Natural gas leakage quantitative detection method and system based on voiceprint

Abstract

The invention relates to the technical field of leakage detection, and discloses a natural gas leakage quantitative detection method and system based on voiceprint, and the method comprises the steps: collecting a vibration signal of a sound wave sensor, determining the position of a leakage point according to a voiceprint array, and obtaining the fluid leakage flow speed of each fluid detection region, preprocessing the fluid leakage flow velocity to obtain the leakage flow velocity, constructing a leakage flow velocity sample for the corresponding fluid detection area according to the leakage flow velocity, and determining the current leakage environment according to the increasing quantity, the decreasing quantity and the negative quantity; acquiring the historical leakage flow rate in the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, judging whether to correct the leakage flow rate according to the comparison result, and determining a leakage correction factor according to the similar result of the leakage flow rate and the historical leakage flow rate data set. And determining a leakage interval according to the leakage type and the final leakage flow velocity. The method has the characteristic of good quantitative detection of leakage.

Description

Technical Field

[0001] The present invention relates to the technical field of leakage detection, and in particular to a natural gas leakage quantitative detection method and system based on voiceprint. Background Art

[0002] With the widespread use of natural gas, the safety of natural gas leakage has become an important issue in public safety and environmental protection. As a clean energy, natural gas has high safety during use, but it is very easy to leak in the gas state. If a leakage accident occurs, it will lead to serious consequences such as explosion and fire, and even endanger life and property safety.

[0003] At present, there are some technical difficulties in the quantitative detection of natural gas leaks. For example, traditional leak detection methods can only identify whether a leak exists or not, but it is difficult to accurately classify the type of leak, and it is impossible to distinguish reasonable levels based on the measured leakage flow rate, which leads to false alarms and missed opportunities for emergency response.

[0004] Therefore, how to provide a natural gas leak quantitative detection method and system based on voiceprint is a technical problem that technicians in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention proposes a natural gas leakage quantitative detection method and system based on voiceprint, aiming to solve the problems in the current quantitative detection technology that it is impossible to quantitatively detect the leakage flow rate according to the environment, reasonably define the leakage range, and classify the leakage type.

[0006] In one aspect, the present invention provides a natural gas leak quantitative detection method based on voiceprint, comprising:

[0007] Collecting the vibration signal of the acoustic sensor and determining the location of the leakage point according to the soundprint array, dividing the leakage point into a plurality of fluid detection areas based on the soundprint array, obtaining the fluid leakage flow rate of each fluid detection area, and pre-processing the fluid leakage flow rate to obtain the leakage flow rate;

[0008] Constructing a leakage flow rate sample for the corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample;

[0009] Counting the number of the increasing samples and recording it as the increasing number, counting the number of the decreasing samples and recording it as the decreasing number, counting the number of the negative samples and recording it as the negative number, and determining the current leakage environment according to the increasing number, the decreasing number and the negative number;

[0010] Obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and determining whether to correct the leakage flow rate according to the comparison result;

[0011] When it is determined that the leakage flow rate is to be corrected, a historical leakage flow rate data set is determined by using a clustering algorithm for the historical leakage flow rate, a leakage correction factor is determined according to similarity between the leakage flow rate and the historical leakage flow rate data set, and the leakage flow rate is corrected according to the leakage correction factor to obtain a final leakage flow rate;

[0012] The leakage type is determined according to the final leakage flow rate, and the leakage interval is determined according to the leakage type and the final leakage flow rate.

[0013] Furthermore, when constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, it includes:

[0014] Counting all the leakage flow rates and determining a standard leakage flow rate corresponding to each fluid detection area;

[0015] Comparing the leakage flow rate with the standard leakage flow rate, if all the leakage flow rates are greater than the standard leakage flow rate, establishing the growth sample for the corresponding fluid detection area;

[0016] If all leakage flow rates are less than or equal to the standard leakage flow rate, then the reduced sample is established for the corresponding fluid detection area;

[0017] If there is one or more leakage flow rates greater than the standard leakage flow rate, and there is one or more leakage flow rates less than or equal to the standard leakage flow rate, then the negative sample is established for the corresponding fluid detection area.

[0018] Further, when determining the current leakage environment according to the increase quantity, the decrease quantity and the negative quantity, it includes:

[0019] The leakage environment value is calculated according to the increase quantity, the decrease quantity and the negative quantity, and the leakage environment value is obtained by the following formula:

[0020]

[0021]

[0022] in, represents the total number of samples, To increase the number, To reduce the number, is a negative quantity, , , is the weight coefficient and , is a leakage environment value, and the current leakage environment is determined according to the leakage environment value.

[0023] Furthermore, when comparing the leakage flow rate with the historical leakage flow rate and determining whether to correct the leakage flow rate according to the comparison result, the method includes:

[0024] When there is a leakage flow rate that is less than the minimum value of the historical leakage flow rate, it is determined to correct the leakage flow rate that is less than the historical leakage flow rate;

[0025] When there is no leakage flow rate less than the minimum value of the historical leakage flow rate, it is determined that the leakage flow rate is not to be corrected, and the average of all leakage flow rates is determined as the final leakage flow rate.

