Gas monitoring method and system with multi-sensor fusion

Abstract

The application provides a multi-sensor fusion gas monitoring method and system, and belongs to the technical field of intelligent monitoring terminals; the method comprises the following steps: configuring a multi-sensor monitoring terminal, and acquiring sensor detection data of a gas equipment use period and a gas equipment non-use period; after the gateway receives the sensor detection data of the gas equipment use period or the non-use period, the gateway performs fusion detection through built-in fusion rules, outputs real-time first leakage risk scores or second slope risk scores, then according to the real-time leakage risk scores, the linkage rules of intelligent home equipment in the gas equipment use period and the non-use period are configured, the gateway drives indoor intelligent home equipment to act or adjusts the amplitude of the action, and corresponding first alarm signals or second alarm signals are sent; in the non-use period of the gas equipment, the gateway also performs self-checking evaluation on the performance of each sensor of the multi-sensor monitoring terminal, and according to the performance degradation, a third alarm signal is sent.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring terminal technology, and in particular to a gas monitoring method and system that integrates multiple sensors. Background Technology

[0002] To better ensure indoor safety, gas monitoring terminals are usually installed in gas-fired kitchens to obtain real-time information on the concentration of natural gas in the environment and to issue audible and visual alarms when the concentration exceeds the standard, thus preventing the continuous rise in indoor gas concentration from causing safety hazards. However, the functions of such terminals are relatively limited.

[0003] To address the issue of limited functionality in existing gas monitoring terminals, some marketed gas monitoring terminals offer features such as actively driving solenoid valves to shut off gas supply and integrating with smart home systems. However, they still suffer from the following shortcomings: 1. They cannot effectively identify combustion characteristics during gas usage, failing to distinguish between gas consumption caused by leaks and gas consumption during cooking or heating, potentially leading to false alarms; 2. As the core of the monitoring terminal, the sensor's performance gradually degrades with use, and existing terminals lack the function of periodically calibrating sensor performance; 3. They do not differentiate between gas usage and non-gas usage scenarios, and the interaction between smart devices and the gas monitoring terminal is not clearly defined, impacting the user experience.

[0004] Therefore, it is essential to provide a gas monitoring method and system that integrates multiple sensors, reliably identifies gas usage periods and non-usage periods, combines the status of multiple sensors to jointly determine the gas leak situation, and periodically calibrates the sensor performance for different purposes to reduce the risk of false alarms and improve user satisfaction. Summary of the Invention

[0005] In view of this, the present invention proposes a gas monitoring method and system that integrates multiple sensors to comprehensively judge the risk and severity of gas leaks based on the usage status of gas equipment, the surrounding environment, and regularly verifies and judges the performance of sensors to reduce the possibility of false alarms.

[0006] On the one hand, the present invention provides a gas monitoring method based on multi-sensor fusion, comprising the following steps: A multi-sensor monitoring terminal is configured to acquire sensor detection data during the usage and non-use periods of gas appliances; the multi-sensor monitoring terminal also communicates with a gateway, which in turn communicates with indoor smart home devices and a cloud server. After receiving the sensor detection data of the gas equipment usage period sent by the multi-sensor monitoring terminal, the gateway performs fusion detection through the built-in fusion rules, outputs the first leakage risk score in real time, and then configures the linkage rules of smart home devices during the gas equipment usage period based on the real-time leakage risk score. The gateway drives the indoor smart home devices to act or adjust the magnitude of the action, and issues the corresponding first alarm signal. During the non-use period of the gas appliance, the gateway receives sensor detection data of the gas appliance during the non-use period sent by the multi-sensor monitoring terminal, outputs a second leakage risk score, configures the linkage rules of smart home devices during the non-use period of the gas appliance according to the second leakage risk score, and issues a corresponding second alarm signal. During non-use periods of gas equipment, the gateway also performs a self-test evaluation of the performance of each sensor in the multi-sensor monitoring terminal, and issues a third alarm signal based on the performance degradation. The cloud server also communicates with the customer's mobile terminal, and real-time risk scores, first alarm signals, second alarm signals, and third alarm signals are also sent to the mobile terminal through the cloud server.

[0007] Based on the above technical solutions, preferably, the multi-sensor monitoring terminal includes a methane sensor, a carbon monoxide sensor, a smoke concentration sensor, a temperature and humidity sensor, a water immersion sensor, and a door and window status sensor. The measured values ​​of the methane sensor, carbon monoxide sensor, smoke concentration sensor, and temperature and humidity sensor are normalized to the range of [0, 1] to obtain the normalized methane sensor measured value. Normalized carbon monoxide sensor measurements Normalized smoke concentration sensor measurement values Normalized temperature change and normalized humidity change Normalized water immersion sensor measurement values Measurement values ​​from door and window status sensors It can be 0 or 1.

