Oil and gas equipment diagnosis and maintenance method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of equipment diagnosis, and provides an oil and gas equipment diagnosis and maintenance method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing of the operation data, obtained in advance, of oil and gas equipment in a preset time period, and obtaining the operation state data; matching the running state data with fault case data pre-stored in a database; when the matching is successful, generating a current maintenance scheme according to the corresponding fault case data, the historical maintenance scheme and the operation state data by using an AI engine; when the matching fails, obtaining a current reconstruction error according to the operation state data, and carrying out future fault state judgment; and utilizing an AI engine to generate a future fault type and a future maintenance scheme according to the future fault state, the operation state data and the historical maintenance scheme. The adaptive capacity of the system to various complex scenes is remarkably enhanced, and estimation, real-time dynamic monitoring and intelligent diagnosis of future faults of the equipment operation state are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of equipment diagnosis, in particular to an oil and gas equipment diagnosis and maintenance method and device, an electronic device and a storage medium. BACKGROUND

[0002] In the global energy pattern, the production activities of oil and gas resources rely on a series of complex and critical industrial equipment, including drilling equipment, pumping units, pipeline systems, and large-scale refining devices, etc. The stable and efficient operation of these equipment is the cornerstone to ensure the continuity, safety and economic benefits of oil and gas production. With the increasing complexity of oil and gas exploitation environment (such as deep sea, desert, high temperature, high coldness, etc.) and the continuous emergence of new technologies (such as fracturing equipment in shale gas exploitation), higher requirements are put forward for the performance and stability of the equipment. Under this background, effective diagnosis and maintenance technology is particularly important, which not only can reduce the production interruption caused by failure, but also can prolong the service life of the equipment and reduce the operating cost.

[0003] In the related art, sensors (such as temperature and pressure sensors) are usually installed at key parts of the equipment, and alarm is triggered by fixed threshold value, but normal fluctuations of equipment parameters or generation occur when the equipment is running under complex working conditions, which can easily lead to false alarm or missed alarm. Or use the average value or standard deviation of the equipment operating parameters as a reference standard, and combine the limited fault case experience to make a judgment, but the judgment standard based on fixed experience and limited data has the problem of not responding in time to new equipment, new technology or special working conditions. And the above methods can only respond passively after the equipment failure occurs, so that the oil and gas equipment often runs with disease, not only accelerates the damage of the equipment and shortens the service life of the equipment, but also increases the maintenance cost and production risk. SUMMARY

[0004] The present application aims to solve at least one of the above technical problems.

[0005] To solve the above problems, the present application provides an oil and gas equipment diagnosis and maintenance method, device, electronic device and storage medium.

[0006] In a first aspect, the present application provides an oil and gas equipment diagnosis and maintenance method, comprising: preprocessing the running data of the oil and gas equipment running in a preset time period to obtain running state data; matching the running state data with the pre-stored fault case data in the database; when the matching is successful, using an AI engine to generate a current maintenance scheme according to the corresponding fault case data, historical maintenance scheme and the running state data; When the matching fails, a current reconstruction error is obtained according to the running state data, a future fault state of the current running state is judged according to the current reconstruction error, and a future fault type and a future maintenance scheme are generated by using the AI engine according to the future fault state, the running state data and the historical maintenance scheme.

[0007] Optionally, the matching of the running state data with the pre-stored fault case data comprises: A current running condition is obtained by using a cluster analysis method to analyze the running state data; A plurality of association rules are obtained by performing rule mining on the pre-stored fault case data; When the current running condition triggers at least two association rules, it is determined that the fault type corresponding to the association rule with the highest confidence degree matches the current running condition successfully; When the current running condition triggers any one of the association rules, it is determined that the fault type corresponding to the association rule matches the current running condition successfully; When the current running condition does not trigger any one of the association rules, it is determined that the matching fails.

[0008] Optionally, the future fault state judgment of the current running state according to the current reconstruction error comprises: If the current reconstruction error is greater than a preset dynamic threshold, it is determined that the current running state has an unknown fault and a warning is given; If the current reconstruction error is less than or equal to the preset dynamic threshold, it is determined that the current running state does not have an unknown fault, a future fault occurrence probability is obtained according to the running state data, and the future fault state is judged based on the future fault occurrence probability.

[0009] Optionally, the future fault state judgment based on the future fault occurrence probability obtained according to the running state data comprises: A Bayesian rule is used to obtain a future fault occurrence probability under each fault level according to the running state data and a preset fault level; The future fault state is obtained according to the fault level corresponding to the maximum future fault occurrence probability.

