Fire monitoring and early warning method, device and system based on Internet of Things, and storage medium

Abstract

The invention provides a fire monitoring and early warning method, device and system based on the Internet of Things, and a storage medium, and belongs to the field of fire monitoring and early warning, and the method comprises the steps: obtaining various types of fire monitoring signals of a target region; performing filtering processing on each type of fire monitoring signals based on the target interference factors to obtain multiple types of target monitoring signals; wherein the target interference factors corresponding to each type of fire monitoring signals are different; fire early warning is carried out on the target area based on the multiple types of target monitoring signals, and the state of the target area is obtained; determining a target control strategy based on the state of the target area, and controlling fire extinguishing equipment in the target area based on the target control strategy. According to the fire monitoring and early warning method, device and system based on the Internet of Things, and the storage medium provided by the invention, the timeliness and accuracy of fire monitoring and early warning can be improved.

Description

Technical Field

[0001] The present invention relates to the field of fire monitoring and early warning, and more specifically, to a fire monitoring and early warning method and device, system, and storage medium based on the Internet of Things. Background Art

[0002] In recent years, the occurrence of fire accidents has brought great threats to people's lives and property safety. With the continuous improvement of public safety needs and the rapid development of modern information technology, traditional fire monitoring methods, such as manual inspections and simple sensor alarms, have been unable to meet the high requirements of modern society for fire warnings. Existing fire monitoring and early warning methods have the problem of untimely response. Summary of the invention

[0003] The purpose of the present invention is to provide a fire monitoring and early warning method and device, system, and storage medium based on the Internet of Things to improve the timeliness and accuracy of fire monitoring and early warning.

[0004] A first aspect of the embodiments of the present disclosure provides a fire monitoring and early warning method based on the Internet of Things, comprising: Obtain multiple types of fire monitoring signals in the target area; Each type of fire monitoring signal is filtered based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor; Performing a fire warning on the target area based on the multiple types of target monitoring signals to obtain the state of the target area; A target control strategy is determined based on the state of the target area, and a fire extinguishing device in the target area is controlled based on the target control strategy.

[0005] A second aspect of the embodiments of the present disclosure provides a fire monitoring and early warning device based on the Internet of Things, including: A signal acquisition module is used to acquire multiple types of fire monitoring signals in the target area; A filtering processing module is used to filter each type of fire monitoring signal based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor; An early warning module, used for providing a fire early warning to the target area based on the multiple types of target monitoring signals, and obtaining the state of the target area; The control module is used to determine a target control strategy based on the state of the target area, and control the fire extinguishing equipment in the target area based on the target control strategy.

[0006] According to a third aspect of an embodiment of the present disclosure, a fire monitoring and early warning system based on the Internet of Things is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned fire monitoring and early warning method based on the Internet of Things when executing the computer program.

[0007] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned fire monitoring and early warning method based on the Internet of Things are implemented.

[0008] The beneficial effects of the fire monitoring and early warning method, device, system, and storage medium based on the Internet of Things provided by the embodiments of the present disclosure are: By acquiring multiple types of fire monitoring signals in the target area, the present disclosure can comprehensively and accurately perceive fire hazards and improve the sensitivity and accuracy of early warning. Filtering is performed on different target interference factors corresponding to different fire monitoring signals, effectively reducing false alarms and missed alarms, and ensuring the reliability of monitoring data. Secondly, the present disclosure can provide fire early warnings for the target area based on the processed multiple types of target monitoring signals, reflecting the status of the area in real time, and winning precious time for timely response measures. By intelligently analyzing the status of the target area, the present disclosure can automatically determine and execute the corresponding target control strategy, thereby achieving precise control of fire extinguishing equipment. Therefore, the present disclosure can improve the timeliness and accuracy of fire monitoring and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flow chart of a fire monitoring and early warning method based on the Internet of Things provided in one embodiment of the present disclosure; Figure 2 A structural block diagram of a fire monitoring and early warning device based on the Internet of Things provided in one embodiment of the present disclosure; Figure 3 A schematic block diagram of a fire monitoring and early warning system based on the Internet of Things provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0011] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0012] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0013] Please refer to Figure 1 , Figure 1 A flow chart of a fire monitoring and early warning method based on the Internet of Things provided in one embodiment of the present disclosure, the method comprising: S101: Acquire multiple types of fire monitoring signals in the target area.

[0014] In this embodiment, the target area is an area that requires fire monitoring, which may be various places where fires are prone to occur or where fire prevention is required, such as residential buildings, office buildings, shopping malls and other buildings; industrial places such as factory workshops and warehouses; public places such as theaters, stadiums and stations; and outdoor areas such as forests.

[0015] The various types of fire monitoring signals are signals collected by setting up multiple different types of sensors in the target area. The various types of fire monitoring signals may include temperature signals, smoke concentration signals, toxic gas concentration signals, and flame signals. The temperature signal can be monitored by a temperature sensor, the smoke concentration signal can be monitored by a smoke sensor, the toxic gas concentration signal can be monitored by a gas sensor, and the flame signal can be detected by a flame sensor. For example, the ambient temperature data monitored in real time by a temperature sensor is a fire monitoring signal.

[0016] In areas where fire monitoring is required, by installing a variety of different types of sensors and applying their respective monitoring principles, corresponding fire-related data information is collected, and multiple different types of signals that can reflect fire-related situations are generated, providing a basic data source for subsequent fire monitoring and early warning work.