[0008] Preferably, after receiving the sensor detection data of the gas appliance usage period sent by the multi-sensor monitoring terminal, the gateway performs fusion detection through built-in fusion rules, outputs a real-time first leakage risk score, and then configures the linkage rules of smart home devices during the gas appliance usage period based on the real-time leakage risk score. The gateway drives the indoor smart home devices to act or adjust the magnitude of the action, and issues a corresponding first alarm signal, including the following: based on the normalized carbon monoxide sensor measurement value. Normalized temperature change and normalized humidity change Calculate the flammability indexCL Combustion Index CL Used to evaluate the completeness of combustion during the use of gas appliances; 1- CL As the non-combustion index and compared with the normalized methane sensor measurement. Normalized smoke concentration sensor measurement values and normalized water immersion sensor measurements The weighted summation yields the first leakage risk score. LRS Further, the leakage acceleration of methane was determined, and the normalized methane sensor measurements were combined. Combustion Index CL First Leakage Risk Score LRS and the acceleration of methane leakage LA Configure the linkage rules for smart home devices during the usage periods of gas appliances.

[0009] More preferably, the linkage rules for smart home devices during the designated gas appliance usage periods include the following: 1) Normalized methane sensor measurement value Not exceeding 0.05, flammability index CL A score greater than 0.4 indicates a first-level leakage risk score. LRS If the methane leakage acceleration is less than 0.3 and the leakage acceleration is less than 0.01, the gateway will only record and save the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed to allow natural ventilation. 2) Normalized methane sensor measurement value In the range (0.05, 0.15), the flammability index CL Less than 0.3, first leakage risk score LRS If the methane leakage acceleration is greater than 0.3 and does not exceed 0.03, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates the first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm prompt; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 30% of the rated maximum speed. 3) Normalized methane sensor measurement value In the range (0.15, 0.25), the flammability index CL Less than 0.2, first leakage risk score LRSIf the methane leakage acceleration is greater than 0.4 and does not exceed 0.08, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates the first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm and closes the gas inlet valve of the gas equipment; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 60% of the rated maximum speed. 4) Normalized methane sensor measurement value >0.25, or the first leakage risk score LRS If the leakage acceleration of methane is greater than 0.6, or greater than 0.08 / min, the multi-sensor monitoring terminal will issue an audible and visual alarm and immediately shut off the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation system will be forced to ventilate at 100% of the rated maximum speed and generate the first alarm signal.

[0010] More preferably, during the non-use period of the gas appliance, the gateway receives sensor detection data of the gas appliance during the non-use period sent by the multi-sensor monitoring terminal, outputs a second leakage risk score, and configures the linkage rules of smart home devices during the non-use period of the gas appliance according to the second leakage risk score, and issues a corresponding second alarm signal, including the following: based on the normalized smoke alarm, the normalized door and window status sensor measurement values ​​of the current period, and the time factor. TF Calculating environmental risk benchmarks ERB Based on environmental risk benchmarks ERB Further calculate the environmental risk amplification factor A env Based on environmental risk amplification factor A env and normalized methane sensor measurements The exponential function and the normalized measurement value of the water immersion sensor The second leakage risk score is obtained by weighted summation of the exponential function. LSI Combined with the second leakage risk score LSI Compared with normalized methane sensor measurements Configure the linkage rules for smart home devices during non-use periods of gas appliances, and issue corresponding second alarm signals.

[0011] More preferably, the linkage rules for smart home devices configured during non-use periods of the gas appliance will issue a corresponding second alarm signal, including the following: 1) Normalized methane sensor measurement value Not exceeding 0.03, or a second leakage risk score. LSI When the value is less than 0.15, the gateway only records and saves the current normalized methane sensor measurement value. Second Leakage Risk Score LSI To maintain the current state of the doors and windows; 2) Normalized methane sensor measurement value In the range (0.03, 0.08), or the second leakage risk score LSI When the value is in the range (0.15, 0.35), a slight methane residue is detected in the monitored area, and the gateway generates a second warning message. 3) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the temperature is in the range (0.35, 0.60), the multi-sensor monitoring terminal issues an audible and visual alarm and generates a second warning message; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 50% of the rated maximum speed. 4) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the value is in the range (0.35, 0.60), the multi-sensor monitoring terminal will issue an audible and visual alarm, generate a second alarm signal, and immediately close the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation facility will be forced to ventilate at 100% of its rated maximum speed.

[0012] More preferably, the time factor TF The value at night is greater than the value during the day.

[0013] Preferably, during the non-use period of the gas equipment, the gateway also performs a self-test evaluation of the performance of each sensor of the multi-sensor monitoring terminal, and issues a third alarm signal based on the performance degradation. The sensor performance self-test is performed at fixed periodic intervals during the non-use period of the gas equipment. The sensor performance self-test process is as follows: by injecting a test signal, feature extraction is performed based on the sensor response curve to obtain the feature vector corresponding to the response curve, and the similarity between the feature vector and the feature reference vector is evaluated to obtain a similarity score; the degradation degree of each element in the feature vector is weighted and accumulated to obtain the comprehensive degradation degree of the sensor; based on the sequence value composed of the most recent similarity scores, a time series-based degradation prediction model is constructed, and combined with the comprehensive degradation degree of the sensor, it is assessed whether the sensor needs to be replaced or the sensor's measurement value needs to be compensated. If the sensor is assessed to need to be replaced, a third warning signal is issued.

[0014] More preferably, the step of injecting a test signal and extracting features from the sensor response curve to obtain the feature vector corresponding to the response curve is to let the feature vector be F={f1, f2, f3, f4, f5, f6, f7}, where f1 is the rising slope of the response curve, f2 is the falling slope of the response curve, f3 is the overshoot percentage, f4 is the steady-state fluctuation coefficient of the response curve, f5 is the curvature integral, f6 is the energy distribution ratio of high-frequency energy to low-frequency energy after the fast Fourier transform of the sensor response curve, and f7 is the signal entropy value.