[0010] Optionally, after the future fault type and the future maintenance scheme are generated according to the future fault state, the running state data and the historical maintenance scheme, the method further comprises: The database is updated according to the running state data, the future fault type and the future maintenance scheme.

[0011] Optionally, the matching the operation state data with the pre-stored fault case data comprises: generating a current operation curve according to the operation state data, setting an upper reference curve and a lower reference curve according to device operation upper and lower limit indicators, and generating a corresponding fault curve according to each of the pre-stored fault case data; when the current operation curve deviates from the range delimited by the upper reference curve and the lower reference curve, obtaining the similarity of the current operation curve and each of the fault curves; if the similarity is greater than or equal to a preset similarity threshold, determining that the fault type of the fault curve corresponding to the highest similarity matches the current operation data successfully; if the similarity is all less than the preset similarity threshold, determining that the matching fails.

[0012] Optionally, the current reconstruction error obtained according to the operation state data comprises: dimensionally reducing the operation state data by using an encoder to obtain a low-dimensional feature vector; reconstructing the low-dimensional feature vector by using a decoder to obtain reconstruction data; obtaining the minimized current reconstruction error based on back propagation according to the operation state data and the reconstruction data.

[0013] In a second aspect, the present application provides an oil and gas equipment diagnosis and maintenance device, comprising: a preprocessing module configured to preprocess operation data of an oil and gas equipment obtained in advance within a preset time period to obtain operation state data; a matching module configured to match the operation state data with fault case data pre-stored in a database; a generating module configured to, when the matching is successful, generate a current maintenance scheme by using an AI engine according to corresponding fault case data, historical maintenance schemes and the operation state data; an estimation module configured to, when the matching fails, obtain a current reconstruction error according to the operation state data, judge a future fault state of a current operation state according to the current reconstruction error, and generate a future fault type and a future maintenance scheme by using the AI engine according to the future fault state, the operation state data and the historical maintenance schemes.

[0014] In a third aspect, the present application provides an electronic device comprising a memory and a processor; the memory is configured to store a computer program; the processor is configured to, when executing the computer program, implement the oil and gas equipment diagnosis and maintenance method of the first aspect.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the oil and gas equipment diagnosis and maintenance method of the first aspect is implemented.

[0016] The oil and gas equipment diagnosis and maintenance method of the present application has the following beneficial effects: the running data of the oil and gas equipment obtained in advance within a preset time period is preprocessed to obtain running state data, which can ensure the purity of the data and provide a data basis for subsequent fault case data matching or future fault judgment. The running state data is matched with the pre-stored fault case data, and the oil and gas equipment operation is preliminarily judged according to the pre-stored fault case data. When the matching is successful, an AI engine is used to quickly generate an adaptive maintenance scheme according to the corresponding fault case data, historical maintenance scheme and running state data, to intelligently adapt the maintenance strategy based on historical experience and real-time state, shorten the maintenance scheme generation cycle, and further shorten the fault response cycle, thereby achieving efficient response to known faults. When the matching fails, the current reconstruction error is obtained according to the running state data, the running state data is deeply mined, the future fault state of the current running state is judged according to the current reconstruction error, the unknown fault state of the oil and gas equipment operation is estimated and judged, and an AI engine is used to generate the future fault type and future maintenance scheme according to the future fault state, running state data and historical maintenance scheme, to cover unknown fault and potential fault scenarios, identify risks in advance and plan maintenance schemes, change passive maintenance to active prevention, comprehensively improve the integrity and foresight of equipment fault diagnosis, and make equipment fault diagnosis more flexible. In the present application, the running state data is matched with the existing fault state, the future fault state is estimated and judged according to the matching result, and an AI engine is introduced to generate the maintenance scheme. The AI engine can quickly adapt to and autonomously optimize the diagnosis process without the need for a large amount of manual intervention, significantly enhance the adaptability of the system to various complex scenarios, and realize real-time and dynamic monitoring and intelligent diagnosis of the equipment running state. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 FIG. 1 is a flowchart of the oil and gas equipment diagnosis and maintenance method of an embodiment of the present application; Figure 2 FIG. 2 is a structural schematic diagram of the oil and gas equipment diagnosis and maintenance system of an embodiment of the present application; Figure 3 FIG. 3 is a structural schematic diagram of the oil and gas equipment diagnosis and maintenance device of an embodiment of the present application; Figure 4 FIG. 4 is a structural schematic diagram of the electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the above objectives, characteristics and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein, but rather, these embodiments are provided so as to more thoroughly and completely understand the present application. It should be understood that the drawings and embodiments of the present application are merely for illustrative purposes, and are not intended to limit the scope of protection of the present application.