[0026] Furthermore, when it is determined that the leakage flow rate is to be corrected, the historical leakage flow rate is determined using a clustering algorithm to determine a historical leakage flow rate data set, including:

[0027] Count all historical leakage flow rates and establish a historical flow rate collection;

[0028] Set the cluster number k=2, randomly select two historical leakage flow rates in the historical flow rate set as the first cluster center and the second cluster center, calculate the distance from each historical leakage flow rate in the historical flow rate set to the first cluster center and the second cluster center, and assign it to the cluster with the closest distance. If the distances are the same, assign it to the cluster where the first cluster center is located.

[0029] According to the allocation result, the cluster where the center of the first cluster is located is determined as a first flow rate cluster, and the cluster where the center of the second cluster is located is determined as a second flow rate cluster;

[0030] The first flow rate cluster and the second flow rate cluster are determined as the historical leakage flow rate dataset.

[0031] Further, when determining the leakage correction factor according to the similarity result between the leakage flow rate and the historical leakage flow rate data set, it includes:

[0032] Obtaining a leakage flow rate that is less than the minimum value of the historical leakage flow rate and recording it as the flow rate to be corrected;

[0033] Calculating the flow velocity similarity according to the mean of the first flow velocity cluster, the mean of the second flow velocity cluster and the flow velocity to be corrected;

[0034] The flow rate similarity is obtained by the following formula:

[0035] ;

[0036] in represents the flow velocity similarity, Indicates the flow rate to be corrected. represents the mean of the first flow rate cluster, represents the mean of the second flow rate cluster, and is the weight coefficient and , determine the leakage correction factor based on the flow velocity similarity.

[0037] Further, when the leakage flow rate is corrected according to the leakage correction factor to obtain a final leakage flow rate, it includes:

[0038] When the flow rate to be corrected exists in all the leakage flow rates, a corrected flow rate is obtained according to the product value of the correction factor and the flow rate to be corrected;

[0039] A validation set is established for all corrected flow rates, a data set is established for the uncorrected leakage flow rate, and a normal distribution model is fitted according to the mean and standard deviation of the data set;

[0040] When the probability that the corrected flow rate exists in the normal distribution model in the validation set is less than 0.3, the corrected flow rate is determined to be an erroneous correction and is removed;

[0041] When the probability of all the corrected flow velocities in the verification set in the normal distribution model is greater than or equal to 0.3, the verification set is determined to be correct;

[0042] All corrected flow rates in the validation set that are determined to be correct and all uncorrected leakage flow rates in the data set are summed and averaged, and the result based on the average is determined as the final leakage flow rate.

[0043] Furthermore, when determining the leakage type according to the final leakage flow rate, it includes:

[0044] Comparing the final leakage flow rate with a first preset leakage flow rate and a second preset leakage flow rate respectively, the first preset leakage flow rate being greater than the second preset leakage flow rate, and determining the leakage type of the leakage point according to the comparison result;

[0045] When the final leakage flow rate is greater than or equal to the first preset leakage flow rate, determining that the leakage type of the leakage point is a serious leakage;

[0046] When the final leakage flow rate is less than the first preset leakage flow rate and greater than or equal to the second preset leakage flow rate, determining that the leakage type of the leakage point is a moderate leakage;

[0047] When the final leakage flow rate is less than the second preset leakage flow rate, it is determined that the leakage type of the leakage point is a slight leakage.

[0048] Further, when determining the leakage interval according to the leakage type and the final leakage flow rate, it includes:

[0049] Obtaining the historical leakage flow rate of the serious leakage and establishing a first interval data set, and obtaining the mean value in the first interval data set to determine a primary leakage interval;

[0050] Obtaining the historical leakage flow rate of the moderate leakage and establishing a second interval data set, and obtaining the mean value in the second interval data set to determine the secondary leakage interval;

[0051] Obtain the historical leakage flow rate of the slight leakage and establish a third interval data set, and obtain the mean value in the third interval data set to determine the third-level leakage interval;

[0052] The leakage interval includes the primary leakage interval, the secondary leakage interval and the tertiary leakage interval;

[0053] The result of the quantitative detection is determined by substituting the final leakage flow rate into the leakage interval.

