Logging instrument fault diagnosis method based on t inspection method and related equipment

Abstract

The invention discloses a logging instrument fault diagnosis method based on a t test method and related equipment, and belongs to the field of petroleum logging downhole instrument fault diagnos.The method is based on a grey relational degree analysis method, grey relational degree data between logging instruments is obtained through calculation of obtained logging data, and then the t test method is used for obtaining the fault diagnosis result of the logging instruments. Constructing a t test model through the grey relational degree data; and finally, diagnosing a to-be-tested sample by using the t test model, and outputting a logging instrument fault diagnosis result. The method is based on a t test method, and whether the instruments are in a fault state is judged through a method for detecting whether the grey correlation degree between the instruments is remarkably changed or not. By adopting the method, the instrument fault can be timely and accurately diagnosed, the instrument can be replaced, and the logging timeliness is improved; according to the method, the problems of poor diagnosis accuracy and low efficiency caused by manual judgment of the faults of the logging instrument are effectively solved.

Description

Technical Field

[0001] The invention belongs to the field of fault diagnosis of petroleum well logging downhole instruments, and in particular relates to a well logging instrument fault diagnosis method based on a t-test method and related equipment. Background Art

[0002] The purpose of oil logging is to efficiently and accurately identify formation information and explore and exploit oil. The response of downhole instruments is a direct reflection of formation characteristics. Logging interpreters make comprehensive interpretations of oil wells based on the response data of various instruments. Although downhole instruments have a certain degree of reliability, instrument failures due to factors such as the harsh downhole environment will cause distorted data output, which will lead to incorrect judgments of the formation and seriously affect the efficiency of logging. Therefore, it is urgent to study whether downhole instruments fail during logging, detect instrument failures in a timely manner, and replace instruments for measurement, so as to ensure logging efficiency and accuracy.

[0003] Failures in downhole instrument components or failures in signal transmission due to environmental interference will directly lead to abnormal logging instrument response data. In the process of analyzing downhole instruments, it is believed that data that cannot reflect the actual response of the formation are distorted data. According to the response form of downhole instruments after failure, distorted data are divided into discrete point distortion faults and continuous segment distortion faults.

[0004] Data distortion failures at discrete points of downhole instruments are common. Usually, this failure is caused by the instability of the instrument due to the influence of the downhole environment. During on-site logging, based on the real-time sampling curve and measurement parameters displayed by the logging surface system, the logging operator can determine whether the downhole instrument has a fault based on the technical specifications of different instruments. If it exceeds the scope of technical specifications, the logging instrument is considered to have a fault. Taking the logging instrument of a certain well as an example, the following judgment criteria are obtained as shown in Table 1:

[0005] Table 1 is the criterion for downhole instrument failure

[0006]

[0007]

[0008] In addition, when a well condition accident occurs during the logging process, such as the instrument getting stuck or encountering obstructions, it can be judged according to the change in the cable tension of the ground system. When the downhole instrument has a continuous response data distortion, the fault situation is more complicated and there are more types of fault data. This is because the structure of the downhole instrument is complex, generally composed of a detector and a subsequent circuit. High temperature and high pressure, electromagnetic interference, etc., which lead to damage to instrument components or aging of components, as well as improper operation by operators, will cause instrument failure. At present, the method for distinguishing such faults relies on the experience of logging interpreters, resulting in poor accuracy and low efficiency of fault diagnosis; and the research on diagnostic methods is almost in a blank stage. The current market has higher and higher requirements for logging efficiency and accuracy, so the demand for fault distinguishing methods for downhole logging instruments has become more and more urgent. Summary of the invention

[0009] In order to overcome the shortcomings of the above-mentioned technology, the present invention provides a logging instrument fault diagnosis method and related equipment based on the t-test method, which can solve the technical problems of poor diagnostic result accuracy and low diagnostic process efficiency of the existing diagnostic method.

