored, potential fault hazards are discovered in a timely manner, and preventive measures are taken to ensure that the fire door is always in good working condition. The environmental assessment information is combined with the fire door status data, and instruction tasks are generated through the intelligent control model, so that the fire door can automatically adjust its operating mode in different environments, thereby improving the flexibility and adaptability of the system. The parameters of the control model are continuously optimized according to the actual feedback information of the fire door, so that the system can continue to learn and improve, continuously improve its own performance over time, and better adapt to complex and changing usage environments.

[0080] In one embodiment, device data and sensor data of the intelligent fire door are acquired in real time, and feature extraction is performed on the sensor data to obtain a feature data set, including:

[0081] Acquire external raw data collected by external sensors deployed around smart fire doors (such as smoke sensors, temperature sensors, flame detectors, airflow sensors, etc.) in real time, and preprocess the collected external raw data, such as removing noise, smoothing signals, and normalizing data range, to obtain usable sensor data.

[0082] The original internal data of the intelligent fire door (such as the opening and closing status of the door, the working status of the motor, etc.) is obtained, and the original internal data is also preprocessed to obtain the device data.

[0083] The sensor data is categorized and integrated according to the preset classification rules to obtain the corresponding environmental data, and the initial features of the sensor data are calculated to obtain a feature data set.

[0084] The environmental data is calculated into environmental types to obtain smoke concentration data, temperature data, flame data and airflow data.

[0085] This embodiment, through real-time collection and preprocessing of sensor data, can quickly and accurately identify environmental changes, thereby accelerating the intelligent fire door's response to emergencies and reducing reaction time. By monitoring environmental conditions (such as smoke concentration, temperature, and the presence of flames) in real time, potential safety hazards can be promptly identified, allowing the fire door to take appropriate measures in the shortest possible time to protect personnel and property.

[0086] In one embodiment, based on the environmental data in the previous embodiment, an evaluation process is performed according to the feature data set to obtain corresponding environmental evaluation information, including:

[0087] The characteristic data set is read to obtain a first smoke concentration change rate, a first temperature rise rate, and a first thermal radiation change amount.

[0088] A deviation analysis is performed on the first smoke concentration change rate based on the smoke concentration data and the airflow data to obtain corresponding smoke deviation data.

[0089] A deviation analysis is performed on the first temperature rise rate based on the temperature data and the airflow data to obtain corresponding temperature deviation data.

[0090] A deviation analysis is performed on the first thermal radiation variation according to the temperature data, the flame data, and the airflow data to obtain corresponding thermal radiation deviation data.

[0091] Determine whether the smoke deviation data, temperature deviation data, and thermal radiation deviation data meet the preset deviation threshold requirements. When any of the smoke deviation data, temperature deviation data, and thermal radiation deviation data do not meet the deviation threshold requirements, perform risk feature calculation based on the smoke deviation data, temperature deviation data, and thermal radiation deviation data combined with environmental data to obtain a corrected feature data set.

[0092] Based on the environmental data and the modified feature data set, the intelligent fire door is evaluated to obtain environmental assessment information. The assessment includes the determination of the safety level, risk trend prediction, etc. The environmental assessment information is output and will serve as the basis for subsequent decision-making.

[0093] This embodiment, through real-time analysis of feature data sets, can quickly and accurately identify environmental changes, thereby accelerating the intelligent fire door's response to emergencies and reducing reaction time. Real-time monitoring of environmental conditions (such as smoke concentration, temperature, and thermal radiation) promptly identifies potential safety hazards, enabling the fire door to take appropriate measures to protect people and property in the shortest possible time. Analysis of deviation data enables intelligent analysis based on environmental data, leading to more accurate decisions. For example, in the event of a fire, the fire door can be quickly closed to prevent its spread. By integrating deviation data with environmental data for analysis, the system can better adapt to varying environmental conditions and provide effective responses even in complex environments.