[0054] Compared with the prior art, the beneficial effect of the present invention is that by collecting vibration signals through acoustic wave sensors and combining with voiceprint array technology, the position of the leakage point can be accurately located, avoiding the inefficiency and errors of traditional methods that rely on manual detection or other sensors for large-scale searches. By detecting the leakage flow rate in real time and classifying and counting the leakage flow rate samples, the current environment can be accurately judged. By comparing with the historical leakage flow rate and correcting the leakage flow rate in combination with the clustering algorithm, the adaptive ability of the quantitative detection process is improved, and it can be automatically adjusted according to different environmental changes, thereby improving the accuracy and reliability of quantitative detection. According to the correction of the final leakage flow rate and the determination of the leakage type, the natural gas leak can be finely classified, thereby providing a detailed leakage interval division. It is helpful to carry out effective leakage prevention and control and emergency response, ensuring that appropriate measures can be taken when a leak occurs to reduce the damage caused by the natural gas leak.

[0055] On the other hand, the present application also provides a natural gas leakage quantitative detection system based on voiceprint, which is used to apply the above-mentioned natural gas leakage quantitative detection method based on voiceprint, including:

[0056] A collection unit, used to collect the vibration signal of the acoustic wave sensor and determine the location of the leakage point according to the soundprint array, divide the leakage point into multiple fluid detection areas based on the soundprint array, obtain the fluid leakage flow rate of each fluid detection area, and pre-process the fluid leakage flow rate to obtain the leakage flow rate;

[0057] A leakage environment unit, used for constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample;

[0058] Counting the number of the increasing samples and recording it as the increasing number, counting the number of the decreasing samples and recording it as the decreasing number, counting the number of the negative samples and recording it as the negative number, and determining the current leakage environment according to the increasing number, the decreasing number and the negative number;

[0059] A judgment unit, used for obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and judging whether to correct the leakage flow rate according to the comparison result;

[0060] a correction unit, when determining to correct the leakage flow rate, using a clustering algorithm to determine a historical leakage flow rate data set for the historical leakage flow rate, determining a leakage correction factor according to a similarity result between the leakage flow rate and the historical leakage flow rate data set, and correcting the leakage flow rate according to the leakage correction factor to obtain a final leakage flow rate;

[0061] The type interval determination unit is used to determine the leakage type according to the final leakage flow rate, and determine the leakage interval according to the leakage type and the final leakage flow rate.

[0062] It is understandable that the above-mentioned natural gas leakage quantitative detection method and system based on voiceprint have the same beneficial effects, which will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0064] Figure 1 A flow chart of a natural gas leakage quantitative detection method based on voiceprint provided in an embodiment of the present invention;

[0065] Figure 2 A structural block diagram of a natural gas leakage quantitative detection system based on voiceprint provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0067] Most of the existing natural gas leak quantitative detection technologies rely on visual detection, gas sensors and infrared detection. These technologies can detect the existence of leaks, but often cannot provide accurate leakage data. Especially in complex environments, it is difficult to quantitatively evaluate the leakage based on the actual environment. This makes it impossible for existing detection methods to provide accurate data support for leakage assessment and emergency response, and it is difficult to provide scientific and reliable basis for decision makers.

[0068] On the one hand, although the existing natural gas leak detection technology can detect the existence of leaks, it cannot provide accurate leakage data. Especially in some specific environments, the leakage data obtained will deviate according to the actual environment. This makes it impossible for the existing detection method to quantitatively detect the leakage flow rate and it is difficult to provide a scientific and reliable basis.

[0069] On the other hand, traditional leak monitoring methods have difficulty accurately classifying leak types and cannot distinguish reasonable levels based on the leak flow rate, resulting in low accuracy and reliability of leak measurement results. For example, for leaks with low leak flow rates, traditional methods cannot accurately determine their types and lead to false alarms, while for more serious leaks, they cannot respond in time and miss the best time for emergency response.

[0070] Therefore, it is necessary to design a quantitative detection method for natural gas leaks based on voiceprint to solve the problems existing in current technology.

[0071] In some embodiments of the present application, see Figure 1 As shown, a natural gas leakage quantitative detection method based on voiceprint includes:

[0072] S100: collecting vibration signals of the acoustic wave sensor and determining the location of the leakage point according to the soundprint array, dividing the leakage point into multiple fluid detection areas based on the soundprint array, obtaining the fluid leakage flow rate of each fluid detection area, and pre-processing the fluid leakage flow rate to obtain the leakage flow rate;

[0073] S200: constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample;

[0074] S300: Count the number of increasing samples and record it as the increasing number, count the number of decreasing samples and record it as the decreasing number, count the number of negative samples and record it as the negative number, and determine the current leakage environment according to the increasing number, the decreasing number and the negative number;

[0075] S400: Obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and determining whether to correct the leakage flow rate according to the comparison result;

[0076] S500: when it is determined that the leakage flow rate is to be corrected, a historical leakage flow rate data set is determined by using a clustering algorithm for the historical leakage flow rate, a leakage correction factor is determined according to similarity between the leakage flow rate and the historical leakage flow rate data set, and the leakage flow rate is corrected according to the leakage correction factor to obtain a final leakage flow rate;

[0077] S600: Determine the leakage type according to the final leakage flow rate, and determine the leakage range according to the leakage type and the final leakage flow rate.