[0019] It should be understood that each of the steps recited in the method embodiments of the present application can be performed in different orders, and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0020] The term "comprising" and variations thereof as used herein are open-ended, that is "including, but not limited to"; the term "based on" is "based, at least in part, on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optional" means "optional in at least some embodiments". Related definitions are given throughout the description. It should be noted that the concepts mentioned in the present application using "first", "second" and the like are merely used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modification of "one" or "multiple" mentioned in the present application is illustrative rather than limiting, and those skilled in the art should understand that, unless otherwise explicitly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between the devices in the embodiments of the present application are merely for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0023] As shown in Figure 1 and Figure 2 The oil and gas equipment diagnosis and maintenance method provided by the embodiments of the present application comprises: Step S1, preprocessing the operation data of the oil and gas equipment obtained in advance within a preset time period to obtain operation state data.

[0024] Specifically, before preprocessing the operational data, the operational data of the oil and gas equipment is acquired first. Various types of sensors can be installed at key components of the oil and gas equipment. For example, temperature sensors can be installed on the engine to acquire temperature data; pressure sensors can be installed at pipelines and pressure vessels to acquire pressure data; and vibration sensors can be installed at locations such as bearing housings of rotating equipment to collect vibration signals. The sensors convert physical quantities into electrical signals, obtaining the operational data, which can be expressed as X = {x1, x2, ..., x...}. i}, where x i This represents the operational data collected by the i-th sensor. It should be noted that operating conditions of oil and gas equipment change during operation, and equipment parameters may fluctuate under complex conditions. Using only the current operational data for equipment diagnosis can easily lead to false alarms or missed alarms. Therefore, it is necessary to obtain operational data over a period of time for subsequent fault diagnosis. The specific preset time period can be set according to the equipment quality (e.g., equipment operational stability, equipment grade). For example, if the oil and gas equipment has high operational stability, the preset time period can be set to a shorter period; if the equipment has low operational stability, the preset time period can be set to a longer period. Obtaining operational data over a longer period provides more data references for fault diagnosis, thereby improving the accuracy of fault diagnosis.

[0025] Preprocessing the operational data includes performing noise reduction, filtering, and normalization on the operational data sequentially to obtain the operational status data.

[0026] Specifically, median filtering or mean filtering is used to remove noise from the operational data, smooth the data, eliminate short-term fluctuations, extract the main trends of the data, and ensure data purity. Digital filtering algorithms are then used to filter the denoised operational data. Operational data with different dimensions are normalized to the same range, making the data from different parameters comparable, resulting in the final processed operational status data.

[0027] Specifically, the acquired operational data may contain noise or other issues, so it is necessary to preprocess the acquired operational data, such as by performing noise reduction and normalization, to ensure the purity of the data and obtain operational status data.

[0028] Step S2: Match the running status data with the fault case data pre-stored in the database.

[0029] Step S3: When a match is successful, use the AI ​​engine to generate the current maintenance plan based on the corresponding fault case data, historical maintenance plans, and the operating status data.

[0030] Specifically, a database is constructed in advance, and the previous failure case data of the oil and gas equipment (including type, corresponding failure operation data when failure occurs) and the corresponding historical maintenance scheme are pre-stored. When the current oil and gas equipment operation state data is obtained, the operation state data can be directly matched with the pre-stored failure case data in the database. For example, the operation state data has high similarity with the failure operation data when the failure type A occurs or the operation data before the failure type A occurs, and then the two are matched successfully, it is determined that the current oil and gas equipment operation or in the future operation exists the failure type A, and then the historical maintenance scheme corresponding to the failure type A is called as the current maintenance scheme to be generated as a reference. After obtaining the corresponding failure type and the corresponding historical maintenance scheme, the corresponding failure type, the corresponding historical maintenance scheme and the operation state data obtained by the current oil and gas equipment operation are input into the pre-trained AI engine, such as GPT model, to generate the adaptive current maintenance scheme for this operation failure, so as to perform targeted maintenance and correction on the oil and gas equipment. It should be noted that before the function AI engine is used to output the current maintenance scheme, the historical operation data of the oil and gas equipment, the failure type and the corresponding historical maintenance scheme are used to train the AI engine, so as to improve the scheme generation accuracy of the AI engine.

[0031] In step S4, when the matching fails, the current reconstruction error is obtained according to the operation state data, the future failure state of the current operation state is judged according to the current reconstruction error, and the future failure type and the future maintenance scheme are generated according to the future failure state, the operation state data and the historical maintenance scheme by using the AI engine.

