A method for monitoring simulated perforation flow rate data in oil and gas wells

Abstract

The present invention relates to the field of data processing technology, and more specifically, to a method for monitoring flow data of simulated perforation of oil and gas wells. The method comprises: collecting flow data and pressure data, obtaining each adjacent data of each flow data, and obtaining the credibility of each adjacent data of each flow data based on the value of each adjacent data; obtaining the final credibility of each adjacent data of each flow data based on the correlation between flow data and pressure data in the same time series and the credibility of the adjacent data of each flow data; obtaining the local reachable density of each flow data based on the final credibility of each adjacent data of each flow data; obtaining the COF value of each flow data based on the local reachable density of each flow data, and obtaining abnormal data based on the COF value of each flow data. The present invention can accurately detect abnormal data and avoid the influence of noise data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a method for monitoring simulated perforation flow rate data of an oil and gas well. Background Art

[0002] The energy industry is facing increasing production cost pressure and market competition. The implementation of advanced monitoring technology will not only help companies maintain their competitive advantage in the fiercely competitive market, but also help promote the sustainable development and resource management of the industry. This solution can help engineers more accurately understand the actual production capacity and output status of oil and gas wells by monitoring flow data. By analyzing the data, wellbore management and production operations can be optimized, oil and gas recovery rates can be maximized, and resource waste can be reduced. Real-time monitoring of abnormal data in flow data can also help predict and prevent potential wellbore problems, such as wellbore collapse, fracturing loss, etc., thereby reducing accidents and safety risks and ensuring the safety of staff and facilities.

[0003] The patent application document currently published with the publication number CN116720753A discloses a method, system and readable storage medium for processing hydrological data. The hydrological data is collected and verified to determine whether the data verification is passed; if not, the abnormal data sequence that fails the verification is subsequently processed; the data points are detected for abnormality using the One-ClassSVM model to obtain a first index value; the data points are detected for abnormality using the EWMA algorithm, the COF algorithm and the IsolationForest algorithm respectively, and the second index value is obtained by weighted summation of the weights of each algorithm; the historical data corresponding to the data point is input into the LSTM model to obtain the fitting value, and the abnormal data point is determined based on the fitting value and the variance of the abnormal data sequence to obtain the third index value; the first index value, the second index value and the third index value are input into the hidden Markov model as observation values ​​to obtain the target abnormal data point.

[0004] When collecting flow data of simulated perforating of oil and gas wells, mechanical vibrations caused by the operation of oil and gas well equipment or the surrounding environment may be transmitted to the flow meter through the structure, causing fluctuations in flow data and thus generating noise data. When using the COF algorithm in the above method to perform abnormal analysis on flow data, when calculating the local reachable density of each flow data, it is obtained by calculating the inverse of the average similarity of the flow data with its most recent multiple flow data. Noise data may exist in the most recent multiple flow data of the flow data, which will make the result of the local reachable density of the flow data inaccurate, and will cause the COF value of the flow data to be inaccurate, which will affect the abnormal analysis results of the flow data. Summary of the invention

[0005] In order to solve the problem that the existence of noise data will greatly affect the similarity measurement between vibration data and cause errors in the identified abnormal data, the present invention proposes a method for monitoring the simulated perforation flow data of oil and gas wells, which includes the following steps:

[0006] Collect flow data and pressure data; obtain adjacent data of each flow data; obtain the credibility of each adjacent data of each flow data, the credibility characterizing the possibility that each adjacent data is noise data; obtain the final credibility of each adjacent data of each flow data, the final credibility being the product of the correlation between each adjacent data of each flow data and the pressure data corresponding to the time series and the credibility of each adjacent data of each flow data;

[0007] Get the local reachable density of each flow data: ; In the formula, represents the local reachable density of the ith flow data; J represents the number of adjacent data; Represents the credibility of the i-th traffic data; Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the sampling time interval between the i-th flow data and its j-th adjacent data; Represents the final credibility of the jth adjacent data of the i-th traffic data; represents the final credibility sum of all adjacent data of the i-th flow data; exp() represents an exponential function with a natural constant as the base; according to the local reachable density, the COF value of each flow data is obtained; according to the COF value of each flow data, the abnormal data is obtained.

