Submarine pipeline abnormal condition real-time detection method, device, equipment and medium

Abstract

The method is mainly applied to the technical field of artificial intelligence. The invention discloses a submarine pipeline abnormal condition real-time detection method, device and equipment and a medium. The method comprises the following steps: monitoring a submarine pipeline through a sensor group to obtain a sensor data set; based on the sensor data set, the mean offset of data in the submarine pipeline monitoring process is determined; updating a preset relaxation parameter according to the mean offset; based on the updated relaxation parameter and the current offset, determining the cumulative sum of the sensor data and taking the cumulative sum as the updated data change trend; and when the updated data change trend is out of the preset value range, determining that the submarine pipeline has an abnormal condition and giving an alarm. According to the method, the detection sensitivity can be effectively improved when the abnormal condition of the submarine pipeline is detected, the detection accuracy is improved, and the probability of false alarm or missing alarm is reduced.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and medium for real-time detection of abnormal conditions in submarine pipelines. Background Art

[0002] Submarine pipelines are key infrastructure for global energy transportation and are widely used in cross-ocean transmission of oil and natural gas. Submarine pipelines are buried under the seabed for a long time and face complex marine environmental challenges, including natural factors such as seawater corrosion, ocean current impact, and deep-sea pressure, which can cause fatigue, cracks, corrosion, damage, or rupture of the pipeline. In order to protect submarine pipelines, it is crucial to detect abnormal conditions of submarine pipelines in a timely manner. Traditional detection technology determines abnormal conditions of submarine pipelines by setting one or more fixed thresholds and comparing the sensor signals of the submarine pipelines with the fixed thresholds. However, the complex and changeable submarine environment can cause continuous changes in the submarine pipeline sensor signals, which can easily cause false alarms or missed alarms when using traditional detection technology for detection. Summary of the invention

[0003] The present invention provides a method, device, equipment and medium for real-time detection of abnormal conditions of submarine pipelines, which can effectively improve detection sensitivity and detection accuracy when detecting abnormal conditions of submarine pipelines, and reduce the probability of false alarms or missed alarms.

[0004] The present invention provides a method for real-time detection of abnormal conditions in a submarine pipeline, the method comprising: The sensor data set is obtained by monitoring the submarine pipeline through the sensor group; Determining a mean shift of data during monitoring of the submarine pipeline based on the sensor data set; updating a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used to adjust monitoring sensitivity when monitoring the submarine pipeline; Based on the updated relaxation parameter and the current offset, determine the cumulative sum of the sensor data as the updated data change trend, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data; When the updated data change trend is outside the preset value range, it is determined that the submarine pipeline is abnormal and an alarm is issued.

[0005] Further, when the updated data change trend is outside the preset value range, determining that the submarine pipeline is abnormal and issuing an alarm includes: If the cumulative sum is a positive cumulative sum, the positive cumulative sum is compared with a preset first alarm threshold, and when the positive cumulative sum is greater than the first alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued; If the cumulative sum is a negative cumulative sum, the negative cumulative sum is compared with a preset second alarm threshold. When the negative cumulative sum is less than the second alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued.

[0006] Furthermore, the method further comprises: When it is determined that the submarine pipeline has an abnormal condition, acquiring an abnormal sensor data set; Based on the time domain features of the abnormal sensor data set, generating a time domain feature matrix; After the time domain feature matrix is ​​input into a preset neural network, the abnormal situation is identified by the neural network and the category of the abnormal situation is output.

[0007] Further, when it is determined that the submarine pipeline is abnormal, obtaining an abnormal sensor data set includes: When it is determined that the submarine pipeline has an abnormal condition, the current time point is recorded as a first time point and new sensor signals are continuously acquired until a second time point is recorded when no abnormal condition is detected in the submarine pipeline; taking each new sensor signal acquired in the time period between the first time point and the second time point as an abnormal signal; The abnormal sensor data set is generated based on a plurality of the abnormal signals.