[0010] In order to achieve the above object, the present invention adopts the following technical solution:

[0011] A method for diagnosing faults of a well logging instrument based on a t-test method, comprising:

[0012] Acquire logging data when the logging instrument is in normal state;

[0013] Based on the grey correlation analysis method, the grey correlation data between logging instruments are obtained according to the logging data;

[0014] Based on the t-test method and the grey relational data, a t-test model is constructed;

[0015] The samples to be tested are input into the t-test model, and the fault diagnosis results of the logging instrument are output.

[0016] Furthermore, the specific steps of obtaining the grey correlation data between logging instruments include:

[0017] The range method is used to normalize the logging data to obtain normalized logging data;

[0018] Based on the grey correlation analysis method, the correlation coefficient is calculated using normalized logging data;

[0019] According to the correlation coefficient, the grey correlation data between logging instruments are calculated.

[0020] Furthermore, the formula for normalizing the logging data using the range method is as follows:

[0021]

[0022] Among them, x i * Represents the normalized value; x i Represents well logging data; x min Indicates the minimum value of logging data; x max Indicates the maximum value of the logging data.

[0023] Furthermore, the correlation coefficient ξ i The calculation formula of (k) is as follows:

[0024]

[0025] in, Indicates the minimum difference between the two poles; represents the maximum difference between the two poles; ρ is the resolution coefficient, and its value range is (0, 1);

[0026] Δ i (k)=|x'0(k)-x i '(k)| represents the relationship between each point on the reference sequence X'0 curve and each comparison sequence X i 'The absolute value difference of each point on the curve; k = 1, 2...n, n is a natural number, representing the number of correlation coefficients.

[0027] Furthermore, the calculation formula of grey relational data is as follows:

[0028]

[0029] Among them, γ i Represents grey relational data.

[0030] Furthermore, the specific steps of constructing the t-test model include:

[0031] According to the t-test method, the QQ diagram is used to perform normality test on the grey relational data, and the t-test model that meets the normality test requirements is output.

[0032] Furthermore, it also includes: inputting the fault diagnosis result of the logging instrument into a fault database;

[0033] The fault database includes a straight section fault model and a jump fault model.

[0034] A well logging instrument fault diagnosis system based on t-test method, used to implement the steps of the well logging instrument fault diagnosis method based on t-test method, characterized in that it includes:

[0035] A data acquisition module, used to acquire the logging data when the logging instrument is in a normal state;

[0036] The correlation analysis module is used to obtain the grey correlation data between the logging instruments based on the logging data based on the grey correlation analysis method;

[0037] Model building module, used to build t-test model based on t-test method and grey relational data;

[0038] The fault diagnosis module is used to input the samples to be tested into the t-test model and output the fault diagnosis results of the logging instrument.

[0039] A device comprising:

[0040] Memory for storing computer programs;

[0041] The processor is used to implement the steps of the above-mentioned logging instrument fault diagnosis method based on the t-test method when executing the computer program.

[0042] A computer-readable storage medium stores a computer program, which is used to implement the steps of the above-mentioned logging instrument fault diagnosis method based on t-test method when executed by a processor.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention provides a method for diagnosing well logging instrument faults based on the t-test method. The method is based on the grey correlation analysis method. The grey correlation data between the well logging instruments is calculated by the acquired well logging data. The t-test method is then used to construct a t-test model through the grey correlation data. Finally, the t-test model is used to diagnose the sample to be tested, and the well logging instrument fault diagnosis result is output. The method is based on the t-test method. Whether the grey correlation between the instruments changes significantly is determined to determine whether the instrument is in a fault state. The method can timely and accurately diagnose instrument faults, replace the instruments, and improve the timeliness of well logging. The method effectively solves the problem of relying on manual judgment of well logging instrument faults, resulting in poor diagnostic accuracy and low efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is the grey correlation degree γ between AC and CNL in the embodiment of the present invention 12 Fitting result graph;

[0046] Figure 2 is the grey correlation degree γ between AC and DEN in the embodiment of the present invention 13 Fitting result graph;