[0094] In one embodiment, status analysis is performed on the device data to obtain corresponding fire door status data, including:

[0095] Read the device data type to obtain the fire door working status and motor status data;

[0096] Analyze the working status data of the fire door to determine whether the fire door is currently open or closed, and obtain the open and closed status of the fire door. Analyze whether the fire door is locked, that is, determine whether the fire door can be opened manually or automatically, and obtain the locked status of the fire door.

[0097] Perform health assessment on motor status data, analyze the motor's working performance, determine whether the motor has abnormalities or potential problems, evaluate whether the motor is working normally, whether there are overheating, abnormal vibration, increased noise, etc., and obtain motor health status information.

[0098] The switch status, lock status and motor health status information are integrated to obtain the fire door status data.

[0099] This embodiment monitors the operating status and motor status of fire doors in real time, enabling rapid and accurate identification of changes in the door's status. This accelerates the intelligent fire door's response to emergencies and reduces reaction time. Real-time monitoring of the fire door's status promptly identifies potential safety hazards, enabling the door to take appropriate measures to protect personnel and property in the shortest possible time. By analyzing fire door status data, the system can perform intelligent analysis based on the current status, enabling more accurate decisions. For example, in the event of a fire, the door can be quickly closed to prevent its spread.

[0100] In one embodiment, the control model performs instruction analysis on the fire door status data to obtain corresponding control instructions, and generates corresponding instruction tasks based on the environmental assessment information and the control instructions, including:

[0101] The input layer of the control model converts the input environmental assessment information and fire door status data into a format that produces an environmental assessment matrix, along with switch status, lock status, and motor health status information. The fire door status data includes the switch status (open or closed), lock status (locked or locked), and motor health status (whether the motor is functioning properly). The environmental assessment matrix contains information about the environment that is directly relevant to the fire door control strategy, such as smoke density and temperature levels.

[0102] The control model's double-layer convolutional layer and recurrent neural network are used to comprehensively analyze the switch status, lock status, and motor health status information to obtain the corresponding instruction features. Instructions are matched based on the instruction features to obtain the corresponding control instructions. The control instructions are then task-analyzed based on the environmental assessment matrix to obtain instruction tasks, which include specific actions such as closing the door, sounding the alarm, and notifying relevant personnel.

[0103] Output the instruction task through the output layer of the control model.

[0104] This embodiment, through real-time monitoring and intelligent analysis, can quickly identify abnormal conditions in the environment (such as a fire) and take immediate action, such as closing the fire door to prevent the fire from spreading. The control model enables more accurate decisions based on historical data and current environmental conditions, avoiding the potential misjudgments and slow responses of traditional manual control or simple rule-driven systems. Furthermore, the control model considers not only the open and closed status of the fire door itself but also its locking status and motor health. This means the system can detect potential problems and take preventative measures before they occur.

[0105] In one embodiment, a control model's double-layer convolutional layer and recurrent neural network are used to comprehensively analyze the switch state, lock state, and motor health status information to obtain corresponding instruction features. Instruction matching is performed based on the instruction features to obtain corresponding control instructions. Task analysis is then performed on the control instructions based on the environmental assessment matrix to obtain instruction tasks, including:

[0106] The switch state and lock state are input into the first convolutional layer, and the switch state and lock state are identified by the first convolutional layer to obtain the corresponding fire door control features;

[0107] The motor health status information is input into the second convolutional layer, and the second convolutional layer performs feature recognition on the motor health status information to obtain the corresponding fire door motor features;

[0108] The environmental assessment matrix is ​​input into the recurrent neural network, and the risk assessment of the environmental assessment matrix is ​​performed through the recurrent neural network to obtain the corresponding environmental risk level;

[0109] The fire door control features and the fire door motor features are input into the activation function layer, and the fire door control features and the fire door motor features are nonlinearly processed by the activation function layer to obtain the corresponding nonlinear fire door control features and nonlinear fire door motor features;

[0110] The nonlinear fire door control features and the nonlinear fire door motor features are input into the pooling layer, and the nonlinear fire door control features and the nonlinear fire door motor features are globally pooled by the pooling layer to obtain the comprehensive instruction pooling features;

[0111] Perform instruction matching on the comprehensive instruction pooling features to obtain the corresponding control instructions, and associate the control instructions with the environmental risk level;

[0112] Perform task analysis on control instructions based on environmental risk levels to obtain instruction tasks.

