Wall-hanging stove gas use safety monitoring and alarming system based on optical sensor

Abstract

The invention relates to the technical field of wall-hanging stoves, in particular to a wall-hanging stove gas safety monitoring and alarming system based on an optical sensor, which comprises a spectrum sensor calibration module, a gas spectrum analysis module, a leakage dynamic monitoring module and a leakage type identification module. According to the invention, by finely adjusting the sensitivity of the spectrum sensor and optimizing the calibration of the optical element, the detection capability on key gases such as methane and carbon monoxide is improved, and the absorption peak value in multi-wavelength spectrum data can be accurately captured and analyzed, so that the system is more accurate in monitoring gas concentration and composition; by analyzing the time sequence distribution of the absorption peak value and combining the correlation between the spectrum peak value and the energy, the system can effectively identify key energy characteristics, the real-time performance of monitoring and the accuracy of early warning are further enhanced, the system monitors the change trend of the energy value in real time and is matched with threshold value setting, and the accuracy of early warning is improved. It is ensured that an alarm is triggered quickly when any abnormal leakage is detected.

Description

Technical Field

[0001] The invention relates to the technical field of wall-mounted boilers, and in particular to a gas safety monitoring and alarm system for wall-mounted boilers based on optical sensors. Background Art

[0002] Wall-mounted boiler technology involves home and commercial heating systems, mainly used for heating and hot water supply by burning natural gas, liquefied petroleum gas or other types of fuel. Wall-mounted boilers are favored for their easy installation, small space occupation, high efficiency and easy operation. The equipment is usually installed on the indoor wall to save space while providing continuous heat output. With the development of technology, more wall-mounted boilers have begun to adopt advanced safety monitoring and fault diagnosis functions to improve the safety and convenience of use.

[0003] Among them, the optical sensor wall-mounted boiler gas safety monitoring and alarm system involves a safety system designed specifically for wall-mounted boilers. Its purpose is to detect gas leaks or other dangerous conditions during the combustion process through optical sensors. The optical sensor can monitor the composition and concentration changes of the gas. Once an abnormal state is detected, the system will automatically trigger an alarm to remind the user to take measures or automatically cut off the gas source, thereby preventing possible fire or explosion accidents.

[0004] Existing technologies are limited to basic temperature or smoke monitoring and lack the ability to dynamically monitor changes in gas composition and concentration, reducing the system's ability to identify and respond to gas leaks in the early stages. In particular, when monitoring multiple gas compositions, the sensitivity and response time of traditional sensors are insufficient to effectively distinguish and alarm low-concentration gas leaks, resulting in delayed response measures in emergencies. Existing systems lack the ability to combine spectral analysis with real-time data processing, resulting in the system being unable to quickly determine the type and specific location of the leak when a leak occurs, increasing potential safety risks and possibly failing to provide sufficient information to take effective safety measures at critical moments, leading to accidents. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a wall-mounted boiler gas safety monitoring and alarm system based on optical sensors.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a gas safety monitoring and alarm system for a wall-mounted boiler based on an optical sensor, the system comprising: The spectral sensor calibration module adjusts the detector sensitivity range based on the spectral sensor parameters to match the spectral absorption characteristics of methane and carbon monoxide. By calibrating the optical components, it analyzes the absorption characteristics of the different wavelength spectra and captures the absorption peaks of multiple wavelengths to obtain the spectral response characteristics. The gas spectrum analysis module extracts the absorption peak of the multi-wavelength spectrum data based on the spectrum response characteristics, analyzes the time series distribution of the absorption peak, and compares the energy deviation of the difference time series according to the relationship between the spectrum peak and energy to obtain energy characteristic data; The dynamic leakage monitoring module monitors the changing trend of the energy value based on the energy characteristic data, determines the abnormal change of the gas concentration, and sets the concentration threshold for real-time monitoring. If data beyond the normal range is detected, the leakage probability is evaluated, and an alarm is triggered to obtain a leakage warning indicator; The leakage type identification module analyzes the time distribution of abnormal signals based on the leakage warning indicators, matches leakage signals of different types, and classifies the signals according to the distribution characteristics of key wavelengths in the spectral data to determine the leakage type and obtain the leakage type identification result.

