Gas leakage early warning method based on infrared detection and application thereof

Abstract

The invention discloses a gas leakage early warning method based on infrared detection and application thereof, and the method comprises the steps: ingeniously carrying out the normalization monitoring of a preset range of a chemical storage region through employing an infrared imaging module, and obtaining the infrared detection data of the temperature distribution condition of the storage region where chemicals are located; according to the scheme, a laser absorption spectrum module is used for detecting the abnormal temperature, the data are collected according to a time sequence to form a data set, and temperature change information is judged on the basis of the data set, so that whether an abnormal temperature condition exists or not is further judged, and an abnormal temperature area is positioned; then, whether abnormal gas components and corresponding concentrations exist in the abnormal temperature area or not is judged, and therefore early warning information is generated in good time; according to the scheme, the operation cost is low, implementation reliability is high, application is flexible, and a reliable and flexible application mechanism basis can be provided for gas leakage safety early warning of chemical storage areas such as wharfs and ports.

Description

Technical Field

[0001] The present invention relates to the fields of terminal hazardous chemicals monitoring and early warning technology, gas leakage monitoring technology and Internet of Things technology, and in particular to a gas leakage early warning method based on infrared detection and its application. Background Art

[0002] As an important transit point for sea and river transportation, port terminals not only play the role of transporting industrial products, but also serve as a transit point for the transportation of chemical raw materials. With the continuous vigorous development of the chemical industry, the demand for raw materials from downstream manufacturers is increasing, which makes the transportation of chemical raw materials and hazardous chemicals gradually present a large-scale trend. When these chemical raw materials and hazardous chemicals are transited at port terminals, they often need to be temporarily stored in specific places for a certain degree of isolation. Especially for some hazardous chemicals that are in the gas phase or may produce gas phase substances, it is very necessary to monitor them closely. The current more common way is to place Sensors are used to prevent or respond to gas leaks based on their monitoring feedback. However, most sensors are deployed in the lower space of the storage area. For the upper or aerial parts, there is often a lack of reliable and routine monitoring methods. This also means that when harmful gas leaks occur, there is a certain lag in the on-site emergency feedback or it is difficult to achieve early warning. At present, some literature has disclosed the use of TDLAS technology to achieve non-contact detection of gas leaks. However, this type of equipment has the problem of high purchase cost, making it difficult to achieve large-scale deployment and routine monitoring in scenes such as port terminals with open terrain and large site areas.

[0003] At present, infrared imaging detection equipment is a common equipment in security applications. It is widely used due to its high cost performance. Based on infrared imaging technology, it can quickly feedback the temperature distribution of the detected area. Gas leakage will present different temperature distributions under infrared imaging due to the objective basis of different gas component distribution. However, there are few literatures that disclose the use of multiple infrared imaging detections to perform joint detection of gas leaks to achieve gas leak warning and tracking; especially when there is overlap in infrared imaging areas, how to fuse temperature data, and how to perform data screening and adaptive adjustment of fusion weights on infrared imaging equipment after long-term use to achieve reliable data output are rare literatures disclosed at present; in addition, how to use infrared imaging technology to judge and monitor gas leaks in a normalized manner, and at the same time, when an abnormality occurs, combined with other detection equipment to achieve reliable early warning of gas leaks is a very positive and practical research topic. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose a gas leakage warning method based on infrared detection and its application, which has fast response, reliable implementation, and can adjust adaptive fusion weights according to the data reliability of infrared imaging equipment and realize efficient judgment and warning of gas leakage.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] A gas leakage early warning method based on infrared detection is used for chemical storage monitoring. A plurality of infrared imaging modules and a plurality of laser absorption spectrum modules are deployed within a preset range of the chemical storage area, and the method comprises:

[0007] S01, performing multi-point detection on a preset range of a chemical storage area at a preset time frequency through an infrared imaging module to obtain the temperature distribution of the chemical storage area and generate infrared detection data;

[0008] S02, collecting infrared detection data in time series to generate an infrared monitoring data set;

[0009] S03, judging the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generating temperature change information;

[0010] S04, determining the abnormal temperature situation and the abnormal temperature area according to the preset conditions based on the temperature change information, and then starting the laser absorption spectrum module to perform spectrum detection on the abnormal temperature area to generate spectrum detection data;

[0011] S05, acquiring spectral detection data, determining the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generating detection feedback information according to the preset conditions;

[0012] S06. Generate early warning information according to preset conditions based on the detection feedback information.

[0013] As a possible implementation mode, further, in the present solution S01, when multiple infrared imaging modules are used to perform multi-point detection on a preset range of a chemical storage area, the detection area ranges of different infrared imaging modules are not exactly the same, and each preset detection point is covered by at least two infrared imaging modules for detection.

[0014] As a preferred implementation option, preferably, this solution S01 includes:

[0015] S011. Perform multi-point detection on a preset range of a chemical storage area at a preset time frequency using multiple infrared imaging modules to generate multiple local infrared data;

[0016] S012, performing spatial standardization and background denoising processing on the multiple local infrared data, so as to align the local infrared data in space and eliminate background noise;

[0017] S013. Use a stitching algorithm to stitch and fuse multiple local infrared data according to overlapping areas of the multiple local infrared data to obtain an image of the temperature distribution of the storage area where the chemicals are located and generate infrared detection data.

