Automatic fire extinguishing garbage can system with hierarchical alarm function

Abstract

The present application relates to the technical field of fire alarm, in particular to an automatic fire extinguishing garbage can system with hierarchical alarm function, which solves the technical problem that the garbage can fire alarm rating is mismatched with the actual burning stage due to the single temperature threshold evaluation mechanism in the prior art. The system is used for: collecting the time sequence data of carbon oxide concentration in the can, smoke concentration in the can, temperature in the can, ambient temperature outside the can, visible fire light intensity in the can and garbage weight in the can; analyzing the carbon oxygen high degree according to the carbon oxide concentration in the can, analyzing the can temperature abnormality degree according to the temperature in the can and the ambient temperature outside the can, and determining the basic representation value of the fire risk in the can; according to the time sequence correlation of the carbon oxygen high degree and the can temperature abnormality degree, the visible fire light intensity in the can, the smoke concentration in the can and the garbage weight in the can, the basic representation value is corrected to obtain the comprehensive evaluation result of the fire risk in the can; determining the corresponding alarm level according to the comprehensive evaluation result.

Description

Technical Field

[0001] This invention relates to the field of fire alarm technology, specifically to an automatic fire-extinguishing trash can system with graded alarm function. Background Technology

[0002] Accelerated urbanization and upgraded consumer spending have driven a year-on-year increase in household waste. The mixed disposal of cigarette butts, used lithium batteries, high-temperature ash, oily rags, and flammable packaging materials has become commonplace. Conventional plastic or metal trash cans only provide basic storage and lack temperature and smoke detection capabilities. The smoldering and smoke-producing stages often go undetected until the fire spreads, posing a significant safety hazard.

[0003] Automatic fire-extinguishing trash cans, as an important component of intelligent sanitation facilities, incorporate multiple application modules such as smoke sensors, temperature sensors, and water bladders. They can be connected to the internet to achieve real-time fire monitoring and autonomous response. Currently, automatic fire-extinguishing trash cans generally adopt a single threshold judgment strategy when performing real-time fire alarm rating. That is, they only judge the alarm based on the real-time temperature inside the can. When the temperature inside the can exceeds a preset threshold, an alarm is triggered and the water bladder is released.

[0004] However, in actual smoldering or open flame scenarios, the concentration changes of smoke and carbon oxides in the bin match the combustion stage. During the smoldering stage, carbon monoxide accumulates in large quantities while the temperature rises slowly, while during the open flame stage, the carbon dioxide concentration rises sharply and the temperature soars. The operating conditions of the trash can also vary significantly at different times. For example, short-term trash disposal can cause a sudden increase in smoke concentration. There is a serious mismatch between the above complex factors and the single temperature threshold judgment strategy, which greatly reduces the accuracy of alarm rating assessment. Consequently, the alarm adaptability of automatic fire extinguishing trash cans in actual operation is low, and the design efficiency of the graded alarm function is constrained. Summary of the Invention

[0005] To address the technical problem of mismatch between fire alarm ratings and actual combustion stages in existing technologies that rely on a single temperature threshold, the present invention aims to provide an automatic fire-extinguishing trash can system with graded alarm functions. The specific technical solution adopted is as follows: Firstly, an automatic fire-extinguishing trash can system with graded alarm function is provided, comprising: a data acquisition module for collecting time-series data on carbon oxide concentration, smoke concentration, temperature, ambient temperature, visible flame intensity, and trash weight inside the can; a risk determination module for analyzing the degree of carbon oxide rise based on the carbon oxide concentration, analyzing the degree of temperature anomaly based on the temperature inside and outside the can, and determining the basic characterization value of the fire risk inside the can; an interference correction module for performing interference correction on the basic characterization value based on the time-series correlation between the degree of carbon oxide rise and the degree of temperature anomaly, the visible flame intensity, smoke concentration, and trash weight inside the can, to obtain a comprehensive assessment result of the fire risk inside the can; and a graded alarm module for determining the corresponding alarm level based on the comprehensive assessment result.

[0006] Based on the above technical solution, in the automatic fire extinguishing trash can system with graded alarm function provided by the present invention, by integrating multi-source information such as carbon oxide concentration, temperature, flame, smoke and trash weight, a coupled evaluation basis for the degree of carbon oxide rise and the degree of trash can temperature anomaly is constructed. Furthermore, time-series correlation, flame deviation, smoke deviation and weight change trend are introduced to correct the interference of the basic characterization value, so that the system can distinguish between smoldering and open flame stages, identify non-combustion temperature anomalies and interference from falsely high trash disposal, and thus output a comprehensive evaluation result that is adapted to the actual combustion stage.

[0007] In conjunction with the first aspect mentioned above, in one possible implementation, the aforementioned risk determination module is specifically used to: analyze the degree of deviation of the carbon oxide concentration inside the tank from the historical peak concentration to obtain a carbon oxide rising factor characterizing the degree of carbon oxide rising; analyze the relative difference between the temperature inside the tank and the ambient temperature outside the tank, as well as the absolute level of the temperature inside the tank, to obtain a tank temperature anomaly factor characterizing the degree of tank temperature anomaly; and couple the carbon oxide rising factor and the tank temperature anomaly factor to obtain a basic characterization value.

[0008] In conjunction with the first aspect above, in one possible implementation, the carbon oxide concentration in the aforementioned barrel includes carbon monoxide concentration and carbon dioxide concentration; the risk determination module is specifically used to: obtain the peak carbon monoxide concentration and peak carbon dioxide concentration under normal operating conditions during historical operations; and determine the carbon oxide maximization factor based on the degree of deviation of the carbon monoxide concentration from the peak carbon monoxide concentration and the degree of deviation of the carbon dioxide concentration from the peak carbon dioxide concentration.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the aforementioned interference correction module is specifically used to: analyze the temporal correlation between the degree of carbon and oxygen increase and the degree of bin temperature anomaly within a preset window, and determine the alarm urgency level by combining the basic characterization values; analyze the deviation of the visible flame intensity inside the bin from the historical average visible flame intensity and the deviation of the smoke concentration inside the bin from the historical average smoke concentration, and determine the operating condition interference correction value by combining the temporal change trend of the waste weight inside the bin within the preset window; and determine the comprehensive evaluation result based on the alarm urgency level and the operating condition interference correction value.

[0010] In conjunction with the first aspect above, in one possible implementation, the aforementioned interference correction module is specifically used to: acquire a first data fluctuation amplitude value of the time-series data on the degree of carbon and oxygen increase within a preset window, and a second data fluctuation amplitude value of the time-series data on the degree of barrel temperature anomaly within a preset window; if either the first data fluctuation amplitude value or the second data fluctuation amplitude value falls within a preset noise floor fluctuation threshold range, then the time-series correlation coefficient between the degree of carbon and oxygen increase and the degree of barrel temperature anomaly is assigned a preset constant; if both the first data fluctuation amplitude value and the second data fluctuation amplitude value exceed the preset noise floor fluctuation threshold range, then the time-series correlation coefficient is determined by a preset correlation analysis algorithm.