[0015] On the other hand, the present invention provides a multi-sensor fusion gas monitoring system for implementing the above-mentioned method, comprising: a multi-sensor monitoring terminal, a gateway, several indoor smart home devices, a cloud server, and a mobile terminal, wherein... The multi-sensor monitoring terminal communicates with the gateway to acquire sensor detection data during the gas equipment's usage period and non-use period, and sends the sensor detection data to the gateway. The gateway receives sensor detection data from the multi-sensor monitoring terminal during the gas appliance usage period. It then performs fusion detection using built-in fusion rules, outputting a real-time first leak risk score. Based on this score, it configures linkage rules for smart home devices during the gas appliance usage period. The gateway then drives the indoor smart home devices to perform actions or adjust the magnitude of those actions, issuing a corresponding first alarm signal. The gateway also receives sensor detection data from the multi-sensor monitoring terminal during non-gas appliance usage periods, outputting a second leak risk score. Based on this score, it configures linkage rules for smart home devices during non-gas appliance usage periods and issues a corresponding second alarm signal. The gateway also communicates with both the indoor smart home devices and the cloud server. Furthermore, the gateway periodically performs self-evaluation of the performance of each sensor in the multi-sensor monitoring terminal, issuing a third alarm signal based on performance degradation. The cloud server communicates with the mobile terminal to send the received real-time risk score, first alarm signal, second alarm signal, and third alarm signal to the mobile terminal.

[0016] The present invention provides a gas monitoring method and system based on multi-sensor fusion, which has the following advantages compared with the prior art: 1. This application provides multi-dimensional data fusion decision-making. Specifically, it integrates data from multiple sensors, such as methane, carbon monoxide, smoke, temperature and humidity, water immersion, and door and window status, and combines this with the usage / non-use periods of gas equipment to construct a comprehensive risk assessment model. This avoids misjudgments caused by environmental interference or drift of a single sensor. The monitoring strategy is clearly distinguished between the usage period and the non-use period of gas equipment, and different risk assessment algorithms and linkage rules are adopted, which is more in line with the actual occurrence scenario of gas leakage risk.

[0017] 2. During the usage period of gas equipment, the combustion completeness of gas equipment is assessed through the combustion index and used as a key factor in risk scoring to better distinguish between normal combustion exhaust gas and abnormal gas leakage, reducing the possibility of false alarms. During the non-use period of gas equipment, an environmental risk benchmark value and an environmental risk amplification coefficient are introduced to eliminate the influence of combustion, making the risk assessment more in line with the actual environment and improving the reliability of the assessment.

[0018] 3. Leakage risk is quantified into specific scores, specific hierarchical linkage thresholds are set, and corresponding linkage measures for smart home devices are taken to achieve intelligent response.

[0019] 4. Regularly perform sensor performance verification. Sensor performance self-testing uses multi-dimensional feature vectors including rise / fall edge slope, overshoot, steady-state fluctuation, curvature integral, spectral energy ratio, and signal entropy value. Through similarity evaluation and decay analysis, the detection values ​​are compensated accordingly or the sensor is recommended to be replaced to improve the protection capability of gas active safety monitoring. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the steps of the gas monitoring method and system based on multi-sensor fusion of the present invention. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] The existing gas monitoring terminals and smart home systems have the following shortcomings in their collaborative operation: 1. They cannot effectively identify combustion characteristics during gas usage, and cannot distinguish whether the gas consumption is due to leakage or cooking / heating, which may lead to false alarms; 2. As the core of the monitoring terminal, the performance of the sensor gradually degrades with use, and the existing monitoring terminals lack the function of regularly calibrating the sensor performance and judging its lifespan; 3. They do not distinguish between situations where gas is used or not, and the linkage content between smart devices and gas monitoring terminals is not subdivided, which affects the user experience.

[0024] In view of this, such as Figure 1 As shown, in one aspect, the present invention provides a gas monitoring method based on multi-sensor fusion, comprising the following steps: S100: Equipped with a multi-sensor monitoring terminal to acquire sensor detection data during gas appliance usage periods and non-use periods; the multi-sensor monitoring terminal also communicates with a gateway, which in turn communicates with indoor smart home devices and a cloud server.

[0025] The multi-sensor monitoring terminal mentioned in step S100 includes a methane sensor, a carbon monoxide sensor, a smoke concentration sensor, a temperature and humidity sensor, a water immersion sensor, and a door and window status sensor. The measured values ​​of the methane sensor, carbon monoxide sensor, smoke concentration sensor, and temperature and humidity sensor are normalized to the range of [0, 1] to obtain the normalized methane sensor measured value. Normalized carbon monoxide sensor measurements Normalized smoke concentration sensor measurement values Normalized temperature change and normalized humidity change Normalized water immersion sensor measurement values Measurement values ​​from door and window status sensors It can be 0 or 1.