[0018] As a preferred implementation option, preferably, in this solution S013, the stitching algorithm is a SIFT algorithm or a SURF algorithm;

[0019] When splicing and fusing, the fusion temperature value of the overlapping area R between different local infrared data is set to T fused,i , which is expressed as:

[0020]

[0021] Among them, T fused,i is the fusion temperature value of the overlapping area R between different local infrared data; T 1,i 、T 2,i are the temperature values ​​of the overlapping position i of the two local infrared data corresponding to the images, and w 1,i 、w 2,i is the temperature weight value of the overlapping position i of the two local infrared data corresponding to the images.

[0022] As a preferred implementation option, preferably, the present solution S013 further includes: collecting actual temperature of overlapping areas of multiple local infrared data at a preset time frequency, and then evaluating the error of the fused temperature value of the spliced ​​fusion area of ​​the multiple local infrared data and adaptively adjusting the temperature weight value of the overlapping position;

[0023] Specifically, it includes the following steps:

[0024] A01. Collect the actual temperature of multiple local infrared data overlapping areas to obtain the actual temperature T i,true ;

[0025] A02. Evaluate the error between the temperature distribution image after splicing and fusion and the real data through the root mean square error RMSE. The formula of the root mean square error RMSE is defined as follows:

[0026]

[0027] Where N represents the number of temperature detection points in the overlapping area; T i,fused It represents the temperature value at the position coordinate point i in the image after the overlapping area is stitched and fused; T i,true Represents the actual temperature value at the position coordinate point i;

[0028] A03. Initial weight setting and temperature fusion formula: For the fusion temperature value of the temperature detection point i in the overlapping area, the initial formula is defined as:

[0029]

[0030] Among them, T 1,i 、T 2,i They respectively represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 respectively represent the initial weights of the temperatures collected by two different infrared imaging modules in the fusion calculation;

[0031] A04. Based on the purpose of minimizing RMSE, the calculation weights of the temperatures collected by different infrared imaging modules are updated using the gradient descent method;

[0032] First, calculate the partial derivative of RMSE with respect to weight w1, and its formula is defined as follows:

[0033]

[0034] Similarly, the partial derivative of RMSE with respect to weight w2 is calculated, and its formula is defined as follows:

[0035]

[0036] Then update the weights, and the formula is defined as follows:

[0037]

[0038]

[0039] Among them, η is the learning rate, which is used to control the step size of weight adjustment, T 1,i 、T 2,i They represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 represent the current weights of the temperatures collected by two different infrared imaging modules in the fusion calculation. represents the updated weight;

[0040] A05. Repeat A02-A04 until RMSE converges to the set threshold or the maximum number of iterations, obtain the updated weights of the temperatures collected by two different infrared imaging modules during the fusion calculation, and then perform weight normalization processing on them. The formula is defined as follows:

[0041]

[0042] in, It represents the update weight of the temperature collected by two different infrared imaging modules during the fusion calculation. Represents the normalized weight value.

[0043] As a preferred implementation option, preferably, this solution S02 includes: collecting infrared detection data S in time series i,tj , generate infrared monitoring data set S, where S = {S 1,t1 , S 2,t2 ……S i,tj}.

[0044] As a preferred implementation option, preferably, this solution S03 includes:

[0045] S031. The historical temperature distribution data of the storage area where the chemicals are located at different time points is used as the first background temperature, and its formula is defined as follows:

[0046]

[0047] Among them, T brackground,i is the background temperature at position i in the storage area where the chemical is located, T hist,i,j is the jth historical temperature data in the historical temperature record data at the storage area position i, and M is the data volume of the historical temperature record data at the storage area position i;

[0048] The average of the temperatures collected by temperature sensors at multiple preset points in the storage area where the chemicals are located is taken as the second background temperature;

[0049] S032. Obtain wind speed data in the storage area where chemicals are located.

[0050] When it is less than or equal to the preset value, the infrared detection data in the infrared monitoring data set is compared with the second background temperature to obtain the real-time temperature difference of the area corresponding to each infrared detection data, and then the real-time temperature difference is collected to generate temperature difference information;

[0051] When it is greater than the preset value, the infrared detection data in the infrared monitoring data set are compared with the first background temperature and the second background temperature to obtain the historical temperature difference and the real-time temperature difference of the area where each infrared detection data is located, and then the historical temperature difference and the real-time temperature difference are collected to generate temperature difference information;

[0052] S033, comparing the infrared detection data at different time points in the infrared monitoring data set to generate temperature fluctuation information of the storage area where the chemicals are located at different time points;

[0053] S034. Gather temperature difference information and temperature fluctuation information to generate temperature change information.

[0054] As a preferred implementation option, preferably, this solution S04 includes:

[0055] S041, obtaining temperature change information, comparing the temperature difference information in the temperature change information with a preset threshold, generating abnormal temperature information when one or more data in the temperature difference information exceeds the preset threshold, and then locating the area where the abnormality exists according to the temperature difference information to determine the temperature abnormal area, and generating abnormal area information;

[0056] S042. According to the abnormal area information, start the laser absorption spectrum module to perform spectrum detection on the temperature abnormal area to generate spectrum detection data.