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, the aforementioned interference correction module is specifically used to: obtain the difference between the intensity of visible flames inside the bin and the historical average intensity of visible flames, and determine the flame release trend value; obtain the ratio of the smoke concentration inside the bin to the historical average smoke concentration, and determine the smoke release trend value; perform linear fitting processing on the time-series data of the weight of garbage inside the bin within a preset window, obtain the absolute value of the slope of the fitted straight line, and determine the disposal disturbance value corresponding to the garbage disposal behavior based on the absolute value of the slope; and determine the working condition interference correction value based on the flame release trend value, the smoke release trend value, and the disposal disturbance value.

[0012] In conjunction with the first aspect above, in one possible implementation, the aforementioned graded alarm module is specifically used to: output a first alarm control signal containing a continuous monitoring instruction if the comprehensive evaluation result falls within the first alarm level range; output a second alarm control signal containing a water bladder coverage activation instruction if the comprehensive evaluation result falls within the second alarm level range; and output a third alarm control signal containing a water bladder coverage activation instruction and an on-site handling prompt instruction if the comprehensive evaluation result falls within the third alarm level range.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the system further includes: a monitoring and observation module for responding to a first alarm control signal and activating video monitoring inside the tank; a water bladder covering execution module for responding to a second or third alarm control signal and executing a water bladder covering action inside the tank; and a communication module for uploading the first, second, or third alarm control signals to the monitoring platform and receiving control commands issued by the monitoring platform.

[0014] In conjunction with the first aspect mentioned above, in one possible implementation, the data acquisition module specifically includes: an aerosol sensing unit for collecting the concentration of carbon oxides and smoke inside the bin; a temperature sensing unit for collecting the temperature inside the bin and the ambient temperature outside the bin; a visible light flame sensing unit for collecting the intensity of visible flame inside the bin; and a weight monitoring unit for collecting the weight of waste inside the bin.

[0015] In conjunction with the first aspect mentioned above, in one possible implementation, the data acquisition module further includes: a preprocessing unit, which is used to clean and preprocess the time-series data, and transmit the preprocessed carbon oxide concentration inside the bin, the temperature inside the bin, and the ambient temperature outside the bin to the risk determination module, and transmit the preprocessed visible flame intensity inside the bin, the smoke concentration inside the bin, and the weight of waste inside the bin to the interference correction module.

[0016] The present invention has the following beneficial effects: By integrating multi-source information such as carbon oxide concentration, temperature, flame, smoke, and waste weight, a coupled assessment basis for the degree of carbon oxide elevation and bin temperature anomaly is constructed. Furthermore, temporal correlation, flame deviation, smoke deviation, and weight change trend are introduced to correct for interference in the basic characterization values. This enables the system to distinguish between smoldering and open flame stages, identify non-combustion temperature anomalies and interference from inflated waste disposal, and thus output a comprehensive assessment result that is compatible with the actual combustion stage. Attached Figure Description

[0017] Figure 1 A system structure diagram of an automatic fire extinguishing trash can system with graded alarm function provided in one embodiment of the present invention; Figure 2 The present invention provides a method flowchart for controlling an automatic fire-extinguishing trash can with graded alarm function, according to an embodiment of the present invention. Detailed Implementation

[0018] The following description, in conjunction with the accompanying drawings, details the specific solution of an automatic fire-extinguishing trash can system with graded alarm function provided by the present invention.

[0019] Please see Figure 1The diagram illustrates a system structure of an automatic fire-extinguishing trash can system with graded alarm function according to an embodiment of the present invention. The automatic fire-extinguishing trash can system with graded alarm function includes: a data acquisition module 1, a risk determination module 2, an interference correction module 3, and a graded alarm module 4.

[0020] Optionally, the automatic fire-extinguishing trash can system with graded alarm function also includes: a monitoring and observation module 5, a water bladder covering execution module 6, and a communication module 7.

[0021] The data acquisition module 1 is located inside and on the outer wall of the trash can body structure. It collects time-series data on the concentration of carbon oxides inside the can, the smoke concentration inside the can, the temperature inside the can, the ambient temperature outside the can, the intensity of visible flame inside the can, and the weight of the trash inside the can. The data acquisition module 1 includes an aerosol sensing unit, a temperature sensing unit, a visible light flame sensing unit, and a weight monitoring unit. The aerosol sensing unit consists of a carbon monoxide sensor, a carbon dioxide sensor, and a particulate smoke sensor, located in the upper middle part of the trash can's interior to detect the gas composition produced by the combustion reaction. The temperature sensing unit consists of an internal temperature sensor and an external ambient temperature sensor, located on the inner side wall of the trash can and in a shaded area on the outside, respectively, to obtain the temperature difference between the inside and outside. The visible light flame sensing unit is located at the top inside the trash can, with its photosensitive surface facing downwards, covering the interior space to capture the radiation of open flame. The weight monitoring unit consists of a pressure sensor array located at the bottom of the trash can, used to sense changes in the total weight of the trash. After completing the acquisition, the data acquisition module 1 cleans and preprocesses the time series data, transmits the preprocessed carbon oxide concentration and temperature data to the risk determination module 2, and transmits the preprocessed fire intensity, smoke concentration and weight time series data to the interference correction module 3.

[0022] The risk assessment module 2 is located inside the waste bin's electrical control box. It analyzes the degree of carbon monoxide (COM) rise and the degree of bin temperature anomaly based on the COM concentration and temperature data from the data acquisition module 1, and determines the basic characterization values ​​for the risk of fire inside the bin. The risk assessment module 2 includes a COM rise determination submodule, a bin temperature anomaly determination submodule, and a basic characterization value determination submodule. Each submodule is implemented by running a preset algorithm through an embedded processor. The COM rise determination submodule compares the real-time carbon monoxide and carbon dioxide concentrations with their corresponding historical peak concentrations to obtain a COM rise factor. The bin temperature anomaly determination submodule jointly analyzes the relative difference between the bin's internal temperature and the external ambient temperature, as well as the absolute level of the bin's internal temperature, to obtain a bin temperature anomaly factor. The basic characterization value determination submodule couples the COM rise factor and the bin temperature anomaly factor to obtain a basic characterization value. The basic characterization value output by the risk assessment module 2 is transmitted to the interference correction module 3 as the baseline input for interference correction.

