A method, device and equipment for identifying abnormal state of oil and gas pipeline network system

Abstract

The embodiments of this specification provide a method, device and equipment for identifying abnormal states of an oil and gas pipeline network system. The method includes: obtaining operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient related to time; dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; respectively determining the behavior pattern data corresponding to the segmented data; the behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; classifying the behavior pattern data to obtain at least one group of clustering data; the clustering data corresponds to a behavior pattern category; identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category. The above method can quickly classify and identify the operation data of the oil and gas pipeline network system, thereby timely and accurately adjusting the operation state of the oil and gas pipeline network to ensure the normal operation of the system.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of oil and gas transmission technology, and in particular, to a method, device and equipment for identifying abnormal conditions of an oil and gas pipeline network system. Background Art

[0002] With the development of oil and gas pipeline network systems and the complexity of the supply and demand market, the construction and optimization of oil and gas pipeline network systems are also ongoing, making the complexity of oil and gas pipeline network systems continue to increase. Rapidly identifying faults in oil and gas pipeline network systems so that they can be handled in a timely manner is of great significance to ensuring the normal operation of oil and gas pipeline network systems.

[0003] In practical applications, for the oil and gas pipeline network system, the corresponding transmission data can be directly obtained, or the corresponding sensor data can be collected by setting corresponding sensors on the oil and gas pipeline network system to collect the operating data of the oil and gas pipeline network system. However, with the increase in the complexity of the oil and gas pipeline network system and the increase in the types of abnormal operating conditions, the difficulty of timely identifying the data corresponding to the fault state from the large amount of operating data, and then determining the fault of the pipeline system based on the identified data is gradually increasing, which has a certain impact on the stable operation of the oil and gas pipeline network system. Therefore, there is an urgent need for a method that can quickly and accurately identify the abnormal state of the oil and gas pipeline network system. Summary of the invention

[0004] The purpose of the embodiments of this specification is to provide a method, device and equipment for identifying abnormal conditions of an oil and gas pipeline network system, so as to solve the problem of how to quickly and accurately identify abnormal conditions of an oil and gas pipeline network system.

[0005] In order to solve the above technical problems, an embodiment of the present specification proposes a method for identifying abnormal states of an oil and gas pipeline network system, comprising: acquiring operating data corresponding to the oil and gas pipeline network system; the operating data corresponds to an autocorrelation coefficient related to time; dividing the operating data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; respectively determining behavior pattern data corresponding to the segmented data; the behavior pattern data is used to represent the operating mode of the oil and gas pipeline network system within a preset time period; classifying the behavior pattern data to obtain at least one group of cluster data; the cluster data corresponds to a behavior pattern category; and identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0006] The embodiment of the present specification also proposes an abnormal state identification device for an oil and gas pipeline network system, including: an operation data acquisition module, used to acquire operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient related to time; a segmented data division module, used to divide the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; a behavior pattern data determination module, used to respectively determine the behavior pattern data corresponding to the segmented data; the behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; a behavior pattern data classification module, used to classify the behavior pattern data to obtain at least one group of cluster data; the cluster data corresponds to a behavior pattern category; an abnormal state identification module, used to identify the abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0007] The embodiment of the present specification also proposes an abnormal state identification device for an oil and gas pipeline network system, comprising a memory and a processor; the memory is used to store computer program instructions; the processor is used to execute the computer program instructions to implement the following steps: obtaining operating data corresponding to the oil and gas pipeline network system; the operating data corresponds to an autocorrelation coefficient related to time; dividing the operating data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; determining the behavior pattern data corresponding to the segmented data respectively; the behavior pattern data is used to represent the operating mode of the oil and gas pipeline network system within a preset time period; classifying the behavior pattern data to obtain at least one group of cluster data; the cluster data corresponds to a behavior pattern category; identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0008] It can be seen from the technical solutions provided by the above embodiments of this specification that after obtaining the operation data of the oil and gas pipeline network system, the embodiments of this specification group the operation data according to the autocorrelation coefficient of the operation data corresponding to time, and then determine the behavior pattern data corresponding to the grouped data, so as to obtain at least one group of cluster data by classification according to the behavior pattern data, so that the corresponding behavior pattern category can be reflected by the cluster data, and finally the abnormal state of the oil and gas pipeline network system can be identified according to the corresponding behavior pattern category. Through the above method, a large amount of operation data can be simplified through classification, and by identifying the behavior pattern category corresponding to the operation data, the abnormal state of the oil and gas pipeline network system can be classified based on the behavior pattern category, so that the operation state of the oil and gas pipeline network can be adjusted in a timely and accurate manner, the normal operation of the oil and gas pipeline network system is realized, and the normal transportation of oil and gas resources is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the embodiments of this specification 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 recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0010] Figure 1 This is a flow chart of a method for identifying abnormal status of an oil and gas pipeline network system according to an embodiment of this specification;

