Family kitchen fire leaving monitoring method and device

Abstract

The invention discloses a method and a device for monitoring fire leaving of a home kitchen, and belongs to the technical field of kitchen monitoring, and the method comprises the following steps: collecting gas data and a human body feedback label, and calculating a meter reading period based on parameters of a stove corresponding to the gas data; filtering the gas data based on the meter reading period to obtain target gas data; based on the target gas data and the human body feedback tag, judging whether a person leaving fire event occurs, if not, updating the gas data and the human body feedback tag based on the time scale, and if yes, generating alarm information based on the target gas data and the human body feedback tag, and giving an alarm; according to the method, the target gas data is obtained by filtering out the non-cooker gas consumption data and the human intervention data in the gas data, and the gas data of the off-person stewing scene and the gas data of the fire off-person scene are taken as two independent sets, so that separation of the off-person stewing scene and the fire off-person scene is realized; and the identification accuracy of the live-fire leaving scene is obviously improved.

Description

Technical Field

[0001] This invention relates to the field of kitchen monitoring technology, specifically to a method and device for monitoring the presence of open flames in a home kitchen when no one is present. Background Technology

[0002] Home cooking, as an inevitable part of daily life, often involves the use of open flames. Open flames are a major cause of kitchen fires and pose a significant safety hazard. Currently, some devices on the market detect when someone is away from the stove. These devices typically determine whether the cook has left the stove area by detecting the heat emitted by the human body, or by using human body sensors to detect the presence of a person in the stove area. They then simply compare the time the person is not in the stove area with a corresponding time threshold to identify whether a "hot stove unattended" incident has occurred. However, in home cooking scenarios, unattended cooking is a very common occurrence, especially since the cooking time and flame size vary depending on the dish being cooked. Therefore, simply judging by the time a person is not in the stove area and the corresponding time threshold often misidentifies unattended cooking as a "hot stove unattended" incident, resulting in low accuracy in identifying such incidents.

[0003] Chinese Patent, Publication No. CN118736756A, Publication Date: October 1, 2024, discloses a kitchen fire early warning system and method, including: a hot work sensor, an oil temperature detection device, a gas detection device, a smoke detection device, a personnel evacuation assistance device, a camera, a buzzer, a PoE switch, an intelligent early warning host, and a cloud platform; the hot work sensor, oil temperature detection device, gas detection device, smoke detection device, personnel evacuation assistance device, and camera are all electrically connected to the intelligent early warning host, the intelligent early warning host is electrically connected to the buzzer, the PoE switch is used to support the network transmission and power supply capabilities of the camera, and the PoE switch is electrically connected to the cloud platform; the personnel evacuation assistance device cooperates with the hot work sensor to generate a leave-of-post signal, and determines whether a hot work evacuation has occurred based on the leave-of-post signal and the warning duration; however, this invention only determines whether a hot work evacuation has occurred based on the leave-of-post signal and the warning duration, which makes it difficult to identify situations where people are cooking without being present, resulting in low accuracy in identifying hot work evacuation scenarios. Summary of the Invention

[0004] The purpose of this invention is to address the problem that existing technologies struggle to identify cooking scenarios where people are away from the stove, leading to low accuracy in identifying such scenarios. This invention proposes a method and device for monitoring cooking scenarios where people are away from the stove in a home kitchen. The method calculates the meter reading cycle based on the stove's parameters and collects gas data and human feedback tags corresponding to the stove. Based on the meter reading cycle, it filters out non-stove gas usage data and human intervention data from the gas data to obtain target gas data. Based on the target gas data and human feedback tags, it determines whether a cooking event has occurred. When a cooking event is detected, it generates and activates a corresponding alarm based on the target gas data and human feedback tags. This invention separates cooking scenarios from cooking scenarios and cooking scenarios where people are away from the stove by filtering out non-stove gas usage data and human intervention data, thus significantly improving the accuracy of identifying cooking scenarios where people are away from the stove.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: Firstly, this application provides a method for monitoring unattended cooking in a home kitchen, comprising the following steps: Collect gas data and human feedback tags, and calculate the meter reading cycle based on the parameters of the stove corresponding to the gas data; Target gas data is obtained by filtering gas data based on the meter reading cycle; Based on the target gas data and human feedback tags, determine whether a hot work incident has occurred. If not, update the gas data and human feedback tags based on the time scale. If so, generate alarm information based on the target gas data and human feedback tags and trigger an alarm.