[0023] The interference correction module 3 is located inside the garbage bin's electrical control box. It uses time-series data on visible flame intensity, smoke concentration, and garbage weight inside the bin from the data acquisition module 1, combined with baseline characterization values ​​from the risk determination module 2, to perform interference correction on the baseline characterization values ​​to obtain a comprehensive assessment result of the fire risk inside the bin. The interference correction module 3 includes a sub-module for determining alarm urgency, a sub-module for determining visual fire characterization values, a sub-module for determining bin operating condition scores, and a sub-module for determining comprehensive assessment results. The alarm urgency determination sub-module analyzes the time-series correlation between the degree of carbon and oxygen increase and the degree of bin temperature anomaly within a preset window, and determines the alarm urgency level based on the baseline characterization values. The visual fire characterization value determination sub-module analyzes the deviation of flame intensity and smoke concentration from historical averages to determine the visual fire characterization value. The bin operating condition score determination sub-module analyzes the time-series trend of garbage weight changes, determines the disposal disturbance value, and combines it with the visual fire characterization value to obtain the operating condition interference correction value. The comprehensive assessment result determination sub-module integrates the alarm urgency determination value and the operating condition interference correction value to obtain the comprehensive assessment result. Interference correction module 3 effectively distinguishes between real combustion and non-combustion temperature anomalies, as well as artificially high interference caused by waste disposal, by introducing interference correction in the dimensions of flame, smoke, and weight. Its comprehensive evaluation results are transmitted to the graded alarm module 4.

[0024] The graded alarm module 4 is located inside the garbage bin's electrical control box. It determines the corresponding alarm level based on the comprehensive evaluation results from the interference correction module 3 and outputs corresponding alarm control signals. The graded alarm module 4 includes an alarm level interval division unit and an alarm control signal generation unit. The alarm level interval division unit divides the numerical range of the comprehensive evaluation results into a first alarm level interval, a second alarm level interval, and a third alarm level interval according to preset rules. The alarm control signal generation unit generates a first alarm control signal containing a continuous monitoring instruction, a second alarm control signal containing a water bag coverage activation instruction, or a third alarm control signal containing a water bag coverage activation instruction and an on-site handling prompt instruction, based on the interval into which the comprehensive evaluation results fall. The first alarm control signal output by the graded alarm module 4 is transmitted to the monitoring and observation module 5, while the second and third alarm control signals are transmitted to the water bag coverage execution module 6. Simultaneously, all alarm control signals at each level are transmitted to the communication module 7, enabling the uploading and remote notification of alarm information.

[0025] The monitoring and observation module 5 is located at the top inside the trash can. It responds to the first alarm control signal from the graded alarm module 4, initiating video monitoring inside the can and continuously transmitting the monitoring images. The monitoring and observation module 5 includes a camera and an image compression and encoding unit. Upon receiving the continuous monitoring instruction in the first alarm control signal, the camera switches from sleep mode to working mode and captures images of the space inside the can at a preset frame rate. The image compression and encoding unit compresses and encodes the captured image data before transmitting it to the communication module 7, which then uploads it to the remote monitoring platform for visual confirmation by operators. The monitoring and observation module 5 provides remote visual verification in low-fire-risk conditions, preventing false alarms caused by occasional sensor fluctuations from directly triggering firefighting actions. The transmitted monitoring image data is remotely shared through the communication module 7.

[0026] The water bladder covering execution module 6 is located on the top and side walls of the trash can. It responds to the second or third alarm control signal from the graded alarm module 4, executing the water bladder covering action inside the can. The water bladder covering execution module 6 includes a water bladder release device, an electrically controlled valve, and status feedback sensors. Upon receiving the water bladder covering start command, the pre-filled water bladder in the release device opens and releases through the electrically controlled valve. Under gravity, the water bladder expands and covers the surface of the trash inside the can, using the oxygen-isolating and cooling effects of water to suppress the fire's development. The status feedback sensors include pressure and position sensors, used to detect whether the water bladder release is successful and transmit the release status feedback to the communication module 7 for reporting to the remote monitoring platform. The water bladder covering execution module 6 achieves automated initial fire suppression response under medium-to-high fire risk conditions, effectively suppressing further fire development. Its execution status is remotely monitored through the communication module 7.

[0027] Communication module 7 is located inside the trash can's electrical control box and is used to enable two-way data communication between the trash can and the remote monitoring platform. Communication module 7 includes an internet communication unit and a wireless transmission unit. The internet communication unit encapsulates and uploads alarm control signals from the graded alarm module 4, monitoring screen data from the monitoring and observation module 5, and status feedback data from the water bladder coverage execution module 6 to the remote monitoring platform. The wireless transmission unit continuously listens for downlink data from the remote monitoring platform, receives control commands issued by the monitoring platform, parses them, and distributes them to various functional modules to achieve remote parameter adjustment and action triggering. As the connection hub between the system and the external monitoring network, communication module 7 enables real-time uploading of alarm information, remote sharing of monitoring screens, and reception and execution of control commands, supporting remote monitoring and manual intervention of the trash can's operating status by operators.

[0028] Through the coordinated operation of the above seven modules, the automatic fire extinguishing trash can system with graded alarm function realizes a complete closed loop from multi-source data acquisition, fire risk assessment, interference correction and adjustment to graded alarm response. At the same time, with the support of monitoring and observation, water bladder coverage and remote communication functions, it forms a full-chain intelligent security system covering data acquisition, risk analysis, intelligent decision-making, automatic execution and remote monitoring.

[0029] Please see Figure 2 The diagram illustrates a method flowchart for controlling an automatic fire-extinguishing trash can with a graded alarm function, according to an embodiment of the present invention. The method includes: S1. Collect time-series data on carbon oxide concentration, smoke concentration, temperature inside the bin, ambient temperature outside the bin, intensity of visible flame inside the bin, and weight of waste inside the bin.

[0030] Carbon oxide concentration refers to the concentration of one or both of the gases produced by the combustion reaction of combustibles inside the bin, namely carbon monoxide and carbon dioxide. Carbon monoxide is a key product of incomplete combustion in the smoldering stage, while carbon dioxide is the main product of complete combustion in the open flame stage. Smoke concentration refers to the concentration of particulate smoke inside the bin, reflecting the content of suspended particulate matter produced by the thermal decomposition or combustion of combustibles, and is an important indicator for early fire warning. Bin internal temperature refers to the real-time temperature of the internal space of the bin, while external ambient temperature refers to the real-time temperature of the external environment surrounding the bin. The two are used together to eliminate the interference of ambient temperature on the judgment of bin internal temperature, distinguishing whether the increase in bin internal temperature is caused by external environmental factors or by an internal fire. Visible flame intensity refers to the intensity of flame radiation detected by the visible light flame sensor inside the bin, used to characterize the energy release level during the open flame release stage. Time-series data of garbage weight refers to the sequence of total garbage weight data collected continuously at fixed time intervals, used to capture the characteristics of weight mutations caused by garbage disposal actions in a short period of time. Time series data refers to a sequence of data arranged in chronological order. Compared with single-point data, it can reflect the dynamic trend of parameter changes over time, providing a data foundation for analyzing the synchronicity and trend of parameter changes.