[0007] The present invention is improved in that the step of obtaining the spectral response characteristics is specifically as follows: Based on the spectral sensor parameters, the sensitivity range of the detector is adjusted to match the spectral absorption characteristics of methane and carbon monoxide, and the adjusted sensitivity parameters are obtained; Based on the adjusted sensitivity parameters, the absorption characteristics of the difference wavelength spectrum are analyzed by calibrating the optical elements, using the formula: ; Calculate the absorption peaks of multiple wavelengths and obtain the spectral response characteristics, where: Indicated in wavelength The spectral response characteristics under is at wavelength The light intensity under is at wavelength The absorption rate under and Respectively represent the starting and ending wavelengths of spectrum detection.

[0008] The present invention is improved in that the analysis steps of the time series distribution of the absorption peak are specifically as follows: Based on the spectral response characteristics, the absorption peak is extracted for the multi-wavelength spectral data, and the occurrence time of each peak is recorded by using a timestamp to obtain a peak data set with a time stamp; The statistical processing of the peak data set with time stamps is performed using the formula: ; Calculate the average time interval between two consecutive absorption peaks , identify regularities and anomalies in time series, and obtain statistical analysis results of time intervals, where and Represent the occurrence time of two consecutive peaks, It is The timestamp of the peak value, It is The timestamp of the peak value, is the total number of peaks; Based on the statistical analysis results of the time intervals, the time series data are smoothed and analyzed, the long-term trend and short-term fluctuation of the absorption peak are mined, and the time series distribution information is obtained.

[0009] The present invention is improved in that the step of acquiring the energy characteristic data is specifically as follows: Based on the time series distribution, according to the relationship between the spectral peak and energy, the formula is adopted: ; Calculate the energy of each spectral peak at the corresponding time point to obtain the energy calculation result, where: Indicates at time The energy represents the absorption coefficient, It's in time The light intensity, represents the monitoring period; Based on the energy calculation result, the deviation between the energy calculation result and the energy at the previous and next time points is analyzed to identify key characteristics of the energy, including periodic fluctuations, sudden increases or decreases, and obtain energy characteristic data.

[0010] The present invention is improved in that the step of judging the abnormal change of gas concentration is specifically as follows: Based on the energy characteristic data, the energy value of each detection point is monitored in real time, and the instantaneous change trend of the energy value is determined by analyzing the continuous readings of the energy value to obtain energy trend data; Based on the energy trend data, the formula is used: ; Analyze the rate and direction of change to obtain the gradient of energy change ,in, represents the differential change in energy, Represents differential change in time; The gradient of the energy change is compared with an abnormality detection standard, and if the energy gradient of consecutive data points exceeds a critical value, it is determined to be an abnormal change in gas concentration.

[0011] The present invention is improved in that the steps of obtaining the leakage warning indicator are specifically as follows: Based on the abnormal changes in the gas concentration, a concentration threshold is set for real-time monitoring, and concentration changes beyond the normal range are identified to obtain an abnormal monitoring log; Based on the abnormal monitoring log, the frequency of occurrence of the over-threshold event is calculated using the formula: ; Assess the severity of the leakage incident and obtain the leakage risk index ,in, is the energy gradient value extracted from the abnormal event, is the weight assigned according to the frequency of event occurrence. is the total number of abnormal events detected; Based on the leakage risk index, if it exceeds a threshold, an early warning is triggered to obtain a leakage early warning indicator.