[0078] Specifically, the vibration generated by the pipeline natural gas leakage is sensed by the acoustic wave sensor, so as to obtain the location information related to the leakage point, improve the accuracy of the detection position, and the specific location of the tiny leakage point can be accurately located through the soundprint array analysis, and the specific location can be divided into multiple fluid detection areas. Through the soundprint propagation time difference, each fluid detection area can be monitored and the fluid leakage flow rate can be obtained, and the fluid leakage flow rate is pre-processed. The pre-processing includes removing the erroneous flow rate and the environmental noise of the fluid leakage flow rate, improving the accuracy of the fluid leakage flow rate, and providing an accurate basis for subsequent leakage analysis and judgment. The leakage flow rate samples are divided into growth samples, reduction samples and negative samples according to their change trends. Growth samples represent a large leakage flow rate, reduction samples represent a small leakage flow rate, and negative samples represent that they will have a certain impact on the samples that process the leakage flow rate. By counting the number of these samples, the current leakage environment state can be judged. The number of growth samples, reduction samples and negative samples is counted, and recorded as the growth number, reduction number and negative number respectively. These numbers will serve as the basic data for analyzing the leakage environment. After the leakage environment is determined, the historical leakage flow rate that is the same as the current leakage environment is collected. The historical leakage flow rate provides a reference benchmark for the leakage flow rate. The comparison process ensures that the current leakage flow rate can accurately reflect the actual leakage situation and avoid quantitative detection errors caused by environmental changes or sensor errors. When it is determined that the leakage flow rate needs to be corrected, the historical leakage flow rate is aggregated through a clustering algorithm to avoid large differences between the historical leakage flow rates obtained. The leakage correction factor is determined based on the historical leakage flow rate data set. The leakage correction factor is a factor used to adjust the leakage flow rate to ensure the accuracy of the final leakage flow rate and provide data support for subsequent leakage type analysis. The leakage type is determined and the leakage interval is determined in combination with the final leakage flow rate to ensure the scientificity and reliability of quantitative leakage detection.

[0079] It is understandable that accurate detection of natural gas leakage flow rate is achieved through the comprehensive use of voiceprint array, fluid detection technology, sample analysis and historical data comparison for correction. It can not only detect leakage, but also dynamically correct the leakage flow rate according to the specific situation, thus providing reliable technical support for the safety monitoring and risk assessment of natural gas leakage.

[0080] In some embodiments of the present application, when constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, it includes: counting all leakage flow rates and determining a standard leakage flow rate corresponding to each fluid detection area, comparing the leakage flow rate with the standard leakage flow rate, if all leakage flow rates are greater than the standard leakage flow rate, then establishing an increase sample for the corresponding fluid detection area, if all leakage flow rates are less than or equal to the standard leakage flow rate, then establishing a decrease sample for the corresponding fluid detection area, if there is one or more leakage flow rates greater than the standard leakage flow rate, and there is one or more leakage flow rates less than or equal to the standard leakage flow rate, then establishing a negative sample for the corresponding fluid detection area.

[0081] It is understandable that the standard leakage flow rate is the industry standard flow rate. According to the comparison results, a corresponding sample is established for each fluid detection area. The increasing sample indicates that the leakage flow rate of the fluid detection area is greater than the standard flow rate, indicating that the leakage is at a relatively high level. The decreasing sample indicates that the leakage flow rate of the fluid detection area is less than or equal to the standard flow rate, indicating that the leakage is at a relatively low level. The negative sample indicates that the leakage is at a relatively low level and may not be expanding or is decreasing. By real-time statistics and comparison of leakage flow rates, the scientific nature of identifying the changing trend of natural gas leakage is improved, the subjective bias of human judgment is avoided, and the accuracy of accurately identifying the current leakage environment is improved.

[0082] In some embodiments of the present application, when determining the current leakage environment according to the increase number, the decrease number and the negative number, it includes: calculating the leakage environment value according to the increase number, the decrease number and the negative number, and the leakage environment value is obtained by the following formula:

[0083]

[0084]

[0085] in, represents the total number of samples, To increase the number, To reduce the number, is a negative quantity, , , is the weight coefficient and , is the leakage environment value, and the current leakage environment is determined according to the leakage environment value.

[0086] It can be understood that the first preset leakage environment, the second preset leakage environment and the third preset leakage environment are preset. ≤0.8, the first preset leakage environment is used as the current leakage environment. When ≤1, the second preset leakage environment is used as the current leakage environment. When 1<n, the third preset leakage environment is used as the current leakage environment. The leakage environment value can be accurately determined according to the increase number, decrease number and negative number, and according to the corresponding preset leakage environment, the current leakage environment can be accurately judged, laying the foundation for subsequent corrections and adjustments.