[0031] In some implementations, an aerosol sensing unit collects the concentrations of carbon oxides and smoke inside the container. This unit includes a carbon monoxide sensor, a carbon dioxide sensor, and a particulate smoke sensor. The carbon monoxide sensor converts the carbon monoxide concentration into an analog electrical signal using an electrochemical detection principle. The carbon dioxide sensor converts the carbon dioxide concentration into an analog electrical signal using a non-dispersive infrared absorption principle. The particulate smoke sensor converts the smoke particle concentration into an analog electrical signal using a light scattering principle. The system synchronously samples the analog electrical signals from the three sensors at a preset sampling frequency, typically once per second. After each sampling, the analog electrical signal is converted into a digital value using an analog-to-digital converter, yielding the current carbon monoxide concentration, carbon dioxide concentration, and smoke concentration values ​​inside the container. This data is then timestamped and saved.

[0032] The system collects the internal temperature and external ambient temperature of the trash can through a temperature sensing unit. This unit includes an internal temperature sensor and an external ambient temperature sensor. The internal temperature sensor is located on the upper-middle part of the inner side wall of the trash can, while the external ambient temperature sensor is located on the outside of the trash can in a shaded area away from heat sources. Both temperature sensors are digital, directly outputting digital temperature values. The system synchronously reads the data from both sensors at a preset sampling frequency, typically set to once per second. After each reading, the system obtains the current internal and external ambient temperature values, marks the data with a timestamp, and saves the data.

[0033] The intensity of visible flames inside the trash can is collected by a visible light flame sensor unit. This unit is positioned at the top inside the trash can, with its photosensitive surface facing downwards, covering the interior space. It outputs an analog electrical signal by detecting the intensity of visible light radiation within the can. The system samples the analog electrical signal from the visible light flame sensor at a preset sampling frequency, typically once per second. After sampling, the analog signal is converted into a digital value using an analog-to-digital converter to obtain the current intensity value of the visible flames inside the can. This value is then marked with a timestamp and saved. Under normal conditions without a fire, the intensity value of the visible flames inside the can reflects the ambient background light intensity.

[0034] The weight of the trash can is collected through a weight monitoring unit. This unit includes a pressure sensor array located at the bottom of the trash can. The array consists of multiple pressure sensors evenly distributed in a grid pattern to sense the total pressure exerted by the trash on the bottom of the can. The system synchronously reads the outputs of each sensor in the pressure sensor array at a preset sampling frequency, typically once per second. After each reading, the pressure values ​​from each sensor are summed to obtain the total weight of the trash in the can at that moment, and this value is then timestamped and saved. The pressure sensor array requires zero-point calibration after installation. Zero-point calibration is performed when the trash can is empty, and the output values ​​of each sensor at this time are recorded as the zero-point reference. Subsequent weight measurements are the difference between the current output value and the zero-point reference.

[0035] Furthermore, the time-series data undergoes cleaning and preprocessing. Specifically, the preprocessing unit executes differentiated cleaning strategies on each received raw time-series data. For data on carbon oxide concentration inside the bin, smoke concentration inside the bin, temperature inside the bin, ambient temperature outside the bin, and visible flame intensity inside the bin, these data types exhibit gradual variation under normal operating conditions. The preprocessing unit sequentially performs timestamp alignment, out-of-range rejection, missing time point interpolation compensation, and smoothing filtering on these data to remove noise from the sensor components themselves and high-frequency environmental disturbances. For data on the weight of waste inside the bin, this data exhibits a step-like abrupt change at the time of waste disposal. This abrupt change is a key information source for identifying waste disposal behavior and assessing its interference with alarm assessment. The preprocessing unit performs timestamp alignment, out-of-range rejection, and missing time point interpolation compensation on the waste weight data inside the bin, and retains its original abrupt change characteristics after completing the aforementioned processing, without performing smoothing filtering. After completing the above cleaning pretreatment, the pretreatment unit transmits the pretreated carbon oxide concentration inside the bin, the temperature inside the bin, and the ambient temperature outside the bin to the risk determination module, and transmits the pretreated visible flame intensity inside the bin, the smoke concentration inside the bin, and the weight of the waste inside the bin to the interference correction module.

[0036] The preset measurement range is set according to the different types of sensors. For example, the preset measurement range of carbon monoxide sensor is 0 to 1000 ppm, the preset measurement range of carbon dioxide sensor is 0 to 5000 ppm, the preset measurement range of smoke concentration sensor is 0 to 1000 μg / m³, the preset measurement range of temperature sensor is -20℃ to 200℃, the preset measurement range of fire intensity sensor is 0 to the preset maximum light intensity value, and the preset measurement range of weight sensor is 0 to the preset maximum load-bearing value. Data that exceeds the corresponding measurement range is judged as invalid data and discarded.

[0037] The missing time interpolation compensation process includes: if a certain type of data is missing at a certain acquisition time, the compensation value for the missing time is calculated using linear interpolation based on the adjacent valid values ​​of the data in the time series. If the number of consecutive missing times exceeds the preset missing threshold, the data in that time period is determined to be invalid and compensation for the data in that time period is stopped.

[0038] S2. Analyze the degree of carbon and oxygen increase based on the carbon oxide concentration inside the barrel, and analyze the degree of barrel temperature anomaly based on the temperature inside the barrel and the ambient temperature outside the barrel to determine the basic characterization value of the fire risk inside the barrel.

[0039] The carbon oxide concentration level refers to how close the concentration of carbon oxides inside the bin is to the highest concentration level recorded during normal operation. It quantifies whether the current carbon oxide concentration is close to or exceeds historical extreme conditions, reflecting whether the combustibles inside the bin have entered the combustion reaction stage and its intensity. The bin temperature anomaly level refers to the abnormal increase in bin temperature relative to ambient temperature. It distinguishes whether the temperature increase is caused by external environmental factors or an internal fire, thus avoiding false alarms caused by external factors such as high temperatures and dry weather. The baseline characterization value characterizes the basic risk level of the current fire situation inside the bin. Its value increases significantly when both carbon oxide and thermodynamic anomalies occur simultaneously, and remains low when only a single dimension is abnormal, thus avoiding bias in the baseline risk assessment caused by occasional drift of a single sensor element or non-fire factors.

