Petroleum workover rig operation monitoring control system

Abstract

The invention relates to the technical field of electric digital data processing, in particular to a petroleum workover rig operation monitoring and control system, which comprises a data acquisition unit, a data processing unit, a data processing unit, a data processing unit and a control unit, and is characterized in that the data acquisition unit is used for acquiring multi-dimensional time sequence data sequences of a plurality of petroleum workover rigs; the abnormity monitoring unit is used for carrying out abnormity monitoring on the time sequence data sequence of each dimension and determining a candidate sequence segment corresponding to the equipment fault; the correlation monitoring unit is used for carrying out fault correlation analysis on the candidate sequence segments of any two dimensions and obtaining the initial fault degree of each dimension according to an analysis result; the fault monitoring unit is used for determining the noise influence corresponding to each dimension of each petroleum workover rig and determining the target fault degree of each petroleum workover rig; and the early warning control unit is used for carrying out fault early warning on each petroleum workover rig according to the target fault degree. According to the invention, electric digital data processing is carried out on the data of the petroleum workover rig, interference caused by noise is eliminated, and the monitoring and control accuracy of the petroleum workover rig is improved.

Description

Technical Field

[0001] The invention relates to the technical field of electric digital data processing, and in particular to an operation monitoring and control system for a petroleum workover rig. Background Art

[0002] An oil well repair rig is a piece of equipment used in oil drilling and well repair operations. It is mainly used to repair and maintain oil wells. It includes drilling equipment, disassembly equipment and reinstallation equipment for oil wells. An oil well repair rig can handle problems in oil wells such as blockage, leakage or equipment damage.

[0003] The monitoring and control of oil well repair rig operation is to identify potential problems of equipment and wellhead by tracking various monitoring data during the operation, such as pressure, temperature, flow, etc., and issue alarms in time to ensure the safety of equipment and wellhead. Since equipment failure and some non-fault factors (such as groundwater level, pressure, sudden impact or vibration generated by the oil well repair rig during operation, instantaneous fluctuation of fluid flow in the wellbore, short-term electrical interference or pulse signal, etc.) will cause abnormal fluctuations in monitoring data, in the existing method of fault judgment by setting thresholds only, if the threshold is set large, when the fluctuation of monitoring data caused by the early stage of the fault is small, the equipment failure cannot be judged; if the threshold is set small, it is easy to alarm and shut down under non-fault factors caused by changes in environmental factors, affecting the efficient operation of the oil well repair rig. The above method of monitoring and controlling the oil well repair rig by setting thresholds has low accuracy and affects the working efficiency of the oil well repair rig. Summary of the invention

[0004] In order to solve the technical problem of how to improve the accuracy of monitoring and control of oil well repair rigs without affecting work efficiency, the purpose of the present invention is to provide an oil well repair rig operation monitoring and control system, and the technical solution adopted is as follows:

[0005] In one embodiment of the present application, a petroleum workover rig operation monitoring and control system is provided, the system comprising:

[0006] A data acquisition unit, used to acquire time series data of multiple dimensions of multiple oil well repairing rigs in the target oil field when they are working;

[0007] An anomaly monitoring unit, used to perform anomaly monitoring on the time series data sequence of each dimension and determine the candidate sequence segment corresponding to the equipment failure;

[0008] A correlation monitoring unit, used to perform fault correlation analysis on candidate sequence segments of any two dimensions, and obtain the initial fault degree of each dimension according to the analysis result;

[0009] A fault monitoring unit, used to determine the noise impact corresponding to each dimension of each oil workover rig, and adjust the initial fault degree according to the noise impact to determine the target fault degree of each oil workover rig;

[0010] The early warning control unit is used to judge the equipment fault of each oil well repairing machine according to the target fault degree, and to issue a fault early warning according to the judgment result.

[0011] In one embodiment of the present application, the multiple dimensions include wellhead pressure, liquid pressure, downhole temperature, and fluid flow rate.

[0012] In one embodiment of the present application, the abnormality monitoring unit includes:

[0013] A first anomaly monitoring subunit is used to obtain an anomaly score of each data in the time series data sequence of each dimension, and determine an abnormal sequence segment according to the anomaly score;

[0014] A second abnormality monitoring subunit is used to compare the data in the abnormal sequence segment with other data in the time series data sequence, and determine the possibility of equipment failure of each abnormal sequence segment according to the comparison result and the abnormality score;

[0015] The third abnormality monitoring subunit is configured to merge a plurality of the abnormal sequence segments according to the equipment failure possibility of each of the abnormal sequence segments to obtain the candidate sequence segment.

[0016] In one embodiment of the present application, comparing the data in the abnormal sequence segment with other data in the time series data sequence includes:

[0017] Acquire a first quantity and a second quantity, wherein the first quantity includes the quantity of data in the abnormal sequence segment and the quantity of other abnormal data in the time series data sequence except the abnormal sequence segment, and the second quantity includes the quantity of data in the abnormal sequence segment and the quantity of all data in the time series data sequence except the abnormal sequence segment;

[0018] A similarity comparison is performed on the first quantity and the second quantity.

[0019] In one embodiment of the present application, the correlation monitoring unit includes:

[0020] A first correlation monitoring subunit is used to calculate a correlation coefficient for candidate sequence segments of any two dimensions in each of the dimensions, where the correlation coefficient is used to indicate the correlation between the candidate sequence segments of any two dimensions;

[0021] The second correlation monitoring subunit is used to obtain the correlation between each dimension and other dimensions based on the correlation coefficient, and obtain the initial fault degree of each dimension according to the correlation between each dimension and other dimensions.

[0022] In one embodiment of the present application, obtaining the initial fault degree of each dimension according to the correlation between each dimension and other dimensions includes:

[0023] The equipment failure probability of the abnormal sequence segment that is the frontmost in the candidate sequence segments is used as the equipment failure probability of the candidate sequence segments;

[0024] Obtaining the time variance of the candidate sequence segments of each dimension;

[0025] The initial fault degree of each dimension is obtained according to the equipment fault probability and time variance of the candidate sequence segments, and the correlation between each dimension and other dimensions.