[0017] For example, in a factory warehouse (target area), in order to monitor fire hazards, different sensors are installed in various corners of the warehouse, between shelves, and near doors and windows. Among them, multiple temperature sensors are installed, which collect temperature data at a certain time (such as every 30 minutes) at the location, and transmit the above temperature values ​​in the form of electrical signals. The above temperature values ​​constitute temperature-type fire monitoring signals; at the same time, smoke concentration sensors are also deployed, which can monitor the content of smoke particles in the air and output corresponding smoke concentration signals in the form of percentage shading rate, etc.; in addition, carbon monoxide gas sensors are also equipped to monitor the concentration of toxic gases such as carbon monoxide in the environment in real time, and the monitored concentration values ​​are used as fire monitoring signals of toxic gas concentration. Through the collaborative work of the above different types of sensors, multiple types of fire monitoring signals in the target area are obtained, which provides basic data for further analysis of whether there is a fire risk in the warehouse.

[0018] S102: Filter each type of fire monitoring signal based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor.

[0019] In this embodiment, the target interference factor is a factor that will have an adverse effect on the fire monitoring signal, causing the fire monitoring signal to fail to accurately reflect the true situation of the fire. The target interference factor has various sources, such as the environment, the characteristics of the device itself, or other surrounding interference sources.

[0020] For example, for temperature monitoring signals, the heat emitted by surrounding heating equipment and temperature fluctuations caused by the ventilation system are all target interference factors of temperature-type fire monitoring signals; and for smoke concentration monitoring signals, dust, water vapor, and smoke generated by human activities are all target interference factors of smoke concentration-type fire monitoring signals.

[0021] Considering that the target interference factors corresponding to each type of monitoring signal are different, the filtering processing methods adopted will also be different.

[0022] Filtering is a signal processing technology that uses filtering algorithms to process the collected original signals (i.e., fire monitoring signals) and remove interference components (i.e., target interference factors) to make the signals smoother, more stable, and closer to a rational state that truly reflects fire-related conditions, thereby improving signal quality and availability.

[0023] In response to each type of fire monitoring signal being a temperature signal: If the target interference factor is the first case, low-pass filtering is used as the filtering method; If the target interference factor is the second case, Kalman filtering is used as the filtering method.

[0024] The first case is that the electromagnetic interference is greater than the electromagnetic interference threshold or the change in the ventilation system is greater than the first change rate. The electromagnetic interference threshold and the first change rate are dynamically adjusted according to the actual scenario. Considering that the temperature signal changes slowly in a fire scenario, the temperature rise caused by the fire is a relatively low-frequency process. Low-pass filtering allows low-frequency signals to pass and attenuates high-frequency signals. When the temperature signal is subject to high-frequency interference, such as rapid temperature fluctuations caused by electromagnetic interference generated by electrical equipment, or temperature jitters caused by rapid changes in the ventilation system, low-pass filtering can effectively remove these high-frequency components and highlight the slow temperature rise trend caused by the fire.

[0025] The second case is that the ambient temperature is greater than the ambient temperature threshold, that is, the ambient temperature changes caused by equipment operation or human activities. Among them, the ambient temperature threshold can be obtained based on experience. Kalman filtering is an optimal estimation filtering method based on a state space model. It recursively estimates the current temperature value and the measured value of the temperature sensor, and combines the temperature change law and the measurement characteristics of the temperature sensor to obtain the optimal filtering result. For example, considering the accuracy error of the temperature sensor itself and the dynamic changes of the environment (that is, the second case), Kalman filtering can combine the above factors to better estimate the true temperature.

[0026] In response to each type of fire monitoring signal is a smoke density signal: If the target interference factor is the third case, median filtering is used as the filtering method; If the target interference factor is the fourth case, the mean filter and the sliding average filter are combined as the filtering processing method.

[0027] The third case is that the number of times the smoke concentration spike signal appears is greater than or equal to 1. Median filtering is a nonlinear filtering method that sorts the smoke concentration data in the window by size and then takes the middle value as the output after filtering. It is very effective in removing impulse noise in the smoke concentration signal. In the actual environment, there may be some instantaneous interference factors, such as someone smoking near the sensor, short-term dust rising, etc. These situations will cause spike signals in the smoke concentration. Median filtering can remove these spikes well without excessively smoothing the real smoke concentration signal like mean filtering.

[0028] The fourth situation is that the random noise interference is greater than the noise interference threshold, such as the electronic noise of the smoke sensor itself or the fluctuation of tiny particles in the surrounding environment. Among them, the noise interference threshold is set dynamically according to the actual scene. The filtering method combining mean filtering and sliding average filtering can effectively smooth the signal, and at the same time can retain the trend of smoke concentration changes to a certain extent. It is more useful for monitoring the slow rise or fall of smoke concentration (such as the gradual accumulation of smoke in the early stage of a fire or the dissipation of smoke in the later stage). The filtering method combining mean filtering and sliding average filtering is to first perform mean filtering, calculate the average value of the smoke concentration data in a certain window to smooth the signal, and then perform sliding average filtering.

[0029] In response to each type of fire monitoring signal is a harmful gas concentration signal: If the target interference factor is the fifth case, adaptive filtering is used as the filtering processing method; If the target interference factor is the sixth case, wavelet filtering is used as the filtering processing method.

[0030] The fifth situation is that the concentration of other harmless gases is greater than the first concentration threshold or the ambient humidity is greater than the humidity threshold. Among them, the first concentration threshold or the humidity threshold is dynamically set according to the actual scenario. Adaptive filtering can automatically adjust the parameters of the filter according to the statistical characteristics of the input signal and the expected signal (or reference signal). For harmful gas concentration signals, due to different fire scenes and environments, the generation and diffusion characteristics of harmful gases are different, and adaptive filtering can adjust the filtering parameters according to actual conditions. Adaptive filtering can automatically optimize the filtering effect according to the real-time changes and interference of harmful gas concentration signals, thereby improving the accuracy of harmful gas concentration monitoring.