[0040] In some implementations, the deviation of the carbon oxide concentration in the bin from historical peak concentrations is analyzed to obtain a carbon oxide susceptibility factor, characterizing the degree of carbon oxide susceptibility. The carbon oxide susceptibility factor is a numerical representation of the degree of carbon oxide susceptibility, obtained by normalizing and comparing the real-time carbon oxide concentration with historical peak concentrations. A higher value indicates that the current carbon oxide concentration is closer to the historical peak, and the greater the fire risk. The historical peak concentration refers to the maximum value of carbon oxide concentration data collected under normal operating conditions of the garbage bin (i.e., continuous monitoring without triggering fire suppression actions, including states where no alarm control signal is output or only the first alarm control signal is output). It serves as a benchmark for judging whether the current parameters are abnormal. The peak value is used instead of the average value because combustion reactions are sudden, and the peak value can reflect the upper limit of parameters under extreme conditions, thus more sensitively capturing the initial concentration surge in a fire. Simultaneously, a preset calibration lower limit is set for the historical peak concentration. This preset calibration lower limit is determined based on the lower limit of the corresponding sensor's range. If the historical peak concentration is lower than the preset calibration lower limit, the preset calibration lower limit is used instead of the historical peak concentration in the deviation calculation. For trash cans with no historical data during the initial use phase, their historical concentration peak values ​​are directly preset as the lower limit of the calibration. Deviation refers to the relative difference between the real-time parameter value and the historical baseline value, quantified by the ratio or difference between the real-time value and the baseline value. A larger deviation indicates a greater deviation from normal operating conditions. Specifically, the carbon oxide concentration in the trash can includes carbon monoxide and carbon dioxide concentrations. The peak values ​​of carbon monoxide and carbon dioxide concentrations under normal operating conditions during historical operations are obtained. Then, the deviation of the carbon monoxide concentration from its peak value is calculated by dividing the current carbon monoxide concentration by the peak value. Similarly, the deviation of the carbon dioxide concentration from its peak value is calculated by dividing the current carbon dioxide concentration by the peak value. The larger of the two deviations is used as the carbon-oxygen maximization factor. Taking the larger value ensures that the concentration deviation signal of the characteristic gas is completely preserved at any stage, avoiding dilution of the effective deviation signal due to the normal low value of the other gas concentration, ensuring sensitive detection of both smoldering and open flame fires.

[0041] By analyzing the relative difference between the temperature inside the container and the ambient temperature outside, as well as the absolute level of the temperature inside the container, a container temperature anomaly factor is obtained to characterize the degree of container temperature anomaly. The relative difference measures the deviation of the temperature inside the container from the ambient temperature. The absolute level refers to the numerical value of the temperature inside the container, used to measure the absolute level of the temperature inside the container. The container temperature anomaly factor is a numerical representation of the degree of container temperature anomaly, determined by both the relative difference between the temperature inside and outside the container and the absolute level of the temperature inside the container. When the temperature inside the container is low and the difference between it and the ambient temperature is small, the value of the container temperature anomaly factor is small, indicating that the temperature inside the container is normal; conversely, when the temperature inside the container is high and much higher than the ambient temperature outside the container, it indicates an abnormally high temperature inside the container. Specifically, the absolute value of the difference between the temperature inside and outside the container is compared with the ambient temperature outside the container. The ratio reflects the relative difference between the temperatures inside and outside the container. Note that to avoid a non-positive denominator, all ambient temperatures need to be converted to Kelvin absolute temperature scale (i.e., Celsius temperature plus 273.15) before the ratio calculation. Then, the temperature inside the tank is normalized to its maximum and minimum values ​​based on the preset range of the temperature sensor. Finally, the product of the relative difference ratio and the normalized temperature inside the tank is used as the tank temperature anomaly factor.

[0042] A baseline characterization value was obtained by coupling the carbon-oxygen rise factor and the bin temperature anomaly factor. Specifically, through multiplicative coupling, the baseline characterization value is larger when both the carbon-oxygen rise factor and the bin temperature anomaly factor rise simultaneously, and smaller when either one is lower, reflecting the coupling logic of mutual verification between the two signals. The higher the baseline characterization value, the higher the basic risk level of the current fire situation inside the garbage bin.

[0043] S3. Based on the temporal correlation between the degree of carbon and oxygen increase and the degree of abnormal temperature in the bin, the intensity of visible flames inside the bin, the smoke concentration inside the bin, and the weight of waste inside the bin, interference correction is performed on the basic characterization values ​​to obtain a comprehensive assessment result of the fire risk inside the bin.

[0044] The temporal correlation between the degree of carbon oxide elevation and the degree of bin temperature anomaly can reflect the characteristic of simultaneous spikes in carbon oxide concentration and temperature during actual combustion, thus distinguishing temperature anomalies caused by non-combustion factors such as cigarette butt contact or hot water spillage. The intensity of visible flame inside the bin can characterize the energy release level during the open flame release phase. The smoke concentration inside the bin can reflect the content of suspended particulate matter produced by the thermal decomposition or combustion of combustibles. The temporal trend of waste weight inside the bin can capture sudden weight changes caused by short-term waste disposal actions, thus identifying interference from artificially high smoke concentrations caused by dust stirred up during waste disposal. The basic characterization values ​​are only initially assessed from the dimensions of carbon oxide concentration and temperature, without considering the aforementioned non-combustion interference factors. Therefore, by introducing temporal correlation to verify the combustion authenticity of the basic characterization values, introducing flame intensity and smoke concentration to correct the fire urgency of the basic characterization values, and introducing waste weight change trends to suppress disposal interference, the verification results, correction results, and suppression results are integrated and corrected with the basic characterization values ​​to obtain a comprehensive assessment result of the in-bin fire risk that can distinguish between real fires and non-combustion interference and is adapted to the actual combustion stage.

[0045] In some implementations, the temporal correlation between the degree of carbon and oxygen increase and the degree of barrel temperature abnormality within a preset window is analyzed, and the alarm urgency is determined by combining the basic characterization values.

[0046] Specifically, the system acquires the first data fluctuation amplitude value of the time-series data on the carbon-oxygen rise level within a preset window, and the second data fluctuation amplitude value of the time-series data on the barrel temperature anomaly level within the preset window. The preset window is typically selected five seconds prior to the current time. The synchronous changes in carbon oxide concentration and temperature during the combustion reaction usually become apparent within seconds, and this duration is sufficient to cover multiple sampling points to ensure statistical validity. For example, the system can calculate the difference between the maximum and minimum values ​​of the carbon-oxygen rise factor within the window as the first data fluctuation amplitude value, and similarly calculate the difference between the maximum and minimum values ​​of the barrel temperature anomaly factor as the second data fluctuation amplitude value. The first data fluctuation amplitude value measures the drastic change in the carbon-oxygen rise level within that time period; the second data fluctuation amplitude value measures the drastic change in the barrel temperature anomaly level within that time period.

[0047] The preset noise floor fluctuation threshold refers to the upper limit of the fluctuation range for determining whether the data is in a relatively static thermodynamic and aerodynamic state. When the data fluctuation range falls within this threshold range, it indicates that the parameter is in an idle state rather than a changing state. For example, the preset noise floor fluctuation threshold can be set to 0.01. That is, if the difference between the maximum and minimum values ​​of the carbon-oxygen rise factor or barrel temperature anomaly factor within the preset window is less than or equal to 0.01, it indicates that the degree of carbon-oxygen rise or barrel temperature anomaly has not changed substantially during this period and is in a static idle state.