[0057] As a preferred implementation option, preferably, this solution S05 includes:

[0058] S051. Acquire spectral detection data, and import it into a trained detection neural network to determine the gas composition and concentration of the corresponding detection area, and generate detection feedback information.

[0059] As a preferred implementation option, preferably, this solution S06 includes:

[0060] S061. Obtain test feedback information and perform data analysis on it.

[0061] When the analyzed information indicates that the gas composition in the detection area is the preset warning component and the concentration exceeds the threshold, a warning instruction and warning information for the warning area are generated, and at the same time, a record of monitoring abnormalities is generated.

[0062] When the analyzed information indicates that the gas composition in the detection area is a non-preset warning component or the gas composition is a preset warning component and the concentration does not exceed the threshold, a record information indicating normal monitoring is generated.

[0063] As a preferred implementation option, preferably, this solution also includes:

[0064] S07, in response to the generation of the early warning information, performing gas plume monitoring on the detection area corresponding to the detection feedback information, and generating gas plume information;

[0065] S08. Tracking the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

[0066] As a preferred implementation option, preferably, this solution S07 includes:

[0067] S071. In response to the generation of the early warning information, an infrared monitoring data set is obtained, and the temperature distribution of the storage area where the chemical is located at different time points is determined according to the infrared detection data in the infrared monitoring data set. Then, in combination with the early warning target area corresponding to the early warning information, the contour of the gas cloud is determined according to the temperature difference between the gas cloud leaked in the storage area of ​​the chemical and the background temperature, and the gas cloud contour information is generated;

[0068] S072. Based on the gas cloud contour information, the gas concentration in the gas cloud contour is detected in real time by a laser absorption spectrum module to determine the gas concentration, or the gas concentration in the gas cloud contour is retrieved from the detection feedback information in S05;

[0069] S073. Based on the gas cloud contour information and the gas concentration within the gas cloud contour obtained in S072, a Gaussian plume model is used to predict and monitor the gas diffusion path and concentration distribution to generate gas plume information of the gas cloud formed by the leaking gas.

[0070] Based on the above, this solution also provides a gas leakage warning system based on infrared detection, which includes:

[0071] There are multiple infrared imaging modules, which are deployed within a preset range of the chemical storage area and are used to perform multi-point detection on the preset range of the chemical storage area at a preset time frequency to obtain the temperature distribution of the chemical storage area and generate infrared detection data;

[0072] The data collection module is used to collect infrared detection data in time series to generate infrared monitoring data sets;

[0073] The data processing module is used to judge the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generate temperature change information; and to determine abnormal temperature conditions and abnormal temperature areas according to preset conditions based on the temperature change information;

[0074] There are multiple laser absorption spectrum modules, which are deployed within a preset range of the storage area of ​​the chemical and are used to perform spectrum detection on the abnormal temperature area according to the abnormal temperature situation and the abnormal temperature area to generate spectrum detection data;

[0075] A data detection module is used to obtain spectral detection data, determine the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generate detection feedback information according to preset conditions;

[0076] The early warning instruction module is used to generate early warning information according to the detection feedback information and the preset conditions;

[0077] The equipment linkage module is used to respond to the generation of early warning information, call the infrared imaging module and the laser absorption spectrum module to monitor the gas plume in the detection area corresponding to the detection feedback information, and generate gas plume information;

[0078] The data processing module also tracks the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

[0079] Based on the above, the present solution also provides a method for monitoring gas leakage of hazardous chemicals at a dock, which applies the above-mentioned gas leakage early warning method based on infrared detection.

[0080] By adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art: the present scheme cleverly uses the infrared imaging module to perform normalized monitoring on the preset range of the chemical storage area, and after data fusion of the infrared data collected by multiple infrared imaging modules, obtains the infrared detection data of the temperature distribution of the storage area where the chemicals are located, which is then aggregated in time series to form a data set, and then compared with the preset background temperature to determine the temperature difference, and the infrared detection data at different time points are compared to obtain temperature fluctuation information, thereby forming temperature change information for further determining whether there is an abnormal temperature situation and locating the abnormal temperature area. On this basis, the present scheme further uses the laser absorption spectrum model to The intervention of the infrared imaging module is used to conduct secondary detection of the abnormal temperature area, and then determine whether there are abnormal gas components and their corresponding concentrations in the abnormal temperature area, so as to generate early warning information in a timely manner; when the early warning occurs, this solution also cleverly monitors the position and trajectory of the gas plume through multiple infrared imaging modules and multiple laser absorption spectrum modules, providing a solid data feedback basis for the intervention of back-end management personnel, and also realizing efficient feedback on the diffusion of leaked gas; under normal circumstances, this solution only requires regular monitoring by infrared imaging modules, which not only has low operating costs, but also has reliable implementation and flexible application. It can provide a reliable and flexible application mechanism basis for gas leakage safety early warning in chemical storage areas such as docks and ports. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0082] Figure 1 This is a brief implementation flow chart of the early warning method of this program;

[0083] Figure 2 It is a two-dimensional illustration of the signal acquisition amplitude when the infrared imaging module and the laser absorption spectrum module mentioned in the early warning method of this scheme are working;