[0008] Furthermore, the method further comprises: Obtain a historical abnormal event data set, wherein each sample in the historical abnormal event data set is provided with a category label; Based on the time domain features of each of the samples, a sample feature matrix is ​​generated, wherein each row of the sample feature matrix represents a sensor, and each column of the sample feature matrix represents a time domain feature; Inputting the sample feature matrix and each of the category labels into a preset Bayesian convolutional neural network model to train the Bayesian convolutional neural network model; When the Bayesian convolutional neural network model is trained, the Bayesian convolutional neural network model is updated by a preset loss function until a trained Bayesian convolutional neural network model is obtained; The trained Bayesian convolutional neural network model is used as the neural network.

[0009] Furthermore, after the time domain feature matrix is ​​input into a preset neural network, the abnormal situation is identified by the neural network, and the category of the abnormal situation is output, including: Inputting the time domain feature matrix into the neural network, wherein the neural network at least includes a convolutional layer, a fully connected layer and an output layer; After the time domain features of the time domain feature matrix are convolved through the convolution layer, a plurality of time feature vectors are obtained; After the multiple time feature vectors are input to the output layer through the fully connected layer, each of the time feature vectors is mapped to a target abnormal event category in multiple abnormal event categories through the output layer to obtain a classification result of the abnormal situation and a confidence interval of the classification result.

[0010] Further, the sensor data set obtained by monitoring the submarine pipeline through the sensor group includes: Acquire the original signal detected by each sensor in the sensor group; Arranging each of the original signals to generate an original signal sequence; Based on the mean and standard deviation of the signal when no abnormal change is detected, the original signal sequence is converted into a standardized signal sequence of standard normal distribution and used as the preprocessed sensor data set.

[0011] The present invention also provides a real-time detection device for abnormal conditions of a submarine pipeline, the device comprising: A collection module, used for monitoring the submarine pipeline through a sensor group to obtain a sensor data set; A first operation module is used to determine a mean offset of data in a process of monitoring the submarine pipeline based on the sensor data set; An updating module, used for updating a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used for adjusting monitoring sensitivity when monitoring the submarine pipeline; A second operation module is used to determine the cumulative sum of the sensor data as the updated data change trend based on the updated relaxation parameter and the current offset, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data; The output module is used to determine that an abnormal situation occurs in the submarine pipeline and issue an alarm when the change trend of the updated data is outside the preset value range.

[0012] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the real-time detection method for abnormal conditions of a submarine pipeline as described in any one of the above items is implemented.

[0013] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the real-time detection method for abnormal conditions of a submarine pipeline as described in any one of the above items is implemented.

[0014] The present invention has at least the following beneficial effects: In the technical solution of the present application, first, by analyzing the sensor data set, the mean offset of the data is calculated, which is helpful to determine whether there is an abnormality in the submarine pipeline. Then, according to the calculated mean offset, the preset relaxation parameter is dynamically updated. The relaxation parameter is the key to adjusting the monitoring sensitivity. By dynamically adjusting the relaxation parameter, the detection sensitivity can be effectively improved, and the probability of false alarms and missed alarms can be reduced, thereby improving the accuracy of detection. Then, based on the updated relaxation parameter and the current offset, the cumulative sum of the sensor data is calculated, which reflects the trend of data changes. Finally, by monitoring the trend of data changes after this update. If this trend exceeds the preset normal value range, it is determined that an abnormal situation has occurred in the submarine pipeline, and an alarm is immediately issued. Through this intelligent monitoring and early warning mechanism, it is helpful to timely discover and deal with abnormal situations of submarine pipelines and ensure the safe operation of submarine pipelines. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0016] Figure 1 is a flowchart of the steps of a method for real-time detection of abnormal conditions in a submarine pipeline; Figure 2 is a flowchart of step S101 in the method for real-time detection of abnormal conditions of submarine pipelines; Figure 3 is another step flow chart of the method for real-time detection of abnormal conditions in submarine pipelines; Figure 4 is a flowchart of step S301 in the method for real-time detection of abnormal conditions of submarine pipelines; Figure 5 is a flowchart of step S303 in the method for real-time detection of abnormal conditions of submarine pipelines; Figure 6 is another step flow chart of the method for real-time detection of abnormal conditions in submarine pipelines; Figure 7 It is a structural schematic diagram of a real-time detection device for abnormal conditions of submarine pipelines; Figure 8 It is a structural diagram of an electronic device. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] Researchers have found that in the relevant technical field, the solutions for detecting abnormal conditions of submarine pipelines are all based on fixed threshold monitoring technology, which detects abnormal changes by setting one or more fixed thresholds. The traditional CUSUM (Cumulative Sum Control Chart) method calculates the cumulative sum of sensor signals and compares it with the set threshold to detect whether the signal deviates from the normal range, thereby identifying abnormal conditions. However, due to the limitation of the threshold, the sensitivity of this method is relatively fixed and is only applicable to detecting simple signal changes, not to detecting complex and changeable submarine environment signal changes. The swing of the submarine environment, such as ocean currents, can cause continuous changes in the submarine pipeline sensor signal. Using the traditional CUSUM for abnormal detection is prone to false alarms or missed alarms. In addition, after detecting an abnormal event of the submarine pipeline, the weights of the traditional neural network used to identify the abnormal event category are fixed values ​​obtained through training. The model calculates the forward propagation through these fixed weights and outputs a single classification result, which cannot provide the uncertainty of the result. In response to the above problems, the technical solution of this application provides the following embodiments to solve the problem.

