Oil and gas pipeline third-party damage monitoring method based on closed amplitude of microwave remote sensing

Abstract

The invention discloses an oil and gas pipeline third-party damage monitoring method based on the closed amplitude of microwave remote sensing, and relates to the field of microwave remote sensing. The closed amplitude oil and gas pipeline third-party damage monitoring method based on microwave remote sensing comprises the following steps: step 1, acquiring a multi-temporal microwave remote sensing SAR data interferogram and amplitude information; 2, calculating closing amplitude information; 3, calculating a covariance matrix of the closed amplitude, and constructing a likelihood function of the covariance matrix; and step 4, adopting regularization likelihood to estimate amplitude changes in two time periods. According to the closed amplitude third-party damage monitoring method based on microwave remote sensing, closed amplitude analysis and L1 regularization likelihood estimation are combined, a closed polygon interference path is constructed, earth surface deformation features are comprehensively extracted from multi-dimensional information, background noise interference is effectively removed through closed amplitude calculation, and the detection accuracy of the closed amplitude third-party damage monitoring method based on microwave remote sensing is improved. And the detection capability on the weak change of the target under the complex terrain condition is obviously improved.

Description

Technical Field

[0001] The invention relates to the technical field of microwave remote sensing, in particular to a method for monitoring third-party damage to oil and gas pipelines based on closure amplitude of microwave remote sensing. Background Art

[0002] As a key infrastructure for energy transportation, the safety of oil and gas pipelines is directly related to national economic development and social stability. However, third-party sabotage (such as illegal construction, mechanical excavation, etc.) is one of the main threats to the safe operation of oil and gas pipelines, which often leads to serious pipeline leakage, environmental pollution and even safety accidents. Traditional monitoring methods usually rely on manual inspections, ground sensors or drone monitoring, but these methods have disadvantages such as limited coverage and susceptibility to environmental influences. With the development of microwave remote sensing technology, it has the advantages of all-weather, all-day monitoring capabilities and wide-area coverage, providing new technical means for third-party sabotage monitoring of oil and gas pipelines.

[0003] In strong clutter and complex interference environments, traditional microwave remote sensing monitoring methods usually rely on a single indicator of amplitude or phase for anomaly detection. However, as a major monitoring indicator, amplitude value is easily lost in complex scenarios due to interference from environmental factors. For example, factors such as terrain undulations, vegetation obstruction, and complex reflections from buildings can cause the amplitude of the target signal to be covered or masked by strong clutter, making it difficult to effectively distinguish between abnormal targets and background signals. Especially under low signal-to-noise ratio conditions, amplitude fluctuations may be completely submerged by background noise, resulting in a high false detection rate and missed detection rate in monitoring results.

[0004] A model for comparing multi-temporal images using a standard logarithmic ratio operator is proposed in existing literature. This method assumes that the surface scattering characteristics conform to the generalized Gaussian distribution, which is used to more accurately describe the conditional probability density of the changed and unchanged areas in the logarithmic ratio image. However, it still has certain limitations for the changes of weak targets in complex terrain or strong clutter environments.

[0005] At the same time, there are existing literatures that use grayscale and texture change information to perform change detection by constructing a double difference image based on intensity and texture information. Although this method effectively combines grayscale and texture information and improves the overall accuracy of change detection, its model that assumes that texture differences can fully reflect the changed area is still limited in complex clutter environments. The above methods are mainly aimed at change detection in specific scenarios. For the diverse third-party sabotage behaviors in the complex environment around oil and gas pipelines, especially the abnormal changes under strong clutter and low signal-to-noise ratio conditions, it is impossible to establish an accurate and universal detection model. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention provides microwave remote sensing, which solves the problem in the prior art that it is impossible to establish an accurate and universal detection model for various third-party destructive behaviors in the complex environment around oil and gas pipelines, especially abnormal changes under strong clutter and low signal-to-noise ratio conditions.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a third-party damage monitoring method for oil and gas pipelines based on closure amplitude of microwave remote sensing, the monitoring method comprising the following steps:

[0008] Step 1: Obtain multi-temporal microwave remote sensing SAR data interference pattern and amplitude information;

[0009] Step 2: Calculate the closure amplitude information;

[0010] Step 3: Calculate the covariance matrix of the closed amplitude and construct its likelihood function;

[0011] Step 4: Use regularized likelihood to estimate the amplitude change between two time periods.