[0012] The present invention is improved in that the matching step of the difference type leakage signal is specifically as follows: Based on the leakage warning indicator, extract the abnormal signal and its corresponding timestamp, and perform timestamp analysis to obtain a time distribution diagram; Based on the time distribution diagram, the collected abnormal signals are classified using the formula: ; Compare with known leakage signal types to identify signals that match the target leakage type, where: Indicates The score of the signal type, Representative in the The signal weight at each time point, It is Class signal The characteristic value at a time point, Represents from 1 to All time points are used to calculate the The sum of the eigenvalues ​​of the class signal at all time points, and is the time point index, is the target time point, All time points, is the total number of time points.

[0013] The present invention is improved in that the steps of obtaining the leakage type identification result are specifically as follows: Preliminarily determining potential leakage types based on distribution characteristics of key wavelengths in the spectral data, including peak values, widths, and intensities of the wavelengths, and obtaining a leakage type candidate set; Signal analysis is performed on the leakage type candidate set, signals of different leakage types are compared and classified, the leakage type is determined, and the leakage identification efficiency is optimized to obtain a leakage type identification result.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by finely adjusting the sensitivity of the spectral sensor and optimizing the calibration of the optical elements, the detection capability of key gases such as methane and carbon monoxide is improved, and the absorption peaks in the multi-wavelength spectral data can be accurately captured and analyzed, making the system more accurate in monitoring gas concentration and composition. By analyzing the time series distribution of the absorption peaks and combining the correlation between the spectral peaks and the energy, the system can effectively identify key energy characteristics, further enhancing the real-time monitoring and the accuracy of early warning. The system monitors the trend of energy value changes in real time, and cooperates with the threshold setting to ensure that the alarm is quickly triggered when any abnormal leakage is detected. It can also accurately identify the type of leakage through the detailed classification of abnormal signals, thereby providing more targeted response measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 is a flow chart for obtaining the spectral response characteristics in the present invention; Figure 3 is an analysis flow chart of the time series distribution of the absorption peak in the present invention; Figure 4 It is a flow chart for obtaining energy characteristic data in the present invention; Figure 5 This is a flow chart for determining abnormal changes in gas concentration in the present invention; Figure 6 A flowchart for obtaining leakage warning indicators in the present invention; Figure 7 is a flow chart of matching leakage signals of different types in the present invention; Figure 8 This is a flow chart for obtaining leakage type identification results in the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0018] Example See also Figure 1 The present invention provides a technical solution: a gas safety monitoring and alarm system for a wall-mounted boiler based on an optical sensor comprises: The spectral sensor calibration module adjusts the detector sensitivity range based on the spectral sensor parameters to match the spectral absorption characteristics of methane and carbon monoxide. By calibrating the optical components, it analyzes the absorption characteristics of the different wavelength spectra and captures the absorption peaks of multiple wavelengths to obtain the spectral response characteristics. The gas spectrum analysis module extracts the absorption peak of multi-wavelength spectrum data based on the spectral response characteristics, analyzes the time series distribution of the absorption peak, and compares the energy deviation of the difference time series according to the relationship between the spectral peak and energy to obtain energy characteristic data; The dynamic leakage monitoring module monitors the changing trend of energy values ​​based on energy characteristic data, determines abnormal changes in gas concentration, and sets concentration thresholds for real-time monitoring. If data beyond the normal range is detected, the leakage probability is evaluated and an alarm is triggered to obtain leakage warning indicators. The leakage type identification module analyzes the time distribution of abnormal signals based on leakage warning indicators, matches leakage signals of different types, and classifies the signals according to the distribution characteristics of key wavelengths in the spectral data to determine the leakage type and obtain the leakage type identification result.

[0019] The spectral response characteristics include absorptivity, band identification information, and reflection characteristics. The energy characteristic data include energy peak, energy distribution, and energy fluctuation. The leakage warning indicators include change rate, threshold range, and alarm status. The leakage type identification results include leakage signal analysis results, leakage location, and leakage material type.

