Savitzky-Golay filtering-based open-type TBM ascending section and stable section distinguishing method

Abstract

The invention provides an open type TBM (Tunnel Boring Machine) ascending section and stable section distinguishing method based on Savitzky-Golay filtering. The method mainly comprises the following four parts: judgment of a complete tunneling period, judgment of a descending section, smoothing processing of data by using a Savitzky-Golay filter, and distinguishing of an ascending section and a stable section. On the basis of the Savitzky-Golay filtering algorithm, the tunneling period data after the descending section is removed are processed to obtain the smooth data, and the smoothed tunneling period data are distinguished from the ascending section and the stable section, so that interference caused by data oscillation is reduced, and the accuracy of the distinguishing method is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of TBM intelligentization, and particularly to a method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering. Background Art

[0002] The construction of an open full-face tunnel boring machine (TBM) has the characteristics of high tunneling efficiency, fast construction, and high comprehensive benefits. Currently, it is an important equipment in the tunneling construction of rail transit, highways, railways, water conservancy projects, etc. TBM construction has become an important direction for tunnel construction in China. When the TBM encounters rock stratum changes or complex geological conditions during construction, geological disasters, TBM jamming, and even casualties are likely to occur.

[0003] With the rise of artificial intelligence, in order to improve the tunneling efficiency of TBM, reduce the probability of construction hazards and abnormal events, and improve the intelligent level of TBM has become a hot direction in the current tunnel construction field. Many experts and scholars have done some research on TBM intelligentization. Currently, there are research results such as establishing a tunneling parameter prediction model based on tunneling data, using a knowledge-data dual-drive to make an intelligent decision-making system, and predicting cutter head wear using cutter head data.

[0004] However, in the current research results, the work is mainly done at the algorithm and model levels. The data generated during TBM construction is complex and has a lot of noise, which is very likely to affect the research results of the data. During the TBM construction process, there are three parts of data: the ascending section, the stable section, and the descending section. The data trends shown by these three parts of data are different, and the meanings represented are also different. If the data is not distinguished, it will affect the judgment of the accuracy of the research results. In the descending section, it is mainly the rapid decline of data caused by shutdown. This part of the data lacks research value, so it is discarded. The main thing is to distinguish the data of the ascending section and the stable section. Distinguishing these two parts of data will bring improvements to models, algorithms, etc. at the data level, and will bring great help to the development of TBM intelligentization. Therefore, it is very important to distinguish the ascending section and the stable section of TBM data.

[0005] However, the existing methods for dividing the ascending section and the stable section of tunnel boring machines are mostly based on simple conditional judgments or mean statistics of the original data, and there are the following problems: the trend of data change is not expressed, which will make the judgment result easily affected by data fluctuations, resulting in inaccurate section division judgment. Summary of the Invention

[0006] In view of the above problems, the present invention is proposed to provide a method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering, which can overcome or at least partially solve the above problems. The method can solve the problem that the existing section division method cannot express the trend of data change, resulting in inaccurate section division judgment, and achieve the purpose of accurately judging the ascending section and the stable section of the tunnel boring machine.

[0007] Specifically, the present invention provides a method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering. The distinguishing method includes the following steps:

[0008] Based on the tunneling data collected by the sensors on the TBM;

[0009] Adopt the complete tunneling cycle judgment method to judge the complete tunneling cycle of the tunneling data;

[0010] Adopt the descending section identification method in the complete tunneling cycle to judge the descending section data, and remove the descending section data from the tunneling cycle data;

[0011] Perform Savitzky-Golay filtering on the tunneling cycle data after removing the descending section data to obtain the smoothed tunneling cycle data F;

[0012] According to the smoothed tunneling cycle data F obtained after the filtering process, first calculate the average value M of the data after each time point n ; According to the obtained average value M n , calculate each average value M n The left derivative L of the new data point n And the right derivative R n , then traverse all points to find the first demarcation point that meets the conditions of the tunneling ascending section and stable section distinguishing method, and this point is the point where the tunneling ascending section and the stable section are distinguished.

[0013] Optionally, performing Savitzky-Golay filtering on the tunneling cycle data after removing the descending section data to obtain the smoothed tunneling cycle data specifically further includes: adopting Savitzky-Golay filtering parameter selection, with a window length of 21 and a polynomial order of 3.

[0014] Optionally, according to the smoothed tunneling cycle data F obtained after the filtering process, first calculate the average value M of the data after each time point n ; According to the obtained average value M n , calculate each average value M n The left derivative L of the new data point n And the right derivative R n, then traverse all points to find the first point that meets the conditions of the distinguishing method for the upward excavation section and the stable section. This point is the one that separates the upward excavation section from the stable section.

[0015] Optionally, use the complete tunneling cycle judgment method to judge the complete tunneling cycle of the tunneling data, which specifically further includes: A complete tunneling cycle refers to the time period from the start state of the TBM to the stop state.

[0016] Judge the start time point of the complete tunneling cycle:

[0017] The total propulsion thrust > 0, the propulsion state = 1, and the total propulsion thrust at the previous time point = 0; if the time point meets the above conditions, then define this time point as the start time point of the complete tunneling cycle.