[0012] Preferably, the formula for calculating the closing amplitude is as follows:

[0013]

[0014] Among them, C amp1234 is the closure amplitude of the SAR data of the four observations, amp ij The amplitude value of the interference data expressing the i-th observation value and the j-th observation value.

[0015] Preferably, the amp ii is a value that varies from 0 to 1, C amp1234 Then there will be an infinite situation, and the further quantitative value is the logarithmic closure amplitude, which is expressed as follows:

[0016]

[0017] The independent closing amplitude and logarithmic closing amplitude are expressed as:

[0018]

[0019] Preferably, the logarithmic closure amplitude is calculated as a linear operator, and a sparse matrix B is set, consisting of 0 and ±1; then the linear operator is described as:

[0020]

[0021] Among them, the matrix B is the design sparse matrix with dimension

[0022] The variable x represents a vector containing rows, where each row corresponds to the logarithm of the interference amplitude, and, represents the logarithmic closed amplitude matrix, containing M indep OK.

[0023] Preferably, the covariance of the logarithmic closed sum amplitude matrix is Therefore, the covariance matrix of y has the following characteristics:

[0024]

[0025] in, Indicates the corresponding variable log(amp ij ) is the variance of

[0026]

[0027] in, is the covariance matrix of the logarithmic closed and amplitude matrix, whose dimension is M indep ×M indep . This covariance matrix must contain a set of non-redundant log-closed sum amplitudes; otherwise, it will result in a rank-deficient and non-invertible matrix.

[0028] Preferably, according to the central limit theorem, when multiple random variables are added, the distribution of their sum tends to be normally distributed, even if the random variables themselves are not necessarily normally distributed. Therefore, when multiple independent terms are linearly combined, the result is usually approximately normally distributed; under this assumption, the logarithmic closed likelihood equation is expressed as:

[0029]

[0030] Among them, The multivariate Gaussian distribution of Represents an estimate of change.

[0031] Preferably, the maximum likelihood method is a common method for parameter estimation. The main purpose of regularization is to reduce the risk of overfitting of the model, thereby enhancing its generalization ability. The generalized likelihood function is defined using L1 regularization, and the formula is as follows:

[0032]

[0033] Where ln L(W) represents the multivariate normal log-likelihood function with L1 regularization;

[0034] W represents a logarithmic amplitude deviation matrix that can capture the error terms related to the closure phase and logarithmic closure amplitude;

[0035] The matrix R represents the covariance matrix of the closed eigenvectors

[0036] The parameter d represents the dimension of the independent closed feature data;

[0037] 1n represents the natural logarithm function, and α is the regularization parameter.

[0038] An electronic device includes a memory and a processor, wherein a computer program is stored in the memory.

[0039] A computer-readable storage medium having a computer program stored thereon.

[0040] The present invention discloses a third-party damage monitoring method for oil and gas pipelines based on microwave remote sensing closure amplitude, which has the following beneficial effects:

[0041] 1. This third-party damage monitoring method for oil and gas pipelines based on microwave remote sensing and closure amplitude combines closure amplitude analysis with L1 regularized likelihood estimation to construct a closed polygon interference path, comprehensively extract surface deformation features from multi-dimensional information, and effectively remove background noise interference through closure amplitude calculation, significantly improving the ability to detect weak changes in targets under complex terrain conditions. Then, L1 regularized likelihood estimation is used to optimize the extraction of amplitude changes in time series, further improving the accuracy and robustness of change detection, and solving the problem that traditional single indicators cannot accurately distinguish between target changes and background interference.

[0042] 2. The closed amplitude monitoring method for oil and gas pipelines based on microwave remote sensing has an algorithm model that is suitable for strong clutter and complex terrain environments, and has the characteristics of high precision and high adaptability. At the same time, through the spatiotemporal analysis of the data, the change information can be projected from the radar coordinate system to the geographic coordinate system to achieve accurate positioning of the destructive behavior. It is widely used in the field of third-party damage monitoring around oil and gas pipelines, providing scientific support and technical guarantee for pipeline safety management. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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.

[0044] Figure 1 The present invention provides a flowchart of the method.

[0045] Figure 2 It is a scene graph implemented by the present invention.