[0031] The sixth case is that the electromagnetic interference is greater than the electromagnetic interference threshold and the environmental fluctuation is less than the first fluctuation threshold (such as the concentration change caused by ventilation). Among them, the first fluctuation threshold is dynamically set according to the actual situation. Wavelet filtering is a time-frequency domain analysis method. By decomposing the signal into wavelet coefficients of different scales, according to the different characteristics of the signal and noise in the wavelet domain, the wavelet coefficients corresponding to the noise are removed, and then the signal is reconstructed. Wavelet filtering can analyze and process signals at different frequencies and time scales, and separate the real change signal of the concentration of harmful gases. It is particularly suitable for processing non-stationary harmful gas concentration signals.

[0032] The target monitoring signal is each type of fire monitoring signal. After filtering, the influence of corresponding interference factors is removed, and the obtained signal can more accurately reflect the actual situation of the fire. This signal can be used for subsequent fire warning, analysis and other work.

[0033] Since different types of fire monitoring signals will be affected by their own specific interference factors during the collection process, it is necessary to analyze the target interference factors that affect each type of fire monitoring signal, and then use a corresponding filtering processing method to process the original fire monitoring signal. Finally, multiple types of target monitoring signals with interference removed and more accuracy are obtained, so that effective fire warning and other operations can be performed based on the above signals in the future.

[0034] S103: Performing a fire warning on the target area based on multiple types of target monitoring signals to obtain the status of the target area.

[0035] In this embodiment, the multiple types of target monitoring signals are a set of signals reflecting the fire-related information of the target area from different dimensions after filtering and removing various interference factors. The target monitoring signals can be monitoring signals such as temperature, smoke concentration, and gas concentration after filtering. Each type of target monitoring signal provides information for fire warning, and the combination can more comprehensively and accurately assess the fire risk.

[0036] Fire warning is to analyze and judge the fire-related information (i.e., multiple types of target monitoring signals) in the target area, and issue an alarm in time before or in the early stages of a fire to remind relevant personnel to take measures to reduce the losses caused by the fire.

[0037] The state of the target area is the state information of the current fire risk level of the target area, which can be divided into normal state, fire warning state and fire occurrence state.

[0038] This embodiment uses processed multi-category target monitoring signals as a basis, and uses algorithms or models to evaluate and judge the fire risk of the target area, so as to determine the current fire-related state of the target area, whether it is normal, there is a fire warning, or a fire has already occurred.

[0039] In this embodiment, a fire warning is performed on the target area based on multiple types of target monitoring signals to obtain the status of the target area, including: If the first target monitoring signal is less than the temperature threshold, and the second target monitoring signal is less than the smoke concentration threshold, the state of the target area is that no fire occurs; If the first target monitoring signal is greater than or equal to the temperature threshold, or the second target monitoring signal is greater than or equal to the smoke concentration threshold, the state of the target area is in the pre-fire stage or the initial stage of fire; If the first target monitoring signal is greater than or equal to the temperature threshold, and the second target monitoring signal is greater than or equal to the smoke concentration threshold, the state of the target area is that a fire has occurred; Among them, the first target monitoring signal is a filtered temperature signal, and the second target monitoring signal is a filtered smoke concentration signal.

[0040] The above judgment rule is one of the analysis and judgment rules for performing fire warning on the target area based on multiple types of target monitoring signals.

[0041] S104: Determine a target control strategy based on the state of the target area, and control fire extinguishing equipment in the target area based on the target control strategy.

[0042] In this embodiment, according to step S103, the state of the target area can be a state such as no fire, pre-fire, initial fire, and fire. The target control strategy is a response measure or operation plan formulated according to the different states of the target area, which is used to effectively control the development of the fire, and may include the control method of the fire extinguishing equipment, the guidance of personnel evacuation, etc. For example, if the state of the target area is the pre-fire and the filtered smoke concentration is greater than or equal to the smoke concentration threshold, the target control strategy is to turn on the ventilation equipment and send security personnel to the suspected fire source area for investigation; if the state of the target area is a fire, the target control strategy is to start the fire sprinkler system to extinguish the fire, and start the fire broadcast to guide personnel to evacuate.

[0043] The fire-fighting equipment in the target area refers to various equipment installed in the target area to extinguish the fire or control the spread of the fire, which may include fire extinguishers, fire hydrant systems, automatic sprinkler systems, gas fire extinguishing systems or fire pumps, etc. Different fire-fighting equipment is suitable for different types of fires and scenarios.

[0044] After the fire-related state of the target area is determined, the response measures for the state are determined according to the pre-established corresponding strategy, that is, the target control strategy. Then, according to the above strategy, the various fire-fighting equipment equipped in the target area is operated and controlled accordingly to achieve the purpose of extinguishing or controlling the fire.

[0045] Assume that a large shopping mall is the target area, which is equipped with a complete fire monitoring and extinguishing system.

[0046] First, status judgment: The fire monitoring system detected that the temperature in multiple areas on a certain floor rose rapidly, the smoke concentration increased sharply, and the concentration of harmful gases exceeded the safety standard, and determined that the shopping mall was in a fire state.