[0020] See also Figure 2 , the specific steps for obtaining the spectral response characteristics are: Based on the spectral sensor parameters, the sensitivity range of the detector is adjusted to match the spectral absorption characteristics of methane and carbon monoxide, and the adjusted sensitivity parameters are obtained; The initial sensitivity range of the detector is determined through experimental comparison. During the adjustment process, the response area of ​​the detector is first scanned and tested. Combined with the known spectral absorption peaks of methane and carbon monoxide, multi-point sampling is performed using the relationship between the absorption characteristics of each band and the change in light intensity. The change trend between the sampling data is analyzed to determine which sensitivity ranges can capture the band areas with significant absorption characteristics. Areas are selected for further testing. By changing the parameter settings of the sensitivity range, the equipment is repeatedly calibrated until the match between the absorption characteristics and the sensitivity range reaches the optimal level. Finally, the optimized sensitivity range is obtained and the relevant parameters are recorded.

[0021] Based on the adjusted sensitivity parameters, the absorption characteristics of the difference wavelength spectrum are analyzed by calibrating the optical components, using the formula: ; Calculate the absorption peaks of multiple wavelengths and obtain the spectral response characteristics, where: Indicated in wavelength The spectral response characteristics under is at wavelength The light intensity under represents the emission intensity of the light source at this wavelength. is at wavelength The absorbance under the wavelength indicates the absorption intensity of the target gas to light at that wavelength. and Respectively represent the starting and ending wavelengths of the spectrum detection, and determine the calculation range. is the specified wavelength, and the spectral response at this wavelength is calculated. is a variable in the formula, representing and Specific wavelength points within the range; The absorption characteristics of the spectra with different wavelengths are further analyzed by adjusting the optical elements. First, the key components in the optical equipment are calibrated to ensure the spectral continuity and intensity stability of the light source output. Then, the changes in light intensity within the range of different wavelengths are measured. By comparing the changes in light absorption rates of different wavelengths, the position and intensity of the absorption peak of each wavelength are extracted, and a spectral absorption characteristic data set is established. In view of the differences in the absorption characteristics of methane and carbon monoxide, the measurement is repeatedly fine-tuned to improve the accuracy of data collection. Finally, the spectral response characteristics are obtained through comprehensive calculation and inductive analysis of the absorption peaks. According to the actual measurement conditions, the spectral response characteristics are set. , , the light intensity measured in this range and absorption rate as follows: At 600nm, , ; At 1700nm, , ; calculate : ; The result shows that the comprehensive index of the spectral response characteristics is 0.071, indicating that within the set wavelength range, the detected comprehensive absorption intensity is low, and it is necessary to adjust the light source intensity or the detector sensitivity to obtain better measurement results.

[0022] See also Figure 3 , the analysis steps of the time series distribution of the absorption peak are as follows: Based on the spectral response characteristics, the absorption peak is extracted for the multi-wavelength spectral data, and the occurrence time of each peak is recorded using a timestamp to obtain a peak data set with a time stamp; Absorption peaks are extracted for multi-wavelength spectral data. The light intensity distribution within a specific wavelength range is captured by a spectrometer, and the data is recorded at multiple wavelengths using a point-by-point scanning method. The specific time point of each detection is marked in the data set using a timestamp, and low-frequency noise interference is removed using filtering technology to ensure the accuracy and significance of the peak. Then, based on the peak judgment conditions, such as light intensity exceeding a certain set threshold, the peak data that meets the conditions is extracted, and a time stamp is attached to each peak. Finally, all peak data that meet the conditions and time stamps are merged to obtain a peak data set with time stamps.

[0023] Statistical processing of the time-stamped peak data set was performed using the formula: ; Calculate the average time interval between two consecutive absorption peaks , identify regularities and anomalies in time series, and obtain statistical analysis results of time intervals, where and Represent the occurrence time of two consecutive peaks, It is The timestamp of the peak value, It is The timestamp of the peak value, is the total number of peaks, is the number of consecutive peak pairs. Since the interval between each pair of peaks is counted only once, the total number of pairs is the total number of peaks minus one. By calculating the time difference between consecutive peaks, recording the time interval of each pair of adjacent peaks, and arranging all time intervals in chronological order to form a sequence, the mean of the time interval is calculated on this basis to reflect the average time distribution between peaks, and the fluctuation in the time interval sequence is analyzed at the same time, for example, the discreteness of the time interval is measured by the standard deviation. If some time intervals are found to deviate significantly from the mean, they are recorded as abnormal data to provide input for subsequent processing. A data set containing five peaks is obtained. The peak times are 1 second, 2 seconds, 4 seconds, 7 seconds, and 11 seconds respectively. Then, in the formula Equal to 5, the time interval can be calculated as follows: ; ; The results indicate that a peak occurs every 2.5 seconds on average, information that is very useful for understanding the reaction rate and dynamic behavior of the sample.