[0046] Figure 3 is the logarithmic closure amplitude in the implementation provided by the present invention.

[0047] Figure 4 is the L1 regularized likelihood estimate in the implementation provided by the present invention.

[0048] Figure 5 It is a result diagram of the logarithmic amplitude change value after regularized likelihood estimation in the implementation provided by the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The embodiment of the present application solves the problem in the prior art that it is impossible to establish an accurate and universal detection model for various third-party sabotage behaviors in complex environments around oil and gas pipelines, especially abnormal changes under strong clutter and low signal-to-noise ratio conditions, by providing a third-party sabotage monitoring method for oil and gas pipelines based on closure amplitude of microwave remote sensing. By combining closure amplitude analysis with L1 regularized likelihood estimation, a closed polygon interference path is constructed, and surface deformation features are comprehensively extracted from multi-dimensional information. The background noise interference is effectively removed by closure amplitude calculation, which significantly improves the detection capability of slight changes in targets under complex terrain conditions. Next, the L1 regularized likelihood estimation is used to optimize the extraction of amplitude changes in time series, further improving the accuracy and robustness of change detection, and solving the problem that the traditional single indicator cannot accurately distinguish between target changes and background interference.

[0051] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0052] The embodiment of the present invention discloses a method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude. Figure 1-5 As shown, the monitoring method comprises the following steps:

[0053] Step 1: Obtain multi-temporal microwave remote sensing SAR data interference pattern and amplitude information;

[0054] Step 2: Calculate the closure amplitude information;

[0055] Step 3: Calculate the covariance matrix of the closed amplitude and construct its likelihood function;

[0056] Step 4: Use regularized likelihood to estimate the amplitude change between two time periods.

[0057] Specifically disclosed, the formula for calculating the closing amplitude is as follows:

[0058]

[0059] Among them, C amp1234 is the closure amplitude of the SAR data of the four observations, amp ij The amplitude value of the interference data expressing the i-th observation value and the j-th observation value.

[0060] Furthermore, the amp ij is a value that varies from 0 to 1, C amp1234 Then there will be an infinite situation, and the further quantitative value is the logarithmic closure amplitude, which is expressed as follows:

[0061]

[0062] The independent closing amplitude and logarithmic closing amplitude are expressed as:

[0063]

[0064] Specifically disclosed, the logarithmic closure amplitude is calculated as a linear operator, and a sparse matrix B is set, consisting of 0 and ±1; then the linear operator is described as:

[0065]

[0066] Among them, the matrix B is the design sparse matrix with dimension

[0067] The variable x represents a vector containing rows, where each row corresponds to the logarithm of the interference amplitude, and, represents the logarithmic closed amplitude matrix, containing M indep OK.

[0068] Furthermore, the covariance of the logarithmic closed sum amplitude matrix is Therefore, the covariance matrix of y has the following characteristics:

[0069]

[0070] in, Indicates the corresponding variable log(amp ij ) is the variance of

[0071]

[0072] in, is the covariance matrix of the logarithmic closed and amplitude matrix, whose dimension is M indep ×M indep. This covariance matrix must contain a set of non-redundant log-closed sum amplitudes; otherwise, it will result in a rank-deficient and non-invertible matrix.

[0073] Furthermore, according to the central limit theorem, when multiple random variables are added together, the distribution of the sum tends to be normally distributed, even if the random variables themselves are not necessarily normally distributed. Therefore, when multiple independent terms are linearly combined, the result is usually approximately normally distributed; under this assumption, the logarithmic closed likelihood equation is expressed as:

[0074]

[0075] Among them, The multivariate Gaussian distribution of Represents an estimate of change.

[0076] Furthermore, the maximum likelihood method is a common method for parameter estimation. The main purpose of regularization is to reduce the risk of overfitting of the model, thereby enhancing its generalization ability. The generalized likelihood function is defined using L1 regularization, and the formula is as follows:

[0077]

[0078] Where ln L(W) represents the multivariate normal log-likelihood function with L1 regularization;

[0079] W represents a logarithmic amplitude deviation matrix that can capture the error terms related to the closure phase and logarithmic closure amplitude;

[0080] The matrix R represents the covariance matrix of the closed eigenvectors

[0081] The parameter d represents the dimension of the independent closed feature data;

[0082] ln represents the natural logarithm function, and α is the regularization parameter.