[0026] Because different sensors have different types and measurement ranges, normalization is necessary. For example, for a methane sensor, the measurement range is 1-100% LEL, where LEL is the lower limit concentration. The normalization formula is: =Measured value / 100, then normalized to the [0, 1] interval; similarly, the measurement range of the smoke concentration sensor is 0-100%, and the normalization formula is... =Concentration measurement value / 100, the normalized value range is [0, 1]; the measurement range of the iron oxide sensor is 1-1000ppm, and the normalization formula is: =min(carbon monoxide concentration / 500,1), the normalized value range is [0,1]; temperature change The measurement range is -10℃ to 60℃, and the normalization formula is: The normalized value range is [0, 1]; humidity change The measurement range is 0-100%RH, and the normalization formula is: The normalized value range is [0, 1]; the measured value of the water immersion sensor is normalized to... =0 or 1, meaning the normalization range is 0 or 1; the measured values ​​of the door and window status sensors are normalized to =0 or 1, meaning the normalization range is 0 or 1. It is evident that normalization eliminates the difference in actual numerical values ​​between different measurements, removing dimensions and facilitating subsequent qualitative analysis and calculations.

[0027] S200: After receiving the sensor detection data of the gas equipment usage period sent by the multi-sensor monitoring terminal, the gateway performs fusion detection through the built-in fusion rules, outputs the first leakage risk score in real time, and then configures the linkage rules of smart home devices during the gas equipment usage period based on the real-time leakage risk score. The gateway drives the indoor smart home devices to act or adjust the magnitude of the action, and issues the corresponding first alarm signal.

[0028] Step S200 includes the following: based on the normalized carbon monoxide sensor measurement value... Normalized temperature change and normalized humidity change Calculate the flammability index CL , , For weight values, The activation function is expressed as follows: , x For the independent variable of the activation function, λ For steepness parameter, θ Activation threshold; Combustion index CL Used to evaluate the completeness of combustion during the use of gas appliances; 1- CL As the non-combustion index and compared with the normalized methane sensor measurement. Normalized smoke concentration sensor measurement values and normalized water immersion sensor measurements The weighted summation yields the first leakage risk score. LRS , , The value is a real number; further, the leakage acceleration of methane is calculated, and the normalized methane sensor measurement is then used as the basis for the calculation. Combustion Index CL First Leakage Risk Score LRS and the acceleration of methane leakage LA , , These are the normalized methane sensor measurements at different sampling times. Configure the linkage rules for smart home devices during the gas appliance usage period for the sampling interval.

[0029] In one embodiment, The values ​​are 2, 3, 1.5, and 5 respectively; kurtosis parameter λ The activation threshold is 8. θ It is 0.6; The values ​​are 0.5, 0.3, 0.1, and 0.1.

[0030] Combustion Index CLBy comprehensively considering changes in carbon monoxide concentration, temperature, and humidity, and employing an activation function, when the independent variable of the activation function is less than the activation threshold... θ At that time, the combustion index CL When the value approaches 0, it is considered that combustion is incomplete or nonexistent. When the independent variable of the activation function exceeds the activation threshold... θ At that time, the combustion index CL A value close to 1 indicates complete combustion of the gas, allowing for quick identification of whether combustion has occurred and avoiding misjudgment. First Leakage Risk Score LRS The system takes into full account the methane content in the air during gas combustion, the possibility of non-combustion, smoke, and the degree of water immersion to comprehensively determine whether a gas leak has occurred.

[0031] The smart home device linkage rules mentioned here for configuring the usage periods of gas appliances include the following: 1) Normalized methane sensor measurement value Not exceeding 0.05, flammability index CL A score greater than 0.4 indicates a first-level leakage risk score. LRS If the methane leakage acceleration is less than 0.3 and the leakage acceleration is less than 0.01, the gateway will only record and save the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed to allow natural ventilation; this indicates that the gas combustion is relatively complete and no alarm will be generated.

[0032] 2) Normalized methane sensor measurement value In the range (0.05, 0.15), the flammability index CL Less than 0.3, first leakage risk score LRS If the methane leakage acceleration is greater than 0.3 and does not exceed 0.03, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates a first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm prompt; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 30% of the rated maximum speed; this situation indicates that there may be a slight risk of leakage, but only the first alarm signal is issued to prompt the user to pay attention to whether the combustion situation is deteriorating.

[0033] 3) Normalized methane sensor measurement value In the range (0.15, 0.25), the flammability indexCL Less than 0.2, first leakage risk score LRS If the methane leakage acceleration is greater than 0.4 and does not exceed 0.08, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates a first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm and closes the gas inlet valve of the gas equipment; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 60% of the rated maximum speed; this situation indicates that there is a certain gas leak, and ventilation is required immediately to prevent the accumulation of harmful gases, and at the same time the first alarm signal is issued.

[0034] 4) Normalized methane sensor measurement value >0.25, or the first leakage risk score LRS If the leakage velocity of methane is greater than 0.6 methane / min, or the leakage acceleration of methane is greater than 0.08 methane / min, the multi-sensor monitoring terminal will issue an audible and visual alarm and immediately shut off the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the doors and windows will be changed, and the ventilation system will be forced to ventilate at 100% of its rated maximum speed, generating the first alarm signal. This situation indicates a large amount of methane leakage or a rapid leakage rate. The gas inlet valve should be shut off immediately, and maximum ventilation should be activated to prevent danger.

[0035] S300: During the non-use period of the gas appliance, the gateway receives sensor detection data of the gas appliance during the non-use period sent by the multi-sensor monitoring terminal, outputs a second leakage risk score, configures the linkage rules of smart home devices during the non-use period of the gas appliance according to the second leakage risk score, and issues a corresponding second alarm signal.