[0047] Second, determine the target control strategy: Based on the fire status, the target control strategy is as follows: Immediately activate the entire shopping center's fire alarm system, including audible and visual alarms, to notify all personnel to evacuate; Fully activate the automatic sprinkler system to spray water on the floor where the fire occurred and adjacent floors to suppress the spread of the fire; Start the fire broadcast and guide customers and staff to evacuate in an orderly manner according to the predetermined evacuation routes; Linked fire pumps to ensure adequate supply of fire-fighting water; Close the fire shutter doors in the fire area to prevent the fire from spreading to other areas.

[0048] Third, control fire extinguishing equipment: When the sprinkler heads in the automatic sprinkler system reach the operating temperature, they quickly open and spray a large amount of water into the fire area, lowering the temperature and extinguishing the flames; The fire pump starts working, pressurizing the water in the fire pool and delivering it to the pipes of each fire hydrant and the automatic sprinkler system to ensure a continuous supply of water for fire fighting; The fire shutter door descends quickly according to the system command, isolating the fire area from other areas to form a fire partition. Through the effective control of these fire-fighting equipment and the coordination of personnel evacuation and other measures, the safety of personnel is guaranteed and property losses are reduced to the greatest extent.

[0049] From the above, it can be concluded that the present disclosure can comprehensively and accurately perceive fire hazards by acquiring multiple types of fire monitoring signals in the target area, thereby improving the sensitivity and accuracy of early warning. Filtering is performed on different target interference factors corresponding to different fire monitoring signals, which effectively reduces false alarms and missed alarms, and ensures the reliability of monitoring data. Secondly, the present disclosure can provide fire early warnings for the target area based on the processed multiple types of target monitoring signals, and reflect the status of the area in real time, which wins precious time for timely response measures. The present disclosure can automatically determine and execute corresponding target control strategies by intelligently analyzing the status of the target area, thereby achieving precise control of fire extinguishing equipment. Therefore, the present disclosure can improve the timeliness and accuracy of fire monitoring early warnings.

[0050] In one embodiment of the present disclosure, each type of fire monitoring signal is filtered based on the target interference factor to obtain multiple types of target monitoring signals, including: Determine the target interference factors corresponding to each type of fire monitoring signal; Determine the filtering algorithm corresponding to each type of fire monitoring signal based on the target interference factor corresponding to the type of fire monitoring signal; Based on the filtering algorithm corresponding to each type of fire monitoring signal, the fire monitoring signal of this type is filtered to obtain multiple types of target monitoring signals.

[0051] In this embodiment, the target interference factors are various factors that have an adverse effect on the fire monitoring signal, causing the signal to fail to accurately reflect the true state of the fire. The target interference factors may be environmental factors, the impact of the operation of peripheral equipment, and other external interference sources, and different types of fire monitoring signals face different interference factors.

[0052] Since different types of fire monitoring will be subject to interference from different sources during the collection process, it is necessary to analyze each type of fire monitoring signal to find out the interference factors that affect the accuracy of this type of signal and cause it to deviate from the actual fire situation, so as to provide a basis for the subsequent clarification of the filtering algorithm.

[0053] After clarifying the specific interference factors (i.e., target interference factors) faced by each type of fire monitoring signal, a filtering algorithm that matches and can effectively remove the interference factors is selected based on the characteristics of the target interference factors, such as the frequency and form of interference, to ensure that the filtered signal can more accurately reflect the actual situation of the fire.

[0054] Using the determined filtering algorithm for each type of fire monitoring signal, the corresponding original fire monitoring signal is actually filtered to remove the interference components, and finally multiple types of optimized target monitoring signals that can more accurately reflect the actual fire situation are obtained, providing a reliable data basis for subsequent fire warning or judgment environments.

[0055] From the above, it can be concluded that this embodiment can significantly improve the accuracy and reliability of fire monitoring by filtering each type of fire monitoring signal based on target interference factors. First, this embodiment accurately identifies the interference factors in various types of monitoring signals, and then selects the filtering algorithm in a targeted manner to effectively remove irrelevant noise and interference and retain key fire feature information. This not only optimizes the signal quality, but also enhances the ability of this embodiment to identify different types of fires, providing a clearer and more accurate signal basis for multi-type target monitoring, which is helpful for early warning and rapid response of fires, and greatly reduces the risk of fires.

[0056] In one embodiment of the present disclosure, the fire monitoring and early warning method based on the Internet of Things further includes: Determine environmental characteristics and signal characteristics; environmental characteristics include ventilation characteristics and combustion material characteristics, and signal characteristics include frequency characteristics and amplitude characteristics; The environmental features and signal features are processed based on the target Q learning algorithm to obtain the filtering parameters; wherein the target Q learning algorithm is a reinforcement learning algorithm based on the value function.

[0057] In one embodiment of the present disclosure, the fire monitoring and early warning method based on the Internet of Things further includes: Determine a state feature vector based on a reference environment feature and a reference signal feature; determining an action space based on reference filter parameters; The value function is determined based on the state feature vector, action space, time characteristics and discount factor.

[0058] In this embodiment, environmental characteristics are some characteristics of the external environment related to fire occurrence and monitoring. Ventilation characteristics may include the quality of ventilation, ventilation direction, wind speed, etc., which will affect the diffusion of smoke, heat and harmful gases generated by the fire; combustion material characteristics are the types of burning materials. Different materials have different combustion characteristics. For example, the speed of fire development, smoke and harmful gases generated when solid, liquid and gas fuels are burned are different. Environmental characteristics are important factors that affect the development of fire and the accuracy of monitoring signals.