[0113] This embodiment uses a double-layer convolutional layer and a recurrent neural network for feature extraction and risk assessment. The system can more quickly identify the state changes of fire doors and the environmental risk level, thereby improving the response speed and accuracy. By utilizing deep learning technology, more accurate decisions can be made based on the control characteristics, motor characteristics, and environmental risk levels of fire doors, ensuring that the optimal control strategy is adopted in various situations. Through the input of the environmental assessment matrix and the assessment of risk levels, its behavior can be flexibly adjusted according to different environmental conditions, thereby better adapting to complex environmental changes. Through the processing of multi-layer neural networks, the system can reduce the probability of false alarms and misoperations, and improve the stability and reliability of the system. By monitoring the health status of the motor in real time, maintenance personnel can discover potential problems earlier and take preventive measures to reduce the occurrence of equipment failures and reduce maintenance costs.

[0114] In one embodiment, the intelligent fire door is controlled according to the instruction task, and feedback information of the intelligent fire door is obtained. The model parameters of the control model are optimized according to the feedback information to obtain the optimized control model, including:

[0115] Control the intelligent fire door to execute the command according to the command task, and obtain the feedback information sent by the intelligent fire door after executing the command task;

[0116] The feedback information is parsed and processed to obtain the current fire door status (whether the fire door is currently closed, locked, etc.), fire door execution result (for example, whether the fire door is successfully closed or opened, etc.), fire door execution time (the time required to execute the instruction) and fire door abnormality (any abnormality or failure encountered during the execution of the instruction).

[0117] Perform expected analysis on the instruction task to obtain the corresponding expected execution threshold. Based on the expected execution threshold, perform error analysis on the current fire door status, fire door execution result, and fire door execution time to obtain the corresponding status error, execution error, and time error. Perform abnormal analysis on the abnormal situation of the fire door to obtain abnormal information data.

[0118] Based on the abnormal information data, the model parameters of the state error, execution error and time error are analyzed to obtain the corresponding optimization parameters;

[0119] The control model is optimized based on the optimization parameters to obtain an optimized control model.

[0120] By analyzing feedback information and optimizing model parameters, this embodiment can better adapt to different environmental conditions and respond effectively even in complex environments. Using feedback information for error analysis can reduce safety hazards caused by false alarms or misoperation and improve the stability and reliability of the system. Real-time device data can help maintenance personnel understand the working status of fire doors, facilitate timely maintenance and inspections, and ensure that the equipment is always in optimal working condition. Through intelligent analysis, the normal state of fire doors can be maintained without the need for emergency measures, avoiding unnecessary energy waste. Feedback information is used for model parameter optimization, which means that the control system will become more intelligent and efficient over time and can better adapt to new challenges that may arise in the future.

[0121] Reference Figure 2 As shown, the present invention also provides an artificial intelligence-based intelligent fire door control system, which is applied to any of the above-mentioned artificial intelligence-based intelligent fire door control methods, including:

[0122] The acquisition module is used to obtain the device data and sensor data of the intelligent fire door in real time, extract features from the sensor data, and obtain a feature data set;

[0123] Analysis module, which is used to evaluate and process environmental data based on feature data sets to obtain corresponding environmental assessment information, and to perform status analysis on equipment data to obtain corresponding fire door status data;

[0124] A processing module is used to transmit environmental assessment information and fire door status data to a preset control model, perform instruction analysis on the fire door status data through the control model, obtain corresponding control instructions, and generate corresponding instruction tasks based on the environmental assessment information and the control instructions;

[0125] The control module is used to control the intelligent fire door according to the instruction task, obtain feedback information of the intelligent fire door, optimize the model parameters of the control model according to the feedback information, and obtain the optimized control model.