[0011] Figure 2 This is a schematic diagram of a pipe network operation data according to an embodiment of this specification;

[0012] Figure 3 A schematic diagram of a classification evaluation result of an embodiment of this specification;

[0013] Figure 4 This is a module diagram of an abnormal state identification device for an oil and gas pipeline network system according to an embodiment of this specification;

[0014] Figure 5 This is a structural diagram of an abnormal state identification device for an oil and gas pipeline network system according to an embodiment of this specification. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0016] In order to solve the above technical problems, the embodiments of this specification propose a method for identifying abnormal status of an oil and gas pipeline network system. The execution subject of the method for identifying abnormal status of an oil and gas pipeline network is an abnormal status identification device of an oil and gas pipeline network, and the abnormal status identification device of an oil and gas pipeline network includes but is not limited to a server, an industrial computer, a PC, a mobile terminal, a wearable device, etc. Figure 1 As shown, the method for identifying abnormal status of the oil and gas pipeline network system may include the following specific steps.

[0017] S110: Acquire operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient related to time.

[0018] The oil and gas pipeline network system can be a system corresponding to the pipeline network for transmitting oil and gas resources. The oil and gas pipeline network system can include not only the pipeline network itself, but also a data acquisition module for collecting the operating status of the oil and gas pipeline network and a control module for controlling the oil and gas pipeline network.

[0019] The oil and gas pipeline network system generates a large amount of operation data during daily operation, and these operation data can be used to reflect the actual status of the oil and gas pipeline network system. The operation data can be data collected by directly setting corresponding sensors in the oil and gas pipeline network system and using the sensors. The operation data can also be data indirectly derived from other acquired data. The specific method for collecting the operation data can be set according to the needs of actual applications, and will not be repeated here.

[0020] Specifically, the operation data may be, for example, the oil and gas resource transmission rate in the oil and gas pipeline network, the temperature, pressure and other state data of the oil and gas pipeline network, the throughput capacity data of the oil and gas pipeline network, etc. In practical applications, the operation data may be set according to the type of data actually collected by the oil and gas pipeline network system, and is not limited to the types of operation data introduced in the above examples, which will not be described in detail here.

[0021] As the complexity of oil and gas pipeline network systems increases, a large amount of operation data will be generated during the daily operation of oil and gas pipeline networks. Manual analysis of these operation data based solely on the experience of operators is obviously unable to effectively process all of these operation data. If the operation data is analyzed based solely on relatively simple hydraulic simulation analysis models set by the experience of operators, it is also difficult to cope with complex and changeable operating conditions. That is, these hydraulic simulation analysis models cannot handle a variety of abnormal conditions, nor can they identify abnormal conditions that are not set. Therefore, there is an urgent need for a method that can effectively and accurately analyze these operation data to determine the abnormal state of the oil and gas pipeline network.

[0022] Therefore, before analyzing the operation data, the operation data may be simplified first, so that the simplified operation data can be analyzed more conveniently and effectively.

[0023] In some embodiments, the operation data may be segmented in combination with a time series. Specifically, the operation data may correspond to an autocorrelation coefficient associated with time. The autocorrelation coefficient is used to indicate the degree of correlation of the operation data in different periods, that is, to distinguish the correlation between the operation data corresponding to different time periods on a certain time series. According to the autocorrelation coefficient, it can be determined that the operation data with certain differences are distinguished, so as to facilitate the subsequent segmentation operation.

[0024] In some embodiments, after the operation data is obtained, the operation data may be preprocessed. The preprocessing may include at least one of outlier processing and dimensionality reduction processing. The outlier processing is used to remove outliers in the operation data, thereby reducing the interference caused by outliers in the operation data. The dimensionality reduction processing is used to reduce the dimension of the operation data, so that the data can be processed more conveniently in the subsequent process.