[0019] Please refer to Figure 1 , Figure 1 It is a flow chart of the steps of a method for real-time detection of abnormal conditions in submarine pipelines.

[0020] This embodiment provides a method for real-time detection of abnormal conditions in a submarine pipeline, including: S101. Monitor a submarine pipeline through a sensor group to obtain a sensor data set.

[0021] S102: Determine a mean offset of data in a process of monitoring a submarine pipeline based on a sensor data set.

[0022] S103. Update a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used to adjust monitoring sensitivity when monitoring a submarine pipeline.

[0023] S104 . Based on the updated relaxation parameter and the current offset, determine the cumulative sum of the sensor data and use it as the updated data change trend, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data.

[0024] S105. When the trend of the updated data change is outside the preset value range, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued.

[0025] In step S101 of some embodiments, the sensor data set is obtained by collecting sensor monitoring data of temperature, strain and vibration in the submarine pipeline. It should be noted that the sensors in the sensor group are distributed at different positions of the submarine pipeline, so the sensor signals at different positions of the submarine pipeline can be collected through these sensors.

[0026] In step S102 of some embodiments, EWMA-based mean offset estimation and EWMA-based adaptive signal offset estimation are adopted. By dynamically smoothing the current signal and the historical signal, the threshold, sensitivity and relaxation value in the detection process are adaptively adjusted, which can better adapt to the continuous changes in the signal caused by factors such as ocean current swings in the marine environment.

[0027] In some embodiments, the method for realizing real-time detection of abnormal conditions of submarine pipelines may be: using an online monitoring and diagnosis system based on ACUSUM to preprocess and analyze the collected monitoring data, and if abnormal changes are detected, extracting the signal area where the abnormal changes occur and sending an abnormal alarm. It should be noted that ACUSUM uses adaptive signal offset estimation based on EWMA, and adaptively adjusts the sensitivity and relaxation value in the detection process by dynamically smoothing the current signal and the historical signal.

[0028] It can be understood that the relaxation parameter determines how sensitive the sensor data is to changes. If the mean shift is large, the relaxation parameter may need to be adjusted to make it more sensitive so that anomalies are detected faster.

[0029] Optionally, the relaxation parameter is adjusted dynamically according to the magnitude of the mean shift. A rule can be set, such as reducing the relaxation parameter by 10% every time the threshold is exceeded by 1%.

[0030] It is understandable that the cumulative sum can be calculated based on the deviation between the collected sensor signal and the mean value of the sensor data. For example, if the value of the collected sensor signal is higher than the mean value of the sensor data, the deviation is added to the positive cumulative sum; conversely, the deviation is added to the negative cumulative sum. In this way, the upward or downward trend of electricity consumption can be monitored. By continuously calculating the cumulative sum, abnormal changes in sensor data can be detected. If the positive cumulative sum continues to increase, it may mean that the sensor data has increased abnormally; if the negative cumulative sum continues to increase, it may mean that the sensor data has decreased abnormally.