[0084] Figure 3 It is a brief schematic diagram of deploying multiple infrared imaging modules and multiple laser absorption spectrum modules within a preset range of a chemical storage area in the early warning method of this scheme;

[0085] Figure 4 This is another brief implementation flow diagram of the early warning method of this scheme;

[0086] Figure 5 This is a brief unit module connection diagram of the early warning system of this scheme. DETAILED DESCRIPTION

[0087] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0088] like Figure 1 As shown, this embodiment provides a gas leakage warning method based on infrared detection, which is used for chemical storage monitoring. A plurality of infrared imaging modules and a plurality of laser absorption spectrum modules are deployed within a preset range of the chemical storage area (reference Figure 3 As shown, where label A is an infrared imaging module, label B is a laser absorption spectrum module, and label C is a brief schematic diagram of the hazardous goods storage area); the gas leakage warning method based on infrared detection in this scheme includes:

[0089] S01, performing multi-point detection on a preset range of a chemical storage area at a preset time frequency through an infrared imaging module to obtain the temperature distribution of the chemical storage area and generate infrared detection data;

[0090] S02, collecting infrared detection data in time series to generate an infrared monitoring data set;

[0091] S03, judging the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generating temperature change information;

[0092] S04, determining the abnormal temperature situation and the abnormal temperature area according to the preset conditions based on the temperature change information, and then starting the laser absorption spectrum module to perform spectrum detection on the abnormal temperature area to generate spectrum detection data;

[0093] S05, acquiring spectral detection data, determining the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generating detection feedback information according to the preset conditions;

[0094] S06. Generate early warning information according to preset conditions based on the detection feedback information.

[0095] In this solution, in order to facilitate data integration and obtain the entire temperature distribution of the storage area where the chemicals are located as much as possible, as a possible implementation method, further, in this solution S01, multiple infrared imaging modules are used to perform multi-point detection on a preset range of the chemical storage area, wherein the detection area ranges of different infrared imaging modules are not exactly the same, and each preset detection point is covered by at least two or more infrared imaging modules for detection.

[0096] In this solution, due to the high deployment cost of the laser absorption spectroscopy module, during routine detection and early warning, infrared monitoring of the chemical storage area is carried out through multiple infrared imaging modules (based on Fourier transform infrared spectroscopy (FTIR) technology), which can not only realize the detection and feedback of the entry of people and animals into the chemical storage area; it can also monitor the ambient temperature distribution of the upper space.

[0097] Combination Figure 2 As shown, in the practical application of the infrared imaging module and the laser absorption spectrum module, the detection range will present a cone shape, that is, when the infrared imaging module and the laser absorption spectrum module are working, the collection range is widely collected in a cone state, rather than being expanded in a single point or two-dimensional fan range. Therefore, when the infrared imaging module performs normal detection on the chemical storage area, it can be reused for the temperature distribution detection in the middle and upper areas of the chemical storage area, that is, when there is a gas leak in the middle or upper part of the chemical storage area or a gas cloud that enters the detection area due to the influence of the wind field, due to the difference in gas composition and concentration of the gas medium to the background area, when the infrared imaging module emits infrared light for detection, there will be a difference in temperature distribution; the detection principle is roughly as follows: different gas molecules have selective absorption characteristics for infrared light of a specific wavelength. When infrared light passes through the air containing these gas molecules, light of a specific wavelength will be absorbed, thereby forming a specific absorption peak on the spectrum. At the same time, gas molecules also emit infrared radiation, which is related to the temperature and concentration of the gas. The infrared imaging module can infer the existence, temperature and concentration of the gas by receiving these emitted infrared radiations. Especially in temperature inference, the infrared imaging module has a relatively mature application, and its reliability has a certain verification basis.

[0098] In the case of combining multiple infrared imaging modules to determine the temperature distribution of the detection area, as a preferred implementation option, preferably, this solution S01 includes:

[0099] S011. Perform multi-point detection on a preset range of a chemical storage area at a preset time frequency using multiple infrared imaging modules to generate multiple local infrared data;

[0100] S012, performing spatial standardization and background denoising processing on the multiple local infrared data, so as to align the local infrared data in space and eliminate background noise;

[0101] S013. Use a stitching algorithm to stitch and fuse multiple local infrared data according to overlapping areas of the multiple local infrared data to obtain an image of the temperature distribution of the storage area where the chemicals are located and generate infrared detection data.

[0102] In terms of data fusion, as a better implementation option, preferably, in this solution S013, the stitching algorithm is a SIFT algorithm or a SURF algorithm;

[0103] When splicing and fusing, the fusion temperature value of the overlapping area R between different local infrared data is set to T fused,i , which is expressed as:

[0104]

[0105] Among them, T fused,i is the fusion temperature value of the overlapping area R between different local infrared data; T 1,i 、T 2,i are the temperature values ​​of the overlapping position i of the two local infrared data corresponding to the images, and w 1,i 、w 2,i is the temperature weight value of the overlapping position i of the two local infrared data corresponding to the images.