[0036] Step S300 includes the following: based on the normalized smoke alarm readings, the normalized door and window status sensor readings for the current time period, and the time factor... TF Calculating environmental risk benchmarks ERB , , δ This refers to the sensitivity index of doors and windows. The weighting is affected by the smoke. q To account for the non-linearity of the smoke's effect, in this embodiment, the time factor... TF The values ​​are higher at night than during the day. Therefore, it is preferable to conduct leak risk assessments at night when gas appliances are not in use, as this provides better sensitivity. Here, the environmental risk benchmark value is... ERB The baseline for risk assessment considers the combined hazards of closed doors and windows and smoke in enclosed spaces.

[0037] In one embodiment, the door and window sensitivity index δ The weight of the smoke effect is 2. It is 0.5. q It is 0.8.

[0038] Then, based on environmental risk benchmarks ERB Further calculate the environmental risk amplification factor A env , , These are linear coefficients and exponential coefficients, respectively; based on the environmental risk amplification factor. A env and normalized methane sensor measurements The exponential function and the normalized measurement value of the water immersion sensor The second leakage risk score is obtained by weighted summation of the exponential function. LSI , , As a weighted term, This is the sensitivity coefficient. This is the nonlinearity coefficient; combined with the second leakage risk score. LSI Compared with normalized methane sensor measurements Configure the linkage rules for smart home devices during off-peak gas appliance usage periods to issue corresponding secondary alarm signals. Pay close attention to methane levels and water immersion indicators in the air during off-peak gas appliance usage periods. This is because the possibility of carbon monoxide and smoke from incomplete gas combustion is very small at these times, and an abnormal increase in methane levels is likely a sign of gas leakage. The water immersion sensor indicator takes into account the possibility of water accumulation or corrosion in gas pipes near basements or manholes, which can cause the sensor to detect leaks. This is also an auxiliary indicator of potential gas pipeline leaks.

[0039] In one embodiment, the linear coefficient and the exponential coefficient are 0.5 and 0.3, respectively; weighting terms The initial values ​​are 0.8 and 0.2; sensitivity coefficient The coefficients are 5 and 8; the nonlinearity coefficient is... The values ​​are 1.2 and 1.0.

[0040] Configure the linkage rules for smart home devices during non-use periods of gas appliances, and issue corresponding second alarm signals, including the following: 1) Normalized methane sensor measurement value Not exceeding 0.03, or a second leakage risk score. LSI When the value is less than 0.15, the gateway only records and saves the current normalized methane sensor measurement value. Second Leakage Risk Score LSIMaintaining the current state of doors and windows indicates that the possibility of indoor methane leakage is very low.

[0041] 2) Normalized methane sensor measurement value In the range (0.03, 0.08), or the second leakage risk score LSI When the value is in the range (0.15, 0.35), a slight methane residue is detected in the monitored area, and the gateway generates a second warning message. 3) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the temperature is in the range (0.35, 0.60), the multi-sensor monitoring terminal issues an audible and visual alarm and generates a second warning message; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 50% of the rated maximum speed. 4) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the value is in the range (0.35, 0.60), the multi-sensor monitoring terminal will issue an audible and visual alarm, generate a second alarm signal, and immediately close the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation facility will be forced to ventilate at 100% of its rated maximum speed.

[0042] S400: During the non-use period of gas equipment, the gateway also performs a self-test evaluation of the performance of each sensor of the multi-sensor monitoring terminal, and issues a third alarm signal based on the performance degradation. The cloud server also communicates with the customer's mobile terminal, and real-time risk scores, first alarm signals, second alarm signals, and third alarm signals are also sent to the mobile terminal through the cloud server.

[0043] During non-use periods of the gas equipment, the gateway also performs self-testing and evaluation of the performance of each sensor in the multi-sensor monitoring terminal. Based on the performance degradation, a third alarm signal is issued. The sensor performance self-test is performed at fixed intervals during non-use periods of the gas equipment. The sensor performance self-testing process is as follows: by injecting test signals, feature extraction is performed based on the sensor response curve to obtain the feature vector corresponding to the response curve. The similarity between the feature vector and the feature reference vector is evaluated to obtain a similarity score. The degradation degree of each element in the feature vector is weighted and accumulated to obtain the overall degradation degree of the sensor. Based on the sequence value composed of the most recent similarity scores, a time series-based degradation prediction model is constructed. Combined with the overall degradation degree of the sensor, it is assessed whether the sensor needs to be replaced or the sensor's measurement value needs to be compensated. If the sensor is assessed to need to be replaced, a third warning signal is issued.

[0044] Specifically, by injecting a test signal, feature extraction is performed based on the sensor response curve to obtain the feature vector corresponding to the response curve. This is done by setting the feature vector to... F ={f1, f2, f3, f4, f5, f6, f7}, where f1 is the rising slope of the response curve, f2 is the falling slope of the response curve, f3 is the overshoot percentage, f4 is the steady-state fluctuation coefficient of the response curve, f5 is the curvature integral, f6 is the energy distribution ratio of high-frequency energy to low-frequency energy after the fast Fourier transform of the sensor response curve, and f7 is the signal entropy value.