[0032] Specifically, if the operation state data and the failure type stored in the database are not successfully matched, it indicates that the current oil and gas equipment operation is in a normal state or there is an unknown failure. Therefore, the current reconstruction error of the operation state data is obtained by using unsupervised technology, the deep features of the oil and gas equipment operation state data are mined, the position failure and / or future failure state are judged, and the unknown failure and / or future failure state are estimated. Then, the trained AI engine (such as the reinforcement learning model DQN (Deep Q-Network)) is used to generate the failure type and the future maintenance scheme of the unknown failure and / or future failure state according to the predicted future failure state, the operation state data and the called historical maintenance scheme for reference, and then the future operation condition of the oil and gas equipment is pre-diagnosed and maintained.

[0033] It should be noted that the AI engine in step S3 and the AI engine in step S4 can use the same AI engine to generate the maintenance scheme, for example, different generation templates (such as GPT models) are set in the AI engine, different maintenance schemes are generated according to different input data, and at the same time, since there is a certain difference between the input data of the AI engine in step S3 and the AI engine in step S4, two different trained AI engines can also be used to generate the maintenance scheme. For example, in step S3, the first AI engine is used to generate the current maintenance scheme, and in step S4, the second AI engine is used to generate the future maintenance scheme and the future fault type. Whether the same AI engine or different AI engines are used to generate the maintenance scheme, a large amount of fault case data and corresponding historical maintenance schemes are used to train the AI engine to improve the accuracy of the scheme generation.

[0034] In the embodiment, the running data of the oil and gas equipment in a preset time period is preprocessed to obtain running state data, which can ensure the purity of the data and provide a data basis for subsequent fault case data matching or future fault judgment. The running state data is matched with the pre-stored fault case data, and the oil and gas equipment operation is preliminarily judged according to the pre-stored fault case data. When the matching is successful, an AI engine is used to quickly generate an adaptive maintenance scheme according to the corresponding fault case data, historical maintenance scheme and running state data, to intelligently adapt the maintenance strategy based on historical experience and real-time state, shorten the maintenance scheme generation cycle, and further shorten the fault response cycle, thereby achieving efficient response to known faults. When the matching fails, the current reconstruction error is obtained according to the running state data, the running state data is deeply mined, the future fault state of the current running state is judged according to the current reconstruction error, the unknown fault state of the oil and gas equipment operation is estimated and judged, and an AI engine is used to generate the future fault type and the future maintenance scheme according to the future fault state, the running state data and the historical maintenance scheme, to cover unknown fault and potential fault scenarios, identify risks in advance and plan maintenance schemes, change passive maintenance to active prevention, and comprehensively improve the integrity and foresight of equipment fault diagnosis, making the equipment fault diagnosis more flexible. In the present application, the running state data is matched with the existing fault state, the future fault state is estimated and judged according to the matching result, and an AI engine is introduced to generate the maintenance scheme. The AI engine can quickly adapt to and autonomously optimize the diagnosis process without a large amount of manual intervention, significantly enhance the adaptability of the system to various complex scenarios, and realize real-time and dynamic monitoring and intelligent diagnosis of the equipment running state.

[0035] Optionally, the matching of the running state data with the pre-stored fault type comprises: performing working condition analysis on the running state data by using a clustering analysis method to obtain a current running working condition; Specifically, a clustering analysis method, such as a K-Means algorithm, is used to cluster the operating state data according to the device operating parameters, production levels, device load states, mining stages, and other related conditions to analyze the operating conditions and obtain the current operating condition. For example, according to device operating experience, the number of clusters is set to 3, and then the operating state data is clustered based on the K-Means algorithm objective function. The current operating condition of the oil and gas equipment is determined according to the data in each cluster. For example, if cluster 1 has the most data after clustering, the operating condition corresponding to cluster 1 is determined as the current operating condition of the oil and gas equipment, that is, the operating condition corresponding to the cluster with the most data is determined as the current operating condition.

[0036] After the clustering analysis method obtains multiple clusters, a feature combination analysis method can be used to identify the typical combination patterns of the data in the clusters to determine the operating condition corresponding to each cluster.