[0047] Figure 3 is the gray correlation degree γ between CNL and DEN in the embodiment of the present invention 23 Combined result graph;

[0048] Figure 4 is the grey correlation degree γ between AC and CNL in the embodiment of the present invention 12 Normal QQ plot of ;

[0049] Figure 5 is the grey correlation degree γ between AC and DEN in the embodiment of the present invention 13 Normal Q-plot of;

[0050] Figure 6 is the gray correlation degree γ between CNL and DEN in the embodiment of the present invention 23 Normal Q-plot of;

[0051] Figure 7 A comparison diagram of fault signals of an acoustic wave instrument in an embodiment of the present invention;

[0052] Figure 8 A comparison diagram of compensated neutron fault signals in an embodiment of the present invention;

[0053] Fig. 9 A comparison diagram of fault signals of a compensation density instrument in an embodiment of the present invention;

[0054] Fig.10 is the grey correlation degree γ between AC and CNL when the instrument is fault-free in the embodiment of the present invention 12 t-test chart;

[0055] Fig.11 is the grey correlation degree γ between AC and DEN when the instrument is fault-free in the embodiment of the present invention 13 t-test chart;

[0056] Fig.12 is the grey correlation degree γ between CNL and DEN when the instrument is fault-free in the embodiment of the present invention 23 t-test chart;

[0057] Fig.13 is the grey correlation degree γ between AC and CNL when the acoustic wave compensation instrument fails in the embodiment of the present invention 12 t-test chart;

[0058] Fig.14 is the grey correlation degree γ between AC and DEN when the acoustic wave compensation instrument fails in the embodiment of the present invention 13 t-test chart;

[0059] Fig.15 is the grey correlation degree γ between CNL and DEN when the acoustic wave instrument fails in the embodiment of the present invention 23 t-test chart;

[0060] Fig.16 is the grey correlation degree γ between AC and CNL when the compensation density instrument fails in the embodiment of the present invention 12 t-test chart;

[0061] Fig.17 is the grey correlation degree γ between AC and DEN when the compensation density instrument fails in the embodiment of the present invention 13 t-test chart;

[0062] Fig.18 is the gray correlation degree γ between CNL and DEN when the density compensation instrument fails in the embodiment of the present invention 23 t-test chart;

[0063] Fig.19 A flow chart of a method for diagnosing well logging instrument faults based on the t-test method provided by the present invention;

[0064] Fig. 20 A schematic diagram of the structure of a well logging instrument fault diagnosis system based on the t-test method provided by the present invention;

[0065] Among them, AC stands for acoustic logging tool; CNL stands for compensated neutron logging tool; DEN stands for compensated density logging tool. DETAILED DESCRIPTION

[0066] The present invention provides a method for diagnosing well logging instrument faults based on the t-test method. Fig.19 As shown, the following steps are included:

[0067] S1: Acquire logging data when the logging instrument is in a normal state;

[0068] S2: Based on the grey correlation analysis method, the grey correlation data between logging instruments are obtained according to the logging data.

[0069] The specific steps of obtaining the grey correlation data between logging instruments include:

[0070] S201: normalizing the logging data using a range method to obtain normalized logging data;

[0071] The formula for normalizing logging data using the range method is as follows:

[0072]

[0073] Among them, x i * Represents the normalized value; x i Represents well logging data; x min Indicates the minimum value of logging data; x max Indicates the maximum value of the logging data.

[0074] S202: Based on the grey correlation analysis method, using the normalized logging data, calculating the correlation coefficient;

[0075] Correlation coefficient ξ i The calculation formula of (k) is as follows:

[0076]

[0077] in, Indicates the minimum difference between the two poles; represents the maximum difference between the two poles; ρ is the resolution coefficient, and its value range is (0, 1);

[0078] Δ i (k)=|x'0(k)-x i '(k)| represents the relationship between each point on the reference sequence X'0 curve and each comparison sequence X i 'The absolute value difference of each point on the curve; k = 1, 2...n, n is a natural number, representing the number of correlation coefficients.