[0059] The signal characteristics are the characteristics of the collected fire monitoring signal itself. The frequency characteristics are the speed of signal change. For example, whether the temperature signal rises slowly or changes sharply. Corresponding to different frequency conditions, it can reflect the rhythm of fire development. The amplitude characteristics are the size of the signal. For example, the concentration value of the smoke concentration signal can reflect the amount of smoke generated by the fire. Analyzing the signal characteristics helps to better grasp the actual situation of the fire and the subsequent processing of the signal.

[0060] The core of the target Q learning algorithm is to learn a value function (i.e., Q function), which represents the expected cumulative reward for taking a certain action in a certain state. When the agent tries different filter parameter adjustments (different actions), if the adjustment makes the fire monitoring signal more accurate, the environment will give a positive reward. The agent updates the value function based on the reward, and slowly learns which filter parameter adjustment is the best to learn the optimal behavior strategy. Finally, this embodiment can clarify how to set the filter parameters to obtain better results when the environmental characteristics and signal characteristics are determined.

[0061] The state feature vector is a vector representation formed by integrating and quantifying environmental features and signal features, which is used to fully describe the current state of the target area. The action space is a set of all possible actions that the agent (i.e., the module that controls the adjustment of filter parameters) can take. The above actions are various adjustment operations on the filter parameters, such as changing the parameter value of the filter algorithm, that is, increasing or decreasing the cutoff frequency, changing the weighting coefficient, etc., or switching different combinations of filter algorithms, etc. The action space defines the optional range of the agent.

[0062] Temporal features refer to some characteristics of fire monitoring signals that change over time or related elements in the time dimension during reinforcement learning, such as the changing trend of signals within a certain period of time, the number of reinforcement learning iterations, the time interval between each action selection and feedback, etc. Temporal features will affect the judgment of the fire status and the calculation of the value function.

[0063] The discount factor is a parameter used in reinforcement learning to calculate cumulative rewards. It is used to measure the importance of future rewards relative to current rewards. The value range is between 0 and 1. A value close to 0 indicates that current rewards are given more attention, while a value close to 1 indicates that future rewards also have a higher weight, which affects the agent's balance between long-term and short-term effects, and thus affects the formation of learning strategies. The value function is used to evaluate the expected cumulative rewards that can be obtained by taking an action in a certain state.

[0064] This embodiment uses the target Q learning algorithm and takes environmental characteristics and signal characteristics as input information. Under the framework of the algorithm, the intelligent agent (i.e., the virtual subject of parameter adjustment) continuously interacts with the environment, where the environment is a real fire monitoring scene and feedback after the filter parameters are adjusted. The value function is updated according to the reward signal fed back by the environment. After multiple iterations, the filter parameters that optimize the quality of the fire monitoring signal are finally determined.

[0065] Specifically, the steps for constructing the value function are as follows: Let the state eigenvector be , ,in Indicates ventilation characteristics, Indicates the characteristics of the burning material. and are the frequency and amplitude characteristics of the temperature signal, and They are the frequency and amplitude characteristics of the smoke concentration signal respectively.

[0066] The action space is , , including various adjustments to the filter parameters, such as changing the cutoff frequency, weighting coefficient, etc., among which, For action.

[0067] The next state is , indicating that in the state Next action The new state entered after the action. This state is determined based on the change of the environment due to the action and the change of the signal due to the adjustment of the filter parameters.

[0068] The time feature is , which can represent the iterative steps in the reinforcement learning process.

[0069] The discount factor is , , used to measure the importance of future rewards.

[0070] The immediate reward is , indicating that in the state Next action The reward is calculated based on the evaluation of the quality of the signal after filtering.

[0071] The expression of the value function is:

[0072] in, For the updated action; is the learning rate, , determines the degree of influence of the newly acquired reward information on the update of the value function. When it is small, the value function is updated slowly and relies more on historical information. When is larger, the new reward information has a greater impact on the value function, and the value function is updated faster.

[0073] The expression of the immediate reward function is:

[0074] in, and are weight coefficients used to adjust the relative importance of temperature signal and smoke concentration signal in reward calculation, and ; is the filtered temperature signal, is the smoke concentration signal after filtering; is the temperature signal under real fire conditions, It is the smoke density signal under real fire conditions.

[0075] The function is used to measure the matching degree of two signals. The expression of the matching function is:

[0076] in, is the number of signal samples, and are the signal and the The value of the sample points.

[0077] It can be concluded from the above that this embodiment realizes intelligent analysis and early warning of fire hazards by identifying environmental features and signal features and combining them with target Q learning algorithm processing. This embodiment can accurately extract the environmental and signal features before the fire, optimize the filtering parameters by reinforcement learning, and improve the accuracy of early warning. At the same time, by constructing the state feature vector, action space and value function, the dynamic adaptability and decision-making ability of this embodiment are further enhanced.

[0078] In one embodiment of the present disclosure, a fire warning is performed on a target area based on multiple types of target monitoring signals to obtain the state of the target area, including: Inputting multiple types of target monitoring signals into the target deep belief network model to determine the state of the target area, the target deep belief network model is trained based on historical monitoring signals and the states corresponding to the historical monitoring signals; Among them, the status of the target area is normal or fire.

[0079] In this embodiment, the target deep belief network model is a trained deep belief network that has learned a large number of historical monitoring signals and the relationship between the corresponding historical states. It can automatically extract features from the monitoring signals, thereby accurately classifying new monitoring signals and determining whether the target area is in a normal state or a fire state.

[0080] The historical monitoring signals are various fire-related signals obtained by monitoring the target area in the past period of time, and have been filtered. The above signals record the state information of the target area at different time points and can be used to train the deep belief network model. The state corresponding to the historical monitoring signal is the actual state of the target area corresponding to each set of historical monitoring signals, that is, normal state or fire state.