[0037] According to the pre-stored fault case data, a rule mining is performed to obtain multiple association rules. Specifically, the pre-stored fault case data, such as parameter records 30 minutes before and after the occurrence of bearing wear, motor overload, and pipeline leakage, is used to extract key feature combinations to obtain multiple association rules using, for example, the Apriori algorithm or the FP-Growth algorithm. For example, rule 1: {vibration anomaly (5-10 m / s²), full load}→bearing wear (confidence 85%), rule 2: {temperature too high (>80℃), current overload (>60A)}→motor overload (confidence 90%), and rule 3: {flow sudden drop (<20m³ / h), pressure fluctuation >10%}→pipeline leakage (confidence 88%) are used for subsequent matching of the current operating condition and the fault case data.

[0038] When the current operating condition triggers at least two association rules, the fault type corresponding to the association rule with the highest confidence is matched successfully with the current operating condition.

[0039] When the current operating condition triggers any association rule, the fault type corresponding to the association rule is matched successfully with the current operating condition.

[0040] Specifically, the running state data corresponding to the current operating condition is matched with the association rules, such as when the running state data is vibration acceleration = 7 m / s² (abnormal), load = 85% (full load), rule 1 is triggered, and the oil and gas equipment may have a bearing wear risk. When the current operating condition triggers multiple rules at the same time, the fault type corresponding to the association rule with the highest confidence is obtained, for example, rule 1 (confidence 85%) and rule 2 (confidence 90%) are triggered at the same time, and the “motor overload” fault type corresponding to rule 2 is matched with the current operating condition successfully, if only one association rule is triggered, the fault type corresponding to the association rule is matched with the current operating condition successfully, realizing the rapid determination of the known fault. Then the triggered rules and the corresponding fault confidence can be displayed in real time through the instrument panel, and the key parameters (such as the red highlighted “motor overload”) are marked with a heat map to assist the operation and maintenance personnel to quickly locate the fault root. Among them, the association rule is in the form of X→Y (X is the former and Y is the latter), the confidence represents the probability of containing Y in the information containing X, for example, rule 1: {vibration anomaly (5-10 m / s²), load full load}→bearing wear (confidence 85%), the confidence is the probability of the fault type being bearing wear under the condition of vibration anomaly (5-10 m / s²) and load full load.

[0041] It should be noted that the fault case data includes a fault type and equipment fault running data in a period corresponding to each fault type.

[0042] When the current operating condition does not trigger any of the association rules, it is determined that the matching fails.

[0043] Specifically, if the current operating condition does not trigger any association rule, it is determined that the matching fails, indicating that the current operating condition is beyond the coverage range of the historical fault cases, and there may be unknown faults or normal operation, and further judgment needs to be made on the future running fault state or unknown fault of the oil and gas equipment.

[0044] Optionally, the future fault state judgment of the current running state according to the current reconstruction error comprises: If the current reconstruction error is greater than a preset dynamic threshold, it is judged that there is an unknown fault in the current running and a warning is given; If the current reconstruction error is less than or equal to the preset dynamic threshold, it is judged that there is no unknown fault in the current running, the future fault occurrence probability is obtained according to the running state data, and the future fault state is judged based on the future fault occurrence probability.

[0045] Specifically, the preset dynamic threshold is a judgment limit that changes in real time with time, data distribution or model output, and can be dynamically adjusted according to various parameters and strategies, for example, according to historical reconstruction errors, mean values in a sliding window, standard deviations, error distributions in different time periods, working conditions (temperature, load), modes (start-up, standby), and recently observed error characteristics. The preset dynamic threshold is adaptively generated based on statistical characteristics of real-time or historical data, model performance, or even the running state of the oil and gas equipment. By comparing the current reconstruction error with the preset dynamic threshold, if the current reconstruction error is greater than the preset dynamic threshold, it is determined that there is an unknown fault in the current operation and a warning is given, thereby achieving early detection of the unknown fault state. If the current reconstruction error is less than or equal to the preset dynamic threshold, it indicates that the oil and gas equipment is currently in a normal operating state and there is no unknown fault. Further, based on the running state data, the future fault occurrence probability of the oil and gas equipment in future operation is estimated and judged to detect potential fault risks of the equipment in advance and prevent future fault states.

[0046] Optionally, the obtaining of the future fault occurrence probability according to the running state data and the judging of the future fault state based on the future fault occurrence probability comprises: According to the running state data and the preset fault level, the future fault occurrence probability under each fault level is obtained by using the Bayes rule.

[0047] Specifically, first, Gaussian distributions are calculated according to the fault levels (for example, normal, minor fault and serious fault) of the running state data and each running state data, and Gaussian distributions P(xi, y) are obtained, where xi represents the i-th running state data and y represents the fault level. Second, the future fault occurrence probability under each fault level is obtained by using the Naive Bayes algorithm, which is represented as: , where P(y|x1, …, x5) represents the future fault occurrence probability, and P(y) represents the prior probability, that is, the proportion of fault level y data in the pre-stored oil and gas equipment running data, such as P(y) = number of normal data / total sample data.