[0006] In this solution, the meter reading cycle is calculated based on the parameters of the stove corresponding to the gas data. This yields the reading time of the gas meter when the stove operates according to the calibrated parameters. Gas data is then filtered based on the meter reading cycle, removing non-stove gas usage data and data subject to human intervention that show significant differences in the data change cycle compared to the stove's gas usage data. For example, data corresponding to a scenario of cooking away from people with a small, continuously linearly changing flow rate is used to obtain target gas data containing only stove gas usage data. This separates the cooking away from people scenario from interfering scenarios such as cooking away from people, effectively improving the accuracy of the gas data. Based on the target gas data and human feedback tags, it is determined whether a cooking away from people event has occurred. When a cooking away from people event is determined, corresponding alarm information is generated and triggered based on the target gas data and human feedback tags, significantly improving the accuracy of identifying cooking away from people scenarios.

[0007] Preferably, the specific process for calculating the meter reading cycle based on the parameters of the stove corresponding to the gas data is as follows: Obtain the minimum power, maximum power, and corresponding gas meter accuracy and gas calorific value of the stove; The longest reading time is calculated based on the minimum power, gas meter accuracy, and gas calorific value. The shortest reading time is calculated based on the maximum power, the gas meter's calculation accuracy, and the gas's calorific value. The longest and shortest read times are combined to obtain the read cycle.

[0008] Preferably, after the meter reading cycle is calculated, the meter reading cycle needs to be adjusted according to the number of burner holes in the stove, specifically as follows: Obtain the number of holes in the stove and set the multi-hole ratio based on the number of holes; The read cycle is adjusted by a fixed ratio based on the aforementioned aperture ratio to obtain the aperture update cycle.

[0009] Preferably, the specific process of obtaining the target gas data by filtering gas data based on the meter reading cycle is as follows: Gas data is processed based on a time scale to obtain a gas sequence; The gas variation is calculated based on the gas sequence, and the gas variance is calculated based on the gas variation. The gas usage type is determined based on the gas consumption change and the meter reading cycle. If the gas consumption change is greater than the minimum value of the meter reading cycle and less than the maximum value of the meter reading cycle, the gas usage type is determined to be gas for stoves; otherwise, the gas usage type is determined to be gas for non-stoves. The combustion type of the corresponding gas data is determined based on the gas variance and variance threshold. If the gas variance is less than the variance threshold, the combustion type of the gas data is determined to be steady combustion. If the gas variance is greater than or equal to the variance threshold, the combustion type of the gas data is determined to be fluctuating combustion. When the gas type is gas for stove and the combustion type is stable combustion, mark the corresponding gas data as the target gas data.

[0010] Preferably, the specific process for determining whether a hot work incident has occurred based on target gas data and human feedback tags is as follows: The number of target gas data is statistically analyzed based on the time scale, and a risk of unattended hot work is determined based on the number of data and a threshold. If the number of data is greater than or equal to the threshold, an unattended hot work risk is determined to have occurred; if the number of data is less than the threshold, no unattended hot work risk is determined to have occurred. When a fire-starting crisis occurs, the feedback time point of the human body feedback tag is extracted, and the fire-starting event is determined based on the feedback time point and the time scale. If the feedback time point is greater than or equal to the minimum value of the time scale and less than or equal to the maximum value of the time scale, it is determined that no fire-starting event has occurred; otherwise, it is determined that a fire-starting event has occurred.