[0026] In one embodiment of the present application, the fault monitoring unit includes:

[0027] A first fault monitoring subunit is used to calculate the degree to which each dimension of each oil well repair machine is affected by the change of groundwater level and pressure to obtain the noise impact;

[0028] A second fault monitoring subunit is used to obtain the fault degree corresponding to each dimension of each oil well repairing machine according to the noise influence corresponding to each dimension of each oil well repairing machine;

[0029] The third fault monitoring subunit is used to obtain a target fault degree of each of the oil well repair rigs according to the fault degree corresponding to each dimension of each of the oil well repair rigs.

[0030] In one embodiment of the present application, the calculating the degree to which each dimension of each oil well repair rig is affected by changes in groundwater level and pressure includes:

[0031] Obtaining the number of oil well repair rigs in the target oil field and the duration of the candidate sequence segments of each dimension;

[0032] Perform similarity calculation on candidate sequence segments corresponding to each of the oil well repair machines and other oil well repair machines in the same dimension to obtain similarity of the candidate sequence segments;

[0033] According to the number of oil well repair rigs in the target oil field, the duration of the candidate sequence segments of each dimension, and the similarity of the candidate sequence segments, the degree to which each dimension is affected by the change in groundwater level and pressure is obtained.

[0034] In one embodiment of the present application, obtaining the fault degree corresponding to each dimension of each oil workover rig according to the noise impact corresponding to each dimension of each oil workover rig includes:

[0035] According to the noise impact corresponding to each dimension of each oil workover rig, the equipment failure impact corresponding to each dimension of each oil workover rig is obtained;

[0036] The initial fault degree is adjusted according to the impact of the equipment fault, so as to obtain the fault degree corresponding to each dimension of each oil workover rig.

[0037] In one embodiment of the present application, the equipment fault judgment is performed on each of the oil workover rigs according to the target fault degree, and a fault warning is performed according to the judgment result, including:

[0038] comparing the target fault degree of each of the oil workover rigs with a preset fault degree threshold;

[0039] When the target fault degree is greater than the preset fault degree threshold, the oil well repair rig with equipment failure is located, and a fault warning is issued to the oil well repair rig with equipment failure.

[0040] The present invention has the following beneficial effects:

[0041] First, a data acquisition unit acquires time series data sequences of multiple dimensions when multiple oil workover rigs in a target oil field are working; then, an abnormality monitoring unit performs abnormality monitoring on the time series data sequences of each dimension to determine the candidate sequence segments corresponding to the equipment fault; then, a correlation monitoring unit performs fault correlation analysis on the candidate sequence segments of any two dimensions, and obtains the initial fault degree of each dimension according to the analysis results; then, a fault monitoring unit determines the noise influence corresponding to each dimension of each oil workover rig, adjusts the initial fault degree according to the noise influence, and determines the target fault degree of each oil workover rig; finally, an early warning control unit performs equipment fault judgment on each oil workover rig according to the target fault degree, and issues a fault early warning according to the judgment result. In the present invention, abnormal monitoring is performed on the time series data sequences of each dimension, and candidate sequence segments of suspected equipment failure can be found; equipment failure usually appears abnormal fluctuations in data of multiple dimensions at the same time. By performing fault correlation analysis on the candidate sequence segments of any two dimensions, the existence of the equipment failure can be further verified, and the performance of the equipment failure in different dimensions can be determined. If abnormal fluctuations appear in multiple dimensions at the same time, and there is a significant correlation between these abnormal fluctuations, then these abnormal fluctuations are more likely to be caused by equipment failure; the noise influence corresponding to each dimension of each oil well repair rig is determined, and the initial fault degree is adjusted according to the noise influence, which can eliminate the interference caused by noise, thereby obtaining a more accurate fault degree assessment result, and improving the accuracy of monitoring and control of the oil well repair rig; in addition, false warning shutdown due to non-equipment failure factors will not occur, thereby ensuring the working efficiency of the oil well repair rig. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A schematic diagram of an implementation environment of a petroleum workover rig operation monitoring and control system provided by an embodiment of the present invention;

[0044] Figure 2 A schematic diagram of the structure of a petroleum workover rig operation monitoring and control system provided by one embodiment of the present invention;

[0045] Figure 3 A schematic diagram of the structure of an abnormality monitoring unit provided by an embodiment of the present invention;

[0046] Figure 4A schematic diagram of the structure of a correlation monitoring unit provided by one embodiment of the present invention;

[0047] Figure 5 A schematic diagram of the structure of a fault monitoring unit provided by an embodiment of the present invention;

[0048] Figure 6 A schematic flow chart of a method for monitoring and controlling an oil well repair rig operation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a petroleum well repair machine operation monitoring and control system proposed according to the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0050] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0051] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0052] The specific scheme of a petroleum workover rig operation monitoring and control system provided by the present invention is described in detail below in conjunction with the accompanying drawings.

[0053] See also Figure 1 , Figure 1 A schematic diagram of an implementation environment of a petroleum workover rig operation monitoring and control system provided by an embodiment of the present invention is shown in FIG. Figure 1As shown, the implementation environment includes a monitoring and control terminal 101 and a data acquisition terminal 102. The monitoring and control terminal 101 may be a terminal device configured with a target client, including but not limited to a laptop computer, tablet computer, handheld computer, PAD, desktop computer, etc. with local computing capabilities. The target client may be a client supporting monitoring and control of oil well repair rig operations, such as a video client, an instant messaging client, a browser client, etc.; the monitoring and control terminal 101 may, but is not limited to, communicate with the data acquisition terminal 102 through a network, and the network may include but is not limited to: a wired network, a wireless network, wherein the wired network includes: a local area network, a metropolitan area network, and a wide area network, and the wireless network includes: Bluetooth, WIFI, and other networks that implement wireless communication. The monitoring and control terminal 101 may, but is not limited to, include a human-computer interaction screen, a processor, and a memory. The human-computer interaction screen may, but is not limited to, be used to display the result of equipment fault judgment on the oil well repair rig. The processor may, but is not limited to, be used to respond to human-computer interaction operations, execute corresponding operations, or generate corresponding instructions.