[0024] Based on the statistical analysis results of the time interval, the time series data is smoothed and analyzed to mine the long-term trend and short-term fluctuation of the absorption peak and obtain the time series distribution information; The time series data is smoothed and analyzed, and the time interval series is smoothed point by point using the sliding average method to reduce random fluctuations in the series. At the same time, long-term trends and short-term fluctuations are monitored by drawing time series graphs. After identifying large fluctuations, the time interval distribution law is deeply explored in combination with statistical indicators such as mean and variance to form a time series information table that reflects the time distribution characteristics of the absorption peak.

[0025] See also Figure 4 , the specific steps for obtaining energy characteristic data are: Based on the time series distribution and the relationship between the spectral peak and energy, the formula is used: ; Calculate the energy of each spectral peak at the corresponding time point to obtain the energy calculation result, where: Indicates at time The energy Represents the absorption coefficient, which is a constant determined based on material properties and spectral characteristics. It's in time The light intensity is measured directly by spectral equipment. represents the monitoring period; First, the peak points are extracted from the spectral data, and the time points corresponding to each peak are recorded to establish a peak-time point array. Then, the energy is calculated by associating the proportional coefficient with the function of wavelength or frequency. The spectral parameters corresponding to each peak point are interpolated to obtain the energy value point by point and form a time series array. Then, the window sliding technology is used to accumulate the energy values ​​in the time series in segments, integrate the total energy values ​​in different time periods, and save the calculation results in a structured form. The observation period For 120 seconds, the light intensity At a specific point in time Measured as , absorption coefficient Obtained through material testing , substitute into the formula to calculate: ; The results show that time The energy at , which provides basic data for further analysis of energy distribution characteristics.

[0026] Based on the energy calculation results, analyze the deviation between the energy at the previous and next time points, identify the key characteristics of energy, including periodic fluctuations, sudden increases or decreases, and obtain energy characteristic data; The energy value at each time point is compared with the energy values ​​at adjacent time points, the energy difference is calculated, and a change rate matrix is ​​established. The time points with significant deviations are screened out by setting a difference threshold, and statistical methods are used to group the significant points into time windows. The variance analysis is performed on the differences within each group to determine the characteristics of periodic fluctuations. At the same time, the sudden increase or decrease points that exceed the threshold are recorded to form an energy feature data set that includes periodic fluctuations and significant change points, providing detailed information for further energy analysis and application.

[0027] See also Figure 5 , the specific steps for judging abnormal changes in gas concentration are: Based on the energy characteristic data, the energy value of each detection point is monitored in real time. By analyzing the continuous readings of the energy value, the instantaneous change trend of the energy value is determined to obtain the energy trend data; The energy value of each detection point is monitored in real time. By recording the energy value every second, the energy value of each time point is associated with its corresponding timestamp and organized into a time series data set. The obvious outliers are removed by noise filters and the collected energy values ​​are interpolated and corrected to generate a continuous energy data stream. By analyzing the time series data stream, the energy difference between adjacent time points is calculated, and the result is associated with the time interval to obtain the trend information of energy value changes over time, and finally determine the instantaneous change trend of the energy value.