[0008] The innovation of the present invention lies in performing weighted summation of the similarities between each flow data and each of its adjacent data according to the final credibility of each adjacent data of each flow data, and taking the inverse of the weighted summation result as the local reachable density of each flow data, which helps to suppress the influence of noise and reduce the influence of noise data on the calculation of local reachable density when performing anomaly analysis in the COF algorithm, so that the result of the local reachable density of the flow data is more accurate, which in turn leads to a more accurate COF value of the flow data and improves the accuracy of the anomaly analysis results of the flow data; further, the final credibility of each adjacent data of each flow data is obtained, and the correlation between the flow data and the pressure data in the same time series is comprehensively considered, so that the possibility of each adjacent data of each flow data being noise can be more comprehensively and accurately evaluated, rather than just based on the changes in the flow data itself.

[0009] Preferably, the obtaining of adjacent data of each flow data includes:

[0010] The number of adjacent data J is preset, and the J flow data before each flow data in the flow data sequence are recorded as the adjacent data of each flow data.

[0011] Preferably, the obtaining of the credibility of each adjacent data of each flow data includes:

[0012] Obtaining adjacent data segments of each adjacent data of each flow data;

[0013] ;

[0014] In the formula, Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the value of the jth adjacent data of the i-th flow data; Represents the average value of all data in the adjacent data segment of the jth adjacent data of the i-th flow data; Represents the interquartile range of the data set composed of all data in the adjacent data segments of the jth adjacent data of the i-th flow data; exp() represents an exponential function with a natural constant as the base; || represents the absolute value symbol.

[0015] The greater the credibility of each adjacent data of each flow data, the less likely it is that each adjacent data of each flow data is noise data, which facilitates correction of the credibility of each adjacent data of each flow data and obtains the final credibility of each adjacent data of each flow data.

[0016] Preferably, the step of obtaining adjacent data segments of each adjacent data of each flow data includes:

[0017] The number of data in the adjacent data segment is preset to Y, and the Y traffic data before each adjacent data of each traffic data in the traffic data sequence constitute the adjacent data segment of each adjacent data of each traffic data.

[0018] Preferably, the obtaining of the final credibility of each adjacent data of each flow data includes:

[0019] Obtain adjacent data segments of pressure data corresponding to each adjacent data of each flow data;

[0020] , where Represents the final credibility of the jth adjacent data of the i-th traffic data; Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the credibility of the pressure data corresponding to the jth adjacent data of the i-th flow data; The number of data whose values ​​in the adjacent data segment of the jth adjacent data of the i-th flow data are smaller than the values ​​of the jth adjacent data; Represents the number of data in the adjacent data segment of the pressure data corresponding to the jth adjacent data of the ith flow data whose value is smaller than the pressure data value corresponding to the jth adjacent data; || represents the absolute value symbol; norm() represents the normalization function.

[0021] The final credibility of each adjacent data of each flow data is obtained by comprehensively considering the correlation between the flow data and the pressure data in the same time series, so as to more comprehensively and accurately evaluate the possibility that each adjacent data of each flow data is noise.

[0022] Preferably, obtaining the COF value of each flow data according to the local reachable density includes:

[0023] The average ratio of the local reachable density of each flow data to all its adjacent data is taken as the local anomaly factor of each flow data. The absolute value of the difference between the local anomaly factor of each flow data and the mean of the local anomaly factors of all its adjacent data is recorded as the outlier value of each flow data. The outlier value of each flow data is linearly normalized to obtain the COF value of each flow data.

[0024] Based on the local reachable density, the COF value of each flow data is obtained more accurately.

[0025] Preferably, obtaining abnormal data according to the COF value of each flow data includes:

[0026] An abnormal threshold T1 is preset. If the COF value of any flow data is greater than the abnormal threshold T1, the flow data is considered abnormal data, and relevant staff are notified to repair the oil and gas well.

[0027] The detected abnormal data is more accurate.

[0028] Preferably, the step of obtaining adjacent data segments of pressure data corresponding to adjacent data of each flow data includes:

[0029] The number of data in the adjacent data segment is preset to Y, and the pressure data corresponding to each adjacent data of each flow data at the sampling moment is recorded as the pressure data corresponding to each adjacent data of each flow data, and the Y pressure data before the pressure data corresponding to each adjacent data of each flow data in the pressure data sequence constitute an adjacent data segment of the pressure data corresponding to each adjacent data of each flow data.