[0054] As an optional method, data can be collected through the data collection terminal 102. The data collection terminal 102 may include, for example: a pressure sensor, a temperature sensor, and a flow sensor. The pressure sensor is used to collect wellhead pressure data and liquid pressure data, the temperature sensor is used to collect downhole temperature data, and the flow sensor is used to collect fluid flow data.

[0055] As an optional manner, the monitoring control terminal 101 may also be a server, which may be a single server, a server cluster composed of multiple servers, or a cloud server. The above is only an example, and this embodiment does not make any limitation to this.

[0056] As an optional method, a petroleum workover rig operation monitoring and control system may be installed on the monitoring and control terminal 101. The structure of the petroleum workover rig operation monitoring and control system is as follows:

[0057] A data acquisition unit, used to acquire time series data of multiple dimensions of multiple oil well repairing rigs in the target oil field when they are working;

[0058] An anomaly monitoring unit, used to perform anomaly monitoring on the time series data sequence of each dimension and determine the candidate sequence segment corresponding to the equipment failure;

[0059] A correlation monitoring unit, used to perform fault correlation analysis on candidate sequence segments of any two dimensions, and obtain the initial fault degree of each dimension according to the analysis result;

[0060] A fault monitoring unit, used to determine the noise impact corresponding to each dimension of each oil workover rig, and adjust the initial fault degree according to the noise impact to determine the target fault degree of each oil workover rig;

[0061] The early warning control unit is used to judge the equipment fault of each oil well repairing machine according to the target fault degree, and to issue a fault early warning according to the judgment result.

[0062] Based on the above method, abnormal monitoring of the time series data sequences in each dimension can be performed to find candidate sequence segments of suspected equipment failures; equipment failures usually appear abnormal fluctuations in data of multiple dimensions at the same time. By performing fault correlation analysis on candidate sequence segments of any two dimensions, the existence of equipment failures can be further verified, and the performance of equipment failures in different dimensions can be determined. If abnormal fluctuations appear in multiple dimensions at the same time, and there is a significant correlation between these abnormal fluctuations, then these abnormal fluctuations are more likely to be caused by equipment failures; determining the noise influence corresponding to each dimension of each oil well repair rig, and adjusting the initial fault degree according to the noise influence, can eliminate the interference caused by noise, thereby obtaining a more accurate fault degree assessment result, and improving the accuracy of monitoring and control of oil well repair rigs; in addition, there will be no false warning shutdown due to non-equipment failure factors, thereby ensuring the working efficiency of the oil well repair rig.

[0063] As an optional example, this embodiment does not limit the host of the above-mentioned oil well repair rig operation monitoring and control system. The above-mentioned oil well repair rig operation monitoring and control system can be installed on the monitoring and control terminal 101. For example, when the monitoring and control terminal 101 is a desktop computer, some or all of the components of the above-mentioned oil well repair rig operation monitoring and control system can be installed on the desktop computer.

[0064] In one embodiment of the present application, a petroleum workover rig operation monitoring and control system is provided. Figure 2 A structural diagram of a petroleum workover rig operation monitoring and control system provided by an embodiment of the present invention, see Figure 2 The oil well repair rig operation monitoring and control system includes the following units 210 to 250:

[0065] The data acquisition unit 210 is used to acquire time series data of multiple dimensions when multiple oil workover rigs in the target oil field are working.

[0066] Among them, multiple dimensions include wellhead pressure, liquid pressure, downhole temperature, and fluid flow rate.

[0067] In a target oil field, there are usually multiple oil workover rigs operating at the same time to improve operating efficiency and shorten development and maintenance cycles. The working data change trends of the same dimension of each oil workover rig usually show certain similarities. Therefore, the multi-dimensional data of each oil workover rig in the target oil field can be collected in real time through sensors to obtain the time series data sequence of each dimension.

[0068] For example, when multiple oil well repairing rigs are working in the target oil field, the above-mentioned data acquisition terminal 102 collects wellhead pressure data and liquid pressure data through pressure sensors, collects downhole temperature data through temperature sensors, and collects fluid flow data through flow sensors. Then, the data acquisition terminal 102 transmits the wellhead pressure data, liquid pressure data, downhole temperature data and fluid flow data to the data acquisition unit 210.

[0069] For example, the monitoring and control system of the oil well repair rig operation will issue a fault warning every hour if the monitoring data of each dimension in the same operation phase of the oil well repair rig does not exceed the normal range. Therefore, the time series data sequence of each dimension is 1 hour long, and data is obtained once per second. For example, the time series data sequence of the wellhead pressure dimension obtained can be Y={Y1,Y2,…,Y 3600}, where Y is the time series data sequence of the wellhead pressure dimension, which is a time series data sequence containing 3600 data; Y1 is the wellhead pressure data obtained in the first second; Y2 is the wellhead pressure data obtained in the second second; Y 3600 This is the wellhead pressure data obtained at the 3600th second.

[0070] The abnormality monitoring unit 220 is used to perform abnormality monitoring on the time series data sequence of each dimension and determine the candidate sequence segment corresponding to the equipment failure.

[0071] Among them, in the well repair operation, the replacement of faulty pipes and rods accounts for about 90%-95% of the entire wellbore maintenance workload, so the embodiments of the present application are mainly aimed at early warning of pumping pipes and pumping rods. The abnormal fluctuations in data caused by pipe and rod failures in oil well repair rig operations are usually continuous, and the abnormal fluctuations in data will not return to normal levels quickly unless the fault is repaired. The sudden impact or vibration generated by the well repair rig during operation, the instantaneous fluctuations in fluid flow in the wellbore, and the short-term electrical interference or pulse signals are all transient noise data. Therefore, the time series data sequences of each dimension are monitored for abnormalities, and long-term and continuous abnormal sequence segments can be merged into candidate sequence segments of suspected equipment failures. However, because the data corresponding to early equipment failures have a small volatility, the candidate sequence segments can include abnormal sequence segments caused by equipment failures, and can also include abnormal sequence segments corresponding to transient noise data caused by non-equipment failure factors.