[0045] The meaning and calculation method of the elements in the feature vector will now be explained.

[0046] The rising slope f1 is the slope of the rising segment from the 10% peak value to the 90% peak value in the left-hand trough-to-peak interval of the response curve. The falling edge slope f2 is the slope of the 90% peak to 10% peak interval of the falling segment in the right-hand peak-to-trough interval of the response curve. Overshoot percentage f3 refers to the percentage by which the peak value of the response curve exceeds the target steady-state value; The steady-state fluctuation coefficient f4 of the response curve is the result of dividing the steady-state standard deviation by the steady-state mean after the response curve has entered the steady-state segment after several periods. The curvature integral f5 characterizes the balance or curvature of the response curve. The smoother the curve, the more stable the sensor response. The more curved the curve, the greater the fluctuation in the sensor response, which may indicate noise, oscillation, or nonlinearity. The curvature integral f5 gradually increases over time, indicating that the sensor response is becoming more unsmooth and the sensor may be experiencing performance degradation. After obtaining the spectrum from the fast Fourier transform of the sensor response curve, the energy distribution ratio f6 of high-frequency energy to low-frequency energy shows an increasing trend, indicating that the high-frequency energy is increasing, the sensor noise is increasing, and the sensor may be faulty. If the ratio remains stable, it indicates that the low-frequency energy always dominates, and the sensor performance is relatively reliable. The signal entropy value f7 characterizes the complexity and predictability of the sensor's output signal and is closely related to the sensor's accuracy, stability, noise level, and information richness.

[0047] Evaluating the similarity between the morphology of the feature vector and the feature reference vector, and obtaining a similarity score, involves defining the reference feature vector: , Let the eigenvector corresponding to the response curve currently measured by the sensor be an element in the reference characteristic vector. , Define a weighted Euclidean distance for each element in the current feature vector. , As weight, and ; Subscript i If the morphological similarity represents 7 distinct elements in the feature vector, then the morphological similarity is... , u This is a scaling factor, with a value ranging from 0.8 to 1.2, used to adjust the sensitivity of morphological similarity.

[0048] Define the degree of decay for each element in the eigenvector: The overall degradation degree is obtained by weighting and summing the degradation degrees of each element. , For element weights, .

[0049] The degradation level of the sensor is defined based on the morphological similarity value and the overall degradation degree obtained above: L1) When the overall degree of decline U <0.1, and morphological similarity S simi When the value is greater than 0.9, the sensor is considered to have stable performance and no degradation has occurred. The degradation trend can be recorded periodically. L2) When the overall decline rate is 0.1 ≤ U <0.3, and morphological similarity ≤0.7 S simi When the value is less than 0.9, the sensor's performance is considered to have slightly degraded. At this point, it is necessary to update the sensor's calibration coefficients and correct the sensor's detection value, so that the corrected sensor detection value is... out new , , This is the calibration coefficient for the sensor, with a value ranging from 0.1 to 0.3; If the sensor calibration coefficient needs to be updated in three consecutive self-test evaluation cycles, and the sensor calibration coefficient gradually increases, then the sensor degradation level should be defined and handled according to the L3 case. L3) When the overall decline rate is 0.3 ≤ U <0.5, and morphological similarity ≤0.5 S simi When the value is less than 0.7, the sensor's performance is considered to have experienced moderate degradation. If the sensor is one of the following: a methane sensor, a carbon monoxide sensor, or a smoke sensor, then the sensor's range is adjusted from 1-100% to 1-75%, which is equivalent to amplifying the normalized sensor measurement value. If the sensor is not one of the following: a methane sensor, a carbon monoxide sensor, or a smoke sensor, then its measurement value is lowered in the first leak risk score. LRS Second Leakage Risk Score LSI The weights in, such as β , ω 1. ω 2. ω 3. ω 4. b Recalculate the overall decay degree and morphological similarity until the requirements of case L2) are met. If the requirements of case L2) cannot be met after multiple adjustments, then proceed according to case L4. L4) When the overall degree of decline U ≥0.5, and morphological similarity S sim When the value is less than 0.5, the sensor's performance is considered to have severely degraded, requiring shutdown and replacement with a new sensor.