[0126] The artificial intelligence-based intelligent fire door control system provided by the present invention can more accurately identify various signs of the early stages of a fire, such as abnormally high temperature and increased smoke concentration, by conducting in-depth analysis of sensor data, thereby achieving early warning and rapid response. By comprehensively analyzing environmental data and feature data sets, the safety status of the current environment can be more accurately assessed, and then a reasonable response strategy can be formulated to ensure that the most appropriate response can be made quickly when a fire occurs. The equipment data is analyzed for status, the working status of the fire door is effectively monitored, potential fault hazards are discovered in a timely manner, and preventive measures are taken to ensure that the fire door is always in good working condition. The environmental assessment information is combined with the fire door status data, and instruction tasks are generated through the intelligent control model, so that the fire door can automatically adjust its operating mode in different environments, thereby improving the flexibility and adaptability of the system. The parameters of the control model are continuously optimized according to the actual feedback information of the fire door, so that the system can continue to learn and improve, continuously improve its own performance over time, and better adapt to the complex and changing use environment.

[0127] Reference Figure 3 As shown, the present invention also provides an intelligent fire door control device based on artificial intelligence, comprising:

[0128] Memory, used to store programs;

[0129] A processor is used to execute the program to implement each step of the artificial intelligence-based intelligent fire door control method described above.

[0130] In this embodiment, the processor and memory may be connected via a bus or other means. The memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive. The processor may be a general-purpose processor, such as a central processing unit, a digital signal processor, an application-specific integrated circuit, or one or more integrated circuits configured to implement the embodiments of the present invention.

[0131] The present invention also provides a storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute any of the above methods.

[0132] It should be noted that, those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the above-described system and each module can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0133] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent fire door control method based on artificial intelligence, characterized in that: include: Acquire device data and sensor data of the intelligent fire door in real time, perform feature extraction on the sensor data, and obtain a feature data set; Evaluate and process the environmental data according to the feature data set to obtain corresponding environmental evaluation information, and perform status analysis on the equipment data to obtain corresponding fire door status data; Transmitting the environmental assessment information and the fire door status data to a preset control model, performing instruction analysis on the fire door status data through the control model to obtain corresponding control instructions, and generating corresponding instruction tasks based on the environmental assessment information and the control instructions; Controlling the intelligent fire door according to the instruction task, obtaining feedback information of the intelligent fire door, and optimizing model parameters of the control model according to the feedback information to obtain the optimized control model; The state analysis of the device data to obtain corresponding fire door state data includes: Read the device data by type to obtain the working status of the fire door and the motor status data; Perform status analysis on the working status of the fire door to obtain the open / close status and lock status of the fire door; Perform health assessment on motor status data to obtain motor health status information; Integrate the switch status, lock status and motor health status information to obtain the fire door status data; The step of performing instruction analysis on the fire door status data by using the control model to obtain corresponding control instructions, and generating corresponding instruction tasks based on the environmental assessment information and the control instructions, includes: Performing format conversion on the input environmental assessment information and the fire door status data through the input layer of the control model to obtain an environmental assessment matrix, the switch status, the locking status, and the motor health status information; Comprehensively analyzing the switch state, the lock state, and the motor health status information through the double-layer convolutional layer and the recurrent neural network of the control model to obtain corresponding instruction features, performing instruction matching based on the instruction features to obtain corresponding control instructions, and performing task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task; Outputting the instruction task through the output layer of the control model; The control model comprehensively analyzes the switch state, the lock state, and the motor health status information through the double-layer convolution layer and the recurrent neural network to obtain corresponding instruction features, performs instruction matching based on the instruction features to obtain corresponding control instructions, and performs task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task, including: Inputting the switch state and the lock state into a first convolutional layer, performing feature recognition on the switch state and the lock state through the first convolutional layer to obtain corresponding fire door control features; Inputting the motor health status information into the second convolutional layer, performing feature recognition on the motor health status information through the second convolutional layer to obtain corresponding fire door motor features; Inputting the environmental assessment matrix into a recurrent neural network, performing risk assessment on the environmental assessment matrix through the recurrent neural network, and obtaining a corresponding environmental risk level; The fire door control features and the fire door motor features are input into the activation function layer, and the fire door control features and the fire door motor features are nonlinearly processed by the activation function layer to obtain the corresponding nonlinear fire door control features and nonlinear fire door motor features; The nonlinear fire door control features and the nonlinear fire door motor features are input into the pooling layer, and the nonlinear fire door control features and the nonlinear fire door motor features are globally pooled by the pooling layer to obtain the comprehensive instruction pooling features; Perform instruction matching on the comprehensive instruction pooling features to obtain the corresponding control instructions, and associate the control instructions with the environmental risk level; Performing task analysis on the control instructions based on the environmental risk level to obtain the instruction tasks.