[0072] The correlation monitoring unit 230 is used to perform fault correlation analysis on candidate sequence segments of any two dimensions, and obtain the initial fault degree of each dimension according to the analysis result.

[0073] Among them, when a pipe failure occurs during an oil well repair rig operation, abnormal fluctuations will occur in the data of each dimension at the same time, and there is a certain correlation (including positive and negative correlation) between the data of each dimension, such as: the wellhead pressure is usually directly related to the downhole liquid pressure (positive correlation); the pipe failure causes poor fluid flow, which may increase the pressure and reduce the flow (negative correlation). Therefore, based on the fact that equipment failures usually cause abnormal fluctuations in data of multiple dimensions at the same time, fault correlation analysis of candidate sequence segments of any two dimensions can further verify the existence of equipment failures and determine the performance of equipment failures in different dimensions. If abnormal fluctuations occur in multiple dimensions at the same time, and there is a significant correlation between these abnormal fluctuations, then these abnormal fluctuations are more likely to be caused by equipment failures.

[0074] The fault monitoring unit 240 is used to determine the noise impact corresponding to each dimension of each oil workover rig, and adjust the initial fault degree according to the noise impact to determine the target fault degree of each oil workover rig.

[0075] Among them, when the groundwater level and pressure of the target oil field change, it will cause abnormal fluctuations in the data of various dimensions during the operation of the oil well repair rig in the oil field. The changes in the groundwater level and pressure of the oil field are usually continuous, so it is necessary to further analyze whether the candidate sequence segments are affected by the changes in groundwater level and pressure.

[0076] For example, the groundwater level and pressure changes in different areas of the same target oil field usually affect each other. This is because the groundwater and oil and gas in the target oil field are usually connected. When the pressure in a certain area changes, it may cause the fluid to move between different areas, thereby affecting the water level and pressure in other areas. That is, there is a certain similarity in the water level and water pressure change trends in each area, and it is a low-probability event that multiple oil well repair rigs in the same target oil field have the same fault at the same time during the operation process. Therefore, it is necessary to determine the noise impact corresponding to each dimension of each oil well repair rig, and adjust the initial fault degree according to the noise impact, so as to eliminate the interference caused by noise, thereby obtaining a more accurate fault degree assessment result, and improving the accuracy of oil well repair rig monitoring and control; in addition, there will be no false warning shutdown due to non-equipment failure factors, which ensures the working efficiency of the oil well repair rig.

[0077] The early warning control unit 250 is used to judge the equipment fault of each oil well repairing machine according to the target fault degree, and to issue a fault early warning according to the judgment result.

[0078] In one embodiment of the present application, see Figure 3 , Figure 3 The following is a schematic diagram of the structure of an abnormality monitoring unit provided by one embodiment of the present invention. Figure 3 As shown, the abnormality monitoring unit 220 includes:

[0079] The first anomaly monitoring subunit 221 is used to obtain an anomaly score of each data in the time series data sequence of each dimension, and determine an abnormal sequence segment according to the anomaly score;

[0080] The second abnormality monitoring subunit 222 is used to compare the data in the abnormal sequence segment with other data in the time series data sequence, and determine the possibility of equipment failure of each abnormal sequence segment according to the comparison result and the abnormality score;

[0081] The third abnormality monitoring subunit 223 is configured to merge a plurality of the abnormal sequence segments according to the equipment failure possibility of each of the abnormal sequence segments to obtain the candidate sequence segment.

[0082] In one embodiment of the present application, comparing the data in the abnormal sequence segment with other data in the time series data sequence includes:

[0083] Acquire a first quantity and a second quantity, wherein the first quantity includes the quantity of data in the abnormal sequence segment and the quantity of other abnormal data in the time series data sequence except the abnormal sequence segment, and the second quantity includes the quantity of data in the abnormal sequence segment and the quantity of all data in the time series data sequence except the abnormal sequence segment;

[0084] A similarity comparison is performed on the first quantity and the second quantity.

[0085] Exemplarily, the first anomaly monitoring subunit 221 can use the LOF algorithm (Local Outlier Factor algorithm) to perform anomaly detection on the time series data sequence of each dimension to obtain an anomaly score (i.e., LOF score) for each data, wherein the higher the anomaly score, the more abnormal the data. Then, the anomaly score is compared with a preset anomaly score threshold, and the adjacent abnormal data corresponding to the anomaly score exceeding the preset anomaly score threshold is used as an abnormal sequence segment.

[0086] Exemplarily, using the LOF algorithm to perform anomaly detection on the time series data sequence of each dimension may include the following steps:

[0087] A. Data preprocessing: cleaning the time series data to remove missing values, noise processing (such as smoothing), etc. Standardization / normalization: standardizing or normalizing the time series data to eliminate the influence of different dimensions;

[0088] B. Setting parameters: In the LOF algorithm, a quantitative parameter needs to be determined, which represents the number of nearest neighbors considered when calculating the local density. The selection of this quantitative parameter should be determined according to the characteristics and distribution of the data; select an appropriate distance measurement method (such as Euclidean distance, Manhattan distance, etc.) to calculate the distance between data points;

[0089] C. Calculate the nearest neighbor distance: For each point in the time series data sequence, calculate its distance to all other points, and find the point whose distance to the point corresponds to the quantity parameter. The point between the point and the point corresponding to the quantity parameter can be called the k distance;

[0090] D. Calculate the reachable distance: For a target point and its neighbor point (a neighbor point is a point whose distance from the target point is less than or equal to the k distance of the target point), calculate the reachable distance from the neighbor point to the target point. The reachable distance is the larger value of the k distance of the neighbor point and the distance between the target point and the neighbor point.

[0091] E: Calculate the local reachability density: The local reachability density of the target point is the reciprocal of the average of the reciprocals of the reachability distances of the nearest neighbors corresponding to the number parameter of the target point;

[0092] F. Calculate LOF score: The LOF score of a target point is the average of the ratio of the local reachability density of its nearest neighbors corresponding to its quantity parameter to the local reachability density of the target point.