[0048] If either the first data fluctuation amplitude value or the second data fluctuation amplitude value falls within the preset noise floor fluctuation threshold range, meaning that no data set has undergone substantial fluctuations within the preset window, it indicates that there is no meaningful synchronicity of change available for analysis during that period. In this case, the correlation analysis algorithm is not invoked, and the time-series correlation coefficient between the carbon-oxygen trend and the barrel temperature anomaly is assigned a preset constant. The preset constant is a fixed substitute value assigned to the time-series correlation when the data is idle, used to avoid division by zero errors caused by calculating the correlation coefficient when the data is not fluctuating. This preset constant is a value much smaller than the minimum effective value of the normal correlation coefficient, for example, 0.01, indicating that there is no meaningful synchronicity between the two changes during that period. If both the first data fluctuation amplitude value and the second data fluctuation amplitude value exceed the preset noise floor fluctuation threshold range, meaning that both data sets have substantial fluctuations within the preset window, the time-series correlation coefficient is determined by the preset correlation analysis algorithm. The preset correlation analysis algorithm can adopt the Pearson correlation coefficient algorithm. This algorithm calculates the ratio of the product of the covariance and the standard deviation of the two sets of data in time series, and outputs a correlation coefficient between -1 and 1. A positive value indicates that the two sets of data change in the same direction, a negative value indicates that they change in opposite directions, and the closer to 1, the stronger the synchronicity of the change.

[0049] The time-series correlation coefficient is compared with a preset positive correlation floor value, and the larger of the two values ​​is taken as the final time-series correlation coefficient. This effectively suppresses negative values ​​to a very small constant close to zero, while retaining the true positive correlation in the same direction, thus accurately distinguishing between combustion fires and non-combustion heat interference. The preset positive correlation floor value is a constant greater than 0 and much smaller than the minimum effective value of the normal positive correlation coefficient, for example, 0.01.

[0050] The average of the final time-series correlation coefficients calculated during the normal operation of the trash can during its historical operation is used as the historical correlation average. For trash cans with no historical data at the initial stage of use, the historical correlation average is preset to 0.5, representing a moderate correlation level, and is continuously updated with actual data after operation. The ratio of the final time-series correlation coefficient to the historical correlation average is calculated to obtain the correlation ratio. A ratio greater than 1 indicates that the synchronicity of changes in the current period is better than the historical average, and the possibility of combustion is relatively high; a ratio less than 1 indicates that the synchronicity of changes in the current period is worse than the historical average, and the possibility of combustion is relatively low. The correlation ratio is then multiplied by the baseline characteristic value to obtain the alarm urgency performance. The alarm urgency performance is higher when the baseline characteristic value is high and the time-series change synchronicity is significantly better than the historical average, indicating that the current fire risk is not only abnormal in static indicators, but also that the dynamic change pattern conforms to combustion characteristics, and the alarm demand is relatively urgent; when the baseline characteristic value is high but the time-series change synchronicity is not significant, the value is suppressed, indicating that the current static anomaly may be due to instantaneous disturbance rather than continuous combustion, and the alarm urgency is low.

[0051] The deviations of the visible flame intensity inside the bin from the historical average visible flame intensity and the deviations of the smoke concentration inside the bin from the historical average smoke concentration are analyzed. Combined with the temporal variation trend of the weight of waste inside the bin within a preset window, the operating condition interference correction value is determined.

[0052] Specifically, the difference between the visible flame intensity inside the bin and the historical average visible flame intensity is obtained to determine the flame release trend value. The historical average visible flame intensity is obtained by taking the arithmetic mean of flame intensity data continuously collected under normal operating conditions (i.e., continuous monitoring without triggering fire extinguishing actions, including states where no alarm control signal is output or only the first alarm control signal is output). For bins with no historical data during the initial use phase, the average flame intensity is preset to the factory calibration value, which is determined based on the average background noise of the visible light flame sensor unit of this model in a standard darkroom environment. The difference between the visible flame intensity value inside the bin and the historical average visible flame intensity is calculated, and this difference is compared with 0. The larger of the two values ​​is taken as the positive flame intensity difference. Then, the positive flame intensity difference is normalized using a maximum value normalization algorithm to obtain the flame release trend value. The maximum value used in the normalization algorithm is set to the difference between the preset maximum light intensity value of the flame intensity sensor and the historical average visible flame intensity. After normalization, the range of the flame emission tendency value is between 0 and 1. When there is no flame in the barrel, the flame emission tendency value is close to 0, and when there is an open flame in the barrel, the flame emission tendency value is close to 1.

[0053] The ratio of the smoke concentration inside the bin to the historical average smoke concentration is used to determine the smoke release trend value. The historical average smoke concentration is obtained by taking the arithmetic mean of the smoke concentration data continuously collected inside the bin under normal operating conditions. Note that a lower limit constraint needs to be imposed on the historical average smoke concentration. If the historical average smoke concentration is less than the factory calibration value, the factory calibration value is used. For trash cans that are initially put into use and have no historical data, the average smoke concentration is preset to the factory calibration value, which is determined based on the average background noise of this model of smoke sensor in a clean air environment. The ratio of the smoke concentration inside the bin to the historical average smoke concentration is calculated as the smoke release trend value. When the smoke concentration is close to the historical average, the ratio is close to 1, indicating that the smoke level is normal; when the smoke concentration is much higher than the historical average, the ratio is significantly greater than 1, indicating that there is abnormal smoke release.

[0054] It should be noted that the flame emission trend value uses difference normalization because the flame intensity is close to the constant ambient background light intensity when there is no fire. Under open flame conditions, the flame intensity produces a significant absolute increase relative to the background light, and difference normalization can intuitively reflect this incremental characteristic. The smoke emission trend value uses a ratio because the smoke concentration fluctuates around the historical average under normal operating conditions. Under fire conditions, the smoke concentration increases exponentially rather than by an absolute increment. The ratio can reflect how many times the current smoke concentration is compared to the normal level, thus more sensitively capturing the relative change in smoke concentration.

[0055] Linear fitting is performed on the time-series data of the weight of garbage in the bins within a preset window to obtain the absolute value of the slope of the fitted line. The disposal disturbance value corresponding to the garbage disposal behavior is then determined based on the absolute value of the slope. Least squares linear fitting is performed on the time-series data of the weight of garbage in the bins within the preset window. The specific algorithm for least squares linear fitting is as follows: using time as the independent variable and weight as the dependent variable, the slope and intercept of the fitted line are determined by minimizing the sum of the squares of the vertical distances from each data point to the fitted line. The slope represents the rate of change of weight over time. The absolute value of the slope of the fitted line is divided by a preset reference mass change rate to obtain the dimensionless disposal disturbance value. The preset reference mass change rate is a preset constant with the same dimension as the slope, which can be set to 1 kg / s, aiming to convert the dimensional slope value into a dimensionless relative rate of change ratio. A larger disposal disturbance value indicates a more drastic weight change and a more obvious garbage disposal action.