[0028] Based on energy trend data, the formula is used: ; Analyze the rate and direction of change to obtain the gradient of energy change ,in, Represents the differential change of energy, that is, the amount of energy change in a short period of time, It represents the differential change of time, that is, the corresponding time interval; In two consecutive time intervals, the energy values ​​collected are and , the time interval is , substitute the data into the formula to calculate: ; The result shows that during this period of time, the energy gradient is , the value reflects the growth rate of energy and is an important indicator for judging the dynamic change trend.

[0029] The gradient of energy change is compared with the abnormal detection standard. If the energy gradient of consecutive data points exceeds the critical value, it is determined to be an abnormal change in gas concentration; By importing the critical threshold range from the anomaly detection standard, each value in the energy change gradient data is compared with the range one by one, and the gradient values ​​exceeding the critical range are marked. The distribution pattern of the abnormal points is determined through statistical analysis of continuous marking time points, and the abnormal marking segments with longer duration are further screened out. The characteristics of the abnormal segments are comprehensively considered to determine whether it is an abnormal change in gas concentration.

[0030] See also Figure 6 , the specific steps for obtaining leakage warning indicators are: Based on abnormal changes in gas concentration, set concentration thresholds for real-time monitoring, identify concentration changes beyond the normal range, and obtain abnormal monitoring logs; The acquired gas concentration data is recorded as a time series in chronological order, and a concentration threshold range is set to distinguish normal values ​​from abnormal values. The gas concentration measured each time is compared with the threshold range. When the concentration is found to exceed the set range, the abnormal point is recorded, and the time, concentration value and corresponding sensor number are stored in the abnormal monitoring log. The abnormal monitoring log must include the deviation between the exceeded concentration value and the normal range, the specific time point of the abnormality and related parameters. The records are then saved in batches for subsequent analysis to form a monitoring log containing records of abnormal concentration changes.

[0031] Based on the abnormal monitoring log, calculate the frequency of occurrence of over-threshold events using the formula: ; Assess the severity of the leakage incident and obtain the leakage risk index ,in, is the energy gradient value extracted from the abnormal event, is the weight assigned according to the frequency of event occurrence. is the total number of abnormal events detected; To calculate the frequency of concentration exceeding threshold events, first group each record in the abnormal monitoring log in chronological order, such as calculating the number of exceeding events by hour or day, and then summarize the overall exceeding frequency according to the number of exceeding events in different time periods. Combined with the deviation between the exceeding concentration value and the normal range, the severity of the leakage event is evaluated and the data is converted into a leakage risk index. In the abnormal monitoring log, the energy gradient values ​​of the extracted abnormal events are , and , and the corresponding event frequency weights are , and , the total number of abnormal events , substitute the value into the formula to calculate: ; The results show that the leakage risk index is , used to quantify the severity of a leak and is an important basis for triggering subsequent warnings.

[0032] Based on the leakage risk index, if it exceeds the threshold, an early warning is triggered to obtain the leakage early warning indicator; Determine whether the risk index exceeds the preset safety threshold, classify the risk index calculated each time through real-time monitoring, mark the risk index that exceeds the threshold as a high-risk state, activate the early warning in the high-risk state, send an alarm notification to the relevant departments, and record the time, location and risk index of the alarm triggering to form a leakage early warning indicator.

[0033] See also Figure 7 , the matching steps of the difference type leakage signal are as follows: Based on the leakage warning indicators, the abnormal signals and their corresponding timestamps are extracted, and the timestamps are analyzed to obtain the time distribution diagram; The abnormal signal data and its corresponding timestamps are extracted. After preliminary screening of the data, invalid records and noise signals are removed. Then, the time series is generated by calculating the fluctuation frequency and intensity changes of the signal in each timestamp. The time series data is sorted in chronological order. Cluster analysis is used to find out the time periods and peak values ​​where the signals frequently appear. Finally, the signal data in each time period is integrated to obtain a time distribution diagram.