[0093] In the above embodiment, the LOF algorithm can capture abnormal points in local areas and is very sensitive to data points that appear normal in the overall data distribution but are significantly different locally. By applying LOF separately in each dimension, it is possible to independently evaluate whether different types of data are abnormal, thereby avoiding the situation where anomalies in a single dimension are masked. The LOF score for each data provides a standard for quantifying the degree of abnormality, which is helpful for subsequent processing and analysis.

[0094] In the above embodiment, adjacent abnormal data points are combined into abnormal sequence segments, taking into account that the fault may not be a problem of a single data point, but is composed of a series of related events; by constructing abnormal sequence segments instead of viewing each abnormal data point in isolation, the false alarm rate caused by accidental factors can be effectively reduced.

[0095] For example, since the abnormal data fluctuation caused by the early failure of the pipe rod may be small, some data can be easily identified as normal data, resulting in the abnormal sequence segment caused by the failure being divided into several segments, so it is necessary to preliminarily obtain the candidate sequence segment. The second abnormal monitoring subunit 222 obtains the equipment failure possibility corresponding to the first abnormal sequence segment starting from the first abnormal sequence segment in the time series data sequence of the i-th dimension. The representation method of the equipment failure possibility corresponding to the first abnormal sequence segment may include:

[0096]

[0097] Among them, R1 is the probability of equipment failure corresponding to the first abnormal sequence segment; S ′ 1 is the first number, which includes the number of data in the first abnormal sequence segment in the time series data sequence of the i-th dimension and the number of all abnormal data thereafter; S1 is the second number, which includes the number of data in the first abnormal sequence segment in the time series data sequence of the i-th dimension and the number of all data thereafter; P 1,avg It is the mean of the anomaly scores of the first anomaly sequence segment and all subsequent data in the time series data sequence of the i-th dimension; norm() is a linear normalization function.

[0098] It should be noted that the data after the pipe rod failure should be abnormal data. Therefore, if the first abnormal sequence segment is the time of the failure, the subsequent data should be abnormal data. ′ 1 should be equal to S1, The closer it is to 1, the 1,avg The larger the value is, the greater the abnormality of the data after the first abnormal sequence segment is, and the more likely it is that there is a possibility of equipment failure. and P 1,avg The product of represents the probability of equipment failure corresponding to the first abnormal sequence segment. The larger R1 is, the more the first abnormal sequence segment is consistent with the abnormality generated when the equipment failure occurs; the smaller R1 is, the more likely the first abnormal sequence segment is an abnormality caused by noise. Similarly, starting from the second abnormal sequence segment, the probability of equipment failure corresponding to the second abnormal sequence segment is obtained, and the probability of equipment failure corresponding to each abnormal sequence segment is obtained in turn to form a fault starting probability sequence {R1, R2, …, R r}, where r is the number of abnormal sequence segments in the time series data sequence of the i-th dimension.

[0099] In the above embodiment, each abnormal sequence segment is assigned a probability value indicating that it represents a device failure, which helps to further distinguish which abnormalities are closer to the device failure.

[0100] Exemplarily, the data in the abnormal sequence segment caused by the fault that is partly mistakenly identified as normal data is often small, and its abnormal score is larger than the real normal data. Therefore, when the third abnormality monitoring subunit 223 gradually merges multiple abnormal sequence segments from back to front in the time series data sequence of the i-th dimension, when merging into an abnormal sequence segment caused by transient noise that is far away from the abnormal sequence segment caused by the equipment fault, it will cause a large change in the possibility of equipment fault. Therefore, in the sequence corresponding to the possibility of equipment fault, the absolute value of the difference between the possibilities of two adjacent equipment faults is calculated from back to front, such as: |R r -R r-1 |、|R r-1 -R r-2 |, ..., when | R j -R j-1 | is greater than the preset threshold of 0.3, the jth abnormal sequence segment and the sequence segment consisting of all subsequent data in the time series data sequence of the i-th dimension are taken as candidate sequence segments, where R r is the probability of equipment failure corresponding to the rth abnormal sequence segment, R r-1 is the probability of equipment failure corresponding to the r-1th abnormal sequence segment, R r-2 is the probability of equipment failure corresponding to the r-2th abnormal sequence segment, R j is the probability of equipment failure corresponding to the jth abnormal sequence segment, R j-1 is the probability of equipment failure corresponding to the j-1th abnormal sequence segment. j The equipment failure probability R of the candidate sequence segment is used as the candidate sequence segment, and the larger R is, the more abnormal the candidate sequence segment is. Using the above method, the candidate sequence segment in the time series data sequence of each dimension can be obtained.

[0101] In the above embodiment, by gradually merging abnormal sequence segments from back to front, the correlation between abnormal sequence segments and the development trend of faults can be better understood; through multi-dimensional data analysis and anomaly detection, the signs of equipment failure are effectively identified and quantified, thereby improving the accuracy and reliability of monitoring and control.

[0102] In one embodiment of the present application, see Figure 4 , Figure 4 FIG. 1 is a schematic diagram of a correlation monitoring unit provided by an embodiment of the present invention. Figure 4 As shown, the correlation monitoring unit 230 includes:

[0103] A first correlation monitoring subunit 231 is used to calculate a correlation coefficient for candidate sequence segments of any two dimensions in each dimension, where the correlation coefficient is used to indicate the correlation between the candidate sequence segments of any two dimensions;

[0104] The second correlation monitoring subunit 232 is used to obtain the correlation between each dimension and other dimensions based on the correlation coefficient, and obtain the initial fault degree of each dimension according to the correlation between each dimension and other dimensions.

[0105] In an embodiment of the present application, obtaining the initial fault degree of each dimension according to the correlation between each dimension and other dimensions includes:

[0106] The equipment failure probability of the abnormal sequence segment that is the frontmost in the candidate sequence segments is used as the equipment failure probability of the candidate sequence segments;

[0107] Obtaining the time variance of the candidate sequence segments of each dimension;

[0108] The initial fault degree of each dimension is obtained according to the equipment fault probability and time variance of the candidate sequence segments, and the correlation between each dimension and other dimensions.