[0011] Preferably, the specific process of generating alarm information based on target gas data and human feedback tags is as follows: extracting the time scale of the target gas data, the feedback time point of the human feedback tag, and the stove hole number corresponding to the target gas data; calculating the unattended time period based on the time scale and feedback time point; Alarm information is obtained by compiling the unattended time period and stove hole number.

[0012] Preferably, after the alarm is completed, it is also necessary to detect the valve switching command, specifically: The valve switching command is cyclically detected based on a preset detection cycle. When a valve switching command is successfully detected, the corresponding valve switching action is executed based on the valve switching command.

[0013] Secondly, this application provides a home kitchen open flame monitoring device, which includes: a data acquisition module, a gateway module and an interaction module; The data acquisition module is used to collect gas data and human body feedback tags, transmit the gas data and human body feedback tags to the gateway module, and execute valve switching actions based on the control commands transmitted by the gateway module. The gateway module determines the unattended hot work event based on gas data and human feedback tags, transmits the data corresponding to the unattended hot work event to the interaction module, and generates control commands based on the valve switching commands transmitted by the interaction module. The interaction module generates alarm information based on the data transmitted by the gateway module, executes alarm actions based on the alarm information, and detects valve switching commands.

[0014] Preferably, the data acquisition module includes: a gas meter unit and a human body sensing unit; The data output terminal of the gas meter unit is electrically connected to the first data input terminal of the gateway module, the controlled terminal of the gas meter unit is electrically connected to the control terminal of the gateway module, and the data output terminal of the human body sensing unit is electrically connected to the second data input terminal of the gateway module.

[0015] Preferably, the interaction module includes: a cloud platform unit and a terminal unit; The first input terminal of the cloud platform unit is electrically connected to the output terminal of the gateway module, the output terminal of the cloud platform unit is electrically connected to the input terminal of the terminal unit, the output terminal of the terminal unit is electrically connected to the second input terminal of the cloud platform unit, and the control terminal of the cloud platform unit is electrically connected to the controlled terminal of the gateway module.

[0016] The beneficial effects of this invention are: This application calculates the meter reading cycle based on the parameters of the stove corresponding to the gas data, obtaining the reading time of the corresponding gas meter when the stove is running at the calibrated power, that is, the meter reading cycle including the longest and shortest reading times; based on the meter reading cycle, gas data is filtered to remove non-stove gas usage data and human intervention data whose data change cycle is significantly different from the stove gas usage data. For example, data corresponding to the scenario of cooking away from people with a small flow rate that changes linearly is obtained to obtain target gas data containing only stove gas usage data, realizing the separation of the cooking away from people scenario and other interfering scenarios, effectively improving the accuracy of gas data; based on the target gas data and human feedback tags, it is determined whether a cooking away from people event has occurred. When a cooking away from people event is determined to have occurred, corresponding alarm information is generated and an alarm is triggered based on the target gas data and human feedback tags, significantly improving the identification accuracy of cooking away from people scenario. Attached Figure Description

[0017] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. The drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0018] Figure 1 A flowchart illustrating a method for monitoring the presence of open flames in a home kitchen. Figure 2 This is a schematic diagram of a household kitchen open flame monitoring device. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only one preferred embodiment of this invention and are only used to explain this invention. They do not limit the scope of protection of this invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0020] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations (or steps) can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it may also have additional steps not included in the figures; the process may correspond to a method, function, procedure, subroutine, subroutine, etc.

[0021] Example 1: like Figure 1 As shown in the figure, this application provides a method for monitoring the absence of people when using open flames in a home kitchen, including the following steps: A1. Collect gas data and human feedback tags, and calculate the meter reading cycle based on the parameters of the stove corresponding to the gas data; A11. Obtain the minimum power, maximum power, and corresponding gas metering accuracy and gas calorific value of the stove; A12. Calculate the longest meter reading time based on the minimum power, gas meter accuracy, and gas calorific value. A13. Calculate the shortest reading time based on the maximum power, gas meter accuracy, and gas calorific value. A14. The longest and shortest reading times are combined to obtain the reading cycle.