[0025] Specifically, at least one of a Hampel filter, a Kalman filter, and a clustering method may be used to remove abnormal values ​​in the operation data.

[0026] Hampel filtering mainly calculates the median of the window composed of sample data and associated sample data for the input sample data, three on each side, and estimates the standard deviation of each sample data to the median using the absolute value of the median. If a sample data differs from the median by more than three standard deviations, the sample data is replaced with the median, thereby removing outliers.

[0027] Kalman filtering seeks the optimal estimate of the system state from the output and input observation data based on the state space representation of the linear system. The system state here is the minimum parameter set that summarizes the effects of all past inputs and disturbances on the system. Knowing the system state can determine the entire behavior of the system together with future inputs and disturbances. Kalman filtering can effectively remove noise signals in operating data.

[0028] The clustering method clusters the data by the correlation between the data. Based on the clustering results, other data that cannot be clustered with other data can be found and these data can be eliminated as abnormal data.

[0029] The above methods can all achieve the removal of outliers. In practical applications, other methods can also be used to remove outliers in the operating data according to specific needs. The methods are not limited to the above examples and will not be described in detail here.

[0030] After removing the abnormal values ​​in the operation data, the data corresponding to the time series that are repeated and unchanged can also be deleted. Within the corresponding collection time, if the values ​​of the operation data are repeated and unchanged, the possibility of abnormality in the operation data is small. In order to reduce the processing volume in the subsequent process, these time series data that are repeated and unchanged can also be eliminated.

[0031] Afterwards, the operating data can be reduced in dimension using the principal component analysis method, which can retain 85% of the original information, thereby improving the data quality while reducing the loss of the data itself, and effectively speeding up the data processing progress in the subsequent process. The specific dimensionality reduction process can be adjusted according to the actual application process, and will not be repeated here.

[0032] After outlier processing and / or dimensionality reduction processing is performed on the operating data, the quality of the operating data can be effectively improved, so that corresponding analysis results can be obtained more efficiently and accurately in the subsequent analysis process.

[0033] S120: Divide the operating data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data correspond to different time frequencies.

[0034] After the operation data is acquired, the operation data can be divided according to the autocorrelation coefficient corresponding to the operation data. Since the autocorrelation coefficient is used to indicate the degree of correlation between data in different time periods, when the autocorrelation coefficient is larger, the correlation between the corresponding two sections of data is smaller. Therefore, the operation data can be divided based on the size of the autocorrelation coefficient.

[0035] Specifically, the maximum value of the autocorrelation coefficients can be selected, and the operation data can be divided based on the time node corresponding to the maximum value. In some embodiments, an autocorrelation threshold can also be pre-set, and autocorrelation coefficients greater than the autocorrelation coefficient threshold can be selected, and the operation data can be divided based on the time nodes corresponding to these selected autocorrelation coefficients.

[0036] The autocorrelation coefficient may be, for example, a coefficient determined by an autocorrelation analysis method. The specific determination process may be set according to the actual application process, and will not be described in detail here.

[0037] After processing the operation data based on the autocorrelation coefficient, the segmented data can also be decomposed using a discrete wavelet transform method, thereby dividing the segmented data of the nonlinear time series into low-frequency sequence data and high-frequency sequence data. The low-frequency sequence data and the high-frequency sequence data correspond to different time frequencies, thereby further distinguishing the data, so that the data can be processed accordingly according to different types of data in the subsequent process.

[0038] In some implementations, when the operation data is preprocessed, the preprocessed operation data may be processed according to the autocorrelation coefficient to obtain at least two groups of segmented data. The specific implementation process may refer to the above introduction and will not be repeated here.

[0039] S130: Determine the behavior pattern data corresponding to the segmented data respectively; the behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period.

[0040] Since the segmented data is only the data obtained after the operation data is divided, the essence of the operation data is not changed, and it is generally difficult to intuitively view the state reflected by the data based on the divided data, so the behavior pattern data corresponding to the segmented data can also be determined, and then subsequent analysis and judgment can be carried out. The behavior pattern data can be used to represent the operation mode of the oil and gas pipeline network system. Operation data can generally only be used to reflect the state of the system at a single moment, and analyzing each moment of the system one by one obviously consumes a lot of workload. Therefore, the behavior pattern data can be used to represent the state of the system within a certain period of time, thereby improving processing efficiency.