[0109] Exemplarily, when the noise data is in an abnormal sequence segment caused by equipment failure, it will affect the correlation between data of different dimensions and reduce the analysis accuracy of the fault data in the candidate sequence segment. Therefore, the first correlation monitoring subunit 231 calculates the correlation coefficient PCCs(i,y) of the candidate sequence segment in the time series data sequence of the i-th and y-th dimensions during any oil well repair machine operation. Among them, the correlation coefficient PCCs(i,y) can be a Pearson correlation coefficient, and the closer PCCs(i,y) is to 1, the positive correlation is represented, and the closer PCCs(i,y) is to -1, the negative correlation is represented.

[0110] For example, when the noise data is close to the abnormal sequence segment caused by the equipment failure and is merged into the candidate sequence segment, the duration of the candidate sequence segments in different dimensions will be inconsistent, which will also reduce the analysis accuracy of the fault data in the candidate sequence segment. Therefore, the second correlation monitoring subunit 232 calculates the initial fault degree of the candidate sequence segment in the time series data sequence of the i-th dimension during any oil well repair machine operation. The initial fault degree can be expressed as:

[0111]

[0112] Where F is the initial fault degree of the candidate sequence segment, R is the possibility of equipment failure of the candidate sequence segment, V is the time variance of the candidate sequence segment in the time series data sequence of all dimensions, n is the number of dimensions in each oil well repair rig operation process, PCCs(i,y) is the correlation coefficient of the candidate sequence segment in the time series data sequence of the i-th and y-th dimensions, and || is the absolute value.

[0113] It should be noted that the larger R is, the more likely it is that the data fluctuation is caused by equipment failure, and equipment failure will cause abnormal fluctuations in data of different dimensions at the same time. Therefore, the duration of candidate sequence segments in time series data sequences of different dimensions should be similar, that is, the smaller V is, the more likely it is a failure, and the abnormal fluctuations of data of different dimensions caused by the failure are correlated. Therefore The larger it is, the more likely it is a device failure, so R, The product of the three represents the initial fault degree. The larger the initial fault degree F is, the more likely it is an equipment fault in the data analysis of a single oil well repair rig operation process. Similarly, in each oil well repair rig operation process, the initial fault degree of the candidate sequence segment in the time series data sequence of each dimension is obtained.

[0114] In one embodiment of the present application, see Figure 5 , Figure 5 A schematic diagram of the structure of a fault monitoring unit provided by an embodiment of the present invention. Figure 5 As shown, the fault monitoring unit 240 includes:

[0115] The first fault monitoring subunit 241 is used to calculate the degree to which each dimension of each oil well repair machine is affected by the change of groundwater level and pressure to obtain the noise impact;

[0116] The second fault monitoring subunit 242 is used to obtain the fault degree corresponding to each dimension of each oil well repairing machine according to the noise impact corresponding to each dimension of each oil well repairing machine;

[0117] The third fault monitoring subunit 243 is used to obtain a target fault degree of each of the oil workover rigs according to the fault degree corresponding to each dimension of each of the oil workover rigs.

[0118] In one embodiment of the present application, calculating the degree to which each dimension of each oil well repair rig is affected by changes in groundwater level and pressure includes:

[0119] Obtaining the number of oil well repair rigs in the target oil field and the duration of the candidate sequence segments of each dimension;

[0120] Perform similarity calculation on candidate sequence segments corresponding to each of the oil well repair machines and other oil well repair machines in the same dimension to obtain similarity of the candidate sequence segments;

[0121] According to the number of oil well repair rigs in the target oil field, the duration of the candidate sequence segments of each dimension, and the similarity of the candidate sequence segments, the degree to which each dimension is affected by the change in groundwater level and pressure is obtained.

[0122] In one embodiment of the present application, according to the noise impact corresponding to each dimension of each oil workover rig, the fault degree corresponding to each dimension of each oil workover rig is obtained, including:

[0123] According to the noise impact corresponding to each dimension of each oil workover rig, the equipment failure impact corresponding to each dimension of each oil workover rig is obtained;

[0124] The initial fault degree is adjusted according to the impact of the equipment fault, so as to obtain the fault degree corresponding to each dimension of each oil workover rig.

[0125] For example, the first fault monitoring subunit 241 calculates the influence degree G of the change of groundwater level and pressure on the i-th dimension during the operation of the x-th oil well repairing machine in the oil field. x,i , impact degree G x,i The representation can be:

[0126]

[0127] Among them, m is the number of oil well repair rigs in the target oil field, H x With H t are respectively the duration of the candidate sequence segments in the time series data sequence of the i-th dimension during the x-th and t-th oil well repair machine operation processes, and DTW(x,t) is the candidate sequence segment similarity of the candidate sequence segments in the time series data sequence of the i-th dimension during the x-th and t-th oil well repair machine operation processes; in this embodiment, the candidate sequence segment similarity is represented by the DTW (Dynamic Time Warping) distance, and the smaller the DTW distance, the more similar the candidate sequence segments are.

[0128] It should be noted that the changes in groundwater level and pressure will cause abnormal fluctuations in the operation of different oil well repairing machines at the same time, and the abnormal fluctuations caused are similar. x -H t The smaller | is, the more abnormal fluctuations occur during the operation of different oil well repair machines, the greater the impact of groundwater level and pressure changes, and the less likely it is a fault; the smaller DTW(x, t) is, the more similar the candidate sequence segments in the time series data sequence of the same dimension in the operation of different oil well repair machines are, the greater the impact of groundwater level and pressure changes, and the less likely it is a fault. Therefore, the normalized inverse proportional value of the product of the two is taken as exp(-|H x -H t |×DTW(x,t)), and then take the average to represent the degree of influence of groundwater level and pressure changes; the degree of influence G x,iThe larger it is, the more likely it is that the abnormal fluctuation is caused by changes in groundwater level and pressure, and the less likely it is that the abnormal fluctuation is caused by equipment failure.

