High-precision identification system and method for stove fire combined with escaped gas detection

Abstract

The present invention discloses a stove fire high-precision identification system and method combined with escaped gas detection, which relates to the technical field of stove fire identification, including: collecting and detecting gas concentration and / or the concentration of waste gas generated after gas combustion and whether there is someone; filtering and normalizing the collected data; establishing an air flow influence coefficient to judge the degree to which the gas concentration and / or the concentration of waste gas generated after gas combustion are affected by air flow; establishing a gas concentration and / or the concentration balance coefficient of waste gas generated after gas combustion to judge whether a "gas combustion" event occurs; further judging the occurrence of the "gas combustion" event; establishing a stove fire identification model, setting the stove fire risk alarm level; setting the stove fire risk warning prompt, and making a risk alarm in time. Effectively improve the ability to identify stove fires and reduce the risk of "fires without people" on the stove.

Description

Technical Field

[0001] The present invention relates to the field of stove fire identification, and in particular to a stove fire high-precision identification system and method combined with escaped gas detection. Background Art

[0002] Modern families generally use various types of gas (natural gas, artificial gas or liquefied petroleum gas, etc.) as kitchen fuel. As a cooking stove, gas stoves are very easy to adjust the firepower, especially suitable for Chinese families with rich cooking skills. Even now that all kinds of electronic cookware are very complete, many people still love to use gas stoves for cooking. Behind the blue flames, there are rich Chinese food and the unique fireworks of Chinese life. However, everything has two sides. Although gas stoves are convenient to use, it is not easy to achieve timing and remote switching operations. In daily life, users often leave the kitchen temporarily during cooking and forget to return to the kitchen in time, which leads to overheating and burning of pots, and even more serious consequences. Therefore, a technology commonly known as "fire leaving people" came into being. This technology is designed to detect the combination of "fire source open" + "unattended" and give an alarm in time. To a certain extent, it can remind users to return to the kitchen in time to avoid accidents.

[0003] With the current technical level, it is very easy to detect whether there is anyone in the kitchen. Designers can use infrared detection, radar sensing or image recognition technology to detect whether there is a person. However, when detecting whether the fire source is turned on, some technical difficulties are encountered. First, the pots on the stove will block the sensor and the flame, so that the sensor cannot detect the flame. Secondly, in many cases, users use a very small flame (commonly known as a slow fire) to cook for a long time, and the flame sensor is also difficult to detect a very small flame. At present, there are some products on the market that use the condition of "the range hood is turned on" to infer that the user is using fire. The core logic of the product is that when it is detected that the range hood is turned on and the person is not in the kitchen, the alarm condition is triggered, thereby issuing an alarm. This method is very dependent on the user's usage habits, because there is no definite correlation between turning on the range hood and turning on the gas stove, so the missed alarm and false alarm rate of such products cannot be guaranteed. Summary of the invention

[0004] In order to solve the above technical problems, a high-precision stove fire identification system and method combined with escaped gas detection are provided. The technical solution solves the problem proposed in the above background technology that the pots on the stove will be blocked between the sensor and the flame, making the sensor unable to detect the flame. Secondly, in many cases, users use a very small flame (commonly known as a low fire) for a long time to cook. The flame sensor finds it difficult to detect a very small flame and is too dependent on the user's usage habits, resulting in a lack of a definite correlation between turning on the range hood and turning on the gas stove, leading to the problem of missed reports and false alarm rates in the judgment of stove fire.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A high-precision method for identifying stove fire combined with escaped gas detection, characterized by comprising:

[0007] Set up a hardware collection device to collect and detect the gas concentration and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone present;

[0008] Based on the normalization formula, the data collected by the hardware acquisition device are filtered and normalized;

[0009] Establish an air flow influence coefficient to determine the degree to which the concentration of gas and / or the concentration of exhaust gas generated after gas combustion is affected by the range hood or air flow;

[0010] According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion to determine whether a "gas combustion" event occurs;

[0011] Further judge the occurrence of the "gas combustion" event based on the detection data of the temperature sensor and / or flow sensor;

[0012] According to the air flow influence coefficient, gas concentration and / or the concentration equilibrium coefficient of exhaust gas generated after gas combustion and the convergence degree of temperature drop rate, a stove fire identification model is established, and the stove fire risk alarm level is set;

[0013] Set up a stove fire risk alarm and issue a risk alarm in a timely manner according to the stove fire risk alarm level.

[0014] Preferably, the hardware gas collection device is provided to collect and detect the concentration of gas and / or the concentration of waste gas generated after gas combustion and whether there is anyone there, specifically including:

[0015] According to the type of gas and the type of waste gas generated after combustion, a gas collection device is provided to collect and detect the concentration of gas and / or the concentration of waste gas generated after gas combustion in real time;

[0016] A human body detection sensor is set up to determine whether there is a human body within the detection range;

[0017] Temperature and flow sensors are provided to collect temperature data within the detection range and gas flow entering the air inlet of the stove in real time;

[0018] An air flow collection device is provided to collect air flow data within the detection range in real time.