[0079] S203: Calculate and obtain grey correlation data between logging tools according to the correlation coefficient.

[0080] The calculation formula of grey relational data is as follows:

[0081]

[0082] Among them, γ i Represents grey relational data.

[0083] S3: Based on the t-test method and grey relational data, a t-test model is constructed.

[0084] The specific steps of constructing the t-test model include:

[0085] According to the t-test method, the QQ diagram is used to perform normality test on the grey relational data, and the t-test model that meets the normality test requirements is output.

[0086] S4: Input the sample to be tested into the t-test model and output the fault diagnosis result of the logging instrument.

[0087] The diagnostic method further includes: inputting the logging instrument fault diagnosis result into a fault database;

[0088] The fault database includes the straight section fault model and the jump fault model. Fig. 20As shown, the present invention also provides a well logging instrument fault diagnosis system based on the t-test method, including: a data acquisition module, used to obtain the well logging data when the well logging instrument is in a normal state; a correlation analysis module, used to obtain the grey correlation data between the well logging instruments based on the grey correlation analysis method according to the well logging data; a model building module, used to construct a t-test model based on the t-test method according to the grey correlation data; a fault diagnosis module, used to input the sample to be tested into the t-test model and output the fault diagnosis result of the well logging instrument.

[0089] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the well logging instrument fault diagnosis method based on the t-test method when executing the computer program.

[0090] When the processor executes the computer program, the above-mentioned steps of t-test-based well logging instrument fault diagnosis are implemented, for example: obtaining well logging data when the well logging instrument is in a normal state; based on the grey correlation analysis method, obtaining grey correlation data between well logging instruments according to the well logging data; based on the t-test method, constructing a t-test model according to the grey correlation data; inputting the sample to be tested into the t-test model, and outputting the well logging instrument fault diagnosis result.

[0091] Alternatively, when the processor executes the computer program, the functions of each module in the above system are realized, for example: a data acquisition module, used to obtain logging data when the logging instrument is in a normal state; a correlation analysis module, used to obtain grey correlation data between logging instruments based on the grey correlation analysis method according to the logging data; a model building module, used to construct a t-test model based on the t-test method according to the grey correlation data; a fault diagnosis module, used to output the fault diagnosis result of the logging instrument by inputting the sample to be tested into the t-test model.

[0092] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete preset functions, and the instruction segments are used to describe the execution process of the computer program in the logging instrument fault diagnosis device based on the t-test method. For example, the computer program can be divided into a data acquisition module, a correlation analysis module, a model building module and a fault diagnosis module; the specific functions of each module are as follows: a data acquisition module, which is used to obtain the logging data of the logging instrument in a normal state; a correlation analysis module, which is used to obtain the gray correlation data between the logging instruments based on the gray correlation analysis method according to the logging data; a model building module, which is used to construct a t-test model based on the t-test method and the gray correlation data; a fault diagnosis module, which is used to input the sample to be tested into the t-test model and output the fault diagnosis result of the logging instrument.

[0093] The well logging instrument fault diagnosis device based on the t-test method can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The well logging instrument fault diagnosis device based on the t-test method may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above is an example of a well logging instrument fault diagnosis device based on the t-test method, and does not constitute a limitation on the well logging instrument fault diagnosis device based on the t-test method, and may include more components than the above, or a combination of certain components, or different components. For example, the well logging instrument fault diagnosis device based on the t-test method may also include input and output devices, network access devices, buses, etc.

[0094] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The processor is the control center of the well logging instrument fault diagnosis based on the t-test method, and uses various interfaces and lines to connect various parts of the whole well logging instrument fault diagnosis device based on the t-test method.

[0095] The memory can be used to store the computer program and / or module, and the processor implements various functions of the t-test-based logging instrument fault diagnosis device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0096] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0097] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the well logging instrument fault diagnosis method based on the t-test method are implemented.