[0041] In some embodiments, the data slope and data mean of the segmented data can be first obtained, and then the behavior pattern data corresponding to the segmented data can be determined based on the data slope and data mean, so as to convert the operating state of the system at each moment into the behavior pattern of the system within a certain period of time.

[0042] In some implementations, when the operation data is decomposed, the behavior pattern data corresponding to the decomposed operation data may also be determined. The specific implementation process may refer to the above introduction and will not be repeated here.

[0043] S140: Classify the behavior pattern data to obtain at least one set of cluster data; the cluster data corresponds to a behavior pattern category.

[0044] Since the behavior pattern data are used to reflect the operating status of the system in different time periods, that is, different behavior pattern data may correspond to different categories of operating status, at least one group of clustering data can be obtained by clustering the behavior pattern data, and the different states of the system can be reflected by the clustering data.

[0045] Specifically, the Euclidean distance between data can be used as a measure of similarity within and between clusters, and the data can be clustered according to the Euclidean distance between the data, so as to classify the data according to the differences between the data when the data categories are not defined.

[0046] In some embodiments, the behavior pattern data can be classified into at least one group of clustered data using the K-Means++ clustering method and / or the Birch clustering method. When implementing the K-Means++ clustering method, for a given sample set, the sample set is divided into K clusters according to the distance between the samples. The points within the cluster are connected as closely as possible, and the distance between clusters is as large as possible, so as to achieve multiple classification of the data. The Birch clustering method BIRCH algorithm uses a tree structure to help us cluster quickly. This tree structure is similar to a balanced B+ tree, and is generally referred to as a clustering feature tree (CF Tree for short), and then a hierarchical method is used to cluster and constrain the corresponding data.

[0047] The above method is only a schematic illustration of the method used when clustering data, and the specific operation process can be set according to the needs of actual applications. In actual applications, other clustering methods can also be used to achieve the above effect, which will not be repeated here.

[0048] After completing the classification of the data, the classification effect can also be evaluated based on the clustering data. For example, the results can be evaluated through DBI, CHI, F test and P value. When CHI is the largest and DBI is the smallest, the number of classifications is optimal and the method used is also the optimal clustering method. When the P value is less than 0.01, it means that there is a high probability that the number of clusters currently assumed is correct.

[0049] The above evaluation method is only an exemplary description. In actual application, other specific methods can also be used to evaluate the classification results. It is not limited to the above example and will not be elaborated here.

[0050] S150: Identifying an abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0051] After the cluster data is acquired, since the cluster data are classified and correspond to corresponding behavior pattern categories, the abnormal state of the oil and gas pipeline network can be identified according to the behavior pattern categories corresponding to different cluster data.

[0052] In some implementations, since different cluster data differ only in similarity, it is not possible to directly determine the category corresponding to each group of cluster data based on the cluster data obtained through classification, that is, it is not possible to directly determine which group specifically reflects the abnormal state of the system. Therefore, after obtaining the cluster data, a classification algorithm can be used to classify each type of data, thereby realizing type recognition of different unknown data.

[0053] Specifically, the ovo (one-to-one) classification principle can be used for multi-class division. The specific implementation process can be to first train at least one classifier model based on the cluster data, and the classifier model is used to classify two groups of cluster data. Since it is necessary to train for any two groups of cluster data, when there are K groups of cluster data, it is necessary to design K(K-1) / 2 classifier models, and use the corresponding cluster data for training in turn. After training, these trained classifier models can be used to classify each group of cluster data in turn, and each classifier model can output a candidate classification result for each group of cluster data. After obtaining multiple candidate classification results corresponding to a group of cluster data, the candidate classification result with the highest frequency of occurrence can be selected as the classification result according to the frequency of occurrence of each classification result in the candidate classification result. The classification result can be used to represent the category of the cluster data. According to the classification result, the abnormal state of the oil and gas pipeline network system can be identified.

[0054] The above classification and identification methods are only specific examples for realizing the identification of abnormal conditions of oil and gas pipeline network systems. In practical applications, other classifier models can also be used to complete the classification of cluster data, thereby realizing the identification of abnormal conditions of oil and gas pipeline network systems. It is not limited to the above examples and will not be repeated here.