[0019] Preferably, the filtering and normalizing of the data collected by the hardware collection device based on the normalization formula specifically includes:

[0020] According to the hardware acquisition device, obtain the gas concentration and / or the concentration of exhaust gas generated after gas combustion, the temperature data within the detection range, the gas flow rate entering the air inlet of the stove, and the air flow data;

[0021] According to the first-order Kalman filter algorithm, the data collected by the hardware acquisition device are filtered and optimized;

[0022] Based on the Min-Max normalization formula, the influence of the magnitude of each data collected by the hardware acquisition device is eliminated to improve the data processing capability;

[0023] The Min-Max normalization formula is:

[0024]

[0025] In the formula, a mm is the normalized data of each data collected by the hardware acquisition device, a is the original data of each data collected by the hardware acquisition device, and a min is the minimum data among the raw data of various data collected by the hardware collection device, a max It is the maximum data among the raw data of various data collected by the hardware collection device.

[0026] Preferably, the establishment of the air flow influence coefficient for determining the degree to which the concentration of the gas and / or the concentration of the exhaust gas generated after the combustion of the gas is affected by the range hood or the air flow specifically includes:

[0027] Based on big data, obtain personalized data values ​​of the ideal gas concentration of the kitchen space under different space conditions and / or the concentration of exhaust gas generated after gas combustion;

[0028] According to the gas collection device, real-time detection data of gas concentration and / or concentration of waste gas generated after gas combustion is obtained;

[0029] According to the air flow collection device, the air flow data within the fugitive gas detection range is obtained in real time;

[0030] Establish an air flow impact sample list based on personalized data values ​​of gas concentration and / or the concentration of exhaust gas generated after gas combustion, gas concentration and / or the concentration of exhaust gas generated after gas combustion, and air flow data;

[0031] According to the air flow impact sample list, establish the air flow impact coefficient to determine the impact of air flow on the gas concentration and / or the concentration of exhaust gas generated after gas combustion;

[0032] Based on big data, the maximum threshold of the air flow influence coefficient is set. When the air flow influence coefficient exceeds the threshold, it means that the air flow seriously affects the gas concentration and / or the concentration detection of the exhaust gas generated after the gas combustion, and the detected data is inaccurate, and a reminder is required. When the air flow influence coefficient does not exceed the threshold, it means that the impact of the air flow is within the acceptable range and the detected data is accurate.

[0033] The air flow influence coefficient expression is:

[0034]

[0035] In the formula, θ i is the air flow influence coefficient affecting the gas concentration and / or the concentration of exhaust gas generated after gas combustion, m is the number of samples in the air flow influence sample list, v i is the air flow velocity value collected for the ith time in the air flow impact sample list, γ i is the ith detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, γ0 is the initial detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, The target data for the concentration of gas and / or the concentration of exhaust gas produced after gas combustion in the sample list of air flow effects, where: is the Euclidean norm.

[0036] Preferably, the establishment of a gas concentration and / or a concentration equilibrium coefficient of exhaust gas generated after gas combustion based on the air flow influence coefficient for determining whether a "gas combustion" event occurs specifically includes:

[0037] Based on big data or actual test results, set boundary thresholds for gas concentration and / or the concentration of exhaust gas generated after gas combustion;

[0038] According to the detection data of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, determine the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data;

[0039] According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion;

[0040] Determine whether a "gas combustion" event occurs based on the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion;

[0041] Determine whether the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is greater than the boundary threshold of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. If so, it indicates a preliminary judgment that a "gas combustion" event has occurred. If not, it indicates that no "gas combustion" event has occurred.

[0042] The gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is expressed as:

[0043]

[0044] Wherein, β is the equilibrium coefficient of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, δ is the equilibrium constant term, and n is the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data.

[0045] Preferably, further judging the occurrence of the "gas combustion" event based on the detection data of the temperature sensor and / or the flow sensor specifically includes:

[0046] According to the detection data of the temperature sensor, further judgment on the occurrence of the "gas combustion" event includes:

[0047] According to the temperature sensor, multiple sets of temperature change data when the gas is burning normally and when the gas is extinguished are obtained;

[0048] Based on big data, the temperature drop rate convergence degree threshold when the gas is burning normally and the temperature drop rate convergence degree threshold when the gas is extinguished are set respectively;

[0049] According to the second-order difference formula, a model of the convergence degree of the temperature drop rate is established, and according to the convergence degree of the temperature drop rate, it is further judged whether a "gas combustion" event occurs;