[0048] The future fault state is obtained according to the fault level corresponding to the maximum future fault occurrence probability.

[0049] Specifically, for all fault levels, the fault level with the maximum future fault occurrence probability is selected as the future fault state of the oil and gas equipment, for example, P(y minor fault|x1, …, x5) is the maximum, and the corresponding fault level (minor fault) is the future fault state.

[0050] Alternatively, according to historical experience values (from device manufacturer manuals, industry specifications, expert experience), limit indicators / safety thresholds (from device physical characteristics or operating limits), statistical analysis estimates (preliminary analysis of limited normal data collected early), relative change thresholds, sensitivity tests / parameter optimization, preset probability thresholds are set. And compare the calculated future failure probability with the preset probability threshold, when a certain failure probability is greater than the preset probability threshold (such as bearing wear probability > 85%), a multi-level early warning mechanism will be automatically triggered.

[0051] Optionally, the oil and gas equipment diagnosis and maintenance method further comprises: Updating the database according to the operating state data, the future failure case type and the future maintenance scheme.

[0052] Specifically, the future failure case type and the future maintenance scheme obtained by the AI engine are bound with the corresponding operating state data and stored in the database to update the pre-stored failure case data in the database, improve the diagnosis efficiency of subsequent oil and gas equipment operation failure, adapt to the changing equipment operating conditions and failure modes, and cope with the increasingly complex equipment diagnosis and maintenance requirements.

[0053] Optionally, the matching of the operating state data with the pre-stored failure type comprises: Generating a current operating curve according to the operating state data, setting an upper reference curve and a lower reference curve according to the upper and lower limit indicators of equipment operation, and generating a corresponding failure curve according to the pre-stored failure case data.

[0054] Specifically, a current operating curve can be generated according to a plurality of operating state data obtained at different time points, and a curve graph can be generated according to the upper reference curve and the lower reference curve, so that the change of the operating state data can be visually determined. A corresponding failure curve is generated according to the pre-stored failure case data, which provides a basis for subsequent failure matching.

[0055] When the current operating curve deviates from the range defined by the upper reference curve and the lower reference curve, the similarity between the current operating curve and the failure curve is obtained.

[0056] Specifically, when the current operating curve deviates from the defined range, for example, the data in the current operating curve at the same time point exceeds the range defined by the upper limit value of the upper reference curve at that time point and the lower limit value of the lower reference curve at that time point, it indicates that the current operation of the oil and gas equipment fluctuates, but it does not necessarily mean that a failure occurs. Therefore, the current operating curve is compared with a plurality of failure curves to obtain the similarity between them.

[0057] The fault type of the fault curve corresponding to the highest similarity greater than or equal to the preset similarity threshold is matched with the current operation data successfully.

[0058] If the similarities are all less than the preset similarity threshold, it is determined that the matching fails.

[0059] Optionally, the preprocessing of the pre-acquired operation data of the oil and gas equipment to obtain the operation state data comprises: Optionally, the obtaining of the current reconstruction error from the operation state data comprises: dimensionality reduction processing of the operation state data by using an encoder to obtain a low-dimensional feature vector; reconstruction of the low-dimensional feature vector by using a decoder to obtain reconstruction data; obtaining of the minimized current reconstruction error based on back propagation according to the operation state data and the reconstruction data.

[0060] Specifically, the high-dimensional operation state data (such as temperature, current, etc.) is compressed into a low-dimensional feature vector by using an encoder, and the operation state data is reconstructed by using a decoder to obtain reconstruction data, wherein the encoder and the decoder can be based on, for example, a Transformer architecture or a CNN architecture, which is not specifically limited herein. Then, the mean square error of the operation state data and the corresponding reconstruction data, i.e., the reconstruction error, is calculated, which is represented as: L = |x - x| | 2 wherein x represents the operation state data, x represents the reconstruction data, and L represents the reconstruction error. The average value of the reconstruction error L of all samples is taken as the optimization target of the training, i.e., the minimization of the reconstruction error. The model adjusts the weights of the encoder and the decoder through back propagation to make the reconstruction error L continuously decrease until the training is stopped when the stop condition is met, to obtain the current reconstruction error, wherein the stop condition can be that the maximum number of iterations is reached, or the global loss fluctuation range is less than a preset threshold.