[0081] The state of the target area is the information of the current fire risk level of the target area, which can be divided into normal state and fire state. Among them, the normal state is no fire; the fire state can be pre-fire, initial fire and fire; the pre-fire is that the fire has not occurred but there are fire hazards; the initial fire is the stage when the fire point has open flames or smoldering, and the fire has not yet fully developed and spread; the fire has spread and has not been contained in time.

[0082] First, the processed multi-class target monitoring signals, such as filtered temperature, smoke concentration, harmful gas concentration and other signals, are input into the trained target deep belief network model. The model is constructed by learning and training a large number of historical monitoring signals and the corresponding actual states (normal or fire). The model uses the signal features and patterns learned during the training process to analyze and process the input current monitoring signal, and finally outputs a judgment result to determine whether the target area is currently in a normal state or a fire state.

[0083] Specifically, the construction process of the target deep belief network model is as follows: First, a large amount of historical data related to fire monitoring in the target area is collected and filtered as the input data set. The target area status corresponding to the above data is the output data set.

[0084] Second, before the collected data is used for model training, it needs to be preprocessed to improve data quality and model training effect. Preprocessing operations include data cleaning, normalization, and feature engineering.

[0085] Third, determine the structure of the deep belief network, including the number of layers, the number of neurons in each layer, and the activation function. The deep belief network is a deep learning model composed of multiple restricted Boltzmann machines (RBMs). The number of neurons in the input layer of the network depends on the number of features of the input data (such as the number of features such as temperature, smoke concentration, harmful gas concentration and its rate of change after preprocessing), and the number of neurons in the output layer is usually determined by the number of target area state categories to be predicted. The number of hidden layers and the number of neurons in each layer need to be adjusted according to the complexity of the data and the performance requirements of the model. The optimal architecture can be determined through experiments and verification. The activation function is used to introduce nonlinear factors so that the network can learn complex signal patterns.

[0086] Fourth, unsupervised pre-training. The training of the deep belief network first performs unsupervised pre-training, using stacked RBMs to learn the intrinsic structure of the data. Each RBM consists of a visible layer and a hidden layer. The RBM is trained in an unsupervised manner through the contrast divergence algorithm or its improved version, so that the hidden layer can learn the abstract features of the input data. Starting from the input layer, the preprocessed fire monitoring signal data is used as the visible layer input of the first RBM. After the training is completed, the output of the hidden layer of the first RBM is used as the visible layer input of the second RBM, and so on. All RBMs are trained layer by layer to provide a good starting point for the parameter initialization of the entire network, which is helpful for subsequent supervised fine-tuning.

[0087] Fifth, supervised fine-tuning. Add a classification layer on top of the deep belief network and input the labeled training data (including historical monitoring signals and corresponding target area status) into the network. Use the back propagation algorithm to fine-tune the parameters of the entire network so that the network can accurately classify whether a fire has occurred. During the fine-tuning process, calculate the cross entropy loss function between the predicted output and the true label, and adjust the network weights and biases according to the value of the loss function to minimize the loss and improve the classification accuracy.

[0088] The expression of the loss function is:

[0089] in, is the total number of samples, is the number of positive samples (fire status), is the number of negative samples (normal state), is the predicted value, is the true value, is the weight coefficient, is the adjustment factor.

[0090] Sixth, model evaluation and optimization. Use an independent validation dataset to evaluate the performance of the deep belief network model. The validation dataset is divided from the original data. Its data distribution and characteristics are similar to those of the training dataset, but it does not participate in the model training process. The evaluation indicators mainly include accuracy, recall rate, etc. The above indicators are used to measure the accuracy of the model in distinguishing between normal and fire states. If the model performance does not meet the requirements, the model needs to be optimized. The optimization methods include adjusting the network architecture, adjusting the training parameters, improving the data preprocessing method, or collecting more high-quality data, and then retraining and evaluating until the model performance achieves satisfactory results.

[0091] It can be concluded from the above that this embodiment improves the accuracy and efficiency of fire warning, and can make full use of historical monitoring signals and their corresponding states for model training, so that the model can accurately identify the state of the target area and distinguish between normal and fire conditions. This embodiment not only helps to discover fire hazards early and reduce fire losses, but also effectively avoids false alarms and missed alarms, and improves the reliability and practicality of early warning.

[0092] In one embodiment of the present disclosure, the fire monitoring and early warning method based on the Internet of Things further includes: In response to the state of the target area being a fire, extracting target features of each type of target monitoring signal; If the reference value of the target feature increases with the first step length, the first level fire warning is taken as the state of the target area; If the reference value of the target feature is increased with a second step length, the second level fire warning is taken as the state of the target area; If the reference value of the target feature increases with the third step length, the third level fire warning is used as the state of the target area; Among them, the first step length is smaller than the second step length, and the second step length is smaller than the third step length.

[0093] In one embodiment of the present disclosure, determining a target control strategy based on a state of a target area includes: If the state of the target area is a first-level fire warning, the first control strategy is used as the target control strategy; If the state of the target area is level 2 fire warning, the second control strategy is used as the target control strategy; If the state of the target area is the third level fire warning, the third control strategy is used as the target control strategy; Among them, the control quantities of fire extinguishing equipment in the first control strategy, the second control strategy and the third control strategy are different.

[0094] In this embodiment, the target feature is a key feature extracted from each type of target monitoring signal that can effectively characterize the fire situation and its development degree. For example, the temperature rise rate in the temperature monitoring signal, the concentration change gradient in the smoke concentration monitoring signal, etc. The reference value of the target feature is a quantified value of the target feature, which is used to measure the degree of fire development. The reference value is compared with the preset step size to determine the corresponding fire warning level.