[0022] Specifically, in the physical scenario of using a single-hole stove, the single-hole stove includes parameters such as minimum combustion power, maximum combustion power, gas meter accuracy, gas calorific value, and gas type. Since different types of gas have different calorific values, the gas calorific value is used directly instead of the gas type; that is, the gas type is not directly used in calculating the meter reading cycle. The longest meter reading time is calculated based on the minimum power, gas meter accuracy, and gas calorific value. The calculation process for the longest meter reading time can be expressed as the longest meter reading formula, which is specifically as follows: In the formula, ST max The longest reading time, HV is the calorific value of the gas, DGT is the metering accuracy of the gas meter, and P is the maximum reading time. min Minimum combustion power; The shortest reading time is calculated based on the maximum power, the gas meter's calculation accuracy, and the gas's calorific value. This calculation process can be expressed as a shortest reading time formula, which is as follows: In the formula, ST min The shortest reading time, HV is the calorific value of the gas, DGT is the metering accuracy of the gas meter, and P is the minimum reading time. max Maximum combustion power; Once the longest and shortest reading times are successfully calculated, the longest reading time is taken as the maximum value of the reading cycle, and the shortest reading time is taken as the minimum value of the reading cycle. The reading cycle is obtained by combining the maximum, minimum, and intermediate values ​​of the maximum and minimum values.

[0023] Furthermore, when the stove is a multi-port stove, both the maximum and minimum combustion power vary exponentially with the number of ports. The corresponding meter reading cycle also varies exponentially with the number of ports. If the meter reading cycle is directly applied to a single-port stove to determine whether a hand-cooking event has occurred, the accuracy of the target gas data will decrease due to this exponential relationship, thus reducing the accuracy of the hand-cooking event determination. To address the issue of poor target gas data accuracy, the meter reading cycle needs to be adjusted according to the number of ports on the stove. Specifically: Obtain the number of holes in the stove and set the multi-hole ratio based on the number of holes; The read cycle is adjusted by a fixed ratio based on the aforementioned aperture ratio to obtain the aperture update cycle. The fixed ratio adjustment can be expressed by a ratio adjustment formula, which is: In the formula, MST is the reading cycle of the multi-hole stove, N is the number of holes in the stove, and ST is the reading time, which includes the shortest reading time and the longest reading time.

[0024] A2. Obtain target gas data by filtering gas data based on the meter reading cycle; A21. Gas data is processed based on time scale to obtain gas sequence; A22. Calculate the gas change based on the gas sequence, and calculate the gas variance based on the gas change. A23. Determine the gas usage type based on the gas change and meter reading cycle. If the gas change is greater than the minimum value of the meter reading cycle and less than the maximum value of the meter reading cycle, then the gas usage type is determined to be stove gas usage; otherwise, the gas usage type is determined to be non-stove gas usage. A24. Based on the gas variance and variance threshold, determine the combustion type of the corresponding gas data. If the gas variance is less than the variance threshold, determine that the combustion type of the gas data is steady combustion. If the gas variance is greater than or equal to the variance threshold, determine that the combustion type of the gas data is fluctuating combustion. A25. When the gas type is gas for stove and the combustion type is stable combustion, mark the corresponding gas data as the target gas data.