[0056] Based on the flame emission trend value, smoke emission trend value, and disposal disturbance value, the operating condition interference correction value is determined. Among them, the flame emission trend value and smoke emission trend value reflect the direct manifestation of the current fire situation inside the bin; the larger these values ​​are, the more sufficient the visual and aerosol evidence of a fire exists inside the bin, and the more the operating environment tends to be a real fire rather than a normal state. The disposal disturbance value reflects whether there has been dust interference caused by recent waste disposal; the larger the value, the more likely the increased smoke concentration originates from disposal dust rather than combustion products, requiring a negative correction to the fire assessment.

[0057] Specifically, the flame emission trend value is summed with a preset fire enhancement constant of 1 to obtain the flame enhancement factor. The flame enhancement factor is then multiplied with the smoke emission trend value to obtain the fire condition visual representation value. In the smoldering state, the physical characteristic is "smoke but no fire," the flame emission trend value is 0, and the flame enhancement factor is 1, thus fully preserving the contribution of the smoke signal in the fire condition visual representation value. In the open flame state, the flame enhancement factor is greater than 1, representing an additional weighted amplification of the fire urgency by the open flame emission, thereby accurately reflecting the fire condition visual representation in both smoldering and open flame stages. A correction intensity coefficient is determined based on the placement disturbance value. The correction intensity coefficient is positively correlated with the placement disturbance value; the larger the placement disturbance value, the larger the correction intensity coefficient, indicating a stronger negative correction is needed to offset the artificially inflated effect of dust interference on smoke concentration assessment. For example, the sum of the placement disturbance value and the preset baseline constant 1 is used as the correction intensity coefficient. The fire condition visual representation value is divided by the correction intensity coefficient to obtain the operating condition interference correction value. When the visual representation of the fire is large and there is no interference from waste disposal, the correction strength coefficient is 1, and the visual representation of the fire is output as is without any distortion. When the visual representation of the fire is large but there is also strong interference from waste disposal, the correction strength coefficient is greater than 1, and the visual representation of the fire is attenuated accordingly, achieving a negative correction effect. The larger the value of the correction strength coefficient, the more severe the interference from waste disposal dust in the current operating condition, and the stronger the correction needs to be applied to suppress the artificially high signal.

[0058] The comprehensive evaluation result is determined based on the alarm urgency level and the operating condition interference correction value. A high alarm urgency level indicates a strong synchronous change in carbon-oxygen concentration and temperature, with a high baseline characteristic value. A large operating condition interference correction value indicates the presence of real fire and smoke with minimal interference from waste disposal. When both conditions are met simultaneously, the comprehensive evaluation result reaches its maximum, accurately identifying a real fire. Conversely, a high alarm urgency level but a small operating condition interference correction value indicates that although carbon-oxygen and temperature change synchronously, the high smoke level may be due to artificially inflated smoke levels caused by waste disposal; this is suppressed to avoid false alarms. Similarly, a low alarm urgency level but a large operating condition interference correction value indicates that although fire and smoke are present, carbon-oxygen and temperature are not synchronized, possibly due to a non-combustion source or external smoke intrusion; this is also suppressed. Based on this, the product of the alarm urgency level and the operating condition interference correction value can be calculated. Then, a preset compression mapping function is used for normalization. The compression mapping function employs the hyperbolic tangent function (tanh function) to force the product result to converge within a closed interval of 0 to 1. The normalized value is then used as the comprehensive evaluation result.

[0059] S4. Determine the corresponding alarm level based on the comprehensive evaluation results.

[0060] The comprehensive assessment result refers to the final quantitative assessment value of the fire risk inside the container after interference correction. Its value ranges from zero to one. The higher the value, the greater the fire risk inside the container, and the higher the alarm level that needs to be triggered. The alarm level refers to the fire risk level determined by dividing the numerical range of the comprehensive assessment result. Different alarm levels correspond to different response measures, thereby achieving differentiated handling of fire risks.

[0061] In some implementations, the numerical range of the comprehensive evaluation results is divided, specifically based on the statistical distribution of the comprehensive evaluation results during the historical operation of the trash cans: Comprehensive evaluation data for at least thirty days is collected from the trash cans under normal operating conditions (i.e., continuous monitoring without triggering fire suppression actions, including states where no alarm control signal is output or only the first alarm control signal is output). The mean and standard deviation of this dataset are calculated. The larger of the sum of the mean and one standard deviation and a preset first semantic threshold is used as the boundary threshold between the first alarm level interval and the second alarm level interval. The larger of the sum of the mean and two standard deviations and a preset second semantic threshold is used as the boundary threshold between the second alarm level interval and the third alarm level interval. The preset first semantic threshold and the preset second semantic threshold are pre-set alarm level boundary benchmark values, corresponding to the empirical boundary levels between low risk and medium risk, and medium risk and high risk, respectively. For example, the preset first semantic threshold can be set to 0.33 and the preset second semantic threshold to 0.67.

[0062] If the comprehensive assessment result falls within the first alarm level range, a first alarm control signal containing a continuous monitoring instruction will be output; if the comprehensive assessment result falls within the second alarm level range, a second alarm control signal containing a water bladder coverage activation instruction will be output; if the comprehensive assessment result falls within the third alarm level range, a third alarm control signal containing a water bladder coverage activation instruction and an on-site handling prompt instruction will be output.

[0063] The continuous monitoring command refers to the instruction included in the alarm control signal to activate the video monitoring inside the tank and continuously transmit the monitoring images. This is used for remote visual confirmation of the tank's internal conditions under low fire risk conditions. The water bladder coverage activation command refers to the instruction included in the alarm control signal to trigger the water bladder release device inside the tank to perform a covering action. This is used for physical isolation and initial fire suppression of the fire source inside the tank under medium to high fire risk conditions. The on-site handling prompt command refers to the instruction included in the alarm control signal to push a prompt message to a preset terminal indicating that manual on-site fire suppression is required. This is used to notify relevant personnel to take further on-site fire suppression measures under the highest fire risk conditions.

[0064] In some implementations, the monitoring module responds to the first alarm control signal and initiates video monitoring inside the tank. The water bladder covering execution module responds to the second or third alarm control signal and executes the water bladder covering action inside the tank. The communication module uploads the first, second, or third alarm control signals to the monitoring platform and receives control commands issued by the monitoring platform.