[0031] Please refer to Figure 2 , Figure 2 It is a flow chart of step S101 in the method for real-time detection of abnormal conditions of submarine pipelines.

[0032] In some embodiments, step S101 includes: S201. Acquire the original signal detected by each sensor in the sensor group.

[0033] S202: Arrange each original signal to generate an original signal sequence.

[0034] S203 . Based on the mean and standard deviation of the signal when no abnormal change is detected, the original signal sequence is converted into a standardized signal sequence of a standard normal distribution and used as the preprocessed sensor data set.

[0035] Optionally, the original signal sequence or the standardized signal sequence includes multiple elements, the number of elements is equal to the number of sensors, and each element represents a signal set collected by a sensor.

[0036] It is understandable that because temperature, strain and vibration sensors are arranged at different locations on the submarine pipeline, the signal scales between sensors at different depths will be different, and the signals need to be preprocessed. By standardizing the signal sequence, the scale differences between different sensors can be eliminated.

[0037] In some embodiments, step S105 includes: If the cumulative sum is a positive cumulative sum, the positive cumulative sum is compared with a preset first alarm threshold. When the positive cumulative sum is greater than the first alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued; if the cumulative sum is a negative cumulative sum, the negative cumulative sum is compared with a preset second alarm threshold. When the negative cumulative sum is less than the second alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued.

[0038] Please refer to Figure 3 , Figure 3 It is another step flow chart of the method for real-time detection of abnormal conditions in submarine pipelines.

[0039] In some embodiments, the method for real-time detection of abnormal conditions in a submarine pipeline further includes: S301. When it is determined that an abnormal situation occurs in a submarine pipeline, an abnormal sensor data set is obtained.

[0040] S302: Generate a time domain feature matrix based on the time domain features of the abnormal sensor data set.

[0041] S303: After the time domain feature matrix is ​​input into a preset neural network, the abnormal situation is identified through the neural network, and the category of the abnormal situation is output.

[0042] In some embodiments, the categories of abnormal conditions that may be identified include seafloor earthquakes, anchor strikes, anchor pulls, and oscillations caused by ocean currents.

[0043] In some embodiments, the time domain feature may be a standard deviation, a peak-to-peak value, or a root mean square value of the sensor signal.

[0044] Please refer to Figure 4 , Figure 4 It is a flow chart of step S301 in the method for real-time detection of abnormal conditions of submarine pipelines.

[0045] In some embodiments, step S301 includes: S401. When it is determined that an abnormal situation occurs in the submarine pipeline, the current time point is recorded as a first time point and new sensor signals are continuously acquired until a second time point is recorded when no abnormal situation occurs in the submarine pipeline.

[0046] S402: taking each new sensor signal acquired in a time period between the first time point and the second time point as an abnormal signal.

[0047] S403: Generate an abnormal sensor data set based on multiple abnormal signals.

[0048] In a specific embodiment, it is determined that an abnormal situation occurs in the submarine pipeline and the time point when the abnormal change occurs is recorded. Moment of time , and continue monitoring. If No abnormal changes are detected within 10 seconds, record this time point , the time interval All sensor signals in the abnormal sensor data set are extracted as the abnormal sensor data set. When executing the step of generating a time domain feature matrix based on the time domain features of the abnormal sensor data set, the time interval is extracted. The time domain features such as standard deviation, peak-to-peak value, and root mean square value of all sensor signals in the system are put into the time domain feature matrix, which is:

[0049] in, is the number of features.

[0050] Please refer to Figure 5 , Figure 5 It is a flow chart of step S303 in the method for real-time detection of abnormal conditions of submarine pipelines.

[0051] In some embodiments, step S303 includes: S501. Input the time domain feature matrix into a neural network, where the neural network at least includes a convolutional layer, a fully connected layer, and an output layer.

[0052] S502: After convolution processing is performed on the time domain features of the time domain feature matrix through a convolution layer, a plurality of time feature vectors are obtained.

[0053] S503. After the multiple time feature vectors are input to the output layer through the fully connected layer, each time feature vector is mapped to a target abnormal event category in the multiple abnormal event categories through the output layer to obtain a classification result of the abnormal situation and a confidence interval of the classification result.