2. The intelligent fire door control method based on artificial intelligence according to claim 1 is characterized in that: The real-time acquisition of device data and sensor data of the intelligent fire door, and feature extraction of the sensor data to obtain a feature data set include: Acquire external raw data collected by external sensors deployed around the intelligent fire door in real time, and pre-process the external raw data to obtain the corresponding sensor data; Obtaining original internal data of the intelligent fire door, and preprocessing the original internal data to obtain the device data; Classifying the sensor data according to a preset classification rule to obtain corresponding environmental data, and performing initial feature calculation on the sensor data to obtain the feature data set; The environmental data is calculated into environmental types to obtain smoke concentration data, temperature data, flame data and airflow number.

3. The intelligent fire door control method based on artificial intelligence according to claim 2 is characterized in that: The performing of evaluation processing according to the feature data set to obtain corresponding environmental evaluation information includes: Reading the characteristic data set to obtain a first smoke concentration change rate, a first temperature rise rate, and a first thermal radiation change amount; Performing a deviation analysis on the first smoke concentration change rate based on the smoke concentration data and the airflow data to obtain corresponding smoke deviation data, performing a deviation analysis on the first temperature rise rate based on the temperature data and the airflow data to obtain corresponding temperature deviation data, and performing a deviation analysis on the first thermal radiation change amount based on the temperature data, the flame data, and the airflow data to obtain corresponding thermal radiation deviation data; Determine whether the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data meet the preset deviation threshold requirements; when any one of the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data does not meet the deviation threshold requirements, calculate the risk characteristics based on the smoke deviation data, the temperature deviation data, and the thermal radiation deviation data in combination with the environmental data to obtain a corrected characteristic data set; An environmental assessment is performed on the intelligent fire door based on the environmental data and the corrected feature data set to obtain the environmental assessment information.

4. The intelligent fire door control method based on artificial intelligence according to claim 1 is characterized in that: The controlling the intelligent fire door according to the instruction task, obtaining feedback information of the intelligent fire door, and optimizing the model parameters of the control model according to the feedback information to obtain the optimized control model includes: Controlling the intelligent fire door to execute the instruction according to the instruction task, and obtaining the feedback information sent by the intelligent fire door after executing the instruction task; Analyze the feedback information to obtain the current fire door status, fire door execution result, fire door execution time and fire door abnormality; Performing an expected analysis on the command task to obtain a corresponding expected execution threshold, performing an error analysis on the current fire door status, fire door execution result, and fire door execution time based on the expected execution threshold to obtain corresponding status error, execution error, and time error, performing an abnormality analysis on the fire door abnormality to obtain abnormal information data; Based on the abnormal information data, the model parameters of the state error, execution error and time error are analyzed to obtain the corresponding optimization parameters; Parameter optimization is performed on the control model based on the optimization parameters to obtain the optimized control model.