[0025] Specifically, the gas data reporting time is defined as a unidirectional increasing sequence to obtain the gas sequence, which can be represented as {st0, st1, ..., st...} n} ， Where st0 is the first reporting time of the gas data, st1 is the second reporting time of the gas data, and st n For the nth time the gas data is reported, the gas change is obtained by subtracting two adjacent data points in the gas sequence, such as Δst = st1 - st0， The Δst represents the gas change. After calculating all gas changes, the gas change is input into the variance formula to calculate the gas variance. That is, the gas sequence is essentially a sequence of gas data reporting times, the gas change is essentially the interval between adjacent reporting times, and the gas variance is essentially the variance of the interval between adjacent reporting times. Secondly, because there are differences between scenarios involving cooking away from people and scenarios involving stewing away from people, the instantaneous gas consumption is high in scenarios involving cooking away from people, and the corresponding gas meter reading changes rapidly in a short time, resulting in a short gas data reporting interval (i.e., a short meter reading cycle). In the case of stewing away from people, the gas consumption is high instantaneously, and the corresponding gas meter reading changes rapidly in a short time, resulting in a short gas data reporting interval (i.e., a short meter reading cycle). The instantaneous gas consumption for cooking scenarios is small, resulting in slow changes in gas meter readings and long reporting intervals for gas data (i.e., long reading cycles). Therefore, when the gas change is greater than 0.9 times the shortest reading time and less than 1.1 times the longest reading time (i.e., the gas change is greater than the minimum reading cycle and less than the maximum reading cycle), the gas data corresponding to this change essentially represents the gas consumption data of the stove requiring the presence of the cook during cooking, such as gas data for scenarios where cooking is done away from the user. Otherwise, when the gas change is less than or equal to 0.9 times the shortest reading time... When the meter reading time is short, or the longest reading time is greater than or equal to 1.1 times the gas change, the gas data corresponding to the gas change is essentially non-stove gas usage data for cooking scenarios where no cook is present, such as gas data for unattended stewing. However, filtering gas data based on a single condition makes it difficult to identify the gas usage type under specific conditions. For example, the gas usage type during high-heat stewing is essentially non-stove gas usage, but high-heat stewing carries the risk of unattended cooking. To accurately identify the gas usage type under specific conditions, gas variance is used as a second filtering condition to identify the combustion type of the corresponding gas data. When the gas variance... When the variance is less than the variance threshold, the combustion type of the gas data is stable combustion of high-heat stewing in the corresponding off-site cooking scenario. When the gas variance is greater than or equal to the variance threshold, it proves that there is human intervention in the corresponding cooking behavior, and the combustion type is fluctuating combustion. After the gas usage type and combustion type are determined, if the gas usage type is stove gas usage and the combustion type is stable combustion, the corresponding gas data is marked as target gas data, the data corresponding to non-stove gas usage or fluctuating combustion are cleared, and the corresponding target gas data is adjusted to continuous time data according to the time scale. For example, when the gas sequence st0 to st n When the corresponding data is cleared, st n+1 As a new ST 0， The target gas data is obtained by adjusting the corresponding data order.

[0026] A3. Based on the target gas data and human feedback tags, determine whether a hot work incident has occurred. If not, update the gas data and human feedback tags based on the time scale. If so, generate alarm information based on the target gas data and human feedback tags and trigger an alarm.

[0027] The specific process for determining whether a hot work incident has occurred based on target gas data and human feedback tags is as follows: The number of target gas data is statistically analyzed based on the time scale, and a risk of unattended hot work is determined based on the number of data and a threshold. If the number of data is greater than or equal to the threshold, an unattended hot work risk is determined to have occurred; if the number of data is less than the threshold, no unattended hot work risk is determined to have occurred. When a fire-starting crisis occurs, the feedback time point of the human body feedback tag is extracted, and the fire-starting event is determined based on the feedback time point and the time scale. If the feedback time point is greater than or equal to the minimum value of the time scale and less than or equal to the maximum value of the time scale, it is determined that no fire-starting event has occurred; otherwise, it is determined that a fire-starting event has occurred.

[0028] Specifically, when the target gas data is determined, the number of data points in the target gas data is statistically analyzed based on the time series, i.e., the number of data points. Based on the number of data points and the corresponding number threshold, it is determined whether a fire hazard has occurred. In this embodiment, the number threshold is set to 3. When the number of data points is greater than or equal to 3, it is determined that a fire hazard has occurred. At this time, the feedback time point of the human feedback tag is extracted, and it is detected whether the feedback time point is within the time scale of the target gas data. If it is, it proves that there are cooks in the stove area during the time period corresponding to the target gas data, and no fire hazard has occurred. If not, i.e., the feedback time point is less than the minimum value of the time scale, or the feedback time point is greater than the maximum value of the time scale, it proves that there is no one in the stove area during the time period corresponding to the target gas data, and a fire hazard has occurred.