[0030] The present invention has the following beneficial effects: the final credibility of each adjacent data of each flow data of the present invention performs weighted summation of the similarities between each flow data and each of its adjacent data, and takes the inverse of the weighted summation result as the local reachable density of each flow data, redefines the method for obtaining the local reachable density, and makes the COF value of each flow data more accurate, which helps to suppress the influence of noise and reduce the influence of noise data points on the calculation of the local reachable density when performing abnormal analysis in the COF algorithm, so that the result of the local reachable density of the flow data is more accurate, thereby improving the abnormal analysis result, and the acquisition of the final credibility of each adjacent data of each flow data is not only based on the change of the flow data itself, but also considers the correlation between the flow data and the pressure data in the same time series, which can more comprehensively and accurately evaluate the possibility that each adjacent data of each flow data is noise. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0032] Figure 1 The present invention is a flowchart of the steps of a method for monitoring simulated perforation flow rate data of an oil and gas well according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0034] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0035] See also Figure 1 , which shows a flow chart of the steps of a method for monitoring simulated perforation flow rate data of an oil and gas well provided by an embodiment of the present invention, the method comprising the following steps:

[0036] S001. Collect flow data and pressure data.

[0037] In an embodiment of the present invention, a flow meter and a pressure sensor are installed on a simulated perforation of an oil and gas well and at the bottom of the oil and gas well. A sampling moment is taken every second, and the two types of data, flow data and pressure data, are collected in turn each time. The data are collected for a total of one hour to obtain flow data and pressure data. The collected flow data are arranged in ascending order according to the sampling moment to obtain a flow data sequence; the collected pressure data are arranged in ascending order according to the sampling moment to obtain a pressure data sequence.

[0038] S002. Obtain each adjacent data of each flow data, and based on the value of each adjacent data, obtain the credibility of each adjacent data of each flow data.

[0039] It should be noted that when collecting flow data of simulated perforating of oil and gas wells, the mechanical vibration of the operation of oil and gas well equipment or the surrounding environment may be transmitted to the flow meter through the structure, causing fluctuations in the flow data, thereby generating noise data. When the existing COF algorithm performs anomaly analysis on flow data, when calculating the local reachable density of each flow data, it is obtained by calculating the inverse of the average similarity between the flow data and its most recent multiple flow data. However, there may be noise data in the most recent multiple flow data of the flow data, which will make the result of the local reachable density of the flow data inaccurate, cause the COF value of the flow data to be inaccurate, and affect the abnormal analysis results of the flow data.

[0040] It should be further explained that it is necessary to analyze the numerical change characteristics of the adjacent data of each adjacent data of each flow data to obtain the credibility of each adjacent data of each flow data. When analyzing this indicator, the more abnormal the numerical performance of each adjacent data of each flow data is, the lower its credibility is, and the greater the discreteness of the data in the adjacent data segment of each adjacent data point of each flow data is, the greater the possibility that the adjacent data belongs to the noise data point, and the lower the credibility is.

[0041] In the embodiment of the present invention, the number of adjacent data J is preset, and the J flow data before each flow data in the flow data sequence are recorded as the adjacent data of each flow data. It should be noted that when the number of J flow data is not satisfied, a 0-filling operation is performed on it. In the embodiment of the present invention, the number of adjacent data J is preset to 20. In other embodiments, the implementer can preset the value of the number of adjacent data J according to the specific implementation situation;

[0042] The number of data in the adjacent data segment is preset to be Y, and the Y flow data before each adjacent data of each flow data in the flow data sequence constitute the adjacent data segment of each adjacent data of each flow data. It should be noted that when the number of Y flow data is not satisfied, a 0-filling operation is performed on it. In the embodiment of the present invention, the number of data in the adjacent data segment is preset to be Y=50. In other embodiments, the implementer can preset the value of the number of data in the adjacent data segment Y according to the specific implementation situation;

[0043] In the embodiment of the present invention, the credibility of each adjacent data of each flow data is obtained:

[0044] ;