[0129] In the above embodiment, by considering the impact of changes in groundwater level and pressure on the operation process of the oil well repair machine, it is possible to more accurately identify which data fluctuations are caused by environmental factors rather than equipment failures; and eliminate the impact of non-fault factors (such as environmental changes) on the data, thereby improving the accuracy of fault detection.

[0130] For example, the second fault monitoring subunit 242 calculates the fault degree F corresponding to the i-th dimension during the x-th oil well repair machine operation. ′ x,i , F ′ x,i The representation can be:

[0131] F ′ x,i =F x,i ×(1-G x,i )

[0132] Among them, F x,i is the fault degree corresponding to the i-th dimension during the x-th oil workover rig operation.

[0133] It should be noted that F x,i The larger the value, the more likely it is that the abnormal fluctuation is caused by equipment failure, and G x,i The larger the value, the more likely the abnormal fluctuation is caused by the change of groundwater level and pressure. Therefore, 1-G x,i With F x,i The product of represents the fault degree corresponding to the i-th dimension, F ′ x,i The larger it is, the more likely it is a device failure in the comprehensive analysis.

[0134] In the above embodiment, by calculating the fault degree corresponding to each dimension, the severity of the equipment fault in the candidate sequence segment can be quantitatively evaluated to provide a basis for subsequent decision-making; quantification of the fault degree can help take measures before the fault worsens to reduce the risk of equipment damage.

[0135] For example, the third fault monitoring subunit 243 calculates the target fault degree Q of the equipment fault in the xth oil well repair machine operation process. x , Q x The representation can be:

[0136]

[0137] Among them, n is the number of dimensions in each oil well repair rig operation process, and norm() is the linear normalization function.

[0138] It should be noted that the fault degree corresponding to the dimension of the candidate sequence segment in the time series data sequence of all dimensions in each oil well repair rig operation process is calculated and then normalized to represent the target fault degree Q of the equipment fault in each oil well repair rig operation process. x , target fault level Q x The larger it is, the more likely it is that there will be equipment failure during the operation of the oil well repair rig.

[0139] In the above embodiment, by calculating the target failure degree of equipment failure, a quantitative risk assessment can be given to help decision makers understand the probability of equipment failure; based on the target failure degree of equipment failure, the accuracy of early warning can be improved and the occurrence of false alarms can be reduced.

[0140] In one embodiment of the present application, according to the target fault degree, equipment fault judgment is performed on each of the oil well repairing machines, and fault warning is performed according to the judgment result, including:

[0141] comparing the target fault degree of each of the oil workover rigs with a preset fault degree threshold;

[0142] When the target fault degree is greater than the preset fault degree threshold, the oil well repair rig with equipment failure is located, and a fault warning is issued to the oil well repair rig with equipment failure.

[0143] For example, if Q x If Q is greater than the preset fault level threshold of 0.8, it is determined that the x-th oil well repair rig has equipment failure during the operation process in the current detection period. x If the value is less than or equal to the preset threshold of 0.8, it is determined that there is no equipment failure in the operation of the x-th oil workover rig during the current detection period. When there is an equipment failure in the operation of the x-th oil workover rig during the current detection period, the x-th oil workover rig is located and the fault warning is issued, and personnel are arranged to conduct a detailed inspection of the x-th oil workover rig to determine whether the oil workover rig needs to be fully shut down for maintenance or partially shut down for maintenance to ensure operation efficiency.

[0144] Based on the above embodiments, it can be known that by performing abnormal monitoring on the time series data sequences of each dimension, candidate sequence segments of suspected equipment failure can be found; equipment failure usually appears abnormal fluctuations in data of multiple dimensions at the same time. By performing fault correlation analysis on the candidate sequence segments of any two dimensions, the existence of the equipment failure can be further verified, and the performance of the equipment failure in different dimensions can be determined. If abnormal fluctuations appear in multiple dimensions at the same time, and there is a significant correlation between these abnormal fluctuations, then these abnormal fluctuations are more likely to be caused by equipment failure; determining the noise influence corresponding to each dimension of each oil well repair rig, and adjusting the initial fault degree according to the noise influence, the interference caused by the noise can be eliminated, thereby obtaining a more accurate fault degree assessment result, and improving the accuracy of monitoring and control of the oil well repair rig; in addition, there will be no false warning shutdown due to non-equipment failure factors, thereby ensuring the working efficiency of the oil well repair rig.

[0145] In one embodiment of the present application, a method for monitoring and controlling an oil well repair rig operation is also provided. Figure 6 A flow chart of a method for monitoring and controlling a petroleum workover rig operation provided by an embodiment of the present invention is shown in FIG. Figure 6 , the method comprises the following steps:

[0146] S610, obtaining time series data of multiple dimensions of multiple oil workover rigs in the target oil field when they are working;

[0147] S620, performing abnormal monitoring on the time series data sequence of each dimension to determine the candidate sequence segment corresponding to the equipment failure;

[0148] S630, performing fault correlation analysis on candidate sequence segments of any two dimensions, and obtaining the initial fault degree of each dimension according to the analysis result;

[0149] S640, determining the noise impact corresponding to each dimension of each oil workover rig, and adjusting the initial fault degree according to the noise impact to determine the target fault degree of each oil workover rig;

[0150] S650: According to the target fault degree, equipment fault judgment is performed on each of the oil workover rigs, and fault warning is issued according to the judgment result.

[0151] By using the oil well repair rig operation monitoring and control method in the embodiment of the present application, the time series data sequences of each dimension are monitored for abnormalities, and candidate sequence segments of suspected equipment failures can be found; equipment failures usually appear abnormal fluctuations in data of multiple dimensions at the same time. By performing fault correlation analysis on candidate sequence segments of any two dimensions, the existence of the equipment failure can be further verified, and the performance of the equipment failure in different dimensions can be determined. If abnormal fluctuations appear in multiple dimensions at the same time, and there is a significant correlation between these abnormal fluctuations, then these abnormal fluctuations are more likely to be caused by equipment failure; determining the noise influence corresponding to each dimension of each oil well repair rig, and adjusting the initial fault degree according to the noise influence, the interference caused by the noise can be eliminated, thereby obtaining a more accurate fault degree assessment result, and improving the accuracy of the monitoring and control of the oil well repair rig; in addition, there will be no false warning shutdown due to non-equipment failure factors, thereby ensuring the working efficiency of the oil well repair rig.