[0055] After determining the specific abnormal state of the oil and gas pipeline network system, the processing strategy corresponding to the abnormal state can be executed, so that the oil and gas pipeline network system can get rid of the corresponding abnormal state and ensure the normal and stable operation of the oil and gas pipeline network system. The specific process of processing the abnormal state of the oil and gas pipeline network can be set according to the needs of actual application, and will not be repeated here.

[0056] A specific scenario example is used to illustrate the abnormal state identification method of the above-mentioned oil and gas pipeline network system. In this scenario example, the TGNET simulation software is used to build an oil and gas pipeline network simulation model, and the model is simulated in steady state and transient state respectively. During the transient simulation process, six common abnormal operation conditions of the pipeline network were simulated, including abnormal oil and gas pipeline supply, abnormal natural gas demand, compressor shutdown, abnormal valve closure, abnormal pressure regulating valve and pipeline leakage. The total duration of the simulation process is 360 hours, and only one abnormal condition occurs in each period of time. It is set to generate a set of oil and gas pipeline network operation data every 3 minutes. The operating status parameters of the oil and gas pipeline network system are as follows: Figure 2 shown.

[0057] According to the abnormal state identification method in the embodiment of this specification, a pipeline network operation data matrix of size 7200×844 is simplified and characterized, and the size of the pipeline network operation data matrix after characterization is 480×28. Based on the simplified data, the perception method of abnormal operation state of complex oil and gas pipeline network system proposed in the present invention is adopted, and the number of clusters k=7 is given. The operation mode of each section of the pipeline network is clustered and analyzed, and the clustering results are evaluated by clustering evaluation indicators DBI, CHI, F test and P value. The clustering evaluation results are as follows: the pseudo F value of k-means++ is [[207.47926878]], the P value of k-means++ is [[1.11022302e-16]], the clustering DBI value of k-means++ is 0.7424885535941589, and the clustering CHI value of k-means++ is 207.47926877872655.

[0058] Since the larger the CHI and the smaller the DBI, the closer the assumed number of clusters is to the actual number of data types, and when the P value is less than 0.01, it means that there is a high probability that the number of clusters we originally assumed is correct. From the above evaluation results, it can be seen that this method is suitable for the abnormal perception of the oil and gas pipeline system simulation data set.

[0059] The above data set is set as the training set, the abnormal change amplitude of the operating condition of the oil and gas pipeline network system is changed, the above steps are repeated to obtain the segmentation label of the data set, and the data set is set as the prediction set. The random forest classification algorithm proposed in the embodiment of this specification is used to classify and identify the abnormal operation mode. The random forest classifier model is trained by the training set and the validation set, and the trained classifier is used for the working condition identification of the prediction set. The classification evaluation result of the prediction set is as follows: Figure 3 shown.

[0060] It can be seen from the classification evaluation results that the recognition accuracy of the prediction set is 97.9%, and this method is suitable for the classification recognition of the prediction data set.

[0061] Based on the introduction of the above embodiments and scenario examples, it can be seen that after the oil and gas pipeline network system abnormal state identification method obtains the operation data of the oil and gas pipeline network system, it groups the operation data according to the autocorrelation coefficient of the operation data corresponding to time, and then determines the behavior pattern data corresponding to the grouped data, thereby classifying the behavior pattern data to obtain at least one group of cluster data, so that the corresponding behavior pattern category can be reflected by the cluster data, and finally the abnormal state of the oil and gas pipeline network system can be identified according to the corresponding behavior pattern category. Through the above method, a large amount of operation data can be simplified through classification, and by identifying the behavior pattern category corresponding to the operation data, the abnormal state of the oil and gas pipeline network system can be classified based on the behavior pattern category, so that the operation state of the oil and gas pipeline network can be adjusted in a timely and accurate manner, the normal operation of the oil and gas pipeline network system is realized, and the normal transportation of oil and gas resources is guaranteed.