[0098] If the integrated module / unit of the well logging instrument fault diagnosis system based on the t-test method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0099] Based on such understanding, the present invention implements all or part of the processes in the above-mentioned method for diagnosing well logging instrument faults based on the t-test method, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method for diagnosing well logging instrument faults based on the t-test method can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or preset intermediate form, etc.

[0100] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0101] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0102] Example

[0103] The present invention will be further described below in conjunction with the embodiments and drawings:

[0104] In order to solve the technical problem mentioned in the background technology that the existing diagnostic methods rely on the experience of logging interpreters, resulting in poor accuracy of diagnostic results and low efficiency of the diagnostic process, this embodiment provides a logging instrument fault diagnosis method based on the t-test method. When the downhole instrument has a fault condition in which continuous response data is distorted, this method can be used to quickly determine the fault of the logging instrument, so that the instrument can be replaced in time for re-logging, thereby improving the logging timeliness.

[0105] This embodiment provides a method for diagnosing well logging instrument faults based on the t-test method, the steps comprising:

[0106] First, the logging data of the logging instrument under normal conditions are collected; according to the grey correlation analysis between the responses of the logging instruments, three instruments with good grey correlation are grouped together, and the grey correlation between each two is calculated respectively; the mean of the grey correlation between the curves is obtained according to statistics, that is, μ0 of the t-test center; samples are selected for verification, and the diagnosis result is obtained according to whether the statistical test quantity T is in the rejection domain.

[0107] The specific steps are:

[0108] The first step is to obtain the logging data of the logging instrument under normal conditions. In this embodiment, the logging data comes from the logging data of 6 wells in a certain area recorded by the ground system of Sichuan Qing Drilling and Well Detection Company. The measuring depth of each well is 3442 meters, and the sampling interval is 0.125m.

[0109] The second step is to obtain the grey correlation data between the logging instruments based on the logging data based on the grey correlation analysis method. This is one of the invention points of the present application: the grey correlation is used to quantitatively analyze the response of the downhole instrument. First, the range method is used to dimensionlessly normalize the original data. The formula is:

[0110]

[0111] Among them, x i * Represents the normalized value; x i Represents well logging data; x min Indicates the minimum value of logging data; x maxIndicates the maximum value of the logging data.

[0112] After dimensionless transformation, the correlation coefficient is calculated. Assume that the reference sequence after processing is X'0 = {x'0(1), x'0(2), ... x'0(n)}, and the comparison sequence is X i '={x i '(1),x'2(2),......x i '(n)}X'0 and X i The correlation coefficient ξ i (k) is:

[0113]

[0114] in, Indicates the minimum difference between the two poles; represents the maximum difference between the two poles; ρ is the resolution coefficient, and its value range is (0, 1);

[0115] Note Δ i (k)=|x'0(k)-x i '(k)| represents the relationship between each point on the reference sequence X'0 curve and each comparison sequence X i 'The absolute value difference of each point on the curve; k = 1, 2...n, n is a natural number, indicating the number of correlation coefficients. The resolution coefficient ρ here is used to reduce the influence of gray correlation distortion, which is caused by the large maximum difference between the two poles. The specific value of ρ can be determined according to the situation, and is usually 0.5.

[0116] Then, the grey correlation data is obtained according to the correlation coefficient, and the grey correlation coefficients are averaged to obtain the grey correlation calculation formula: Among them, γ i Represents grey relational data.

[0117] Here, according to the different physical parameters measured by the downhole instruments, this embodiment uses the natural gamma curve, the acoustic wave curve and the resistivity curve as reference series to obtain the grey correlation degree with the response curves of other instruments. The calculation results are shown in Table 2 below:

[0118] Table 2 shows the grey correlation of the logging tool response curve

[0119]

[0120] From the above grey correlation values, it can be seen that the compensated density and compensated neutron logging curves have the best grey correlation with the compensated acoustic wave curve, indicating that their dynamic change processes are similar. Similarly, the dynamic change processes of microsphere focusing, shallow lateral and deep lateral curves are similar, and the dynamic change processes of natural gamma and natural potential and acoustic waves are similar.