[0029] Secondly, the specific process of generating alarm information based on target gas data and human body feedback tags is as follows: Extract the time scale of the target gas data, the feedback time point of the human body feedback tag, and the stove hole number corresponding to the target gas data; calculate the unmanned time period based on the time scale and feedback time point; Alarm information is obtained by compiling the unattended time period and stove hole number.

[0030] Specifically, if the feedback time point is less than the minimum value of the time scale, the difference between the maximum value of the time scale and the feedback time point is used to obtain the unmanned time period. If the feedback time point is greater than the maximum value of the time scale, the difference between the feedback time point and the minimum value of the time scale is used to obtain the unmanned time period. The unmanned time period and the stove hole number are then combined to obtain alarm information.

[0031] In addition, after the alarm in step A3 is completed, it is also necessary to check the valve switching command, specifically: The valve switching command is cyclically detected based on a preset detection cycle. When a valve switching command is successfully detected, the corresponding valve switching action is executed based on the valve switching command.

[0032] Specifically, the cook can identify the stove and time of the unattended cooking incident based on the unattended time period and stove hole number in the alarm information, and then decide whether to shut off the gas valve and issue a valve opening / closing command to instruct the corresponding household kitchen unattended cooking monitoring device to perform the corresponding valve opening / closing action, shut off the stove, and eliminate the safety hazards caused by unattended cooking incidents.

[0033] Secondly, such as Figure 2 As shown in the figure, this application embodiment also provides a home kitchen open flame monitoring device, including: a data acquisition module, a gateway module and an interaction module; The data acquisition module is used to collect gas data and human body feedback tags, transmit the gas data and human body feedback tags to the gateway module, and execute valve switching actions based on the control commands transmitted by the gateway module. The gateway module determines the unattended hot work event based on gas data and human feedback tags, transmits the data corresponding to the unattended hot work event to the interaction module, and generates control commands based on the valve switching commands transmitted by the interaction module. The interaction module generates alarm information based on the data transmitted by the gateway module, executes alarm actions based on the alarm information, and detects valve switching commands.

[0034] Preferably, the data acquisition module includes: a gas meter unit and a human body sensing unit; The data output terminal of the gas meter unit is electrically connected to the first data input terminal of the gateway module, the controlled terminal of the gas meter unit is electrically connected to the control terminal of the gateway module, and the data output terminal of the human body sensing unit is electrically connected to the second data input terminal of the gateway module.

[0035] Preferably, the interaction module includes: a cloud platform unit and a terminal unit; The first input terminal of the cloud platform unit is electrically connected to the output terminal of the gateway module, the output terminal of the cloud platform unit is electrically connected to the input terminal of the terminal unit, the output terminal of the terminal unit is electrically connected to the second input terminal of the cloud platform unit, and the control terminal of the cloud platform unit is electrically connected to the controlled terminal of the gateway module.