[0087] In some embodiments of the present application, when comparing the leakage flow rate with the historical leakage flow rate and judging whether to correct the leakage flow rate based on the comparison result, it includes: when there is a leakage flow rate that is less than the minimum value of the historical leakage flow rate, judging to correct the leakage flow rate that is less than the historical leakage flow rate; when there is no leakage flow rate that is less than the minimum value of the historical leakage flow rate, judging not to correct the leakage flow rate, and taking the average of all the leakage flow rates to determine the final leakage flow rate.

[0088] It is understandable that the historical leakage flow rate provides a reference benchmark, representing a reasonable range of all leakage flow rates that occur. By comparing with the historical leakage flow rate, abnormal situations can be effectively identified. For example, the leakage flow rate is abnormally high or low. This situation may be caused by human detection operation errors and fluctuations in environmental conditions. The historical leakage flow rate is the result of long-term accumulation in actual use, and can reflect the actual performance of the leakage flow rate. Comparison with the historical leakage flow rate avoids the subsequent accumulation of errors and causes errors in quantitative detection results. When there is a leakage flow rate that is less than the minimum value of the historical leakage flow rate, it means that there is a large deviation in the measured leakage flow rate, which needs to be corrected to the correct flow rate range. Taking the average of all leakage flow rates avoids accidental errors in the obtained leakage flow rate and improves the accuracy of quantitative detection.

[0089] In some embodiments of the present application, when it is determined that the leakage flow rate should be corrected, the historical leakage flow rate is determined by a clustering algorithm, including: counting all historical leakage flow rates and establishing a historical flow rate set, setting the cluster number k=2, randomly selecting two historical leakage flow rates in the historical flow rate set as the first cluster center and the second cluster center, calculating the distance to the first cluster center and the second cluster center for each historical leakage flow rate in the historical flow rate set, and assigning them to the cluster closest to the distance. When the distances are the same, they are assigned to the cluster where the first cluster center is located. According to the assignment result, the cluster where the first cluster center is located is determined as the first flow rate cluster, and the cluster where the second cluster center is located is determined as the second flow rate cluster. The first flow rate cluster and the second flow rate cluster are determined as the historical leakage flow rate data set.

[0090] It can be understood that the clustering algorithm can automatically divide the historical leakage flow rate and accurately obtain the data distribution in the historical leakage flow rate, so as to accurately judge the difference between the leakage flow rate and the historical leakage flow rate. The clustering algorithm simplifies the historical leakage flow rate and processes the data of each cluster separately, avoiding the errors caused by relying solely on judgment.

[0091] In some embodiments of the present application, when determining the leakage correction factor according to the similarity results between the leakage flow rate and the historical leakage flow rate data set, it includes: obtaining the leakage flow rate that is less than the minimum value of the historical leakage flow rate and recording it as the flow rate to be corrected, and calculating the flow rate similarity according to the mean of the first flow rate cluster, the mean of the second flow rate cluster and the flow rate to be corrected, and the flow rate similarity is obtained by the following formula:

[0092] ;

[0093] in represents the flow velocity similarity, Indicates the flow rate to be corrected. represents the mean of the first flow rate cluster, represents the mean of the second flow rate cluster, and is the weight coefficient and , determine the leakage correction factor based on the flow velocity similarity.

[0094] It is understandable that when When it is greater than or equal to 0.6, the first preset leakage correction factor is used as the correction factor of the flow rate to be corrected. When it is less than 0.6, the second preset leakage correction factor is used as the correction factor of the flow rate to be corrected. The first preset leakage correction factor is greater than the second preset leakage correction factor. The flow rate similarity shows the similarity with the historical leakage flow rate. The different historical leakage flow rates are combined, and the corresponding preset leakage correction factor is selected according to the flow rate similarity, thereby realizing dynamic optimization of the flow rate to be corrected and ensuring the accuracy of quantitative detection.

[0095] In some embodiments of the present application, when the leakage flow rate is corrected according to the leakage correction factor to obtain the final leakage flow rate, it includes: when there is a flow rate to be corrected among all the leakage flow rates, the corrected flow rate is obtained according to the product of the correction factor and the flow rate to be corrected, a verification set is established for all the corrected flow rates, and a data set is established for the leakage flow rates that have not been corrected, a normal distribution model is fitted according to the mean and standard deviation of the data set, when the probability of the existence of a corrected flow rate in the verification set in the normal distribution model is less than 0.3, the corrected flow rate is determined to be an erroneous correction and is removed, when the probability of all the corrected flow rates in the verification set in the normal distribution model is greater than or equal to 0.3, the verification set is determined to be correct, all the corrected flow rates in the correct verification set and all the uncorrected leakage flow rates in the data set are summed and averaged, and the result based on the mean is determined as the final leakage flow rate.