[0121] The next step is another invention point of this application, that is, to statistically analyze the mean value of the gray correlation of each logging instrument when it is working normally. Taking the first group of compensated sound waves, compensated neutrons and compensated density as an example, a method is used to test whether the gray correlation between the instrument to be diagnosed and the other two instruments has changed significantly.

[0122] Since the condition for the t-test method to be applied is that the population obeys the normal distribution, the present invention performs a normality test on the grey correlation between instruments. Therefore, it is first necessary to perform a normality test on the grey correlation between logging instruments. The source of the data is the logging data of 6 wells in a certain area recorded by the ground system of Sichuan Qing Drilling and Logging Company. In order to make the whole representative, representative logging data with a total depth of 3,000 meters are screened. The logging sampling interval is 0.125 meters per sampling point, that is, 8 data points are collected for one meter (a total of 24,000 data points). The invention uses data at a depth of one meter to calculate a grey correlation value, so there are a total of 3,000 grey correlation values. By using the data analysis software SPSS for analysis, the grey correlation γ is obtained. 12 , γ 13 , γ 23 The fitting results of .

[0123] According to the normality test method, a QQ graph (probability graph) is drawn to test the normality of the above grey relational data. The QQ graph tests the distribution of data by comparing the quantiles of sample data with the known distribution. Its scatter plot corresponds to the QQ graph of normal distribution, with the quantiles of the standard normal distribution as the horizontal coordinate and the standardized value (xm) / std (where m is the mean and std is the standard deviation) as the vertical coordinate. 12 , γ 13 , γ 23 The QQ graph of Figure 4 , Figure 5 and Figure 6 As shown in the figure, it can be seen that the scatter plot is approximately near a straight line, which means that the distribution of the data is a normal distribution.

[0124] The fitting results show that Figure 1 , Figure 2 and Figure 3 As shown, the grey relational degree γ 12 , γ 13 , γ 23 They obey the normal distribution, and their means are 0.6301, 0.6361, and 0.6302 respectively.

[0125] Finally, a verification sample (sample to be tested) is selected for verification. Since the logging data does not contain data when there is a fault, in this embodiment, the response change when a fault occurs is simulated according to the common fault data representation form, and a data model (fault database) of the well logging instrument response data distortion fault is established based on this.

[0126] These include:

[0127] ① Straight section: When the sensor at the front end of the instrument is damaged or the insulation impedance of its sensitive core decreases, its sensitivity will decrease, and the response value of the logging instrument will show a straight section. Considering that the logging instrument will be interfered by the formation magnetic field during data acquisition and transmission through long cables. This interference will add a random noise to the signal. That is, the final expression is:

[0128] Y out =C+δ

[0129] Among them, Y OUT represents the final output result, C represents the instrument fault signal, and δ is random white noise;

[0130] ② Jumping: When the high voltage of the instrument is unstable, the power supply is unstable, or the sensitive detector of the instrument is damaged, the curve will jump violently, and the actual output signal value is several times the true value. This data distortion fault model can be expressed by the following formula:

[0131] Y out =βx(t)+δ

[0132] Among them, Y OUT Represents the final output result, X (t) represents the fault signal of the instrument, β represents the multiple, and δ is the random white noise.

[0133] In order to further illustrate the changes in the logging instrument response data caused by the above fault types, in this embodiment, it is assumed that in the straight section, the fault signal C is the average of the maximum and minimum values ​​of the instrument response signal. For the abnormal beating, the gain coefficient is generally different. It is generally assumed that the gain coefficient is between 0.7 and 1.4, so = 0.7 + 0.7 *, which is a random number between 0 and 1. Take δ as a white noise with a mean of 0, a variance of 1, and an amplitude of 1% of the maximum value of the instrument response signal. That is:

[0134]

[0135] like Figure 7 , Figure 8 and Fig. 9 As shown, according to the above fault model, the fault distortion data of the compensated acoustic wave instrument, the compensated neutron instrument and the compensated density instrument are simulated respectively.