[0095] The first step length, the second step length, and the third step length are all values ​​used to classify different fire warning levels, which can be change rates. Level 1 fire warning, level 2 fire warning, and level 3 fire warning are all fire warning levels of different severity divided according to the change of the target feature reference value. Level 1 fire warning is for the early stage of fire, level 2 fire warning is for the early stage of fire, and level 3 fire warning is for the occurrence of fire.

[0096] The calculation formula for the first step length is:

[0097] The calculation formula for the second step length is:

[0098] The calculation formula for the third step length is:

[0099] in, is the initial moment, is the time corresponding to the first step length calculation, is the time corresponding to the second step calculation, is the time corresponding to the third step calculation, , , All are The change rate of the target monitoring signal, For the Target monitoring signal, For the The value of a target monitoring signal at the corresponding moment.

[0100] The target control strategy is a pre-set response plan based on the different fire warning states of the target area, which may include the control strategy for fire extinguishing equipment, with the purpose of effectively controlling the fire. Different warning states correspond to different control strategies to adapt to different degrees of fire risks. The first control strategy, the second control strategy, and the third control strategy correspond to the specific control strategies for the first, second, and third level fire warning states, respectively. The above strategies differ in terms of the number of fire extinguishing equipment controlled and the control method.

[0101] This embodiment determines the fire warning level of the target area according to the change of the quantitative reference value of the extracted target feature. If the reference value of the target feature increases according to the first smaller step length, it means that the fire has not yet occurred but there is a fire hazard. At this time, the state of the target area is determined to be a first-level fire warning; when the increase in the reference value reaches the second step length, it indicates that the fire state is from the beginning of open flame or smoldering at the ignition point to the stage where the fire has not yet fully developed and spread. At this time, the state is upgraded to a second-level fire warning; if the reference value increases rapidly with a larger third step length, it means that the fire situation is serious and the fire is developing rapidly. The state of the target area is determined to be a third-level fire warning.

[0102] Once the fire warning status of the target area is determined, it is necessary to adopt corresponding control strategies to deal with the fire according to different status. If the target area is in the first-level fire warning status, it means that the fire risk is relatively low. At this time, the relatively conservative first control strategy is adopted, and the first number of fire-fighting equipment needs to be activated; when the status is upgraded to the second-level fire warning, the fire risk increases, and the second number of fire-fighting equipment needs to be activated, that is, the second control strategy is adopted; if it reaches the third-level fire warning, it means that the fire situation is critical and all available fire-fighting equipment needs to be fully activated, that is, the third control strategy is adopted. The main difference between different control strategies lies in the number and intensity of control over fire-fighting equipment. Among them, the first number is less than the second number.

[0103] It can be concluded from the above that this embodiment realizes a refined assessment of fire risk by real-time monitoring of the state of the target area and graded warning according to the increase step of the target feature reference value. Different levels of fire warning correspond to different control strategies, ensuring a rapid response in the early stage of the fire and effectively curbing the spread of the fire. In addition, this embodiment adjusts the control quantity of the fire extinguishing equipment to match the fire warning level, which not only ensures the fire extinguishing efficiency but also avoids excessive use of resources.

[0104] Corresponding to the fire monitoring and early warning method based on the Internet of Things in the above embodiment, Figure 2 This is a structural block diagram of a fire monitoring and early warning device based on the Internet of Things provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The fire monitoring and early warning device 20 based on the Internet of Things includes: a signal acquisition module 21, a filtering and processing module 22, an early warning module 23 and a control module 24.

[0105] Among them, the signal acquisition module 21 is used to obtain multiple types of fire monitoring signals in the target area; A filtering processing module 22 is used to filter each type of fire monitoring signal based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor; The early warning module 23 is used to provide a fire early warning for the target area based on multiple types of target monitoring signals to obtain the state of the target area; The control module 24 is used to determine a target control strategy based on the state of the target area, and control the fire extinguishing equipment in the target area based on the target control strategy.

[0106] In one embodiment of the present disclosure, the filtering processing module 22 is specifically configured to: Determine the target interference factors corresponding to each type of fire monitoring signal; Determine the filtering algorithm corresponding to each type of fire monitoring signal based on the target interference factor corresponding to the type of fire monitoring signal; Based on the filtering algorithm corresponding to each type of fire monitoring signal, the fire monitoring signal of this type is filtered to obtain multiple types of target monitoring signals.

[0107] In one embodiment of the present disclosure, the fire monitoring and early warning device 20 based on the Internet of Things further includes: a filtering parameter determination module; A filter parameter determination module is used to determine environmental characteristics and signal characteristics; environmental characteristics include ventilation characteristics and combustion material characteristics, and signal characteristics include frequency characteristics and amplitude characteristics; The environmental features and signal features are processed based on the target Q learning algorithm to obtain the filtering parameters; wherein the target Q learning algorithm is a reinforcement learning algorithm based on the value function.

[0108] In one embodiment of the present disclosure, the fire monitoring and early warning device 20 based on the Internet of Things further includes: a value function determination module; A value function determination module, used for determining a state feature vector based on a reference environment feature and a reference signal feature; determining an action space based on reference filter parameters; The value function is determined based on the state feature vector, action space, time characteristics and discount factor.

[0109] In one embodiment of the present disclosure, the early warning module 23 is specifically used to: Inputting multiple types of target monitoring signals into the target deep belief network model to determine the state of the target area, the target deep belief network model is trained based on historical monitoring signals and the states corresponding to the historical monitoring signals; Among them, the status of the target area is normal or fire.