[0061] Further, the oil and gas equipment diagnosis and maintenance method can be used for Figure 2The system shown is executed. The system comprises a data collector, a fault analyzer and an AI engine. The oil and gas equipment is connected with the data collector, the data collector is used for collecting the operation data of the oil and gas equipment and pre-processing the operation data to obtain the operation state data. The data collector is connected with the fault analyzer and the AI engine respectively, and is used for continuously inputting the historical operation state data and the operation state data at the current operation time into the fault analyzer for fault analysis, and inputting the operation state data at the current operation time into the AI engine. The output of the fault analyzer is connected with the AI engine, and the fault analyzer is used for matching the fault case data and predicting the future fault probability according to the historical operation state data and the operation state data at the current operation time, and inputting the matching result and / or the future fault probability and the historical operation state data obtained into the AI engine, and the AI engine is used for generating a maintenance scheme according to the operation state data at the current operation time and the output of the fault analyzer, so that the operating personnel adjusts the maintenance of the oil and gas equipment according to the maintenance scheme. The data collector, the fault analyzer and the AI engine form an intelligent learning and optimization fault diagnosis and maintenance learning closed loop, and each structure cooperates with each other and is dynamically optimized, so that the system can continuously evolve and adapt to the changing equipment operation condition and fault mode.

[0062] As shown in Figure 3 The oil and gas equipment diagnosis and maintenance device 400 provided by the embodiment of the present application comprises: A preprocessing module 310 is configured to pre-process the operation data of the oil and gas equipment in a preset time period to obtain operation state data. A matching module 320 is configured to match the operation state data with the pre-stored fault case data in a database. A generating module 330 is configured to, when the matching is successful, generate a current maintenance scheme according to the corresponding fault case data, historical maintenance schemes and the operation state data by using an AI engine. An estimating module 340 is configured to, when the matching fails, obtain a current reconstruction error according to the operation state data, judge a future fault state of the current operation state according to the current reconstruction error, and generate a future fault type and a future maintenance scheme according to the future fault state, the operation state data and the historical maintenance schemes by using the AI engine.

[0063] As shown in Figure 4 The electronic device 400 provided by the embodiment of the present application comprises a memory 410 and a processor 420; the memory 410 is configured to store a computer program; and the processor 420 is configured to, when the computer program is executed, implement the oil and gas equipment diagnosis and maintenance method as described above.

[0064] Or, an electronic device 400, comprising a memory 410 and a processor 420 coupled to the memory 410;The memory 410 is configured to store a computer program;The processor 420 is configured to execute the following operations when executing the computer program: The running data of the oil and gas equipment obtained in advance in the preset time period is preprocessed to obtain running state data; The running state data is matched with the pre-stored fault case data in the database; When the matching is successful, the AI engine is used to generate the current maintenance scheme according to the corresponding fault case data, the historical maintenance scheme and the running state data; When the matching fails, the current reconstruction error is obtained according to the running state data, the future fault state of the current running state is judged according to the current reconstruction error;The AI engine is used to generate the future fault type and the future maintenance scheme according to the future fault state, the running state data and the historical maintenance scheme.

[0065] The embodiment of the application provides a kind of computer readable storage medium, the storage medium is stored with computer program, when the computer program is executed by processor, the oil and gas equipment diagnosis and maintenance method as described above is realized.

[0066] Or, a non-volatile computer readable storage medium, the storage medium is stored with computer program, when the computer program is executed by processor, the processor executes the following operations: The running data of the oil and gas equipment obtained in advance in the preset time period is preprocessed to obtain running state data; The running state data is matched with the pre-stored fault case data in the database; When the matching is successful, the AI engine is used to generate the current maintenance scheme according to the corresponding fault case data, the historical maintenance scheme and the running state data; When the matching fails, the current reconstruction error is obtained according to the running state data, the future fault state of the current running state is judged according to the current reconstruction error;The AI engine is used to generate the future fault type and the future maintenance scheme according to the future fault state, the running state data and the historical maintenance scheme.

[0067] An electronic device 400, which can be a server or a client of the present application, will now be described, which is an example of a hardware device that can be applied to aspects of the present application. The electronic device 400 is intended to represent various forms of digital electronic computer devices such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device 400 can also represent various forms of mobile devices such as personal digital assistants, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0068] The electronic device 400 includes a computing unit that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). Various programs and data required for device operation can also be stored in the RAM. The computing unit, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0069] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), or a random access memory (RAM). In this application, the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application. In addition, the functional units in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0070] Although the present application is disclosed as above, the protection scope of the present application is not limited to this. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, and these changes and modifications will fall within the protection scope of the present application.