[0050] When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is burning normally, it indicates that a "gas burning" event has occurred. When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is extinguished, it indicates that no "gas burning" event has occurred, and it is determined that there is a gas leak, and an early warning is issued;

[0051] The temperature drop rate convergence degree algorithm is:

[0052]

[0053] Where τ is the convergence degree of temperature drop rate, ε is the error balance term, s is the number of temperature data collected by the temperature sensor, and T i is the ith temperature data collected by the temperature sensor, μ is the mean value of the temperature data collected by the temperature sensor, T i-1 The i-1th temperature data collected by the temperature sensor;

[0054] According to the detection data of the flow sensor, further judgment on the occurrence of the "gas combustion" event includes:

[0055] The gas flow rate entering the air inlet of the stove is collected according to the gas flow rate sensor arranged at the air inlet of the stove;

[0056] If the gas flow entering the air inlet of the stove is outside the normal combustion flow range, it means that no "gas combustion" event has occurred, it is determined that there is a gas leak, and an early warning is issued. If the gas flow entering the air inlet of the stove is within the normal combustion flow range, it means that a "gas combustion" event has definitely occurred.

[0057] Preferably, the stove fire identification model is established according to the air flow influence coefficient, the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion and the convergence degree of the temperature drop rate, and the stove fire risk alarm level is set, which specifically includes:

[0058] A stove fire identification model is established based on the air flow influence coefficient, gas concentration and / or the concentration equilibrium coefficient of exhaust gas generated after gas combustion and the degree of convergence of the temperature drop rate;

[0059] Based on big data, the stove fire risk alarm level is set according to the feedback of the stove fire identification model;

[0060] According to the detection results of the human body detection sensor and the stove fire recognition model, it is determined whether a "fire away from the person" incident has occurred.

[0061] Preferably, the setting of the stove fire risk warning prompt and timely issuing of the risk warning according to the stove fire risk warning level specifically includes:

[0062] If a "gas combustion" event is confirmed and no one is in front of the stove, a "fire start, leave no one" warning will be issued;

[0063] If a "gas burning" event is confirmed and someone is detected in front of the stove, no warning will be issued;

[0064] If it is determined that no "gas combustion" event has occurred, no warning will be issued.

[0065] Furthermore, the present invention proposes a stove fire high-precision identification system combined with escaped gas detection, which is used to implement the stove fire high-precision identification method combined with escaped gas detection as described above, including:

[0066] A data acquisition module, which is used to set a hardware acquisition device to collect and detect the concentration of gas and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone;

[0067] A data optimization module, wherein the data optimization module is used to filter and normalize various data collected by the hardware collection device based on a normalization formula;

[0068] A stove fire identification module, which is used to establish an air flow influence coefficient to determine the degree to which the gas concentration and / or the concentration of waste gas generated after gas combustion is affected by the range hood or air flow; based on the air flow influence coefficient, establish a gas concentration and / or a concentration balance coefficient of waste gas generated after gas combustion to determine whether a "gas combustion" event has occurred; based on the detection data of the temperature sensor and / or the flow sensor, further determine the occurrence of the "gas combustion" event; based on the air flow influence coefficient, the gas concentration and / or the concentration balance coefficient of waste gas generated after gas combustion and the degree of convergence of the temperature drop rate, establish a stove fire identification model, and set the stove fire risk alarm level;

[0069] The risk alarm module is used to set a stove fire risk alarm prompt and make a risk alarm in time according to the stove fire risk alarm level.

[0070] Preferably, the stove fire identification module specifically includes:

[0071] An air flow influencing unit, the air flow influencing unit is used to establish an air flow influencing coefficient, which is used to determine the degree to which the concentration of the gas and / or the concentration of the exhaust gas generated after the combustion of the gas is affected by the range hood or the air flow;

[0072] A "gas combustion" initial judgment unit, which is used to establish a gas concentration and / or a concentration equilibrium coefficient of exhaust gas generated after gas combustion according to an air flow influence coefficient, so as to judge whether a "gas combustion" event occurs;

[0073] A "gas combustion" determination unit, the "gas combustion" determination unit is used to further determine the occurrence of a "gas combustion" event based on detection data of a temperature sensor and / or a flow sensor;

[0074] A stove fire identification model unit is used to establish a stove fire identification model and set the stove fire risk alarm level according to the air flow influence coefficient, gas concentration and / or concentration equilibrium coefficient of exhaust gas generated after gas combustion and the degree of convergence of temperature drop rate.