[0136] Under the same formation conditions, 4000 normal sample data points were selected, and the grey correlation degree was calculated in groups of 8 data points (one meter depth), and 500 grey correlation degrees were obtained. They were divided into 50 groups of 10 as samples for t-test, that is, 50 independent samples were obtained, and the sample capacity of each group was 10. Fault data was simulated based on the normal sample data.

[0137] The specific implementation process is as follows:

[0138] The following three situations will be discussed separately. Fig.10 —18, the solid line represents the maximum value of the t-test rejection region, and the area below it is the rejection region with a significance level of 0.05.

[0139] Case 1: All instruments are normal:

[0140] According to the 4000 well logging data points selected above, 500 grey correlation degrees are obtained, which are divided into 10 groups to obtain 50 independent samples, and the t-test statistic is calculated.

[0141] H0: All instruments are fault-free, and the gray correlation between instruments is μ i ≥μ0 i=1,2,3.

[0142] H1: A certain instrument is faulty, and there is a gray correlation between the instruments μ i <μ0 i=1,2,3.

[0143] like Fig.10 and Fig.11 As shown in the figure, when there is no instrument failure, there are 50 independent samples with statistical test values ​​below the threshold line, which are the 8th and 41st sample points, that is, they are within the rejection region of the t-test.

[0144] It should be noted that, through comparison with the adjacent formation curves and based on the comprehensive interpretation of the logging interpreters, it is known that due to the presence of a thick gas layer (about 10 meters) in the formation, the acoustic time difference curve has a sharp "fluctuating" frequency jump phenomenon, which reduces the gray correlation between it and the compensated neutron and compensated density curves, resulting in misjudgment. This shows that formation factors will affect the judgment of faults and there is a possibility of misjudgment. However, through case one, it can be seen that, except in extremely special cases, the accuracy and efficiency of this method for diagnosing well logging instrument faults are extremely high.

[0145] Case 2: Compensation acoustic wave instrument failure

[0146] The same 4000 logging data points as above were selected, and the data straight segment fault was simulated based on the first 2000 data of the selected compensation acoustic wave instrument response sample, and the data jump fault was simulated based on the last 2000 data, and the grey correlation degree between them and the other two instruments was calculated respectively. 500 grey correlation degrees were obtained, which were divided into 10 groups, 50 independent samples were obtained, and the t-test statistics were calculated.

[0147] H0: All instruments are fault-free, and the gray correlation between instruments is μ i ≥μ0i=1,2,3.

[0148] H1: A certain instrument is faulty, and there is a gray correlation between the instruments μ i <μ0i=1,2,3.

[0149] according to Fig.13 , Fig.14 and Fig.15 It can be seen that when the sonic instrument fails, Fig.13 , 14 In the t-test, all 50 independent test samples are within the rejection region of the t-test. Fig.15 , there has been a significant change, that is, the mean value of the grey correlation between the acoustic logging tool and the compensated density logging tool and the compensated neutron logging tool has changed significantly, so it is believed that the acoustic tool has failed.

[0150] Case 3: Failure of the compensation density instrument

[0151] The same 4000 logging data points as above were selected, and the data flat segment fault was simulated on the basis of the first 2000 data of the selected compensation density instrument response sample, and the data jump fault was simulated on the basis of the last 2000 data, and the grey correlation degree between them and the other two instruments was calculated respectively. 500 grey correlation degrees were obtained, which were divided into 10 groups, 50 independent samples were obtained, and the t-test statistic was calculated.

[0152] H0: All instruments are fault-free, and the gray correlation between instruments is μ i ≥μ0i=1,2,3.

[0153] H1: A certain instrument is faulty, and there is a gray correlation between the instruments μ i <μ0i=1,2,3.