Claims

1. A method for diagnosing and maintaining oil and gas equipment, characterized in that, include: Preprocess the pre-acquired operating data of oil and gas equipment within a preset time period to obtain operating status data; The operational status data is matched with the fault case data pre-stored in the database; When a match is successful, the AI ​​engine is used to generate the current maintenance plan based on the corresponding fault case data, historical maintenance plans, and the operational status data. When a match fails, the current reconstruction error is obtained based on the running status data, and the future fault status is determined based on the current reconstruction error. Using the AI ​​engine, future fault types and future maintenance plans are generated based on the future fault states, the operational status data, and the historical maintenance plans.

2. The method for diagnosing and maintaining oil and gas equipment according to claim 1, characterized in that, The step of matching the operating status data with pre-stored fault case data includes: Cluster analysis is used to analyze the operating status data to obtain the current operating conditions. Based on the pre-stored fault case data, rule mining is performed to obtain multiple association rules; When the current operating condition triggers at least two of the association rules, it is determined that the fault type corresponding to the association rule with the highest confidence level is successfully matched with the current operating condition. When the current operating condition triggers any of the association rules, the fault type corresponding to the association rule is successfully matched with the current operating condition; If none of the association rules are triggered under the current operating conditions, the matching is deemed to have failed.

3. The method for diagnosing and maintaining oil and gas equipment according to claim 1, characterized in that, The step of determining the future fault state of the current operating state based on the current reconstruction error includes: If the current reconstruction error is greater than the preset dynamic threshold, it is determined that there is an unknown fault in the current operating state and an early warning is issued; If the current reconstruction error is less than or equal to the preset dynamic threshold, it is determined that there is no unknown fault in the current operating state. The probability of future fault occurrence is obtained based on the operating state data, and the future fault state is determined based on the probability of future fault occurrence.

4. The method for diagnosing and maintaining oil and gas equipment according to claim 3, characterized in that, The step of obtaining the probability of future failures based on the operating status data and determining the future failure state based on the probability of future failures includes: Using Bayes' theorem, the probability of future failures under each of the aforementioned fault levels is obtained based on the operational status data and preset fault levels. The future fault state is obtained based on the fault level corresponding to the highest probability of the future fault occurring.

5. The method for diagnosing and maintaining oil and gas equipment according to claim 1, characterized in that, After generating the future fault type and future maintenance plan based on the future fault state, the operating status data, and the historical maintenance plan, the method further includes: The database is updated based on the operational status data, the future fault types, and the future maintenance plan.

6. The method for diagnosing and maintaining oil and gas equipment according to claim 1, characterized in that, The step of matching the operating status data with pre-stored fault case data includes: The current operating curve is generated based on the operating status data. The upper and lower reference curves are set according to the upper and lower limit indicators of equipment operation. The corresponding fault curves are generated according to the pre-stored fault case data. When the current operating curve deviates from the defined range of the upper reference curve and the lower reference curve, the similarity between the current operating curve and each of the fault curves is obtained; If there is a similarity greater than or equal to a preset similarity threshold, then the fault type of the fault curve corresponding to the highest similarity is determined to be a successful match with the current running data; If all the similarities are less than the preset similarity threshold, the matching is deemed to have failed.

7. The method for diagnosing and maintaining oil and gas equipment according to claim 1, characterized in that, The process of obtaining the current reconstruction error based on the operating status data includes: The operating status data is reduced in dimensionality using an encoder to obtain a low-dimensional feature vector. The low-dimensional feature vector is reconstructed using a decoder to obtain reconstructed data; Based on the running status data and the reconstruction data, the minimized current reconstruction error is obtained through backpropagation.

8. A diagnostic and maintenance device for oil and gas equipment, characterized in that, include: The preprocessing module is used to preprocess the pre-acquired operating data of oil and gas equipment within a preset time period to obtain operating status data; The matching module is used to match the running status data with the fault case data pre-stored in the database; The generation module is used to generate the current maintenance plan by using the AI ​​engine when a match is successful, based on the corresponding fault case data, historical maintenance plans and the operation status data. The prediction module is used to obtain the current reconstruction error based on the running status data when the matching fails, and to make a future fault state judgment on the current running status based on the current reconstruction error. Using the AI ​​engine, future fault types and future maintenance plans are generated based on the future fault states, the operational status data, and the historical maintenance plans.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the oil and gas equipment diagnosis and maintenance method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the oil and gas equipment diagnosis and maintenance method as described in any one of claims 1 to 7.

Images