[0054] In a specific embodiment, the trained neural network includes 2 convolutional layers, 2 fully connected layers and an output layer. The first convolutional layer uses a 3*3 convolution kernel for convolution operation, where the weight of the convolution kernel is Incorporating Bayesian theorem, weights Modeled from a fixed value as a probability distribution, where the weight The initial value is set to normal distribution. The second convolution layer uses a 2*2 convolution kernel. The flattened time feature vector passes through two fully connected layers and enters the output layer. The output layer uses the Softmax activation function to map the time feature vector to different abnormal event categories. After two convolution and pooling layers, the time feature vector is used as the input of the fully connected layer.

[0055] Exemplarily, a specific embodiment is used to illustrate that the output layer maps each time feature vector to a target abnormal event category in multiple abnormal event categories to obtain the classification results of the abnormal situation and the confidence interval of the classification results. In this embodiment, it is assumed that there are 3 sensors, each of which collects 100 data points in 1 second, forming a feature vector containing 300 elements. This vector is the time feature vector, which captures the dynamic changes of the machine in this second. These time feature vectors are input into a fully connected layer of a neural network. The role of the fully connected layer is to convert these high-dimensional data points into a smaller and easier to handle representation. For example, the fully connected layer may convert a vector of 300 elements into a vector containing 10 elements, which is a compressed representation of the original data and retains the most important information. Next, the vector of 10 elements is passed to the output layer. The task of the output layer is to map these feature vectors to different abnormal event categories. Suppose we have 5 possible abnormal events. The output layer will have a softmax layer containing 5 neurons, each neuron corresponding to an abnormal event category. The softmax layer calculates the probability of each category and outputs a probability distribution, indicating which abnormal event category the model believes the current sensor data is most likely to belong to. For example, if the model believes that the current abnormal state is most likely an overheated state, then the probability of the corresponding overheated neuron output will be the highest. In addition, this embodiment can also provide a confidence interval for the classification result. This confidence interval is obtained through the uncertainty estimate of the model. If the confidence interval is very narrow, it means that the classification result is very certain; if the confidence interval is very wide, it means that the classification result is not very certain, and more data or manual intervention may be required to confirm.

[0056] It can be understood that this embodiment uses a Bayesian convolutional neural network model to convert weight parameter estimates into probability distribution estimates. By sampling the weight distribution, the uncertainty of the classification results is quantified. The model outputs not only the abnormal event category results, but also the confidence interval of the results, allowing staff to make more cautious maintenance decisions based on the confidence interval of the results.

[0057] Please refer to Figure 6 , Figure 6 It is another step flow chart of the method for real-time detection of abnormal conditions in submarine pipelines.

[0058] In some embodiments, the method for real-time detection of abnormal conditions in a submarine pipeline further includes: S601. Obtain a historical abnormal event data set, where each sample in the historical abnormal event data set is provided with a category label.

[0059] S602: Generate a sample feature matrix based on the time domain feature of each sample, where each row of the sample feature matrix represents a sensor, and each column of the sample feature matrix represents a time domain feature.

[0060] S603: Input the sample feature matrix and each category label into a preset Bayesian convolutional neural network model to train the Bayesian convolutional neural network model.

[0061] S604: When training the Bayesian convolutional neural network model, the Bayesian convolutional neural network model is updated by a preset loss function until a trained Bayesian convolutional neural network model is obtained.

[0062] S605: Using the trained Bayesian convolutional neural network model as a neural network.

[0063] In a specific embodiment, the process of training the Bayesian convolutional neural network model is: The Bayesian convolutional neural network model is trained using the laboratory's existing submarine pipeline abnormal event data set (including anchor damage events and natural events). Each data sample has multiple time domain features (such as peak-to-peak value, standard deviation, etc.), and a category label is assigned to each data sample. The time domain features of the abnormal event data set are combined into a feature matrix, which is used as the input of the Bayesian convolutional neural network model together with the category label. The dimension of the feature matrix is ​​6*6, and the number of rows represents a temperature sensor, a vibration sensor, and four strain sensors, and the number of columns is the peak-to-peak value, variance, root mean square value, mean, standard deviation, and skewness. The Bayesian convolutional neural network quantifies the uncertainty of the model's classification of noise data by sampling the weight distribution. The classification result output by the model is accompanied by a variance, which represents the model's confidence interval for the classification result. The loss function is calculated and used for the back propagation of the model, and finally the trained Bayesian convolutional neural network model is obtained. The loss function formula is as follows:

[0064] The loss function combines KL divergence and log likelihood, where is the approximate posterior distribution, is the prior distribution, is the likelihood function, is the sampling times.