[0062] Based on the above-mentioned oil and gas pipeline network system abnormal state identification method, this specification also proposes an embodiment of an oil and gas pipeline network system abnormal state identification device. The oil and gas pipeline network system abnormal state identification device can be arranged on the oil and gas pipeline network system abnormal state identification device. Figure 4 As shown, the pore throat structure dynamic characterization device specifically includes:

[0063] The operation data acquisition module 410 is used to acquire operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient associated with time;

[0064] A segmented data division module 420 is used to divide the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data correspond to different time frequencies;

[0065] The behavior pattern data determination module 430 is used to respectively determine the behavior pattern data corresponding to the segmented data; the behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period;

[0066] A behavior pattern data classification module 440 is used to classify the behavior pattern data to obtain at least one set of cluster data; the cluster data corresponds to a behavior pattern category;

[0067] The abnormal state identification module 450 is used to identify the abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0068] Based on the above-mentioned oil and gas pipeline network system abnormal state identification method, the embodiment of this specification also provides an oil and gas pipeline network system abnormal state identification device. Figure 5 As shown, the oil and gas pipeline network system abnormal state identification device may include a memory and a processor.

[0069] In this embodiment, the memory may be implemented in any appropriate manner. For example, the memory may be a read-only memory, a mechanical hard disk, a solid-state hard disk, or a USB flash drive, etc. The memory may be used to store computer program instructions.

[0070] In this embodiment, the processor can be implemented in any appropriate manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc. The processor can execute the computer program instructions to implement the following steps: obtaining operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient associated with time; dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; determining the behavior pattern data corresponding to the segmented data respectively; the behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; classifying the behavior pattern data to obtain at least one group of cluster data; the cluster data corresponds to a behavior pattern category; identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category.

[0071] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0072] It can be known from the above description of the implementation mode that the technicians in this field can clearly understand that the present specification can be implemented by means of software plus the necessary first hardware platform. Based on such an understanding, the technical solution of the present specification can be essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present specification or some parts of the embodiments.

[0073] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0074] This specification can be used in many first or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0075] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0076] Although the present specification is described through embodiments, those skilled in the art will appreciate that there are many modifications and changes to the present specification without departing from the spirit of the present specification, and it is intended that the appended claims include these modifications and changes without departing from the spirit of the present specification.

Claims

1. A method for identifying abnormal conditions of an oil and gas pipeline network system, characterized in that: include: Acquiring operation data corresponding to the oil and gas pipeline network system; the operation data corresponding to the autocorrelation coefficient associated with time; Dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data correspond to different time frequencies; respectively determining the behavior pattern data corresponding to the segmented data; The behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; Classifying the behavior pattern data to obtain at least one set of cluster data; the cluster data corresponds to a behavior pattern category; Identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category; wherein, after determining the abnormal state of the oil and gas pipeline network system, executing a processing strategy corresponding to the abnormal state to make the oil and gas pipeline network system get rid of the corresponding abnormal state; Among them, obtaining the operation data corresponding to the oil and gas pipeline network system includes: using TGNET simulation software to build an oil and gas pipeline network simulation model, and performing steady-state and transient simulations on the model respectively. In the transient simulation process, a preset number of abnormal operation conditions of the pipeline network are simulated, and the total duration of the simulation process is a first preset duration, and only one abnormal condition occurs in each period of time, and a set of oil and gas pipeline network operation data is generated every second preset duration; Wherein, respectively determining the behavior pattern data corresponding to the segmented data includes: Obtaining a data slope and a data mean of the segmented data; The behavior pattern data corresponding to the segmented data is determined according to the data slope and the data mean.

2. The method according to claim 1, characterized in that Dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient includes: dividing the operation data into low-frequency data and high-frequency data based on the autocorrelation coefficient; Accordingly, before dividing the operating data into low-frequency data and high-frequency data based on the autocorrelation coefficient, the method further includes: The operating data is preprocessed; the preprocessing includes at least one of outlier processing and dimensionality reduction processing.

3. The method according to claim 2, characterized in that The preprocessing of the operation data includes: Using at least one of a Hampel filter, a Kalman filter, and a clustering method to remove outliers in the operating data; Delete the running data corresponding to the time series that are repeated and unchanged; Using principal component analysis to reduce the dimension of the operating data; Accordingly, the operation data is divided into at least two groups of segmented data according to the autocorrelation coefficient, including: The dimension-reduced operating data is divided into at least two groups of segmented data based on the autocorrelation coefficient.

4. The method according to claim 1, characterized in that The step of dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient comprises: Determine the autocorrelation coefficient corresponding to the operation data by using an autocorrelation analysis method; The operation data is divided into at least two groups of segmented data according to a comparison result between the autocorrelation coefficient and the autocorrelation threshold.