[0096] It is understandable that by introducing a correction factor to correct the flow rate to be corrected, the accuracy of the final leakage flow rate can be effectively improved. By verifying the corrected flow rate based on the normal distribution model, the probability of the normal distribution model can be appropriately adjusted to ensure that the corrected flow rate conforms to the normal leakage flow rate distribution law, ensuring the rationality and accuracy of the corrected flow rate, thereby improving the accuracy of the verification set and reducing the interference of abnormal conditions on the final leakage flow rate. All the corrected flow rates in the correct verification set are added to all the uncorrected leakage flow rates and the average is taken, which improves the comprehensiveness of quantitative detection and prevents the loss of a certain data from interfering with the accuracy of obtaining the final leakage flow rate.

[0097] In some embodiments of the present application, when determining the leakage type based on the final leakage flow rate, it includes: comparing the final leakage flow rate with a first preset leakage flow rate and a second preset leakage flow rate respectively, the first preset leakage flow rate is greater than the second preset leakage flow rate, and determining the leakage type of the leakage point according to the comparison result; when the final leakage flow rate is greater than or equal to the first preset leakage flow rate, the leakage type of the leakage point is determined to be a severe leakage; when the final leakage flow rate is less than the first preset leakage flow rate and greater than or equal to the second preset leakage, the leakage type of the leakage point is determined to be a moderate leakage; when the final leakage flow rate is less than the second preset leakage flow rate, the leakage type of the leakage point is determined to be a slight leakage.

[0098] It can be understood that by comparing the final leakage flow rate with the two preset leakage flow rates, the leakage type of the leakage point can be determined, which reduces the deviation of human judgment and clearly shows the natural gas leakage situation, improves the efficiency of quantitative leakage detection, and provides an accurate scientific basis for emergency treatment of natural gas leaks.

[0099] In some embodiments of the present application, when determining the leakage interval based on the leakage type and the final leakage flow rate, it includes: obtaining the historical leakage flow rate of severe leakage and establishing a first interval data set, obtaining the mean in the first interval data set to determine the first-level leakage interval; obtaining the historical leakage flow rate of moderate leakage and establishing a second interval data set, obtaining the mean in the second interval data set to determine the second-level leakage interval; obtaining the historical leakage flow rate of mild leakage and establishing a third interval data set, obtaining the mean in the third interval data set to determine the third-level leakage interval, the leakage interval includes a first-level leakage interval, a second-level leakage interval and a third-level leakage interval, and the result of quantitative detection is determined by substituting the final leakage flow rate into the leakage interval.

[0100] It can be understood that the minimum value of the first-level leakage interval is the mean of the first interval data set, and the maximum value is four times its mean, the minimum value of the second-level leakage interval is the mean of the second interval data set, and the maximum value is the mean of the first interval data set, the minimum value of the third-level leakage interval is the mean of the third interval data set, and the maximum value is the mean of the second interval data set. Substituting the final leakage flow rate into the leakage interval, it can be clearly determined that the leakage level is several, which provides an accurate and scientific basis for the quantitative detection of natural gas leaks. The standardized processing method of quantitative detection improves the degree of automation of natural gas leak detection, reduces manual intervention and subjective judgment, and quantitative detection provides scientific, precise and efficient support for natural gas leak monitoring and emergency response.

[0101] In summary, the beneficial effects of the present invention are: by collecting vibration signals through acoustic wave sensors and combining them with voiceprint array technology, the position of the leakage point can be accurately located, avoiding the inefficiency and errors of traditional methods that rely on manual detection or other sensors for large-scale searches. By detecting the leakage flow rate in real time and classifying and counting the leakage flow rate samples, the current environment can be accurately judged. By comparing with the historical leakage flow rate and correcting the leakage flow rate in combination with the clustering algorithm, the adaptive ability of the quantitative detection process is improved, and it can be automatically adjusted according to different environmental changes, thereby improving the accuracy and reliability of quantitative detection. According to the correction of the final leakage flow rate and the determination of the leakage type, the natural gas leak can be finely classified, which is helpful for effective leakage prevention and control and emergency response, ensuring that appropriate measures can be taken when a leak occurs to reduce the damage caused by the natural gas leak.

[0102] In another preferred embodiment based on the above embodiment, refer to Figure 2 As shown, this embodiment provides a natural gas leakage quantitative detection system based on voiceprint, which is used to apply the above-mentioned natural gas leakage quantitative detection method based on voiceprint, including:

[0103] A collection unit, used to collect the vibration signal of the acoustic wave sensor and determine the location of the leakage point according to the soundprint array, divide the leakage point into multiple fluid detection areas based on the soundprint array, obtain the fluid leakage flow rate of each fluid detection area, and pre-process the fluid leakage flow rate to obtain the leakage flow rate;

[0104] A leakage environment unit, used for constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample;

[0105] Count the number of increasing samples and record it as the increasing number, count the number of decreasing samples and record it as the decreasing number, count the number of negative samples and record it as the negative number, and determine the current leakage environment according to the increasing number, decreasing number and negative number;

[0106] A judgment unit, used for obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and judging whether to correct the leakage flow rate according to the comparison result;

[0107] The correction unit, when determining to correct the leakage flow rate, uses a clustering algorithm to determine a historical leakage flow rate data set based on the historical leakage flow rate, determines a leakage correction factor based on similarity between the leakage flow rate and the historical leakage flow rate data set, and corrects the leakage flow rate based on the leakage correction factor to obtain a final leakage flow rate;

[0108] The type interval determination unit is used to determine the leakage type according to the final leakage flow rate, and determine the leakage interval according to the leakage type and the final leakage flow rate.