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] By analyzing the degree to which the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is affected by the range hood or air flow, an air flow influence coefficient is established to determine the degree of influence of air flow on the detection of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. Secondly, through the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion, according to the value of the concentration balance coefficient of the gas concentration and / or the exhaust gas generated after the gas combustion, an initial judgment is made on whether a "gas combustion" event has occurred. Furthermore, by setting the degree of convergence of the temperature drop rate, by determining the degree of convergence of the temperature drop rate, a final determination is made on whether a "gas combustion" event has occurred. Finally, according to the air flow influence coefficient, the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion and the degree of convergence of the temperature drop rate, a stove fire recognition model is established. Combined with the detection results of the human body detection sensor, it is determined whether a "fire is in motion and no one is around" event has occurred, thereby achieving the purpose of effectively improving the ability to recognize stove fires and reducing the risk of "fire is in motion and no one is around" on stoves. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 A flow chart of a method for high-precision identification of stove fire combined with escaped gas detection according to the present invention;

[0078] Figure 2 A flow chart for establishing an air flow influence coefficient for judging the degree to which the concentration of gas and / or the concentration of exhaust gas generated after gas combustion is affected by the range hood or air flow in the present invention;

[0079] Figure 3 The flow chart of the present invention is to establish a gas concentration and / or a concentration balance coefficient of exhaust gas generated after gas combustion based on the air flow influence coefficient, which is used to determine whether a "gas combustion" event occurs;

[0080] Figure 4 This is a flow chart of further judging the occurrence of a "gas combustion" event based on detection data of a temperature sensor and / or a flow sensor according to the present invention. DETAILED DESCRIPTION

[0081] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0082] Reference Figure 1 As shown, a high-precision method for identifying stove fire combined with escaped gas detection includes:

[0083] Set up a hardware collection device to collect and detect the gas concentration and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone present;

[0084] Based on the normalization formula, the data collected by the hardware acquisition device are filtered and normalized;

[0085] Establish an air flow influence coefficient to determine the degree to which the concentration of gas and / or the concentration of exhaust gas generated after gas combustion is affected by the range hood or air flow;

[0086] According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion to determine whether a "gas combustion" event occurs;

[0087] Further judge the occurrence of the "gas combustion" event based on the detection data of the temperature sensor and / or flow sensor;

[0088] According to the air flow influence coefficient, gas concentration and / or the concentration equilibrium coefficient of exhaust gas generated after gas combustion and the convergence degree of temperature drop rate, a stove fire identification model is established, and the stove fire risk alarm level is set;

[0089] Set up a stove fire risk alarm and issue a risk alarm in a timely manner according to the stove fire risk alarm level.

[0090] It can be explained that during the combustion process of the gas stove, due to the amount of oxygen entering the stove damper and the maintenance level of the stove, it is basically impossible to achieve 100% combustion. There is always a certain proportion of gas that will escape and diffuse into the surrounding space environment. This solution analyzes the degree to which the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is affected by the range hood or air flow, and establishes an air flow influence coefficient to determine the degree of influence of air flow on the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. Secondly, through the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion, according to the gas concentration and / or the exhaust gas generated after the gas combustion The concentration balance coefficient value of the exhaust gas is used to make an initial judgment on whether a "gas combustion" event has occurred. Secondly, by setting the convergence degree of the temperature drop rate, and judging the convergence degree of the temperature drop rate, a final determination is made on whether a "gas combustion" event has occurred. Finally, a stove fire recognition model is established based on the air flow influence coefficient, gas concentration and / or the concentration balance coefficient of the exhaust gas generated after gas combustion and the convergence degree of the temperature drop rate. Combined with the detection results of the human body detection sensor, it is possible to judge whether a "fire is active and no one is around" event has occurred, thereby achieving the purpose of effectively improving the stove fire recognition ability and reducing the risk of "fire is active and no one is around" on the stove.

[0091] Reference Figure 2 As shown, the establishment of the air flow influence coefficient for determining the degree to which the concentration of the gas and / or the concentration of the exhaust gas generated after the combustion of the gas is affected by the range hood or the air flow specifically includes:

[0092] Based on big data, obtain personalized data values ​​of the ideal gas concentration of the kitchen space under different space conditions and / or the concentration of exhaust gas generated after gas combustion;

[0093] According to the gas collection device, real-time detection data of gas concentration and / or concentration of waste gas generated after gas combustion is obtained;

[0094] According to the air flow collection device, the air flow data within the fugitive gas detection range is obtained in real time;

[0095] A sample list of air flow impacts is established based on personalized data values ​​of gas concentration and / or the concentration of exhaust gas generated after gas combustion, gas concentration and / or the concentration of exhaust gas generated after gas combustion, and air flow data;

[0096] According to the air flow impact sample list, establish the air flow impact coefficient to determine the impact of air flow on the gas concentration and / or the concentration of exhaust gas generated after gas combustion;

[0097] Based on big data, the maximum threshold of the air flow influence coefficient is set. When the air flow influence coefficient exceeds the threshold, it means that the air flow seriously affects the gas concentration and / or the concentration detection of the exhaust gas generated after the gas combustion, and the detected data is inaccurate, and a reminder is required. When the air flow influence coefficient does not exceed the threshold, it means that the impact of the air flow is within the acceptable range and the detected data is accurate.