[0045] In the formula, Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the value of the jth adjacent data of the i-th flow data; Represents the average value of all data in the adjacent data segment of the jth adjacent data of the i-th flow data; represents the interquartile range of a data set composed of all data in the adjacent data segments of the jth adjacent data of the i-th flow data; it should be noted that the acquisition of the interquartile range is a prior art, and in the present invention, it will not be described in detail; exp() represents an exponential function with a natural constant as the base; The larger the value is, the greater the deviation between the numerical performance of the jth adjacent data of the jth flow data and the average numerical value of all data in the adjacent data segment of the jth adjacent data is, and the greater the possibility that the jth adjacent data of the ith flow data belongs to a noise data point is, and the smaller the credibility is; The larger the value is, the greater the dispersion of the data in the adjacent data segments of the jth adjacent data of the i-th flow data is, that is, the greater the fluctuation or dispersion of the data in the middle part is; then it can be said that the possibility of outliers in the data set is greater, that is, the possibility of the jth adjacent data of the i-th flow data being affected by noise is greater, and its credibility is smaller;

[0046] S003. According to the correlation between flow data and pressure data in the same time series and the credibility of the adjacent data of each flow data, the final credibility of each adjacent data of each flow data is obtained; according to the final credibility of each adjacent data of each flow data, the local reachable density of each flow data is obtained.

[0047] It should be noted that the analysis of the credibility of each adjacent data of each flow data is based on the numerical performance of each adjacent data of each flow data. It is known that when the flow data is collected, there may be real abnormal flow data caused by the state of the oil and gas well itself, which is similar to the noise data in numerical performance. Therefore, it is not accurate enough to judge whether each adjacent data is affected by noise only by the credibility of each adjacent data of each flow data. At the same time, according to the scene investigation in this scenario, it can be known that there is usually a certain correlation between pressure data and flow data. The change in pressure can affect the output and flow rate of the fluid. Specifically, because the oil and gas enter the wellbore from the formation is driven by the oil and gas pressure in the formation, if the formation pressure at the bottom of the well is higher than the pressure in the wellbore, the oil and gas will naturally flow to the low-pressure area, that is, enter the wellbore. Therefore, the greater the bottom hole pressure means a greater natural driving force, making it easier for oil and gas to enter the wellbore, and the flow data of the simulated perforation of the oil and gas well will be greater; therefore, the present invention needs to analyze the correlation between each adjacent data of each flow data and the corresponding pressure data at the sampling time, and then combine the credibility of each adjacent data of each flow data to obtain the final credibility of each adjacent data of each flow data.

[0048] In the embodiment of the present invention, the pressure data corresponding to each adjacent data of each flow data at the sampling time is recorded as the pressure data corresponding to each adjacent data of each flow data;

[0049] The Y pressure data preceding the pressure data corresponding to each adjacent data of each flow data in the pressure data sequence constitute an adjacent data segment of the pressure data corresponding to each adjacent data of each flow data;

[0050] Get the final credibility of each adjacent data of each traffic data:

[0051] ;

[0052] In the formula, Represents the final credibility of the jth adjacent data of the i-th traffic data; Represents the credibility of the jth adjacent data of the i-th traffic data; represents the credibility of the pressure data corresponding to the jth adjacent data of the i-th flow data. It should be noted that the method for obtaining the credibility of the pressure data corresponding to the jth adjacent data of the i-th flow data is consistent with the method for obtaining the credibility of each adjacent data of each flow data; The number of data whose values ​​in the adjacent data segment of the jth adjacent data of the i-th flow data are smaller than the values ​​of the jth adjacent data; represents the number of data whose values ​​in the adjacent data segment of the pressure data corresponding to the jth adjacent data of the i-th flow data are smaller than the pressure data value corresponding to the jth adjacent data; || represents the absolute value symbol; norm() represents the normalization function; The larger the value of , the greater the credibility of the jth adjacent data of the i-th traffic data, and the greater its final credibility will be; Represents the correlation between the jth adjacent data of the i-th flow data and its corresponding pressure data. The larger its value is, the less likely it is that the jth adjacent data of the i-th flow data belongs to noise data, and the greater its final credibility is. The difference in the number of data whose values ​​between the jth adjacent data of the i-th flow data and its corresponding pressure data in their respective adjacent data segments are smaller than the values ​​between the jth adjacent data and its corresponding pressure data. The smaller the value, the greater the correlation between the jth adjacent data of the i-th flow data and its corresponding pressure data, indicating that the possibility that the jth adjacent data of the i-th flow data is noise data is small, and its final credibility is greater; The larger the value of is, the more accurate the correlation between the jth adjacent data of the analyzed i-th flow data and its corresponding pressure data is.