[0065] It can be understood that using the Bayesian convolutional neural network model, the model generates a series of prediction results by modeling the probability distribution of weights, sampling different weights from the current weight distribution, and estimating the uncertainty of the prediction through the mean and variance of these results.

[0066] The technical solution of the present application also provides a specific embodiment of a method for real-time detection of abnormal conditions in submarine pipelines using an online monitoring and diagnosis system based on ACUSUM.

[0067] In this embodiment, before performing abnormality detection on the monitoring data signal, the initial parameters of the system are first set:

[0068] in is the initial value of the exponentially weighted moving average operator (EWMA), and They are the initial values ​​of the upstream ACUSUM statistic and the downstream ACUSUM statistic, respectively.

[0069] After the system parameters are initialized, the monitoring data signal sequence is extracted ,in is the number of sensors.

[0070] Convert the signal to a dimensionless standard normal distribution and obtain a standardized sequence The normalization formula is:

[0071] in, and Respectively represent the mean and standard deviation of the signal when there is no abnormal change.

[0072] Perform an EWMA-based mean shift estimation on the standardized signal , the EWMA operator formula for estimating mean shift is as follows:

[0073] in is a smoothing constant used to control the weight of the latest data.

[0074] based on The signal mean offset is used to update the relaxation parameter of ACUSUM. Relaxation parameter Calculated by the following formula:

[0075] in is the preset minimum offset, is the mean shift estimate of the signal at the current moment.

[0076] Based on the signal offset, the cumulative sums in the positive and negative directions are calculated separately to detect abnormal rising or falling trends in the signal.

[0077] The positive ACUSUM statistics are as follows:

[0078] The negative ACUSUM statistics are as follows:

[0079] The positive or negative cumulative sum exceeds the set alarm threshold When the system issues an abnormal alarm, the threshold The value is set according to historical data. The specific trigger conditions are as follows:

[0080] During the period of time when the abnormal signal is detected, the signal continues to be monitored.

[0081] The data of the area where abnormal changes occur is analyzed and the matrix is ​​extracted. Finally, the extracted feature matrix is ​​imported into the abnormal event recognition system based on the Bayesian convolutional neural network for abnormal event recognition, to determine the category of the abnormal event and send the corresponding diagnosis result.

[0082] It can be understood that in this embodiment, the online monitoring and diagnosis system based on ACUSUM and the submarine pipeline anomaly detection and classification system based on the Bayesian convolutional neural network use the online monitoring and diagnosis system to monitor the changes in the submarine pipeline sensor signals in real time, detect the abnormal changes in the signals in time and make abnormal alarms, and extract the features in the abnormal change area as the input of the submarine pipeline anomaly detection and classification system, and finally obtain the abnormal event category result and the confidence interval of the result as the abnormal event alarm result. This embodiment makes up for the shortcomings of the monitoring technology based on fixed thresholds that it is difficult to detect abnormal changes in complex and changeable marine environment signals, and the traditional neural network cannot provide the uncertainty of the results.

[0083] In any of the above embodiments, first, by analyzing the sensor data set, the mean offset of the data is calculated, which is helpful to determine whether the submarine pipeline is abnormal. Then, according to the calculated mean offset, the system will dynamically update the preset relaxation parameters. The relaxation parameters are the key to adjusting the monitoring sensitivity. By dynamically adjusting the relaxation parameters, the detection sensitivity can be effectively improved, the probability of false alarms and missed alarms can be reduced, and the accuracy of detection can be improved. Then, based on the updated relaxation parameters and the current offset, the cumulative sum of the sensor data is calculated, which reflects the trend of data changes. Finally, by monitoring the trend of data changes after the update. If this trend exceeds the preset normal value range, the system will determine that the submarine pipeline has an abnormal situation and immediately issue an alarm. Through this intelligent monitoring and early warning mechanism, it is helpful to timely discover and deal with abnormal situations of submarine pipelines and ensure the safe operation of submarine pipelines.