[0036] Specifically, the gas meter unit includes a smart gas meter with a smart valve, capable of transmitting gas readings to the gateway module in real time, and also capable of receiving control commands from the gateway module and executing valve switching actions based on the control commands; the human body sensing unit includes a human body sensing device, such as microwave radar, millimeter-wave radar, or a time camera, which can detect whether there is a cook in the area corresponding to the stove. During the identification of unattended cooking events, the human body sensing data closest to the start time of the gas data on the time scale is marked as a human body feedback tag, and the human body feedback tag is dynamically updated based on the time scale of the human body sensing data. For example, if a human body feedback tag exists, and another human body sensing data closer to the start time of the gas data appears on the time scale, or new human body sensing data appears within the corresponding time period of the gas data, then the human body sensing data is marked as a new human body feedback tag; the gateway module includes a smart gateway, which can receive gas data and human body feedback tags, and determine whether an unattended cooking event has occurred based on the gas data and human body feedback tags; The cloud platform unit is essentially a cloud platform capable of processing data from devices and users, specifically data between the cook and the stove, enabling human-machine interaction. When a cooking event occurs while the cook is away from the stove, the gateway module transmits the corresponding data to the cloud platform. Based on this data, the cloud platform generates an alarm message and notifies the cook via software push or SMS. The terminal unit can be a smartphone, tablet, or smartwatch, or other mobile device with internet and communication capabilities. The mobile device receives data transmitted from the cloud platform to notify the cook. Simultaneously, the cook can send valve switching commands to the cloud platform via the mobile device. Upon receiving the valve switching command, the cloud platform transmits it to the gateway module. The gateway module generates control commands based on the valve switching commands to control the smart gas meter to perform the valve switching action. Both the smart gas meter and the human body sensing device can wirelessly connect to the gateway module via Bluetooth, StarFlash, or Zigbee. The gateway module wirelessly connects to the cloud platform via Ethernet, Wi-Fi, or cellular networks, and the cloud platform wirelessly connects to the mobile device via Ethernet, Wi-Fi, or cellular networks.

[0037] This embodiment has at least the following substantial effects: This embodiment calculates the meter reading cycle based on the parameters of the stove corresponding to the gas data, obtaining the reading time of the corresponding gas meter when the stove is running at the calibrated power, that is, the meter reading cycle including the longest and shortest reading times; based on the meter reading cycle, gas data is filtered to remove non-stove gas usage data and human intervention data whose data change cycle is significantly different from the stove gas usage data. For example, data corresponding to the scenario of cooking away from people with a small flow rate that changes linearly is obtained to obtain target gas data containing only stove gas usage data, realizing the separation of the cooking away from people scenario and other interfering scenarios, effectively improving the accuracy of gas data; based on the target gas data and human feedback tags, it is determined whether a cooking away from people event has occurred. When a cooking away from people event is determined to have occurred, corresponding alarm information is generated and an alarm is triggered based on the target gas data and human feedback tags, significantly improving the identification accuracy of cooking away from people scenario.

[0038] The above-described specific embodiments are preferred embodiments of the present invention and are not intended to limit the specific scope of the present invention. The scope of the present invention includes, but is not limited to, these specific embodiments. All equivalent changes made in accordance with the shape, structure, and method of the present invention are within the protection scope of the present invention.

Claims

1. A method for monitoring the absence of people when using open flames in a household kitchen, characterized in that, Includes the following steps: Collect gas data and human feedback tags, and calculate the meter reading cycle based on the parameters of the stove corresponding to the gas data; Target gas data is obtained by filtering gas data based on the meter reading cycle; Based on the target gas data and human feedback tags, determine whether a hot work incident has occurred. If not, update the gas data and human feedback tags based on the time scale. If so, generate alarm information based on the target gas data and human feedback tags and trigger an alarm.

2. The method for monitoring the absence of people at a home kitchen during cooking according to claim 1, characterized in that, The specific process for calculating the meter reading cycle based on the parameters of the stove corresponding to the gas data is as follows: Obtain the minimum power, maximum power, and corresponding gas meter accuracy and gas calorific value of the stove; The longest reading time is calculated based on the minimum power, gas meter accuracy, and gas calorific value. The shortest reading time is calculated based on the maximum power, the gas meter's calculation accuracy, and the gas's calorific value. The longest and shortest read times are combined to obtain the read cycle.

3. The method for monitoring the absence of people at a home kitchen during cooking according to claim 2, characterized in that, After the meter reading cycle is calculated, the meter reading cycle needs to be adjusted according to the number of burner holes in the stove. Specifically: Obtain the number of holes in the stove and set the multi-hole ratio based on the number of holes; The read cycle is adjusted by a fixed ratio based on the aforementioned aperture ratio to obtain the aperture update cycle.