[0109] It can be understood that through the collaborative work of multiple units, efficient and accurate leakage detection and correction of pipeline natural gas leaks are achieved. The acquisition unit uses acoustic wave sensors and voiceprint arrays to locate the leakage point, and obtains the leakage flow rate through preprocessing to ensure the accuracy of data collection. The leakage environment unit generates different types of samples according to the leakage flow rate, and analyzes the current leakage environment through statistical quantity to provide a basis for subsequent judgment. The judgment unit judges whether the leakage flow rate needs to be corrected by comparing with the historical leakage flow rate, which further enhances the reliability of quantitative detection. The correction unit uses a clustering algorithm to converge the historical leakage flow rate and adjust the leakage flow rate to ensure the accuracy of the final leakage flow rate. The type interval determination unit divides the leakage type and interval according to the final leakage flow rate, which improves the accuracy and scientificity of quantitative detection of natural gas leaks.

[0110] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0111] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A natural gas leakage quantitative detection method based on voiceprint, characterized in that: include: Collecting the vibration signal of the acoustic sensor and determining the location of the leakage point according to the soundprint array, dividing the leakage point into a plurality of fluid detection areas based on the soundprint array, obtaining the fluid leakage flow rate of each fluid detection area, and pre-processing the fluid leakage flow rate to obtain the leakage flow rate; Constructing a leakage flow rate sample for the corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample; Counting the number of the increasing samples and recording it as the increasing number, counting the number of the decreasing samples and recording it as the decreasing number, counting the number of the negative samples and recording it as the negative number, and determining the current leakage environment according to the increasing number, the decreasing number and the negative number; Obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and determining whether to correct the leakage flow rate according to the comparison result; When it is determined that the leakage flow rate is to be corrected, a historical leakage flow rate data set is determined by using a clustering algorithm for the historical leakage flow rate, a leakage correction factor is determined according to similarity between the leakage flow rate and the historical leakage flow rate data set, and the leakage flow rate is corrected according to the leakage correction factor to obtain a final leakage flow rate; The leakage type is determined according to the final leakage flow rate, and the leakage interval is determined according to the leakage type and the final leakage flow rate.

2. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 1 is characterized in that: When constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, it includes: Counting all the leakage flow rates and determining a standard leakage flow rate corresponding to each fluid detection area; Comparing the leakage flow rate with the standard leakage flow rate, if all the leakage flow rates are greater than the standard leakage flow rate, establishing the growth sample for the corresponding fluid detection area; If all leakage flow rates are less than or equal to the standard leakage flow rate, then the reduced sample is established for the corresponding fluid detection area; If there is one or more leakage flow rates greater than the standard leakage flow rate, and there is one or more leakage flow rates less than or equal to the standard leakage flow rate, then the negative sample is established for the corresponding fluid detection area.

3. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 1 is characterized in that: When determining the current leakage environment according to the increase number, the decrease number and the negative number, it includes: The leakage environment value is calculated according to the increase quantity, the decrease quantity and the negative quantity, and the leakage environment value is obtained by the following formula: ； ； in, represents the total number of samples, To increase the number, To reduce the number, is a negative quantity, , , is the weight coefficient and , is a leakage environment value, and the current leakage environment is determined according to the leakage environment value.

4. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 1, characterized in that: When comparing the leakage flow rate with the historical leakage flow rate and determining whether to correct the leakage flow rate according to the comparison result, the method includes: When there is a leakage flow rate that is less than the minimum value of the historical leakage flow rate, it is determined to correct the leakage flow rate that is less than the historical leakage flow rate; When there is no leakage flow rate less than the minimum value of the historical leakage flow rate, it is determined that the leakage flow rate is not to be corrected, and the average of all leakage flow rates is determined as the final leakage flow rate.

5. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 4 is characterized in that: When it is determined that the leakage flow rate is to be corrected, the historical leakage flow rate is determined by using a clustering algorithm to determine a historical leakage flow rate data set, including: Count all historical leakage flow rates and establish a historical flow rate collection; Set the cluster number k=2, randomly select two historical leakage flow rates in the historical flow rate set as the first cluster center and the second cluster center, calculate the distance from each historical leakage flow rate in the historical flow rate set to the first cluster center and the second cluster center, and assign it to the cluster with the closest distance. If the distances are the same, assign it to the cluster where the first cluster center is located. According to the allocation result, the cluster where the center of the first cluster is located is determined as a first flow rate cluster, and the cluster where the center of the second cluster is located is determined as a second flow rate cluster; The first flow rate cluster and the second flow rate cluster are determined as the historical leakage flow rate dataset.

6. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 5 is characterized in that: When determining the leakage correction factor according to the similarity results between the leakage flow rate and the historical leakage flow rate data set, it includes: Obtaining a leakage flow rate that is less than the minimum value of the historical leakage flow rate and recording it as the flow rate to be corrected; Calculating the flow velocity similarity according to the mean of the first flow velocity cluster, the mean of the second flow velocity cluster and the flow velocity to be corrected; The flow rate similarity is obtained by the following formula: ； in represents the flow velocity similarity, Indicates the flow rate to be corrected. represents the mean of the first flow rate cluster, represents the mean of the second flow rate cluster, and is the weight coefficient and , determine the leakage correction factor based on the flow velocity similarity.

7. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 6 is characterized in that: When the leakage flow rate is corrected according to the leakage correction factor to obtain a final leakage flow rate, the method includes: When the flow rate to be corrected exists in all the leakage flow rates, a corrected flow rate is obtained according to the product value of the correction factor and the flow rate to be corrected; A validation set is established for all corrected flow rates, a data set is established for the uncorrected leakage flow rate, and a normal distribution model is fitted according to the mean and standard deviation of the data set; When the probability that the corrected flow rate exists in the normal distribution model in the validation set is less than 0.3, the corrected flow rate is determined to be an erroneous correction and is removed; When the probability of all the corrected flow velocities in the verification set in the normal distribution model is greater than or equal to 0.3, the verification set is determined to be correct; All corrected flow rates in the validation set that are determined to be correct and all uncorrected leakage flow rates in the data set are summed and averaged, and the result based on the average is determined as the final leakage flow rate.

8. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 1, characterized in that: When determining the leakage type according to the final leakage flow rate, it includes: Comparing the final leakage flow rate with a first preset leakage flow rate and a second preset leakage flow rate respectively, the first preset leakage flow rate being greater than the second preset leakage flow rate, and determining the leakage type of the leakage point according to the comparison result; When the final leakage flow rate is greater than or equal to the first preset leakage flow rate, determining that the leakage type of the leakage point is a serious leakage; When the final leakage flow rate is less than the first preset leakage flow rate and greater than or equal to the second preset leakage flow rate, determining that the leakage type of the leakage point is a moderate leakage; When the final leakage flow rate is less than the second preset leakage flow rate, it is determined that the leakage type of the leakage point is a slight leakage.

9. The method for quantitative detection of natural gas leakage based on voiceprint according to claim 8, characterized in that: When determining the leakage interval according to the leakage type and the final leakage flow rate, it includes: Obtaining the historical leakage flow rate of the serious leakage and establishing a first interval data set, and obtaining the mean value in the first interval data set to determine a primary leakage interval; Obtaining the historical leakage flow rate of the moderate leakage and establishing a second interval data set, and obtaining the mean value in the second interval data set to determine the secondary leakage interval; Obtain the historical leakage flow rate of the slight leakage and establish a third interval data set, and obtain the mean value in the third interval data set to determine the third-level leakage interval; The leakage interval includes the primary leakage interval, the secondary leakage interval and the tertiary leakage interval; The final leakage flow rate is substituted into the leakage interval to determine the result of the quantitative detection.

10. A natural gas leakage quantitative detection system based on voiceprint, used for applying the natural gas leakage quantitative detection method based on voiceprint according to any one of claims 1 to 9, characterized in that: include: A collection unit, used to collect the vibration signal of the acoustic wave sensor and determine the location of the leakage point according to the soundprint array, divide the leakage point into multiple fluid detection areas based on the soundprint array, obtain the fluid leakage flow rate of each fluid detection area, and pre-process the fluid leakage flow rate to obtain the leakage flow rate; A leakage environment unit, used for constructing a leakage flow rate sample for a corresponding fluid detection area according to the leakage flow rate, wherein the leakage flow rate sample includes an increase sample, a decrease sample and a negative sample; Counting the number of the increasing samples and recording it as the increasing number, counting the number of the decreasing samples and recording it as the decreasing number, counting the number of the negative samples and recording it as the negative number, and determining the current leakage environment according to the increasing number, the decreasing number and the negative number; A judgment unit, used for obtaining a historical leakage flow rate of the same leakage environment as the current leakage environment, comparing the leakage flow rate with the historical leakage flow rate, and judging whether to correct the leakage flow rate according to the comparison result; a correction unit, when determining to correct the leakage flow rate, using a clustering algorithm to determine a historical leakage flow rate data set for the historical leakage flow rate, determining a leakage correction factor according to a similarity result between the leakage flow rate and the historical leakage flow rate data set, and correcting the leakage flow rate according to the leakage correction factor to obtain a final leakage flow rate; The type interval determination unit is used to determine the leakage type according to the final leakage flow rate, and determine the leakage interval according to the leakage type and the final leakage flow rate.

Images