[0106] Due to external factors such as the location and service life of the infrared imaging module, different detection errors and performance fluctuations may occur after long-term operation. In order to improve the reliability of data fusion in the overlapping area of ​​data detection, a more accurate, stable and reliable image of the temperature distribution of the storage area can be obtained; as a preferred implementation option, preferably, this solution S013 also includes: collecting actual temperatures in the overlapping area of ​​multiple local infrared data at a preset time frequency, and then evaluating the error of the fused temperature value of the spliced ​​fusion area of ​​multiple local infrared data and adaptively adjusting the temperature weight value of the overlapping position; specifically, it includes the following steps:

[0107] A01. Collect the actual temperature of multiple local infrared data overlapping areas to obtain the actual temperature T i,true ;

[0108] A02. Evaluate the error between the temperature distribution image after splicing and fusion and the real data through the root mean square error RMSE. The formula of the root mean square error RMSE is defined as follows:

[0109]

[0110] Where N represents the number of temperature detection points in the overlapping area; T i,fused It represents the temperature value at the position coordinate point i in the image after the overlapping area is stitched and fused; T i,true Represents the actual temperature value at the position coordinate point i;

[0111] A03. Initial weight setting and temperature fusion formula: For the fusion temperature value of the temperature detection point i in the overlapping area, the initial formula is defined as:

[0112]

[0113] Among them, T 1,i 、T 2,i They respectively represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 respectively represent the initial weights of the temperatures collected by two different infrared imaging modules in the fusion calculation;

[0114] A04. Based on the purpose of minimizing RMSE, the calculation weights of the temperatures collected by different infrared imaging modules are updated using the gradient descent method.

[0115] First, calculate the partial derivative of RMSE with respect to weight w1, and its formula is defined as follows:

[0116]

[0117]

[0118] Similarly, the partial derivative of RMSE with respect to weight w2 is calculated, and the formula is defined as follows:

[0119]

[0120] Then update the weights, and the formula is defined as follows:

[0121]

[0122] Among them, η is the learning rate, which is used to control the step size of weight adjustment, T 1,i 、T 2,iThey represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 represent the current weights of the temperatures collected by two different infrared imaging modules in the fusion calculation. represents the updated weight;

[0123] A05. Repeat A02-A04 until RMSE converges to the set threshold or the maximum number of iterations, obtain the updated weights of the temperatures collected by two different infrared imaging modules during the fusion calculation, and then perform weight normalization processing on them. The formula is defined as follows:

[0124]

[0125] in, It represents the update weight of the temperature collected by two different infrared imaging modules during the fusion calculation. Represents the normalized weight value.

[0126] Through the above data fusion, the present solution can obtain a temperature distribution diagram of the storage area where the chemicals are located in step S01, which can assist the back-end monitoring personnel in making a preliminary judgment on the on-site situation.

[0127] In order to improve the reliability of data collection, when the infrared imaging module generates detection data, this solution also marks the data generation time stamp accordingly, so as to improve the time consistency when fusing data from multiple infrared imaging modules and avoid the abnormal situation of local data time inconsistency.

[0128] In terms of data collection, by collecting data in time series, the flexibility of background processing can be improved, and the lag or time difference misleading caused by data integration can be avoided; as a better implementation option, preferably, this solution S02 includes: collecting infrared detection data S in time series i,tj , generate infrared monitoring data set S, where S = {S 1,t1 , S 2,t2 ……S i,tj}.

[0129] Regarding the determination of the background temperature, since the environment of the storage area is located in the dock area, it is mostly open on all sides. In order to improve the reference reliability and avoid abnormal warnings caused by short-term sudden changes in environmental conditions, as a better implementation option, preferably, this solution S03 includes:

[0130] S031. The historical temperature distribution data of the storage area where the chemicals are located at different time points is used as the first background temperature, and its formula is defined as follows:

[0131]

[0132] Among them, T brackground,i is the background temperature at location i in the storage area where the chemical is located, T hist,u,j is the jth historical temperature data in the historical temperature record data at the storage area position i, and M is the data volume of the historical temperature record data at the storage area position i;

[0133] The average of the temperatures collected by temperature sensors at multiple preset points in the storage area where the chemicals are located is taken as the second background temperature;

[0134] S032. Obtain wind speed data in the storage area where chemicals are located.

[0135] When it is less than or equal to the preset value, the infrared detection data in the infrared monitoring data set is compared with the second background temperature to obtain the real-time temperature difference of the area corresponding to each infrared detection data, and then the real-time temperature difference is collected to generate temperature difference information;

[0136] When it is greater than the preset value, the infrared detection data in the infrared monitoring data set are compared with the first background temperature and the second background temperature to obtain the historical temperature difference and the real-time temperature difference of the area where each infrared detection data is located, and then the historical temperature difference and the real-time temperature difference are collected to generate temperature difference information;

[0137] S033, comparing the infrared detection data at different time points in the infrared monitoring data set to generate temperature fluctuation information of the storage area where the chemicals are located at different time points;

[0138] S034. Gather temperature difference information and temperature fluctuation information to generate temperature change information.