[0084] Please refer to Figure 7 , Figure 7 It is a structural schematic diagram of a real-time detection device for abnormal conditions of submarine pipelines.

[0085] This embodiment also provides a real-time detection device for abnormal conditions of submarine pipelines, including: A collection module 701 is used to monitor the submarine pipeline through a sensor group to obtain a sensor data set; A first operation module 702 is used to determine a mean offset of data in a process of monitoring a submarine pipeline based on a sensor data set; An updating module 703, used to update a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used to adjust the monitoring sensitivity when monitoring the submarine pipeline; A second operation module 704 is used to determine the cumulative sum of the sensor data as the updated data change trend based on the updated relaxation parameter and the current offset, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data; The output module 705 is used to determine that an abnormal situation occurs in the submarine pipeline and issue an alarm when the trend of the updated data change is outside the preset value range.

[0086] It will be appreciated by those skilled in the art that all or some of the steps and devices in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. As is well known to those skilled in the art, communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0087] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0088] An embodiment of the present application further provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing any of the above methods for real-time detection of abnormal conditions in a submarine pipeline when executing the computer program.

[0089] refer to Figure 8 , Figure 8 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes: The processor 801 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 802 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device or a random access memory (RAM). The memory 802 can store operating devices and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 802, and the processor 801 calls and executes the real-time detection method for abnormal conditions of submarine pipelines in the embodiment of this application; Input / output interface 803, used to implement information input and output; The communication interface 804 is used to realize the communication interaction between the device and other devices. The communication can be realized through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.); A bus 805 that transmits information between various components of the device (e.g., the processor 801, the memory 802, the input / output interface 803, and the communication interface 804); The processor 801 , the memory 802 , the input / output interface 803 and the communication interface 804 are connected to each other in communication within the device via a bus 805 .

[0090] It can be understood that the contents of the above method embodiments are all applicable to the electronic device embodiment, the functions specifically implemented by the electronic device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0091] An embodiment of the present application further provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the real-time detection method for abnormal conditions of submarine pipelines as described in any one of the above specific embodiments.

[0092] The embodiment of the present application also discloses a computer program product, including a computer program or a computer instruction, wherein the computer program or the computer instruction is stored in a computer-readable storage medium, and a processor of a computer device reads the computer program or the computer instruction from the computer-readable storage medium, and the processor executes the computer program or the computer instruction, so that the computer device executes the real-time detection method for abnormal conditions of submarine pipelines as described in any of the previous embodiments.

[0093] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0094] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. It should be understood that in the present application, "at least one (item)" refers to one or more, and "a plurality" refers to two or more.

[0095] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0097] In addition, each functional unit in each embodiment of the present application 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.

[0098] 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 application is essentially 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 to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0099] Although the description of the present application has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims by reference to the attached claims, taking into account the prior art, so as to effectively cover the intended scope of the present application. In addition, the above description of the present application is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present application that have not yet been foreseen may still represent equivalent changes to the present application.

Claims

1. A method for real-time detection of abnormal conditions in submarine pipelines, characterized in that: The method comprises: The sensor data set is obtained by monitoring the submarine pipeline through the sensor group; Determining a mean shift of data during monitoring of the submarine pipeline based on the sensor data set; updating a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used to adjust monitoring sensitivity when monitoring the submarine pipeline; Based on the updated relaxation parameter and the current offset, determine the cumulative sum of the sensor data as the updated data change trend, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data; When the updated data change trend is outside the preset value range, it is determined that the submarine pipeline is abnormal and an alarm is issued.

2. A method for real-time detection of abnormal conditions of submarine pipelines according to claim 1, characterized in that: When the updated data change trend is outside the preset value range, determining that the submarine pipeline is abnormal and issuing an alarm includes: If the cumulative sum is a positive cumulative sum, the positive cumulative sum is compared with a preset first alarm threshold, and when the positive cumulative sum is greater than the first alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued; If the cumulative sum is a negative cumulative sum, the negative cumulative sum is compared with a preset second alarm threshold. When the negative cumulative sum is less than the second alarm threshold, it is determined that an abnormal situation occurs in the submarine pipeline and an alarm is issued.