5. The method according to claim 1, characterized in that After dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient, the method further includes: Dividing the segmented data into low-frequency sequence data and high-frequency sequence data by using a discrete wavelet transform method; Accordingly, the step of respectively determining the behavior pattern data corresponding to the segmented data includes: The behavior pattern data corresponding to the low-frequency sequence data and the high-frequency sequence data are determined respectively.

6. The method according to claim 1, characterized in that The classifying the behavior pattern data to obtain at least one set of cluster data includes: The behavior pattern data is classified into at least one group of cluster data using a K-Means clustering method and / or a Birch clustering method.

7. The method according to claim 1, characterized in that Identifying the abnormal state of the oil and gas pipeline network system based on the behavior pattern category includes: Training at least one classifier model based on the cluster data; the classifier model is used to classify two groups of cluster data; Using the at least one classifier model to classify the cluster data respectively to obtain corresponding candidate classification results; Determine the candidate classification result with the highest frequency as the classification result; the classification result is used to represent the category of the clustering data; Based on the classification result, the abnormal state of the oil and gas pipeline network system is identified.

8. An abnormal state identification device for an oil and gas pipeline network system, characterized in that: include: An operation data acquisition module, used to acquire operation data corresponding to the oil and gas pipeline network system; The operation data corresponds to an autocorrelation coefficient associated with time; A segmented data division module, used for dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data correspond to different time frequencies; A behavior pattern data determination module, used to respectively determine the behavior pattern data corresponding to the segmented data; The behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; A behavior pattern data classification module, used for classifying the behavior pattern data to obtain at least one set of cluster data; the cluster data corresponds to a behavior pattern category; an abnormal state identification module, used to identify the abnormal state of the oil and gas pipeline network system based on the behavior pattern category; wherein, after determining the abnormal state of the oil and gas pipeline network system, executing a processing strategy corresponding to the abnormal state to make the oil and gas pipeline network system get rid of the corresponding abnormal state; Among them, obtaining the operation data corresponding to the oil and gas pipeline network system includes: using TGNET simulation software to build an oil and gas pipeline network simulation model, and performing steady-state and transient simulations on the model respectively. In the transient simulation process, a preset number of abnormal operation conditions of the pipeline network are simulated, and the total duration of the simulation process is a first preset duration, and only one abnormal condition occurs in each period of time, and a set of oil and gas pipeline network operation data is generated every second preset duration; Wherein, respectively determining the behavior pattern data corresponding to the segmented data includes: Obtaining a data slope and a data mean of the segmented data; The behavior pattern data corresponding to the segmented data is determined according to the data slope and the data mean.

9. An abnormal state identification device for an oil and gas pipeline network system, comprising a memory and a processor; The memory is used to store computer program instructions; The processor is used to execute the computer program instructions to implement the following steps: obtaining operation data corresponding to the oil and gas pipeline network system; the operation data corresponds to an autocorrelation coefficient associated with time; dividing the operation data into at least two groups of segmented data according to the autocorrelation coefficient; the segmented data corresponds to different time frequencies; and determining the behavior pattern data corresponding to the segmented data respectively; The behavior pattern data is used to represent the operation mode of the oil and gas pipeline network system within a preset time period; the behavior pattern data is classified to obtain at least one group of cluster data; the cluster data corresponds to a behavior pattern category; based on the behavior pattern category, the abnormal state of the oil and gas pipeline network system is identified; wherein, After determining the abnormal state of the oil and gas pipeline network system, executing the processing strategy corresponding to the abnormal state, so that the oil and gas pipeline network system can get rid of the corresponding abnormal state; wherein, obtaining the operation data corresponding to the oil and gas pipeline network system includes: using TGNET simulation software to build an oil and gas pipeline network simulation model, and respectively performing steady-state and transient simulations on the model, in the transient simulation process, simulating a preset number of abnormal pipeline network operation conditions, the total duration of the simulation process is a first preset duration, and only one abnormal condition occurs in each period of time, and setting a group of oil and gas pipeline network operation data to be generated every second preset duration; wherein, respectively determining the behavior pattern data corresponding to the segmented data includes: obtaining the data slope and data mean of the segmented data; The behavior pattern data corresponding to the segmented data is determined according to the data slope and the data mean.

Images