[0034] Based on the time distribution graph, the collected abnormal signals are classified using the formula: ; Compare with known leakage signal types to identify signals that match the target leakage type, where: Indicates The scores of the signal types are used to distinguish and confirm different leakage signal types. Representative in the The signal weight at each time point, It is Class signal The characteristic value at a time point, Represents from 1 to All time points are used to calculate the The sum of the eigenvalues ​​of the class signal at all time points, and is the time point index, is the target time point, All time points, is the total number of time points; There are 3 signal types ( ), at 4 time points ( ) is measured, and the weight at each time point ( ) has different settings according to the importance of time, such as [0.1, 0.3, 0.4, 0.2], the characteristic value of a certain signal type at a time point ( ) is [20, 40, 30, 10], then the sum of the eigenvalues ​​of this signal type at all time points is 100, and the score is calculated: ; ; The result shows that based on the set weights and the characteristic values ​​at each time point, the total score of this signal type is 0.28, and the score is used to compare with other types to identify the signal type.

[0035] See also Figure 8 ,The specific steps for obtaining the leakage type identification results are: Based on the distribution characteristics of key wavelengths in the spectral data, including the peak value, width and intensity of the wavelength, the potential leakage type is preliminarily determined to obtain a candidate set of leakage types; The spectral signal data recorded by the spectral acquisition equipment is used to extract the spectral curves at different time points, and the signal analysis tool is used to separate the eigenvalues ​​of the key wavelengths in the spectrum. The frequency distribution, amplitude and time evolution of each eigenvalue are calculated, and the corresponding relationship between the eigenvalue and the leakage type is established. The main abnormal wavelengths in the spectrum are preliminarily associated with the leakage type, and potential leakage types that meet the abnormal conditions are screened out to obtain a candidate set of leakage types.

[0036] Perform signal analysis on the leakage type candidate set, compare and classify the signals of different leakage types, determine the leakage type, and optimize the leakage identification efficiency to obtain the leakage type identification result; By comparing the characteristic signals of candidate leakage types, matching and analyzing the spectral characteristic values ​​of each candidate type, combining the distribution and intensity of abnormal signals, and using the proportion of high-frequency signals and low-frequency signals in the time series for classification, the unique signal patterns of different types of leakage are further quantified. The signal classification efficiency is optimized and the processing logic is adjusted based on the analysis results to generate classification decision data, and the leakage type is determined based on this to obtain the leakage type identification result.

[0037] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A gas safety monitoring and alarm system for wall-mounted boilers based on optical sensors, characterized in that: The system comprises: The spectral sensor calibration module adjusts the detector sensitivity range based on the spectral sensor parameters to match the spectral absorption characteristics of methane and carbon monoxide. By calibrating the optical components, it analyzes the absorption characteristics of the different wavelength spectra and captures the absorption peaks of multiple wavelengths to obtain the spectral response characteristics. The gas spectrum analysis module extracts the absorption peak of the multi-wavelength spectrum data based on the spectrum response characteristics, analyzes the time series distribution of the absorption peak, and compares the energy deviation of the difference time series according to the relationship between the spectrum peak and energy to obtain energy characteristic data; The dynamic leakage monitoring module monitors the changing trend of the energy value based on the energy characteristic data, determines the abnormal change of the gas concentration, and sets the concentration threshold for real-time monitoring. If data beyond the normal range is detected, the leakage probability is evaluated, and an alarm is triggered to obtain a leakage warning indicator; The leakage type identification module analyzes the time distribution of abnormal signals based on the leakage warning indicators, matches leakage signals of different types, and classifies the signals according to the distribution characteristics of key wavelengths in the spectral data to determine the leakage type and obtain the leakage type identification result.

2. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The steps for obtaining the spectral response characteristics are specifically as follows: Based on the spectral sensor parameters, the sensitivity range of the detector is adjusted to match the spectral absorption characteristics of methane and carbon monoxide, and the adjusted sensitivity parameters are obtained; Based on the adjusted sensitivity parameters, the absorption characteristics of the difference wavelength spectrum are analyzed by calibrating the optical elements, using the formula: ； Calculate the absorption peaks of multiple wavelengths and obtain the spectral response characteristics, where: Indicated in wavelength The spectral response characteristics under is at wavelength The light intensity under is at wavelength The absorption rate under and Respectively represent the starting and ending wavelengths of spectrum detection.

3. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The analysis steps of the time series distribution of the absorption peak are specifically as follows: Based on the spectral response characteristics, the absorption peak is extracted for the multi-wavelength spectral data, and the occurrence time of each peak is recorded by using a timestamp to obtain a peak data set with a time stamp; The statistical processing of the peak data set with time stamps is performed using the formula: ； Calculate the average time interval between two consecutive absorption peaks , identify regularities and anomalies in time series, and obtain statistical analysis results of time intervals, where and Represent the occurrence time of two consecutive peaks, It is The timestamp of the peak value, It is The timestamp of the peak value, is the total number of peaks; Based on the statistical analysis results of the time intervals, the time series data are smoothed and analyzed, the long-term trend and short-term fluctuation of the absorption peak are mined, and the time series distribution information is obtained.

4. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The steps for obtaining the energy characteristic data are specifically as follows: Based on the time series distribution, according to the relationship between the spectral peak and energy, the formula is adopted: ； Calculate the energy of each spectral peak at the corresponding time point to obtain the energy calculation result, where: Indicates at time The energy represents the absorption coefficient, It's in time The light intensity, represents the monitoring period; Based on the energy calculation result, the deviation between the energy calculation result and the energy at the previous and next time points is analyzed to identify key characteristics of the energy, including periodic fluctuations, sudden increases or decreases, and obtain energy characteristic data.

5. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The steps of determining the abnormal change of gas concentration are specifically as follows: Based on the energy characteristic data, the energy value of each detection point is monitored in real time, and the instantaneous change trend of the energy value is determined by analyzing the continuous readings of the energy value to obtain energy trend data; Based on the energy trend data, the formula is used: ； Analyze the rate and direction of change to obtain the gradient of energy change ,in, represents the differential change in energy, Represents differential change in time; The gradient of the energy change is compared with an abnormality detection standard, and if the energy gradient of consecutive data points exceeds a critical value, it is determined to be an abnormal change in gas concentration.

6. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The steps for obtaining the leakage warning indicator are specifically as follows: Based on the abnormal changes in the gas concentration, a concentration threshold is set for real-time monitoring, and concentration changes beyond the normal range are identified to obtain an abnormal monitoring log; Based on the abnormal monitoring log, the frequency of occurrence of the over-threshold event is calculated using the formula: ； Assess the severity of the leakage incident and obtain the leakage risk index ,in, is the energy gradient value extracted from the abnormal event, is the weight assigned according to the frequency of event occurrence. is the total number of abnormal events detected; Based on the leakage risk index, if it exceeds a threshold, an early warning is triggered to obtain a leakage early warning indicator.

7. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The matching steps of the difference type leakage signal are specifically as follows: Based on the leakage warning indicator, extract the abnormal signal and its corresponding timestamp, and perform timestamp analysis to obtain a time distribution diagram; Based on the time distribution diagram, the collected abnormal signals are classified using the formula: ； Compare with known leakage signal types to identify signals that match the target leakage type, where: Indicates The score of the signal type, Representative in the The signal weight at each time point, It is Class signal The characteristic value at a time point, Represents from 1 to All time points are used to calculate the The sum of the eigenvalues ​​of the class signal at all time points, and is the time point index, is the target time point, All time points, is the total number of time points.

8. The optical sensor-based wall-mounted boiler gas safety monitoring and alarm system according to claim 1 is characterized in that: The steps for obtaining the leakage type identification result are specifically as follows: Preliminarily determining potential leakage types based on the distribution characteristics of key wavelengths in the spectral data, including the peak value, width and intensity of the wavelength, and obtaining a leakage type candidate set; Signal analysis is performed on the leakage type candidate set, signals of different leakage types are compared and classified, the leakage type is determined, and the leakage identification efficiency is optimized to obtain a leakage type identification result.

Images