[0065] Based on the above technical solution, by integrating multi-source information such as carbon oxide concentration, temperature, flame, smoke, and waste weight, a coupled evaluation basis for the degree of carbon oxide rise and bin temperature anomaly is constructed. Furthermore, time-series correlation, flame deviation, smoke deviation, and weight change trend are introduced to correct the interference of the basic characterization value, enabling the system to distinguish between smoldering and open flame stages, identify non-combustion temperature anomalies and interference from inflated waste disposal, thereby outputting a comprehensive evaluation result that is compatible with the actual combustion stage.

Claims

1. An automatic fire-extinguishing trash can system with graded alarm function, characterized in that, include: The data acquisition module is used to collect time-series data on carbon oxide concentration, smoke concentration, temperature, ambient temperature, visible light intensity, and weight of waste inside the bin. The risk determination module is used to analyze the degree of carbon and oxygen increase based on the carbon oxide concentration inside the barrel, analyze the degree of barrel temperature anomaly based on the temperature inside the barrel and the ambient temperature outside the barrel, and determine the basic characterization value of the fire risk inside the barrel. The interference correction module is used to perform interference correction on the basic characterization values ​​based on the temporal correlation between the degree of carbon and oxygen increase and the degree of barrel temperature anomaly, the intensity of visible fire inside the barrel, the smoke concentration inside the barrel, and the weight of garbage inside the barrel, to obtain a comprehensive assessment result of the fire risk inside the barrel. The graded alarm module is used to determine the corresponding alarm level based on the comprehensive evaluation results.

2. The automatic fire-extinguishing trash can system with graded alarm function according to claim 1, characterized in that, The risk determination module is specifically used for: By analyzing the degree of deviation of the carbon oxide concentration in the barrel from the historical peak concentration, a carbon oxide maximization factor characterizing the degree of carbon oxide maximization is obtained. By analyzing the relative difference between the temperature inside the barrel and the ambient temperature outside the barrel, as well as the absolute level of the temperature inside the barrel, a barrel temperature anomaly factor characterizing the degree of barrel temperature anomaly is obtained. The basic characterization value is obtained by coupling the carbon-oxygen uptrend factor with the barrel temperature anomaly factor.

3. The automatic fire-extinguishing trash can system with graded alarm function according to claim 2, characterized in that, The concentration of carbon oxides in the barrel includes both carbon monoxide and carbon dioxide concentrations; the risk determination module is specifically used for: Acquire the peak carbon monoxide and carbon dioxide concentrations under normal operating conditions during historical operations; The carbon-oxygen maximization factor is determined based on the degree of deviation of the carbon monoxide concentration from the peak value of the carbon monoxide concentration and the degree of deviation of the carbon dioxide concentration from the peak value of the carbon dioxide concentration.

4. The automatic fire-extinguishing trash can system with graded alarm function according to claim 1, characterized in that, The interference correction module is specifically used for: The temporal correlation between the degree of carbon and oxygen increase and the degree of barrel temperature abnormality within the preset window is analyzed, and the alarm urgency is determined by combining the basic characterization values. The deviation of the visible flame intensity inside the bin from the historical average visible flame intensity and the deviation of the smoke concentration inside the bin from the historical average smoke concentration are analyzed, and the working condition interference correction value is determined by combining the temporal change trend of the garbage weight inside the bin within the preset window. The comprehensive evaluation result is determined based on the alarm urgency level and the operating condition interference correction value.

5. The automatic fire-extinguishing trash can system with graded alarm function according to claim 4, characterized in that, The interference correction module is specifically used for: The time-series data of the carbon-oxygen level increase is obtained within the preset window as a first data fluctuation amplitude value, and the time-series data of the barrel temperature anomaly level is obtained within the preset window as a second data fluctuation amplitude value. If either the first data fluctuation amplitude value or the second data fluctuation amplitude value falls within the preset bottom noise fluctuation threshold range, then the time-series correlation coefficient between the carbon and oxygen tendency and the barrel temperature anomaly is assigned a preset constant value. If both the first data fluctuation amplitude value and the second data fluctuation amplitude value exceed the preset noise floor fluctuation threshold range, the time-series correlation coefficient is determined by a preset correlation analysis algorithm.

6. The automatic fire-extinguishing trash can system with graded alarm function according to claim 4, characterized in that, The interference correction module is specifically used for: The difference between the intensity of visible flames inside the barrel and the historical average intensity of visible flames is obtained to determine the flame release trend value; Obtain the ratio of the smoke concentration inside the barrel to the historical average smoke concentration to determine the smoke release trend value; The time-series data of the weight of garbage in the bin within the preset window is linearly fitted to obtain the absolute value of the slope of the fitted line, and the garbage disposal disturbance value corresponding to the garbage disposal behavior is determined based on the absolute value of the slope. The operating condition interference correction value is determined based on the flame emission trend value, the smoke emission trend value, and the release disturbance value.

7. The automatic fire-extinguishing trash can system with graded alarm function according to claim 1, characterized in that, The hierarchical alarm module is specifically used for: If the comprehensive evaluation result falls within the first alarm level range, then a first alarm control signal containing a continuous monitoring instruction is output. If the comprehensive evaluation result falls within the second alarm level range, a second alarm control signal containing a water bladder coverage activation command will be output. If the comprehensive evaluation result falls within the third alarm level range, a third alarm control signal containing a water bladder coverage activation command and an on-site handling prompt command will be output.

8. The automatic fire-extinguishing trash can system with graded alarm function according to claim 7, characterized in that, Also includes: The monitoring and observation module is used to respond to the first alarm control signal and start video monitoring inside the tank; The water bladder covering execution module is used to respond to the second alarm control signal or the third alarm control signal to perform the water bladder covering action inside the tank. The communication module is used to upload the first alarm control signal, the second alarm control signal, or the third alarm control signal to the monitoring platform, and to receive control commands issued by the monitoring platform.

9. The automatic fire-extinguishing trash can system with graded alarm function according to claim 1, characterized in that, The data acquisition module specifically includes: An aerosol sensing unit is used to collect the concentration of carbon oxides and the concentration of smoke inside the barrel. A temperature sensing unit is used to collect the temperature inside the barrel and the ambient temperature outside the barrel. A visible light flame sensing unit is used to collect the intensity of visible flames inside the barrel; A weight monitoring unit is used to collect the weight of the waste inside the bin.

10. The automatic fire-extinguishing trash can system with graded alarm function according to claim 9, characterized in that, The data acquisition module further includes: The preprocessing unit is used to clean and preprocess the time-series data, and transmit the preprocessed carbon oxide concentration inside the bin, the temperature inside the bin, and the ambient temperature outside the bin to the risk determination module, and transmit the preprocessed visible light intensity inside the bin, the smoke concentration inside the bin, and the weight of waste inside the bin to the interference correction module.

Images