[0139] In step S03 of this solution, two background temperatures are established, one of which is the average temperature calculated based on historical temperature data, and the other is determined by taking the average of data collected in real time by temperature sensors at multiple preset points in the storage area where the chemicals are located. On this basis, based on different situations, the infrared detection data in the infrared monitoring data set is compared with one or both of the two background temperatures to calculate the temperature difference, so that the data obtained by the back-end management personnel is more comprehensive and has reference value, especially the temperature in the historical temperature data with a decreased confidence in the event of an abnormal environment can be compensated by the real-time temperature on site.

[0140] In terms of temperature judgment, as a better implementation option, preferably, this solution S04 includes:

[0141] S041, obtaining temperature change information, comparing the temperature difference information in the temperature change information with a preset threshold, generating abnormal temperature information when one or more data in the temperature difference information exceeds the preset threshold, and then locating the area where the abnormality exists according to the temperature difference information to determine the temperature abnormal area, and generating abnormal area information;

[0142] S042. According to the abnormal area information, start the laser absorption spectrum module to perform spectrum detection on the temperature abnormal area to generate spectrum detection data.

[0143] In order to improve the response efficiency of the system, as a better implementation option, preferably, this solution S05 includes:

[0144] S051. Acquire spectral detection data, and import it into a trained detection neural network to determine the gas composition and concentration of the corresponding detection area, and generate detection feedback information.

[0145] This solution cleverly introduces a detection neural network, which is used to judge the spectral detection data obtained by the laser absorption spectrum module. It not only has high detection efficiency, but also relies on a large amount of training data to provide efficient and stable data detection for gas leaks at chemical sites, and provides a computing power basis for on-site staff to intervene early. The training logic and principle of the detection neural network are existing technologies, so they will not be described in detail here.

[0146] In terms of the generation of early warning information, as a better implementation option, preferably, this solution S06 includes:

[0147] S061. Obtain test feedback information and perform data analysis on it.

[0148] When the analyzed information indicates that the gas composition in the detection area is the preset warning component and the concentration exceeds the threshold, a warning instruction and warning information for the warning area are generated, and at the same time, a record of monitoring abnormalities is generated.

[0149] When the analyzed information indicates that the gas composition in the detection area is a non-preset warning component or the gas composition is a preset warning component and the concentration does not exceed the threshold, a record information indicating normal monitoring is generated.

[0150] Based on the above examples, as another extended implementation plan of this solution, combined with Figure 4 As shown, in order to improve the efficiency of tracking chemical gas leakage and provide more reliable data feedback for back-end management personnel, as a better implementation option, preferably, this solution also includes:

[0151] S07, in response to the generation of the early warning information, performing gas plume monitoring on the detection area corresponding to the detection feedback information, and generating gas plume information;

[0152] S08. Tracking the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

[0153] Among them, this solution can use the infrared imaging module and the laser absorption spectrum module to track the diffusion trajectory and contour of the gas cloud generated by the leaked gas in real time; thereby providing data and hardware basis for the construction and collection of gas plume information. As a better implementation option, preferably, this solution S07 includes:

[0154] S071. In response to the generation of the early warning information, an infrared monitoring data set is obtained, and the temperature distribution of the storage area where the chemical is located at different time points is determined according to the infrared detection data in the infrared monitoring data set. Then, in combination with the early warning target area corresponding to the early warning information, the contour of the gas cloud is determined according to the temperature difference between the gas cloud leaked in the storage area of ​​the chemical and the background temperature, and the gas cloud contour information is generated;

[0155] S072. Based on the gas cloud contour information, the gas concentration in the gas cloud contour is detected in real time by a laser absorption spectrum module to determine the gas concentration, or the gas concentration in the gas cloud contour is retrieved from the detection feedback information in S05;

[0156] S073. Based on the gas cloud contour information and the gas concentration within the gas cloud contour obtained in S072, a Gaussian plume model is used to predict and monitor the gas diffusion path and concentration distribution to generate gas plume information of the gas cloud formed by the leaking gas.

[0157] In this scheme, the Gaussian plume model can be further used to predict and monitor the diffusion path and concentration distribution of gases, especially for the chemical storage area at the terminal, which, as an open space, has a more convenient site advantage in gas diffusion tracking.

[0158] Based on the above, the infrared detection-based gas leakage warning method described in this scheme can be applied to the hazardous chemical gas leakage monitoring method at the wharf.

[0159] Combination Figure 5 As shown, based on the above, this solution also provides a gas leakage warning system based on infrared detection, which includes:

[0160] There are multiple infrared imaging modules, which are deployed within a preset range of the chemical storage area and are used to perform multi-point detection on the preset range of the chemical storage area at a preset time frequency to obtain the temperature distribution of the chemical storage area and generate infrared detection data;

[0161] The data collection module is used to collect infrared detection data in time series to generate infrared monitoring data sets;

[0162] The data processing module is used to judge the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generate temperature change information; and to determine abnormal temperature conditions and abnormal temperature areas according to preset conditions based on the temperature change information;

[0163] There are multiple laser absorption spectrum modules, which are deployed within a preset range of the storage area of ​​the chemical and are used to perform spectrum detection on the abnormal temperature area according to the abnormal temperature situation and the abnormal temperature area to generate spectrum detection data;

[0164] A data detection module is used to obtain spectral detection data, determine the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generate detection feedback information according to preset conditions;

[0165] The early warning instruction module is used to generate early warning information according to the detection feedback information and the preset conditions;