[0050] On the other hand, the present invention provides a multi-sensor fusion gas monitoring system for implementing the above-mentioned method, comprising: a multi-sensor monitoring terminal, a gateway, several indoor smart home devices, a cloud server, and a mobile terminal, wherein... The multi-sensor monitoring terminal communicates with the gateway to acquire sensor detection data during the gas equipment's usage period and non-use period, and sends the sensor detection data to the gateway. The gateway receives sensor detection data from the multi-sensor monitoring terminal during the gas appliance usage period. It then performs fusion detection using built-in fusion rules, outputting a real-time first leak risk score. Based on this score, it configures linkage rules for smart home devices during the gas appliance usage period. The gateway then drives the indoor smart home devices to perform actions or adjust the magnitude of those actions, issuing a corresponding first alarm signal. The gateway also receives sensor detection data from the multi-sensor monitoring terminal during non-gas appliance usage periods, outputting a second leak risk score. Based on this score, it configures linkage rules for smart home devices during non-gas appliance usage periods and issues a corresponding second alarm signal. The gateway also communicates with both the indoor smart home devices and the cloud server. Furthermore, the gateway periodically performs self-evaluation of the performance of each sensor in the multi-sensor monitoring terminal, issuing a third alarm signal based on performance degradation. The cloud server communicates with the mobile terminal to send the received real-time risk score, first alarm signal, second alarm signal, and third alarm signal to the mobile terminal.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A gas monitoring method using multi-sensor fusion, characterized in that, Includes the following steps: A multi-sensor monitoring terminal is configured to acquire sensor detection data during the usage and non-use periods of gas appliances; the multi-sensor monitoring terminal also communicates with a gateway, which in turn communicates with indoor smart home devices and a cloud server. After receiving the sensor detection data of the gas equipment usage period sent by the multi-sensor monitoring terminal, the gateway performs fusion detection through the built-in fusion rules, outputs the first leakage risk score in real time, and then configures the linkage rules of smart home devices during the gas equipment usage period based on the real-time leakage risk score. The gateway drives the indoor smart home devices to act or adjust the magnitude of the action, and issues the corresponding first alarm signal. During the non-use period of the gas appliance, the gateway receives sensor detection data of the gas appliance during the non-use period sent by the multi-sensor monitoring terminal, outputs a second leakage risk score, configures the linkage rules of smart home devices during the non-use period of the gas appliance according to the second leakage risk score, and issues a corresponding second alarm signal. During non-use periods of gas equipment, the gateway also performs a self-test evaluation of the performance of each sensor in the multi-sensor monitoring terminal, and issues a third alarm signal based on the performance degradation. The cloud server also communicates with the customer's mobile terminal, and real-time risk scores, first alarm signals, second alarm signals, and third alarm signals are also sent to the mobile terminal through the cloud server.

2. The gas monitoring method using multi-sensor fusion according to claim 1, characterized in that, The multi-sensor monitoring terminal includes a methane sensor, a carbon monoxide sensor, a smoke concentration sensor, a temperature and humidity sensor, a water immersion sensor, and a door and window status sensor. The measured values ​​from the methane sensor, carbon monoxide sensor, smoke concentration sensor, and temperature and humidity sensor are normalized to the range [0, 1] to obtain the normalized methane sensor measurement value. Normalized carbon monoxide sensor measurements Normalized smoke concentration sensor measurement values Normalized temperature change and normalized humidity change Normalized water immersion sensor measurement values Measurement values ​​from door and window status sensors It can be 0 or 1.

3. The gas monitoring method using multi-sensor fusion according to claim 2, characterized in that, After receiving sensor detection data from the multi-sensor monitoring terminal regarding the usage period of the gas appliances, the gateway performs fusion detection using built-in fusion rules, outputs a real-time first leakage risk score, and then configures the linkage rules for smart home devices during the gas appliance usage period based on the real-time leakage risk score. The gateway drives the indoor smart home devices to perform actions or adjust the magnitude of their actions, and issues a corresponding first alarm signal, including the following: based on the normalized carbon monoxide sensor measurement value... Normalized temperature change and normalized humidity change Calculate the flammability index CL Combustion Index CL Used to evaluate the completeness of combustion during the use of gas appliances; 1- CL As the non-combustion index and compared with the normalized methane sensor measurement. Normalized smoke concentration sensor measurement values and normalized water immersion sensor measurements The weighted summation yields the first leakage risk score. LRS Further, the leakage acceleration of methane was determined, and the normalized methane sensor measurements were combined. Combustion Index CL First Leakage Risk Score LRS and the acceleration of methane leakage LA Configure the linkage rules for smart home devices during the usage periods of gas appliances.

4. The gas monitoring method using multi-sensor fusion according to claim 3, characterized in that, The linkage rules for smart home devices during the designated gas appliance usage periods include the following: 1) Normalized methane sensor measurement value Not exceeding 0.05, flammability index CL A score greater than 0.4 indicates a first-level leakage risk score. LRS If the methane leakage acceleration is less than 0.3 and the leakage acceleration is less than 0.01, the gateway will only record and save the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS ; If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed to allow natural ventilation. 2) Normalized methane sensor measurement value In the range (0.05, 0.15), the flammability index CL Less than 0.3, first leakage risk score LRS If the methane leakage acceleration is greater than 0.3 and does not exceed 0.03, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates the first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm prompt; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 30% of the rated maximum speed. 3) Normalized methane sensor measurement value In the range (0.15, 0.25), the flammability index CL Less than 0.2, first leakage risk score LRS If the methane leakage acceleration is greater than 0.4 and does not exceed 0.08, the gateway not only records and saves the current normalized methane sensor measurement value. Combustion Index CL and the first leakage risk score LRS It also generates the first alarm signal, and the multi-sensor monitoring terminal issues an audible and visual alarm and closes the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation system will be turned on to force ventilation at 60% of the rated maximum speed. 4) Normalized methane sensor measurement value >0.25, or the first leakage risk score LRS If the leakage acceleration of methane is greater than 0.6, or greater than 0.08 / min, the multi-sensor monitoring terminal will issue an audible and visual alarm and immediately shut off the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation system will be forced to ventilate at 100% of the rated maximum speed and generate the first alarm signal.