5. An artificial intelligence-based intelligent fire door control system, applied to the artificial intelligence-based intelligent fire door control method according to any one of claims 1 to 4, characterized in that: include: An acquisition module is used to acquire device data and sensor data of the intelligent fire door in real time, perform feature extraction on the sensor data, and obtain a feature data set; An analysis module, configured to evaluate and process the environmental data based on the feature data set to obtain corresponding environmental evaluation information, and to perform status analysis on the device data to obtain corresponding fire door status data; a processing module, the processing module being configured to transmit the environmental assessment information and the fire door status data to a preset control model, perform instruction analysis on the fire door status data through the control model to obtain corresponding control instructions, and generate corresponding instruction tasks based on the environmental assessment information and the control instructions; a control module, the control module being configured to control the intelligent fire door according to the instruction task, obtain feedback information of the intelligent fire door, and optimize model parameters of the control model according to the feedback information to obtain the optimized control model; The state analysis of the device data to obtain corresponding fire door state data includes: Read the device data by type to obtain the working status of the fire door and the motor status data; Perform status analysis on the working status of the fire door to obtain the open / close status and lock status of the fire door; Perform health assessment on motor status data to obtain motor health status information; Integrate the switch status, lock status and motor health status information to obtain the fire door status data; The step of performing instruction analysis on the fire door status data by using the control model to obtain corresponding control instructions, and generating corresponding instruction tasks based on the environmental assessment information and the control instructions, includes: Performing format conversion on the input environmental assessment information and the fire door status data through the input layer of the control model to obtain an environmental assessment matrix, the switch status, the locking status, and the motor health status information; Comprehensively analyzing the switch state, the lock state, and the motor health status information through the double-layer convolutional layer and the recurrent neural network of the control model to obtain corresponding instruction features, performing instruction matching based on the instruction features to obtain corresponding control instructions, and performing task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task; Outputting the instruction task through the output layer of the control model; The control model comprehensively analyzes the switch state, the lock state, and the motor health status information through the double-layer convolution layer and the recurrent neural network to obtain corresponding instruction features, performs instruction matching based on the instruction features to obtain corresponding control instructions, and performs task analysis on the control instructions based on the environmental assessment matrix to obtain the instruction task, including: Inputting the switch state and the lock state into a first convolutional layer, performing feature recognition on the switch state and the lock state through the first convolutional layer to obtain corresponding fire door control features; Inputting the motor health status information into the second convolutional layer, performing feature recognition on the motor health status information through the second convolutional layer to obtain corresponding fire door motor features; Inputting the environmental assessment matrix into a recurrent neural network, performing risk assessment on the environmental assessment matrix through the recurrent neural network, and obtaining a corresponding environmental risk level; The fire door control features and the fire door motor features are input into the activation function layer, and the fire door control features and the fire door motor features are nonlinearly processed by the activation function layer to obtain the corresponding nonlinear fire door control features and nonlinear fire door motor features; The nonlinear fire door control features and the nonlinear fire door motor features are input into the pooling layer, and the nonlinear fire door control features and the nonlinear fire door motor features are globally pooled by the pooling layer to obtain the comprehensive instruction pooling features; Perform instruction matching on the comprehensive instruction pooling features to obtain the corresponding control instructions, and associate the control instructions with the environmental risk level; Performing task analysis on the control instructions based on the environmental risk level to obtain the instruction tasks.

6. An intelligent fire door control device based on artificial intelligence, characterized in that: include: Memory, used to store programs; A processor is used to execute the program to implement the various steps of the artificial intelligence-based intelligent fire door control method as described in any one of claims 1 to 4.

7. A storage medium, characterized in that: Computer instructions are stored, and the computer instructions are used to make a computer execute the method according to any one of claims 1 to 4.

Images