[0166] The equipment linkage module is used to respond to the generation of early warning information, call the infrared imaging module and the laser absorption spectrum module to monitor the gas plume in the detection area corresponding to the detection feedback information, and generate gas plume information;

[0167] The data processing module also tracks the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

[0168] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0169] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0170] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A gas leakage early warning method based on infrared detection, used for chemical storage monitoring, wherein a plurality of infrared imaging modules and a plurality of laser absorption spectrum modules are deployed within a preset range of the chemical storage area, characterized in that: It includes: S01, performing multi-point detection on a preset range of a chemical storage area at a preset time frequency through an infrared imaging module to obtain the temperature distribution of the chemical storage area and generate infrared detection data; S02, collecting infrared detection data in time series to generate an infrared monitoring data set; S03, judging the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generating temperature change information; S04, determining the abnormal temperature situation and the abnormal temperature area according to the preset conditions based on the temperature change information, and then starting the laser absorption spectrum module to perform spectrum detection on the abnormal temperature area to generate spectrum detection data; S05, acquiring spectral detection data, determining the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generating detection feedback information according to the preset conditions; S06. Generate early warning information according to preset conditions based on the detection feedback information.

2. The infrared detection-based gas leakage early warning method according to claim 1, characterized in that: In S01, when multiple infrared imaging modules are used to perform multi-point detection on a preset range of the chemical storage area, the detection areas of different infrared imaging modules are not exactly the same, and each preset detection point is covered by at least two infrared imaging modules for detection.

3. The infrared detection-based gas leakage early warning method according to claim 2, characterized in that: S01 includes: S011. Perform multi-point detection on a preset range of a chemical storage area at a preset time frequency using multiple infrared imaging modules to generate multiple local infrared data; S012, performing spatial standardization and background denoising processing on the multiple local infrared data, so as to align the local infrared data in space and eliminate background noise; S013. Use a stitching algorithm to stitch and fuse multiple local infrared data according to overlapping areas of the multiple local infrared data to obtain an image of the temperature distribution of the storage area where the chemicals are located and generate infrared detection data.

4. The infrared detection-based gas leakage early warning method according to claim 3, characterized in that: In S013, the stitching algorithm is a SIFT algorithm or a SURF algorithm; When splicing and fusing, the fusion temperature value of the overlapping area R between different local infrared data is set to T fused,i , which is expressed as: Among them, T fused,i is the fusion temperature value of the overlapping area R between different local infrared data; T 1,i , T 2,i are the temperature values ​​of the overlapping position i of the two local infrared data corresponding to the images, and w 1,i 、w 2,i is the temperature weight value of the overlapping position i of the two local infrared data corresponding to the images; S013 also includes: collecting actual temperature of overlapping areas of the multiple local infrared data at a preset time frequency, and then evaluating the error of the fused temperature value of the spliced ​​fusion area of ​​the multiple local infrared data and adaptively adjusting the temperature weight value of the overlapping position; It includes the following steps: A01. Collect the actual temperature of multiple local infrared data overlapping areas to obtain the actual temperature T i,true ; A02. Evaluate the error between the temperature distribution image after splicing and fusion and the real data through the root mean square error RMSE. The formula of the root mean square error RMSE is defined as follows: Where N represents the number of temperature detection points in the overlapping area; T i,fused It represents the temperature value at the position coordinate point i in the image after the overlapping area is stitched and fused; T i,true Represents the actual temperature value at the position coordinate point i; A03. Initial weight setting and temperature fusion formula: For the fusion temperature value of the temperature detection point i in the overlapping area, the initial formula is defined as: Among them, T 1,i , T 2,i They respectively represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 respectively represent the initial weights of the temperatures collected by two different infrared imaging modules in the fusion calculation; A04. Based on the purpose of minimizing RMSE, the calculation weights of the temperatures collected by different infrared imaging modules are updated using the gradient descent method; First, calculate the partial derivative of RMSE with respect to weight w1, and its formula is defined as follows: Similarly, the partial derivative of RMSE with respect to weight w2 is calculated, and its formula is defined as follows: Then update the weights, and the formula is defined as follows: Among them, η is the learning rate, which is used to control the step size of weight adjustment, T 1,i 、T 2,i They represent the temperature values ​​of position i collected by two different infrared imaging modules in their overlapping collection areas; w1 and w2 represent the current weights of the temperatures collected by two different infrared imaging modules in the fusion calculation. represents the updated weight; A05. Repeat A02-A04 until RMSE converges to the set threshold or the maximum number of iterations, obtain the updated weights of the temperatures collected by two different infrared imaging modules during the fusion calculation, and then perform weight normalization processing on them. The formula is defined as follows: in, It represents the update weight of the temperature collected by two different infrared imaging modules during the fusion calculation. Represents the normalized weight value.