[0053] It should be noted that when the existing COF algorithm performs anomaly analysis on flow data, when calculating the local reachable density of each flow data, it is obtained by calculating the inverse of the average similarity between the flow data and its multiple nearest flow data. There may be noise data in the multiple nearest flow data of the flow data, which will make the result of the local reachable density of the flow data inaccurate. It is known that the final credibility of each adjacent data of each flow data has been obtained through the above. In this step, it is necessary to perform weighted summation of the similarity between each flow data and each of its adjacent data according to the final credibility of each adjacent data of each flow data, and take the inverse of the weighted summation result as the local reachable density of each flow data; wherein, the greater the final credibility of each adjacent data of each flow data, the more accurate the similarity between each flow data and the adjacent data.

[0054] In this embodiment of the present invention, the local reachable density of each flow data is obtained:

[0055] ;

[0056] In the formula, represents the local reachable density of the ith flow data; J represents the number of adjacent data; represents the credibility of the ith flow data. It should be noted that the method for obtaining the credibility of flow data is consistent with the method for obtaining the credibility of the adjacent data of flow data; Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the sampling time interval between the i-th flow data and its j-th adjacent data; Represents the final credibility of the jth adjacent data of the i-th traffic data; Represents the final credibility sum of all adjacent data of the i-th traffic data; It represents the similarity between the ith flow data and its jth adjacent data, which is obtained by analyzing the credibility difference between the ith flow data and its jth adjacent data and the sampling time interval. The smaller the credibility difference between the ith flow data and its jth adjacent data and the sampling time interval, the greater the similarity. It represents the weight of the jth neighboring data of the ith flow data when calculating the local reachable density of the ith flow data. The greater the final credibility of the jth neighboring data of the ith flow data, the greater the weight of the jth neighboring data of the ith flow data.

[0057] S004. Obtain the COF value of each flow data according to the local reachable density of each flow data, and obtain abnormal data according to the COF value of each flow data.

[0058] It should be noted that the COF value of each flow data is obtained according to the local reachable density of each flow data, and the larger the COF value of any flow data, the greater the degree of abnormality of the flow data, so the abnormal data is obtained according to the COF value of each flow data.

[0059] In an embodiment of the present invention, an abnormality threshold T1 is preset, and the average ratio of the local reachable density of each flow data to all its adjacent data is taken as the local abnormality factor of each flow data. The absolute value of the difference between the local abnormality factor of each flow data and the mean of the local abnormality factors of all its adjacent data is recorded as the abnormal value of each flow data. The abnormal value of each flow data is linearly normalized to obtain the COF value of each flow data. If the COF value of any flow data is greater than the abnormality threshold T1, the flow data is abnormal data. All abnormal data are obtained, and relevant staff are notified to repair the oil and gas wells. In an embodiment of the present invention, the abnormality threshold T1 is preset to 0.6. In other embodiments, the implementers can preset the value of the abnormality threshold T1 according to the specific implementation situation.

[0060] The final credibility of each adjacent data of each flow data of the present invention performs weighted summation of the similarities between each flow data and each of its adjacent data, and takes the inverse of the weighted summation result as the local reachable density of each flow data, redefines the method for obtaining the local reachable density, and helps to suppress the influence of noise and reduce the influence of noise data on the calculation of the local reachable density when performing anomaly analysis in the COF algorithm, so that the result of the local reachable density of each flow data is more accurate, thereby improving the anomaly detection result, and the acquisition of the final credibility of each adjacent data of each flow data is not only based on the change of the flow data itself, but also considers the correlation between the flow data and the pressure data in the same time series, so that the possibility of each adjacent data of each flow data being noise can be more comprehensive and accurate.