[0083] An electronic device comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, a method for monitoring oil and gas pipeline third-party damage based on microwave remote sensing closure amplitude is implemented.

[0084] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for monitoring third-party damage to an oil and gas pipeline based on closure amplitude using microwave remote sensing.

[0085] By constructing the interference path of closed polygons through closed amplitude analysis, multi-dimensional information is processed comprehensively to effectively distinguish the difference between surface deformation characteristics and background interference. This solves the problem of insufficient detection of weak target changes under low signal-to-noise ratio and complex terrain conditions, thus failing to achieve the high-precision monitoring demand for third-party sabotage around oil and gas pipelines in actual applications.

[0086] The solution of the present invention is to use multi-dimensional microwave remote sensing data for interference processing, remove background noise by constructing a closed amplitude, and realize the amplitude change information in each time period through L1 regularized likelihood estimation. This method effectively solves the problem that traditional single indicators (such as amplitude and phase) are difficult to distinguish weak change characteristics under strong clutter and complex terrain conditions, thereby realizing accurate identification and effective monitoring of weak changes in the surrounding areas of oil and gas pipelines. At the same time, the introduction of closed amplitude enhances the adaptability of the algorithm to strong clutter environments, and L1 regularized likelihood estimation can further optimize the change extraction ability of time series data, and improve the accuracy and robustness of change detection. Through this method, it is possible to effectively monitor and accurately locate third-party sabotage behaviors around oil and gas pipelines, providing scientific support for pipeline safety management.

[0087] Example:

[0088] The experiment used five sets of Sentinel-1A and Sentinel-1B VV polarization microwave remote sensing data (descending orbit mode) from January 12, 2020 to February 11, 2020. In the experiment, the amplitude change characteristics of the target area were extracted through the interference path closure analysis of multi-phase data, and the accuracy and robustness of change detection were enhanced by combining L1 regularization, and finally the accurate monitoring of third-party sabotage behavior of oil and gas pipelines was achieved. The present invention is further described in detail below with regard to specific implementation methods.

[0089] Step 1: Obtain multi-temporal microwave remote sensing SAR data interference pattern and amplitude information;

[0090] First, collect multi-temporal Sentinel-1A and Sentinel-1B SAR data in VV polarization mode to ensure that the temporal and spatial resolution meets monitoring requirements. Figure 2 The study area where the implementation was demonstrated is in the suburban area.

[0091] Second, the raw SAR data is preprocessed, including registration, geocoding, interference, and filtering operations, to generate multi-temporal interferograms and their amplitude information.

[0092] Step 2: Calculate the closure amplitude information

[0093] First, at least one set of SAR data containing four observations is selected to construct an interferometric path of an independent closed polygon.

[0094] Second, calculate the closure amplitude value C of each independent path according to the closure amplitude formula amp1234 , C amp1345 , C amp1235 , C amp1345 , C amp2345 .

[0095] Third, the closure amplitude results are logarithmically transformed to obtain the logarithmic closure amplitude to enhance the ability to resist complex background noise. Figure 3 5 independent groups of logarithmic closure amplitudes to display.

[0096] Step 3: Calculate the covariance matrix of the closed amplitude and construct its likelihood function

[0097] Among them, the logarithmic closure amplitude covariance matrix is ​​constructed, and the sparse matrix B is set; the closure amplitude covariance matrix Perform feature analysis to ensure that the matrix is ​​non-redundant and reversible, and meets the linear model assumptions; construct the likelihood function based on the covariance matrix Used to evaluate the changing pattern of closure amplitude.

[0098] Step 4: Use regularized likelihood to estimate the amplitude change between two time periods

[0099] First, based on the closed amplitude covariance matrix, L1 regularization is used to optimize the likelihood function to reduce the risk of model overfitting.

[0100] The L1 norm regularization parameter is set to 0.5 here. In the estimation of the change of logarithmic closure amplitude, the present invention uses the likelihood estimation method of L1 norm regularization to model and calculate the three logarithmic closure amplitude parameters (x1, x2, x3) to obtain the estimated value and its distribution characteristics (such as Figure 4 As shown in the right probability density graph, the color gradient (from dark blue to yellow) indicates the distribution characteristics of different probability densities. The red straight line is the final estimate, and its position indicates the convergence characteristics of the estimation process. As shown in the probability density graph on the right, each logarithmic closure amplitude estimation result is displayed as a distribution center point, and the upper and lower error ranges are used. ± indicates the uncertainty range.