5. The infrared detection-based gas leakage early warning method according to any one of claims 1 to 4, characterized in that: S02 includes: collecting infrared detection data S in time series i,tj , generate infrared monitoring data set S, where S = {S 1,t1 , S 2,t2 ……S i,tj }; S03 includes: S031. The historical temperature distribution data of the storage area where the chemicals are located at different time points is used as the first background temperature, and its formula is defined as follows: Among them, T brackground,i is the background temperature at position i in the storage area where the chemical is located, T hist , i,j is the jth historical temperature data in the historical temperature record data at the storage area position i, and M is the data volume of the historical temperature record data at the storage area position i; The average of the temperatures collected by temperature sensors at multiple preset points in the storage area where the chemicals are located is taken as the second background temperature; S032. Obtain wind speed data in the storage area where chemicals are located. When it is less than or equal to the preset value, the infrared detection data in the infrared monitoring data set is compared with the second background temperature to obtain the real-time temperature difference of the area corresponding to each infrared detection data, and then the real-time temperature difference is collected to generate temperature difference information; When it is greater than the preset value, the infrared detection data in the infrared monitoring data set are compared with the first background temperature and the second background temperature to obtain the historical temperature difference and the real-time temperature difference of the area where each infrared detection data is located, and then the historical temperature difference and the real-time temperature difference are collected to generate temperature difference information; S033, comparing the infrared detection data at different time points in the infrared monitoring data set to generate temperature fluctuation information of the storage area where the chemicals are located at different time points; S034. Gather temperature difference information and temperature fluctuation information to generate temperature change information.

6. The infrared detection-based gas leakage early warning method according to claim 5, characterized in that: S04 includes: S041, obtaining temperature change information, comparing the temperature difference information in the temperature change information with a preset threshold, generating abnormal temperature information when one or more data in the temperature difference information exceeds the preset threshold, and then locating the area where the abnormality exists according to the temperature difference information to determine the temperature abnormal area, and generating abnormal area information; S042. According to the abnormal area information, start the laser absorption spectrum module to perform spectrum detection on the temperature abnormal area to generate spectrum detection data; S05 includes: S051. Acquire spectral detection data and import it into a trained detection neural network to determine the gas composition and concentration of the corresponding detection area and generate detection feedback information; S06 includes: S061. Obtain test feedback information and perform data analysis on it. When the analyzed information indicates that the gas composition in the detection area is the preset warning component and the concentration exceeds the threshold, a warning instruction and warning information for the warning area are generated, and at the same time, a record of monitoring abnormalities is generated. When the analyzed information indicates that the gas composition in the detection area is a non-preset warning component or the gas composition is a preset warning component, and the concentration does not exceed the threshold, a record information indicating normal monitoring is generated.

7. The infrared detection-based gas leakage early warning method according to claim 6, characterized in that: Also includes: S07, in response to the generation of the early warning information, performing gas plume monitoring on the detection area corresponding to the detection feedback information, and generating gas plume information; S08. Tracking the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

8. The infrared detection-based gas leakage early warning method according to claim 7, characterized in that: S07 includes: S071. In response to the generation of the early warning information, an infrared monitoring data set is obtained, and the temperature distribution of the storage area where the chemical is located at different time points is determined according to the infrared detection data in the infrared monitoring data set. Then, in combination with the early warning target area corresponding to the early warning information, the contour of the gas cloud is determined according to the temperature difference between the gas cloud leaked in the storage area of ​​the chemical and the background temperature, and the gas cloud contour information is generated; S072. Based on the gas cloud contour information, the gas concentration in the gas cloud contour is detected in real time by a laser absorption spectrum module to determine the gas concentration, or the gas concentration in the gas cloud contour is retrieved from the detection feedback information in S05; S073. Based on the gas cloud contour information and the gas concentration within the gas cloud contour obtained in S072, a Gaussian plume model is used to predict and monitor the gas diffusion path and concentration distribution to generate gas plume information of the gas cloud formed by the leaking gas.

9. A gas leakage warning system based on infrared detection, characterized in that: It includes: There are multiple infrared imaging modules, which are deployed within a preset range of the chemical storage area and are used to perform multi-point detection of the preset range of the chemical storage area at a preset time frequency to obtain the temperature distribution of the chemical storage area and generate infrared detection data; The data collection module is used to collect infrared detection data in time series to generate infrared monitoring data sets; A data processing module is used to judge the infrared detection data in the infrared monitoring data set according to preset conditions to determine the temperature change in the storage area where the chemicals are located and generate temperature change information; and determining abnormal temperature conditions and abnormal temperature areas according to preset conditions based on the temperature change information; There are multiple laser absorption spectrum modules, which are deployed within a preset range of the storage area of ​​the chemical and are used to perform spectrum detection on the abnormal temperature area according to the abnormal temperature situation and the abnormal temperature area to generate spectrum detection data; A data detection module is used to obtain spectral detection data, determine the gas composition and / or concentration of the corresponding detection area according to preset conditions, and then generate detection feedback information according to preset conditions; The early warning instruction module is used to generate early warning information according to the detection feedback information and the preset conditions; The equipment linkage module is used to respond to the generation of early warning information, call the infrared imaging module and the laser absorption spectrum module to monitor the gas plume in the detection area corresponding to the detection feedback information, and generate gas plume information; The data processing module also tracks the diffusion position of the gas cloud leaking from the storage area of ​​the chemical according to the gas plume information.

10. A method for monitoring gas leakage of hazardous chemicals at a dock, characterized by: The invention is applied to the gas leakage early warning method based on infrared detection as claimed in any one of claims 1 to 8.

Images