[0061] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring simulated perforation flow rate data in oil and gas wells, It is characterized in that include: Collect flow data and pressure data; Acquiring adjacent data of each flow data, including: presetting the number of adjacent data J, recording the J flow data before each flow data in the flow data sequence as adjacent data of each flow data; acquiring the credibility of each adjacent data of each flow data, wherein the credibility represents the possibility that each adjacent data is noise data; acquiring the final credibility of each adjacent data of each flow data, wherein the final credibility is obtained by multiplying the correlation between each adjacent data of each flow data and the pressure data in the corresponding time series and the credibility of each adjacent data of each flow data; Get the local reachable density of each flow data: ; In the formula, represents the local reachable density of the ith flow data; J represents the number of adjacent data; Represents the credibility of the i-th traffic data; Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the sampling time interval between the i-th flow data and its j-th adjacent data; Represents the final credibility of the jth adjacent data of the i-th traffic data; represents the final credibility sum of all adjacent data of the i-th flow data; exp() represents an exponential function with a natural constant as the base; according to the local reachable density, the COF value of each flow data is obtained; according to the COF value of each flow data, the abnormal data is obtained.

2. A method for monitoring simulated perforation flow rate data of an oil and gas well according to claim 1, It is characterized in that The obtaining of the credibility of each adjacent data of each flow data includes: Obtaining adjacent data segments of each adjacent data of each flow data; ； In the formula, Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the value of the jth adjacent data of the i-th flow data; Represents the average value of all data in the adjacent data segment of the jth adjacent data of the i-th flow data; Represents the interquartile range of the data set composed of all data in the adjacent data segments of the jth adjacent data of the i-th flow data; exp() represents an exponential function with a natural constant as the base; || represents the absolute value symbol.

3. A method for monitoring simulated perforation flow rate data of oil and gas wells according to claim 2, It is characterized in that The step of obtaining adjacent data segments of adjacent data of each flow data includes: The number of data in the adjacent data segment is preset to Y, and the Y traffic data before each adjacent data of each traffic data in the traffic data sequence constitute the adjacent data segment of each adjacent data of each traffic data.

4. The method for monitoring simulated perforation flow rate data of an oil and gas well according to claim 1, It is characterized in that The final credibility of each adjacent data of each flow data is obtained, including: Obtain adjacent data segments of pressure data corresponding to each adjacent data of each flow data; , where Represents the final credibility of the jth adjacent data of the i-th traffic data; Represents the credibility of the jth adjacent data of the i-th traffic data; Represents the credibility of the pressure data corresponding to the jth adjacent data of the i-th flow data; The number of data whose values ​​in the adjacent data segment of the jth adjacent data of the i-th flow data are smaller than the values ​​of the jth adjacent data; Represents the number of data in the adjacent data segment of the pressure data corresponding to the jth adjacent data of the ith flow data whose value is smaller than the pressure data value corresponding to the jth adjacent data; || represents the absolute value symbol; norm() represents the normalization function.

5. The method for monitoring simulated perforation flow rate data of an oil and gas well according to claim 1, It is characterized in that The obtaining of the COF value of each flow data according to the local reachable density includes: The average ratio of the local reachable density of each flow data to all its adjacent data is taken as the local anomaly factor of each flow data. The absolute value of the difference between the local anomaly factor of each flow data and the mean of the local anomaly factors of all its adjacent data is recorded as the outlier value of each flow data. The outlier value of each flow data is linearly normalized to obtain the COF value of each flow data.

6. The method for monitoring simulated perforation flow rate data of an oil and gas well according to claim 1, It is characterized in that The abnormal data is obtained according to the COF value of each flow data, including: An abnormal threshold T1 is preset. If the COF value of any flow data is greater than the abnormal threshold T1, the flow data is considered abnormal data, and relevant staff are notified to repair the oil and gas well.

7. A method for monitoring simulated perforation flow rate data of oil and gas wells according to claim 4, It is characterized in that The step of obtaining adjacent data segments of pressure data corresponding to adjacent data of each flow data includes: The number of data in the adjacent data segment is preset to Y, and the pressure data corresponding to each adjacent data of each flow data at the sampling moment is recorded as the pressure data corresponding to each adjacent data of each flow data, and the Y pressure data before the pressure data corresponding to each adjacent data of each flow data in the pressure data sequence constitute an adjacent data segment of the pressure data corresponding to each adjacent data of each flow data.

Images