[0152] The specific embodiments of the oil well repair rig operation monitoring and control method in the present application can refer to the examples shown in the above-mentioned oil well repair rig operation monitoring and control system, which will not be repeated here in this example.

[0153] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0154] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A monitoring and control system for oil well repair rig operation, characterized in that: The system comprises: A data acquisition unit, used to acquire time series data of multiple dimensions of multiple oil well repairing rigs in the target oil field when they are working; An anomaly monitoring unit, used to perform anomaly monitoring on the time series data sequence of each dimension and determine the candidate sequence segment corresponding to the equipment failure; A correlation monitoring unit, used to perform fault correlation analysis on candidate sequence segments of any two dimensions, and obtain the initial fault degree of each dimension according to the analysis result; A fault monitoring unit, used to determine the noise impact corresponding to each dimension of each oil workover rig, and adjust the initial fault degree according to the noise impact to determine the target fault degree of each oil workover rig; The early warning control unit is used to judge the equipment fault of each oil well repairing machine according to the target fault degree, and to issue a fault early warning according to the judgment result.

2. The oil well repair rig operation monitoring and control system according to claim 1, characterized in that: The multiple dimensions include wellhead pressure, liquid pressure, downhole temperature, and fluid flow rate.

3. The oil well repair rig operation monitoring and control system according to claim 1, characterized in that: The abnormality monitoring unit comprises: A first anomaly monitoring subunit is used to obtain an anomaly score of each data in the time series data sequence of each dimension, and determine an abnormal sequence segment according to the anomaly score; A second abnormality monitoring subunit is used to compare the data in the abnormal sequence segment with other data in the time series data sequence, and determine the possibility of equipment failure of each abnormal sequence segment according to the comparison result and the abnormality score; The third abnormality monitoring subunit is configured to merge a plurality of the abnormal sequence segments according to the equipment failure possibility of each of the abnormal sequence segments to obtain the candidate sequence segment.

4. The oil well repair rig operation monitoring and control system according to claim 3, characterized in that: Comparing the data in the abnormal sequence segment with other data in the time series data sequence includes: Acquire a first quantity and a second quantity, wherein the first quantity includes the quantity of data in the abnormal sequence segment and the quantity of other abnormal data in the time series data sequence except the abnormal sequence segment, and the second quantity includes the quantity of data in the abnormal sequence segment and the quantity of all data in the time series data sequence except the abnormal sequence segment; A similarity comparison is performed on the first quantity and the second quantity.

5. The oil well repair rig operation monitoring and control system according to claim 3, characterized in that: The correlation monitoring unit comprises: A first correlation monitoring subunit is used to calculate a correlation coefficient for candidate sequence segments of any two dimensions in each of the dimensions, where the correlation coefficient is used to indicate the correlation between the candidate sequence segments of any two dimensions; The second correlation monitoring subunit is used to obtain the correlation between each dimension and other dimensions based on the correlation coefficient, and obtain the initial fault degree of each dimension according to the correlation between each dimension and other dimensions.

6. The oil well repair rig operation monitoring and control system according to claim 5, characterized in that: The obtaining of the initial fault degree of each dimension according to the correlation between each dimension and other dimensions includes: The equipment failure probability of the abnormal sequence segment that is the frontmost in the candidate sequence segments is used as the equipment failure probability of the candidate sequence segments; Obtaining the time variance of the candidate sequence segments of each dimension; The initial fault degree of each dimension is obtained according to the equipment fault probability and time variance of the candidate sequence segments, and the correlation between each dimension and other dimensions.

7. The oil well repair rig operation monitoring and control system according to claim 6, characterized in that: The fault monitoring unit comprises: A first fault monitoring subunit is used to calculate the degree to which each dimension of each oil well repair machine is affected by the change of groundwater level and pressure to obtain the noise impact; A second fault monitoring subunit is used to obtain the fault degree corresponding to each dimension of each oil well repairing machine according to the noise influence corresponding to each dimension of each oil well repairing machine; The third fault monitoring subunit is used to obtain a target fault degree of each of the oil well repair rigs according to the fault degree corresponding to each dimension of each of the oil well repair rigs.

8. The oil well repair rig operation monitoring and control system according to claim 7, characterized in that: The calculating the degree to which each dimension of each oil well repairing machine is affected by the change of groundwater level and pressure includes: Obtaining the number of oil well repair rigs in the target oil field and the duration of the candidate sequence segments of each dimension; Perform similarity calculation on candidate sequence segments corresponding to each of the oil well repair machines and other oil well repair machines in the same dimension to obtain similarity of the candidate sequence segments; According to the number of oil well repair rigs in the target oil field, the duration of the candidate sequence segments of each dimension, and the similarity of the candidate sequence segments, the degree to which each dimension is affected by the change in groundwater level and pressure is obtained.

9. The oil well repair rig operation monitoring and control system according to claim 7, characterized in that: The step of obtaining the fault degree corresponding to each dimension of each oil well repairing machine according to the noise impact corresponding to each dimension of each oil well repairing machine includes: According to the noise impact corresponding to each dimension of each oil workover rig, the equipment failure impact corresponding to each dimension of each oil workover rig is obtained; The initial fault degree is adjusted according to the impact of the equipment fault, so as to obtain the fault degree corresponding to each dimension of each oil workover rig.

10. The oil well repair rig operation monitoring and control system according to claim 1, characterized in that: According to the target fault degree, the equipment fault judgment is performed on each oil workover rig, and a fault warning is performed according to the judgment result, including: comparing the target fault degree of each of the oil workover rigs with a preset fault degree threshold; When the target fault degree is greater than the preset fault degree threshold, the oil well repair rig with equipment failure is located, and a fault warning is issued to the oil well repair rig with equipment failure.

Images