5. The gas monitoring method using multi-sensor fusion according to claim 3, characterized in that, During the non-use period of the gas appliance, the gateway receives sensor detection data from the multi-sensor monitoring terminal, outputs a second leakage risk score, and configures the linkage rules for smart home devices during the non-use period based on the second leakage risk score, issuing a corresponding second alarm signal, including the following: based on the normalized smoke alarm values, normalized door and window status sensor measurements, and time factor of the current time period. TF Calculating environmental risk benchmarks ERB Based on environmental risk benchmarks ERB Further calculate the environmental risk amplification factor A env ; Based on environmental risk amplification factor A env and normalized methane sensor measurements The exponential function and the normalized measurement value of the water immersion sensor The second leakage risk score is obtained by weighted summation of the exponential function. LSI ; Combined with the second leakage risk score LSI Compared with normalized methane sensor measurements Configure the linkage rules for smart home devices during non-use periods of gas appliances, and issue corresponding second alarm signals.

6. The gas monitoring method using multi-sensor fusion according to claim 5, characterized in that, The configuration rules for the linkage of smart home devices during non-use periods of the gas appliance will issue a corresponding second alarm signal, including the following: 1) Normalized methane sensor measurement value Not exceeding 0.03, or a second leakage risk score. LSI When the value is less than 0.15, the gateway only records and saves the current normalized methane sensor measurement value. Second Leakage Risk Score LSI To maintain the current state of the doors and windows; 2) Normalized methane sensor measurement value In the range (0.03, 0.08), or the second leakage risk score LSI When the value is in the range (0.15, 0.35), a slight methane residue is detected in the monitored area, and the gateway generates a second warning message. 3) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the temperature is in the range (0.35, 0.60), the multi-sensor monitoring terminal issues an audible and visual alarm and generates a second warning message; if the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status is changed, and the ventilation facility is turned on to force ventilation at 50% of the rated maximum speed. 4) Normalized methane sensor measurement value In the range (0.08, 0.15], or the second leakage risk score LSI When the range is (0.35, 0.60), the multi-sensor monitoring terminal will issue an audible and visual alarm, generate a second alarm signal, and immediately shut off the gas inlet valve of the gas equipment. If the normalized measurement value of the door and window status sensor is 1 at this time, the opening degree of the door and window status will be changed, and the ventilation system will be forced to ventilate at 100% of its rated maximum speed.

7. The gas monitoring method using multi-sensor fusion according to claim 5, characterized in that, The time factor TF The value at night is greater than the value during the day.

8. A gas monitoring method using multi-sensor fusion according to claim 2, characterized in that, During the non-use periods of the gas equipment, the gateway also performs a self-test evaluation of the performance of each sensor in the multi-sensor monitoring terminal. Based on the performance degradation, a third alarm signal is issued. The sensor performance self-test is performed at fixed intervals during the non-use periods of the gas equipment. The sensor performance self-test process is as follows: by injecting a test signal, feature extraction is performed based on the sensor response curve to obtain the feature vector corresponding to the response curve. The similarity between the feature vector and the feature reference vector is evaluated to obtain a similarity score. The degradation degree of each element in the feature vector is weighted and accumulated to obtain the comprehensive degradation degree of the sensor. Based on the sequence value composed of the most recent similarity scores, a time series-based degradation prediction model is constructed. Combined with the comprehensive degradation degree of the sensor, it is assessed whether the sensor needs to be replaced or the sensor's measurement value needs to be compensated. If the sensor needs to be replaced, a third warning signal is issued.

9. A gas monitoring method using multi-sensor fusion according to claim 8, characterized in that, The step of injecting a test signal and extracting features from the sensor response curve to obtain the corresponding feature vector is to let the feature vector be F={f1, f2, f3, f4, f5, f6, f7}, where f1 is the rising slope of the response curve, f2 is the falling slope of the response curve, f3 is the overshoot percentage, f4 is the steady-state fluctuation coefficient of the response curve, f5 is the curvature integral, f6 is the energy distribution ratio of high-frequency energy to low-frequency energy after the fast Fourier transform of the sensor response curve, and f7 is the signal entropy value.

10. A multi-sensor fusion gas monitoring system, used to implement the method according to any one of claims 1-9, characterized in that, include: The system includes a multi-sensor monitoring terminal, a gateway, several indoor smart home devices, a cloud server, and a mobile terminal. The multi-sensor monitoring terminal communicates with the gateway to acquire sensor detection data during the gas equipment's usage period and non-use period, and sends the sensor detection data to the gateway. The gateway receives sensor detection data from the multi-sensor monitoring terminal during the gas appliance usage period. It then performs fusion detection using built-in fusion rules, outputting a real-time first leak risk score. Based on this score, it configures linkage rules for smart home devices during the gas appliance usage period. The gateway then drives the indoor smart home devices to perform actions or adjust the magnitude of those actions, issuing a corresponding first alarm signal. The gateway also receives sensor detection data from the multi-sensor monitoring terminal during non-gas appliance usage periods, outputting a second leak risk score. Based on this score, it configures linkage rules for smart home devices during non-gas appliance usage periods and issues a corresponding second alarm signal. The gateway also communicates with both the indoor smart home devices and the cloud server. Furthermore, the gateway periodically performs self-evaluation of the performance of each sensor in the multi-sensor monitoring terminal, issuing a third alarm signal based on performance degradation. The cloud server communicates with the mobile terminal to send the received real-time risk score, first alarm signal, second alarm signal, and third alarm signal to the mobile terminal.

Images