[0110] In one embodiment of the present disclosure, the early warning module 23 is further configured to: In response to the state of the target area being a fire, extracting target features of each type of target monitoring signal; If the reference value of the target feature increases with the first step length, the first level fire warning is taken as the state of the target area; If the reference value of the target feature is increased with a second step length, the second level fire warning is taken as the state of the target area; If the reference value of the target feature increases with the third step length, the third level fire warning is used as the state of the target area; Among them, the first step length is smaller than the second step length, and the second step length is smaller than the third step length.

[0111] In one embodiment of the present disclosure, the control module 24 is specifically configured to: If the state of the target area is the first level fire warning, the first control strategy is used as the target control strategy; If the state of the target area is level 2 fire warning, the second control strategy is used as the target control strategy; If the state of the target area is the third level fire warning, the third control strategy is used as the target control strategy; Among them, the control quantities of fire extinguishing equipment in the first control strategy, the second control strategy and the third control strategy are different.

[0112] See also Figure 3 , Figure 3 This is a schematic block diagram of a fire monitoring and early warning system based on the Internet of Things provided by an embodiment of the present disclosure. Figure 3 The fire monitoring and early warning system 300 based on the Internet of Things in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other through a communication bus 305. The memory 304 is used to store computer programs, and the computer programs include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. Among them, the processor 301 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 24 are shown.

[0113] It should be understood that in the embodiments of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0114] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0115] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0116] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the fire monitoring and early warning method based on the Internet of Things provided in the embodiments of the present disclosure, and can also execute the implementation methods of the fire monitoring and early warning system based on the Internet of Things described in the embodiments of the present disclosure, which will not be repeated here.

[0117] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0118] The computer-readable storage medium may be an internal storage unit of the fire monitoring and early warning system based on the Internet of Things in any of the aforementioned embodiments, such as a hard disk or memory of the system. The computer-readable storage medium may also be an external storage device of the system, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the system. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the system and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the system. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0119] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0120] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0121] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0123] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0124] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A fire monitoring and early warning method based on the Internet of Things, characterized in that: include: Obtain multiple types of fire monitoring signals in the target area; Each type of fire monitoring signal is filtered based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor; Performing a fire warning on the target area based on the multiple types of target monitoring signals to obtain the state of the target area; A target control strategy is determined based on the state of the target area, and a fire extinguishing device in the target area is controlled based on the target control strategy.

2. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 1, characterized in that: The filtering process is performed on each type of fire monitoring signal based on the target interference factor to obtain multiple types of target monitoring signals, including: Determine the target interference factors corresponding to each type of fire monitoring signal; Determine the filtering algorithm corresponding to each type of fire monitoring signal based on the target interference factor corresponding to the type of fire monitoring signal; The fire monitoring signals of each type are filtered based on the filtering algorithm corresponding to the type of fire monitoring signals to obtain the multiple types of target monitoring signals.

3. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 2 is characterized in that: Also includes: Determining environmental characteristics and signal characteristics; the environmental characteristics include ventilation characteristics and combustion material characteristics, and the signal characteristics include frequency characteristics and amplitude characteristics; The environmental features and the signal features are processed based on a target Q learning algorithm to obtain filtering parameters; wherein the target Q learning algorithm is a reinforcement learning algorithm based on a value function.

4. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 3 is characterized in that: Also includes: Determine a state feature vector based on a reference environment feature and a reference signal feature; determining an action space based on reference filter parameters; A value function is determined based on the state feature vector, the action space, a time feature and a discount factor.

5. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 1, characterized in that: The performing fire warning on the target area based on the multiple types of target monitoring signals to obtain the state of the target area includes: Inputting multiple types of target monitoring signals into a target deep belief network model to determine the state of the target area, wherein the target deep belief network model is trained based on historical monitoring signals and states corresponding to the historical monitoring signals; The state of the target area is normal or fire.

6. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 5, characterized in that: Also includes: In response to the state of the target area being a fire, extracting target features of each type of target monitoring signal; If the reference value of the target feature increases with the first step length, the first level fire warning is taken as the state of the target area; If the reference value of the target feature is increased by a second step length, a second level fire warning is used as the state of the target area; If the reference value of the target feature increases with a third step length, a third level fire warning is used as the state of the target area; The first step length is smaller than the second step length, and the second step length is smaller than the third step length.

7. The fire monitoring and early warning method based on the Internet of Things as claimed in claim 6, characterized in that: The determining of the target control strategy based on the state of the target area includes: If the state of the target area is the first level fire warning, the first control strategy is used as the target control strategy; If the state of the target area is the second level fire warning, the second control strategy is used as the target control strategy; If the state of the target area is the third level fire warning, the third control strategy is used as the target control strategy; The first control strategy, the second control strategy and the third control strategy have different control quantities of fire extinguishing equipment.

8. A fire monitoring and early warning device based on the Internet of Things, characterized in that: include: A signal acquisition module is used to acquire multiple types of fire monitoring signals in the target area; A filtering processing module is used to filter each type of fire monitoring signal based on the target interference factor to obtain multiple types of target monitoring signals; wherein each type of fire monitoring signal corresponds to a different target interference factor; An early warning module, used for providing a fire early warning to the target area based on the multiple types of target monitoring signals, and obtaining the state of the target area; The control module is used to determine a target control strategy based on the state of the target area, and control the fire extinguishing equipment in the target area based on the target control strategy.

9. A fire monitoring and early warning system based on the Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images