3. The method for real-time detection of abnormal conditions of submarine pipelines according to claim 1, characterized in that: The method further comprises: When it is determined that the submarine pipeline has an abnormal condition, acquiring an abnormal sensor data set; Based on the time domain features of the abnormal sensor data set, generating a time domain feature matrix; After the time domain feature matrix is ​​input into a preset neural network, the abnormal situation is identified by the neural network and the category of the abnormal situation is output.

4. A method for real-time detection of abnormal conditions of submarine pipelines according to claim 3, characterized in that: When it is determined that the submarine pipeline is abnormal, obtaining an abnormal sensor data set includes: When it is determined that the submarine pipeline has an abnormal condition, the current time point is recorded as a first time point and new sensor signals are continuously acquired until a second time point is recorded when no abnormal condition is detected in the submarine pipeline; taking each new sensor signal acquired in the time period between the first time point and the second time point as an abnormal signal; The abnormal sensor data set is generated based on a plurality of the abnormal signals.

5. The method for real-time detection of abnormal conditions of submarine pipelines according to claim 3 is characterized in that: The method further comprises: Obtain a historical abnormal event data set, wherein each sample in the historical abnormal event data set is provided with a category label; Based on the time domain features of each of the samples, a sample feature matrix is ​​generated, wherein each row of the sample feature matrix represents a sensor, and each column of the sample feature matrix represents a time domain feature; Inputting the sample feature matrix and each of the category labels into a preset Bayesian convolutional neural network model to train the Bayesian convolutional neural network model; When the Bayesian convolutional neural network model is trained, the Bayesian convolutional neural network model is updated by a preset loss function until a trained Bayesian convolutional neural network model is obtained; The trained Bayesian convolutional neural network model is used as the neural network.

6. A method for real-time detection of abnormal conditions of submarine pipelines according to claim 3, characterized in that: After the time domain feature matrix is ​​input into a preset neural network, the abnormal situation is identified by the neural network, and the category of the abnormal situation is output, including: Inputting the time domain feature matrix into the neural network, wherein the neural network at least includes a convolutional layer, a fully connected layer and an output layer; After the time domain features of the time domain feature matrix are convolved through the convolution layer, a plurality of time feature vectors are obtained; After the multiple time feature vectors are input to the output layer through the fully connected layer, each of the time feature vectors is mapped to a target abnormal event category in multiple abnormal event categories through the output layer to obtain a classification result of the abnormal situation and a confidence interval of the classification result.

7. A method for real-time detection of abnormal conditions of submarine pipelines according to claim 1, characterized in that: The sensor data set obtained by monitoring the submarine pipeline through the sensor group includes: Acquire the original signal detected by each sensor in the sensor group; Arranging each of the original signals to generate an original signal sequence; Based on the mean and standard deviation of the signal when no abnormal change is detected, the original signal sequence is converted into a standardized signal sequence of standard normal distribution and used as the preprocessed sensor data set.

8. A real-time detection device for abnormal conditions of submarine pipelines, characterized in that: The device comprises: A collection module, used for monitoring the submarine pipeline through a sensor group to obtain a sensor data set; A first operation module is used to determine a mean offset of data in a process of monitoring the submarine pipeline based on the sensor data set; An updating module, used for updating a preset relaxation parameter according to the mean shift, wherein the relaxation parameter is a parameter used for adjusting monitoring sensitivity when monitoring the submarine pipeline; A second operation module is used to determine the cumulative sum of the sensor data as the updated data change trend based on the updated relaxation parameter and the current offset, wherein the current offset is the difference between the current sensor signal and the mean of the sensor data; The output module is used to determine that an abnormal situation occurs in the submarine pipeline and issue an alarm when the change trend of the updated data is outside the preset value range.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the real-time detection method for abnormal conditions of submarine pipelines according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the real-time detection method for abnormal conditions of submarine pipelines according to any one of claims 1 to 7 is implemented.

Images