[0154] like Fig.16 , Fig.17 and Fig.18 Except for special cases, as shown in the attached Fig.16 In the case of a gas layer, the test statistics of several sample points are within the rejection region. Fig.17 and Fig.18 As shown in the figure, all 50 independent test samples are within the threshold of the t-test, and the attached Fig.16 The statistical test quantities in are not within the rejection region, that is, the mean of the grey correlation degree between the compensated density logging tool and the acoustic logging tool and the compensated neutron logging tool has changed significantly, so it is believed that the compensated density logging tool has failed.

[0155] In summary, this embodiment provides a method for diagnosing well logging instrument faults based on the t-test method. This method can effectively solve the problems of poor diagnostic accuracy and low diagnostic process efficiency of existing diagnostic methods. This method is based on the gray correlation analysis method. The gray correlation data between the well logging instruments is calculated by the acquired well logging data, and then the t-test method is used to construct a t-test model through the gray correlation data. Finally, the t-test model is used to diagnose the sample to be tested and output the well logging instrument fault diagnosis result. This method is based on the t-test method. It determines whether the instrument is in a fault state by checking whether the gray correlation between the instruments has changed significantly. This method can timely and accurately diagnose instrument faults, replace the instruments, and improve the timeliness of well logging.

[0156] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. A method for diagnosing well logging instrument faults based on t-test method, characterized in that: include: Acquire logging data when the logging instrument is in normal state; Based on the grey correlation analysis method, the grey correlation data between logging instruments are obtained according to the logging data; Based on the t-test method and the grey relational data, a t-test model is constructed; The samples to be tested are input into the t-test model, and the fault diagnosis results of the logging instrument are output.

2. A method for diagnosing well logging instrument faults based on t-test method according to claim 1, characterized in that: The specific steps of obtaining the grey correlation data between logging instruments include: The range method is used to normalize the logging data to obtain normalized logging data; Based on the grey correlation analysis method, the correlation coefficient is calculated using normalized logging data; According to the correlation coefficient, the grey correlation data between logging instruments are calculated.

3. A method for diagnosing well logging instrument faults based on t-test method according to claim 2, characterized in that: The formula for normalizing logging data using the range method is as follows: Among them, x i * Represents the normalized value; x i Represents well logging data; x min Indicates the minimum value of logging data; x max Indicates the maximum value of the logging data.

4. A method for diagnosing well logging instrument faults based on t-test method according to claim 3, characterized in that: Correlation coefficient ξ i The calculation formula of (k) is as follows: in, Indicates the minimum difference between the two poles; represents the maximum difference between the two poles; ρ is the resolution coefficient, and its value range is (0, 1); Δ i (k)=|x'0(k)-x' i (k)| represents the relationship between each point on the reference sequence X'0 curve and each comparison sequence X' i The absolute value difference of each point on the curve; k = 1, 2...n, n is a natural number, representing the number of correlation coefficients.

5. A method for diagnosing well logging instrument faults based on t-test method according to claim 4, characterized in that: The calculation formula of grey relational data is as follows: Among them, γ i Represents grey relational data.

6. A method for diagnosing well logging instrument faults based on t-test method according to claim 1, characterized in that: The specific steps to construct a t-test model include: According to the t-test method, the QQ diagram is used to perform normality test on the grey relational data, and the t-test model that meets the normality test requirements is output.

7. A method for diagnosing well logging instrument faults based on t-test method according to claim 1, characterized in that: Also includes: Inputting the fault diagnosis results of the logging instrument into the fault database; The fault database includes a straight section fault model and a jump fault model.

8. A well logging instrument fault diagnosis system based on t-test method, used to implement the steps of the well logging instrument fault diagnosis method based on t-test method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module, used to acquire the logging data when the logging instrument is in a normal state; The correlation analysis module is used to obtain the grey correlation data between the logging instruments based on the logging data based on the grey correlation analysis method; Model building module, used to build t-test model based on t-test method and grey relational data; The fault diagnosis module is used to input the samples to be tested into the t-test model and output the fault diagnosis results of the logging instrument.

9. A device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the logging instrument fault diagnosis method based on the t-test method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the well logging instrument fault diagnosis method based on the t-test method as described in any one of claims 1 to 7.

Images