[0101] For example: The central estimate is 0.09, and the error range is a percentage indicating that the estimated uncertainty is [5.24%, 5.10%]. The L1 regularization method can estimate the change of the logarithmic closure amplitude under sparse data and complex background noise conditions. Although the estimated values ​​are close to zero, the error range is large, indicating the existence of slight changes in the surface in the area. This result verifies the applicability of the logarithmic closure amplitude calculation method in the present invention under complex conditions.

[0102] Second, Figure 5 Finally, the logarithmic amplitude changes of the two time periods in the time series are extracted through optimal parameter estimation to ensure the accuracy and robustness of the change extraction. The results of each path combination show that the changes in the middle part of the target area are more significant, especially in paths 15 and 12, where the distribution of red and blue areas is more concentrated, indicating that the logarithmic amplitude changes at these locations may be related to actual sabotage behavior. The green area is the area with smaller changes and where the background noise is effectively suppressed.

[0103] Implementation Effect

[0104] Through the above steps, the present invention can accurately extract weak surface deformation features under complex terrain and low signal-to-noise ratio conditions, and realize high-precision monitoring and accurate positioning of third-party sabotage behaviors around oil and gas pipelines. The method has strong noise resistance, adaptability and robustness, and provides reliable technical support and scientific basis for the safety management of oil and gas pipelines.

[0105] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A third-party damage monitoring method for oil and gas pipelines based on microwave remote sensing closure amplitude, characterized in that: The monitoring method comprises the following steps: Step 1: Obtain multi-temporal microwave remote sensing SAR data interference pattern and amplitude information; Step 2: Calculate the closure amplitude information; Step 3: Calculate the covariance matrix of the closed amplitude and construct its likelihood function; Step 4: Use regularized likelihood to estimate the amplitude change between two time periods.

2. The method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude according to claim 1 is characterized in that: The formula for calculating the closing amplitude is as follows: Among them, C amp1234 is the closure amplitude of the SAR data of the four observations, amp ij The amplitude value of the interference data expressing the i-th observation value and the j-th observation value.

3. The method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude according to claim 2 is characterized in that: The amp ij is a value that varies from 0 to 1, C amp1234 Then there will be an infinite situation, and the further quantitative value is the logarithmic closure amplitude, which is expressed as follows: The independent closing amplitude and logarithmic closing amplitude are expressed as:

4. The method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude according to claim 3 is characterized in that: The logarithmic closure amplitude is calculated as a linear operator, and a sparse matrix B is set, consisting of 0 and ±1; then the linear operator is described as: Among them, the matrix B is the design sparse matrix with dimension The variable x represents a vector containing rows, where each row corresponds to the logarithm of the interference amplitude, and, Represents a logarithmic closed amplitude matrix, containing M indep OK.

5. The method for monitoring third-party damage to oil and gas pipelines based on closure amplitude of microwave remote sensing according to claim 4 is characterized in that: The covariance of the log-closed and amplitude matrices is Therefore, the covariance matrix of y has the following characteristics: in, Indicates the corresponding variable log(amp ij ) is the variance of in, is the covariance matrix of the logarithmic closed and amplitude matrix, whose dimension is M indep ×M indep .

6. The method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude according to claim 5 is characterized in that: When multiple independent terms are linearly combined, the result is usually approximately normally distributed; its log-closed likelihood equation is expressed as: Among them, The multivariate Gaussian distribution of Represents an estimate of change.

7. The method for monitoring third-party damage to oil and gas pipelines based on microwave remote sensing closure amplitude according to claim 1 is characterized in that: The generalized likelihood function is defined using L1 regularization, and the formula is as follows: Where lnL(W) represents the multivariate normal log-likelihood function with L1 regularization; W represents a logarithmic amplitude deviation matrix that can capture the error terms related to the closure phase and logarithmic closure amplitude; The matrix R represents the covariance matrix of the closed eigenvectors The parameter d represents the dimension of the independent closed feature data; ln represents the natural logarithm function, and α is the regularization parameter.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the program.

9. A computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images