[0098] The air flow influence coefficient expression is:

[0099]

[0100] In the formula, θ i is the air flow influence coefficient affecting the gas concentration and / or the concentration of exhaust gas generated after gas combustion, m is the number of samples in the air flow influence sample list, v i is the air flow velocity value collected for the ith time in the air flow impact sample list, γ i is the ith detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, γ0 is the initial detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, The target data for the concentration of gas and / or the concentration of exhaust gas produced after gas combustion in the sample list of air flow effects, where: is the Euclidean norm.

[0101] It can be explained that the fugitive concentration of gas or the fugitive concentration of waste gas produced after gas combustion will be affected by the range hood or air flow, resulting in smaller data collected by the gas collection device, causing inaccurate results of stove fire identification, thus affecting the final risk judgment of "stove fire away from people". This scheme establishes an air flow influence coefficient by analyzing that the air flow velocity is inversely proportional to the fugitive concentration of gas or the fugitive concentration of waste gas produced after gas combustion, and judges the influence of air flow on the gas concentration and / or the concentration detection of waste gas produced after gas combustion based on the air flow influence coefficient, thereby correcting the gas concentration and / or the concentration data of waste gas produced after gas combustion collected by the gas collection device based on the air flow influence coefficient, thereby improving the accuracy and reliability of the data.

[0102] Reference Figure 3 As shown, the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion are established according to the air flow influence coefficient to determine whether a "gas combustion" event occurs, specifically including:

[0103] Based on big data or actual test results, set boundary thresholds for gas concentration and / or the concentration of exhaust gas generated after gas combustion;

[0104] According to the detection data of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, determine the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data;

[0105] According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion;

[0106] Determine whether a "gas combustion" event occurs based on the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion;

[0107] Determine whether the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is greater than the boundary threshold of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. If so, it indicates a preliminary judgment that a "gas combustion" event has occurred. If not, it indicates that no "gas combustion" event has occurred.

[0108] The gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is expressed as:

[0109]

[0110] Wherein, β is the equilibrium coefficient of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, δ is the equilibrium constant term, and n is the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data.

[0111] It can be explained that during the combustion process of the gas stove, due to the amount of oxygen entering the stove damper and the maintenance level of the stove, it is basically impossible to achieve 100% combustion. There is always a certain proportion of gas that will escape and diffuse into the surrounding space environment. As the gas burns, the gas concentration within the detection range and / or the concentration of the exhaust gas generated after the gas combustion will appear to continuously exceed the target data of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. This scheme establishes a concentration balance coefficient of the gas concentration and / or the exhaust gas generated after the gas combustion, and determines whether a "gas combustion" event has occurred based on the gas concentration and / or the exhaust gas concentration detection value after a period of time, thereby achieving an initial judgment on whether a "gas combustion" event has occurred.

[0112] Reference Figure 4 As shown, the further determination of the occurrence of the "gas combustion" event based on the detection data of the temperature sensor and / or the flow sensor specifically includes:

[0113] According to the detection data of the temperature sensor, further judgment on the occurrence of the "gas combustion" event includes:

[0114] According to the temperature sensor, multiple sets of temperature change data when the gas is burning normally and when the gas is extinguished are obtained;

[0115] Based on big data, the temperature drop rate convergence degree threshold when the gas is burning normally and the temperature drop rate convergence degree threshold when the gas is extinguished are set respectively;

[0116] According to the second-order difference formula, a model of the convergence degree of the temperature drop rate is established, and according to the convergence degree of the temperature drop rate, it is further judged whether a "gas combustion" event occurs;

[0117] When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is burning normally, it indicates that a "gas burning" event has occurred. When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is extinguished, it indicates that no "gas burning" event has occurred, and it is determined that there is a gas leak, and an early warning is issued;

[0118] The temperature drop rate convergence degree algorithm is:

[0119]

[0120] Where τ is the convergence degree of temperature drop rate, ε is the error balance term, s is the number of temperature data collected by the temperature sensor, and T i is the ith temperature data collected by the temperature sensor, μ is the mean value of the temperature data collected by the temperature sensor, T i-1 The i-1th temperature data collected by the temperature sensor;

[0121] According to the detection data of the flow sensor, further judgment on the occurrence of the "gas combustion" event includes:

[0122] The gas flow rate entering the air inlet of the stove is collected according to the gas flow rate sensor arranged at the air inlet of the stove;

[0123] If the gas flow entering the air inlet of the stove is outside the normal combustion flow range, it means that no "gas combustion" event has occurred, it is determined that there is a gas leak, and an early warning is issued. If the gas flow entering the air inlet of the stove is within the normal combustion flow range, it means that a "gas combustion" event has definitely occurred.