4. The method for monitoring the absence of people at a household kitchen during cooking according to claim 1, characterized in that, The specific process of obtaining the target gas data by filtering gas data based on the meter reading cycle is as follows: Gas data is processed based on a time scale to obtain a gas sequence; The gas variation is calculated based on the gas sequence, and the gas variance is calculated based on the gas variation. The gas usage type is determined based on the gas consumption change and the meter reading cycle. If the gas consumption change is greater than the minimum value of the meter reading cycle and less than the maximum value of the meter reading cycle, the gas usage type is determined to be gas for stoves; otherwise, the gas usage type is determined to be gas for non-stoves. The combustion type of the corresponding gas data is determined based on the gas variance and variance threshold. If the gas variance is less than the variance threshold, the combustion type of the gas data is determined to be steady combustion. If the gas variance is greater than or equal to the variance threshold, the combustion type of the gas data is determined to be fluctuating combustion. When the gas type is gas for stove and the combustion type is stable combustion, mark the corresponding gas data as the target gas data.

5. The method for monitoring the absence of people at a home kitchen during cooking according to claim 1, characterized in that, The specific process for determining whether a hot work incident has occurred based on target gas data and human feedback tags is as follows: The number of target gas data is statistically analyzed based on the time scale, and a risk of unattended hot work is determined based on the number of data and a threshold. If the number of data is greater than or equal to the threshold, an unattended hot work risk is determined to have occurred; if the number of data is less than the threshold, no unattended hot work risk is determined to have occurred. When a fire-starting crisis occurs, the feedback time point of the human body feedback tag is extracted, and the fire-starting event is determined based on the feedback time point and the time scale. If the feedback time point is greater than or equal to the minimum value of the time scale and less than or equal to the maximum value of the time scale, it is determined that no fire-starting event has occurred; otherwise, it is determined that a fire-starting event has occurred.

6. The method for monitoring the absence of people at a home kitchen during cooking according to claim 1, characterized in that, The specific process for generating alarm information based on target gas data and human body feedback tags is as follows: Extract the time scale of the target gas data, the feedback time point of the human body feedback tag, and the stove hole number corresponding to the target gas data; calculate the unmanned time period based on the time scale and feedback time point; Alarm information is obtained by compiling the unattended time period and stove hole number.

7. The method for monitoring the absence of people at a household kitchen during cooking according to claim 1, characterized in that, After the alarm is completed, it is also necessary to check the valve switching command, specifically: The valve switching command is cyclically detected based on a preset detection cycle. When a valve switching command is successfully detected, the corresponding valve switching action is executed based on the valve switching command.

8. A household kitchen open flame monitoring device, characterized in that, It includes: a data acquisition module, a gateway module, and an interaction module; The data acquisition module is used to collect gas data and human body feedback tags, transmit the gas data and human body feedback tags to the gateway module, and execute valve switching actions based on the control commands transmitted by the gateway module. The gateway module determines the unattended hot work event based on gas data and human feedback tags, transmits the data corresponding to the unattended hot work event to the interaction module, and generates control commands based on the valve switching commands transmitted by the interaction module. The interaction module generates alarm information based on the data transmitted by the gateway module, executes alarm actions based on the alarm information, and detects valve switching commands.

9. A household kitchen open flame monitoring device according to claim 8, characterized in that, The data acquisition module includes: a gas meter unit and a human body sensing unit; The data output terminal of the gas meter unit is electrically connected to the first data input terminal of the gateway module, the controlled terminal of the gas meter unit is electrically connected to the control terminal of the gateway module, and the data output terminal of the human body sensing unit is electrically connected to the second data input terminal of the gateway module.

10. A household kitchen open flame monitoring device according to claim 8, characterized in that, The interaction module includes: a cloud platform unit and a terminal unit; The first input terminal of the cloud platform unit is electrically connected to the output terminal of the gateway module, the output terminal of the cloud platform unit is electrically connected to the input terminal of the terminal unit, the output terminal of the terminal unit is electrically connected to the second input terminal of the cloud platform unit, and the control terminal of the cloud platform unit is electrically connected to the controlled terminal of the gateway module.

Images