[0124] It can be explained that, according to the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion, it can be judged that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion has a certain degree of leakage phenomenon, and the risk of "gas combustion" event or gas leakage can be preliminarily judged based on this phenomenon, but whether there is a "gas combustion" event cannot be accurately confirmed. Therefore, this scheme determines the risk of "gas combustion" event or gas leakage by judging the convergence degree of the temperature drop rate within the detection range, so as to determine whether there is a "gas combustion" event. When the "gas combustion" event occurs, due to the nature of the combustion, the convergence degree of the temperature drop rate is relatively slow. For the judgment of gas leakage, due to the cessation of "gas combustion", the temperature within the detection range will drop significantly within a period of time, and the convergence degree of the temperature drop rate is relatively fast, so that it is well determined whether there is a "gas combustion" event through the convergence degree of the temperature drop rate. Secondly, by monitoring the gas flow entering the air inlet of the stove, whether it is in the normal combustion flow range, further auxiliary verification of the gas leakage or "gas combustion" event is performed, so as to ensure the accuracy of the judgment of the "gas combustion" event.

[0125] Further, based on the same inventive concept as the above-mentioned stove fire high-precision identification method combined with escaped gas detection, the present solution also proposes a stove fire high-precision identification system combined with escaped gas detection, including:

[0126] A data acquisition module, which is used to set a hardware acquisition device to collect and detect the concentration of gas and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone;

[0127] A data optimization module, wherein the data optimization module is used to filter and normalize various data collected by the hardware collection device based on a normalization formula;

[0128] A stove fire identification module, which is used to establish an air flow influence coefficient to determine the degree to which the gas concentration and / or the concentration of waste gas generated after gas combustion is affected by the range hood or air flow; based on the air flow influence coefficient, establish a gas concentration and / or a concentration balance coefficient of waste gas generated after gas combustion to determine whether a "gas combustion" event has occurred; based on the detection data of the temperature sensor and / or the flow sensor, further determine the occurrence of the "gas combustion" event; based on the air flow influence coefficient, the gas concentration and / or the concentration balance coefficient of waste gas generated after gas combustion and the degree of convergence of the temperature drop rate, establish a stove fire identification model, and set the stove fire risk alarm level;

[0129] A risk alarm module, which is used to set a risk alarm for stove fire, and to issue a risk alarm in a timely manner according to the stove fire risk alarm level;

[0130] The stove fire identification module specifically includes:

[0131] An air flow influencing unit, the air flow influencing unit is used to establish an air flow influencing coefficient, which is used to determine the degree to which the concentration of the gas and / or the concentration of the exhaust gas generated after the combustion of the gas is affected by the range hood or the air flow;

[0132] A "gas combustion" initial judgment unit, which is used to establish a gas concentration and / or a concentration equilibrium coefficient of exhaust gas generated after gas combustion according to an air flow influence coefficient, so as to judge whether a "gas combustion" event occurs;

[0133] A "gas combustion" determination unit, the "gas combustion" determination unit is used to further determine the occurrence of a "gas combustion" event based on detection data of a temperature sensor and / or a flow sensor;

[0134] A stove fire identification model unit is used to establish a stove fire identification model and set the stove fire risk alarm level according to the air flow influence coefficient, gas concentration and / or concentration equilibrium coefficient of exhaust gas generated after gas combustion and the degree of convergence of temperature drop rate.

[0135] In summary, the advantages of the present invention are: this solution effectively improves the ability to identify fire on the stove and reduces the risk of "the stove is on and the person is not around".

[0136] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A high-precision method for identifying stove fire combined with escaped gas detection, characterized in that: include: Set up a hardware collection device to collect and detect the gas concentration and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone present; Based on the normalization formula, the data collected by the hardware acquisition device are filtered and normalized; Establish an air flow influence coefficient to determine the degree to which the concentration of gas and / or the concentration of exhaust gas generated after gas combustion is affected by the range hood or air flow; According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion to determine whether a "gas combustion" event occurs; Further judge the occurrence of "gas combustion" event based on the detection data of temperature sensor and / or flow sensor; According to the air flow influence coefficient, gas concentration and / or the concentration equilibrium coefficient of exhaust gas generated after gas combustion and the convergence degree of temperature drop rate, a stove fire identification model is established, and the stove fire risk alarm level is set; Set up a stove fire risk alarm and issue a risk alarm in a timely manner according to the stove fire risk alarm level.

2. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 1, characterized in that: The hardware gas collection device is provided to collect and detect the concentration of gas and / or the concentration of waste gas generated after gas combustion and whether there is someone there, specifically including: According to the type of gas and the type of waste gas generated after combustion, a gas collection device is provided to collect and detect the concentration of gas and / or the concentration of waste gas generated after gas combustion in real time; A human body detection sensor is set up to determine whether there is a human body within the detection range; Temperature and flow sensors are provided to collect temperature data within the detection range and gas flow entering the air inlet of the stove in real time; An air flow collection device is provided to collect air flow data within the detection range in real time.

3. A high-precision identification method for stove fire combined with escaped gas detection according to claim 2, characterized in that: The filtering and normalization processing of the various data collected by the hardware collection device based on the normalization formula specifically includes: According to the hardware acquisition device, obtain the gas concentration and / or the concentration of exhaust gas generated after gas combustion, the temperature data within the detection range, the gas flow rate entering the air inlet of the stove, and the air flow data; According to the first-order Kalman filter algorithm, the data collected by the hardware acquisition device are filtered and optimized; Based on the Min-Max normalization formula, the influence of the magnitude of each data collected by the hardware acquisition device is eliminated to improve the data processing capability; The Min-Max normalization formula is: In the formula, a mm is the normalized data of each data collected by the hardware acquisition device, a is the original data of each data collected by the hardware acquisition device, and a min is the minimum data among the raw data of various data collected by the hardware collection device, a max It is the maximum data among the raw data of various data collected by the hardware collection device.

4. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 3, characterized in that: The establishment of the air flow influence coefficient for determining the degree to which the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is affected by the range hood or the air flow specifically includes: Based on big data, obtain personalized data values ​​of the ideal gas concentration of the kitchen space under different space conditions and / or the concentration of exhaust gas generated after gas combustion; According to the gas collection device, real-time detection data of gas concentration and / or concentration of waste gas generated after gas combustion is obtained; According to the air flow collection device, the air flow data within the fugitive gas detection range is obtained in real time; Establish an air flow impact sample list based on personalized data values ​​of gas concentration and / or the concentration of exhaust gas generated after gas combustion, gas concentration and / or the concentration of exhaust gas generated after gas combustion, and air flow data; According to the air flow impact sample list, establish the air flow impact coefficient to determine the impact of air flow on the gas concentration and / or the concentration of exhaust gas generated after gas combustion; Based on big data, the maximum threshold of the air flow influence coefficient is set. When the air flow influence coefficient exceeds the threshold, it means that the air flow seriously affects the gas concentration and / or the concentration detection of the exhaust gas generated after the gas combustion, and the detected data is inaccurate, and a reminder is required. When the air flow influence coefficient does not exceed the threshold, it means that the impact of the air flow is within the acceptable range and the detected data is accurate. The air flow influence coefficient expression is: In the formula, θ i is the air flow influence coefficient affecting the gas concentration and / or the concentration of exhaust gas generated after gas combustion, m is the number of samples in the air flow influence sample list, v i is the air flow velocity value collected for the ith time in the air flow impact sample list, γ i is the ith detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, γ0 is the initial detection data of the gas concentration and / or the exhaust gas concentration after gas combustion in the air flow impact sample list, The target data for the concentration of gas and / or the concentration of exhaust gas produced after gas combustion in the sample list of air flow effects, where: is the Euclidean norm.

5. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 4, characterized in that: The establishment of a gas concentration and / or a concentration equilibrium coefficient of exhaust gas generated after gas combustion based on the air flow influence coefficient for determining whether a "gas combustion" event occurs specifically includes: Based on big data or actual test results, set boundary thresholds for gas concentration and / or the concentration of exhaust gas generated after gas combustion; According to the detection data of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, determine the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data; According to the air flow influence coefficient, establish the gas concentration and / or the concentration balance coefficient of the exhaust gas generated after the gas combustion; Determine whether a "gas combustion" event has occurred based on the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion; Determine whether the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is greater than the boundary threshold of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion. If so, it indicates a preliminary judgment that a "gas combustion" event has occurred. If not, it indicates that no "gas combustion" event has occurred. The gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion is expressed as: Wherein, β is the equilibrium coefficient of the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion, δ is the equilibrium constant term, and n is the number of times that the gas concentration and / or the concentration of the exhaust gas generated after the gas combustion is continuously greater than the target data.

6. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 5, characterized in that: Further judging the occurrence of the "gas combustion" event based on the detection data of the temperature sensor and / or the flow sensor specifically includes: According to the detection data of the temperature sensor, further judgment on the occurrence of the "gas combustion" event includes: According to the temperature sensor, multiple sets of temperature change data when the gas is burning normally and when the gas is extinguished are obtained; Based on big data, the temperature drop rate convergence degree threshold when the gas is burning normally and the temperature drop rate convergence degree threshold when the gas is extinguished are set respectively; According to the second-order difference formula, a model of the convergence degree of the temperature drop rate is established, and according to the convergence degree of the temperature drop rate, it is further judged whether a "gas combustion" event occurs; When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is burning normally, it indicates that a "gas burning" event has occurred. When the convergence degree of the temperature drop rate approaches the convergence degree threshold of the temperature drop rate when the gas is extinguished, it indicates that no "gas burning" event has occurred, and it is determined that there is a gas leak, and an early warning is issued; The temperature drop rate convergence degree algorithm is: Where τ is the convergence degree of temperature drop rate, ε is the error balance term, s is the number of temperature data collected by the temperature sensor, and T i is the ith temperature data collected by the temperature sensor, μ is the mean value of the temperature data collected by the temperature sensor, T i-1 The i-1th temperature data collected by the temperature sensor; According to the detection data of the flow sensor, further judgment on the occurrence of the "gas combustion" event includes: The gas flow rate entering the air inlet of the stove is collected according to the gas flow rate sensor arranged at the air inlet of the stove; If the gas flow entering the air inlet of the stove is outside the normal combustion flow range, it means that no "gas combustion" event has occurred, it is determined that there is a gas leak, and an early warning is issued. If the gas flow entering the air inlet of the stove is within the normal combustion flow range, it means that a "gas combustion" event has definitely occurred.

7. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 6, characterized in that: The stove fire identification model is established according to the air flow influence coefficient, the gas concentration and / or the concentration equilibrium coefficient of the exhaust gas generated after the gas combustion and the convergence degree of the temperature drop rate, and the stove fire risk alarm level is set, which specifically includes: A stove fire identification model is established based on the air flow influence coefficient, gas concentration and / or the concentration equilibrium coefficient of exhaust gas generated after gas combustion and the degree of convergence of the temperature drop rate; Based on big data, the stove fire risk alarm level is set according to the feedback of the stove fire identification model; According to the detection results of the human body detection sensor and the stove fire recognition model, it is determined whether there is a "fire away from the person" incident.

8. A high-precision method for identifying stove fire combined with escaped gas detection according to claim 7, characterized in that: The setting of the stove fire risk warning prompt and timely issuing of risk warnings according to the stove fire risk warning level specifically include: If a "gas burning" event is confirmed and no one is in front of the stove, a "fire start, leave no one" warning will be issued; If a "gas burning" event is confirmed and someone is detected in front of the stove, no warning will be issued; If it is determined that no "gas combustion" event has occurred, no warning will be issued.

9. A high-precision identification system for stove fire combined with escaped gas detection, used to implement the high-precision identification method for stove fire combined with escaped gas detection as claimed in any one of claims 1 to 8, characterized in that: include: A data acquisition module, which is used to set a hardware acquisition device to collect and detect the concentration of gas and / or the concentration of exhaust gas generated after gas combustion and whether there is anyone; A data optimization module, wherein the data optimization module is used to filter and normalize various data collected by the hardware collection device based on a normalization formula; A stove fire identification module, which is used to establish an air flow influence coefficient to determine the degree to which the gas concentration and / or the concentration of waste gas generated after gas combustion is affected by the range hood or air flow; based on the air flow influence coefficient, establish a gas concentration and / or a concentration balance coefficient of waste gas generated after gas combustion to determine whether a "gas combustion" event has occurred; based on the detection data of the temperature sensor and / or the flow sensor, further determine the occurrence of the "gas combustion" event; based on the air flow influence coefficient, the gas concentration and / or the concentration balance coefficient of waste gas generated after gas combustion and the degree of convergence of the temperature drop rate, establish a stove fire identification model, and set the stove fire risk alarm level; The risk alarm module is used to set a stove fire risk alarm prompt and make a risk alarm in time according to the stove fire risk alarm level.

10. A stove fire high-precision identification system combined with escaped gas detection according to claim 9, characterized in that: The stove fire identification module specifically includes: An air flow influencing unit, the air flow influencing unit is used to establish an air flow influencing coefficient, which is used to determine the degree to which the concentration of the gas and / or the concentration of the exhaust gas generated after the combustion of the gas is affected by the range hood or the air flow; A "gas combustion" initial judgment unit, which is used to establish a gas concentration and / or a concentration equilibrium coefficient of exhaust gas generated after gas combustion according to an air flow influence coefficient, and is used to judge whether a "gas combustion" event occurs; A "gas combustion" determination unit, the "gas combustion" determination unit is used to further determine the occurrence of a "gas combustion" event based on detection data of a temperature sensor and / or a flow sensor; A stove fire identification model unit is used to establish a stove fire identification model and set the stove fire risk alarm level according to the air flow influence coefficient, gas concentration and / or concentration equilibrium coefficient of exhaust gas generated after gas combustion and the degree of convergence of temperature drop rate.

Images