[0018] Starting from the start time point of the complete tunneling cycle, judge the end time point of the complete tunneling cycle:

[0019] The total propulsion thrust = 0, the propulsion state = 0; if there is a subsequent time point that meets the conditions, then this time point is the end time point of the complete tunneling cycle.

[0020] For the time between the start time point and the end time point of the complete tunneling cycle, including both end time points, it is a complete tunneling cycle.

[0021] Optionally, use the downward section identification method in the complete tunneling cycle, which specifically further includes:

[0022] Downward section identification method: In a complete tunneling cycle, the stage from the first time point when the propulsion state = 0 (indicating tunneling stop) to the end of the complete tunneling cycle belongs to the downward section.

[0023] Optionally, the tunneling data includes: total propulsion thrust, propulsion state, and time.

[0024] A method for distinguishing the upward section and the stable section of an open TBM based on Savitzky-Golay filtering of the present invention has the following beneficial effects:

[0025] 1. The existing methods for distinguishing the upward section and the stable section do not express the trend of data changes, which will make the distinguishing results easily affected by data oscillations, resulting in inaccurate distinguishing. The present invention uses Savitzky-Golay filtering to smooth the data. Based on the smoothed data, it is easier to find the trend of data changes. Then, by judging the changes in the left and right derivatives and the degree of closeness to the overall value, the dividing points of the upward section and the stable section are found, reducing the interference caused by data oscillations and improving the accuracy of the method for distinguishing the upward section and the stable section of the tunnel boring machine.

[0026] 2. By combining the discrimination method with data mining technology or a data processing system, it is possible to automatically and batch-discriminate the ascending section and the stable section of TBM data, providing support for the work of integrating data sets in the intelligent research of TBM, improving the efficiency of TBM data processing, and providing data support for researching algorithms to improve construction efficiency and reduce construction risks.

[0027] Those skilled in the art will better understand the above and other objects, advantages, and features of the present invention from the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Some specific embodiments of the present invention will be described in detail hereinafter with reference to the accompanying drawings in an exemplary but not restrictive manner. The same reference numerals in the drawings denote the same or similar components or parts. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings:

[0029] Figure 1 is a flowchart of a method for discriminating the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention;

[0030] Figure 2 is the original data of a complete tunneling cycle of a method for discriminating the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention;

[0031] Figure 3 is the smoothed data after Savitzky-Golay filtering processing of a method for discriminating the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention;

[0032] Figure 4 is the boundary result diagram of the tunneling ascending section and the stable section of a method for discriminating the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] The following refers to Figures 1 to 4A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention will be described. In the description of this embodiment, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of these features, that is, include one or more of these features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined. When a certain feature "includes or contains" a certain or certain features it covers, unless otherwise specifically described, this indicates that other features are not excluded and other features may be further included.

[0034] In addition, in the description of this embodiment, the first feature being "above" or "below" the second feature may include the first and second features being in direct contact, or may include the first and second features not being in direct contact but being in contact through additional features therebetween. That is, in the description of this embodiment, the first feature being "above", "over", and "on" the second feature includes the first feature being directly above and obliquely above the second feature, or merely indicating that the first feature has a higher horizontal height than the second feature. The first feature being "under", "beneath", or "below" the second feature may be the first feature being directly below or obliquely below the second feature, or merely indicating that the first feature has a lower horizontal height than the second feature.

[0035] In the description of this embodiment, the description with reference to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

[0036] Figure 1 is a flowchart of a method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to an embodiment of the present invention, as Figure 1 shown, and with reference to Figures 2 to 4 , an embodiment of the present invention provides a method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering. The distinguishing method includes the following steps:

[0037] S1: Based on the tunneling data collected by the sensors on the TBM;

[0038] S2: Use the complete tunneling cycle judgment method to judge the complete tunneling cycle of the tunneling data;

[0039] S3: Use the identification method for the descending section in the complete tunneling cycle to judge the descending section data and remove the descending section data from the tunneling cycle data;

[0040] S4: Perform Savitzky-Golay filtering on the tunneling cycle data after removing the descending section data to obtain the smoothed tunneling cycle data F;

[0041] S5: According to the smoothed tunneling cycle data F obtained after the filtering process, first calculate the average value M of the data after each time point n ; According to the obtained average value M n , calculate the left derivative L n and the right derivative R n of the new data point for each average value M n , and then traverse all points to find the first demarcation point that meets the conditions of the tunneling rising section and stable section discrimination method, and this point is the demarcation point between the tunneling rising section and the stable section.

[0042] In some embodiments of the present invention, as Figure 1 shown, in step S1, the tunneling data includes: total thrust of propulsion, propulsion state, and time.

[0043] In some embodiments of the present invention, as Figures 1 to 2 shown, in step S2, it specifically further includes: a complete tunneling cycle refers to the time period from the start state to the stop state of the TBM;

[0044] During TBM construction, combined with the changing characteristics of the tunneling data, from the rising section to the stable section, the tunneling parameter with a gradually increasing trend and relatively stable is the total thrust of propulsion. Combining with the tunneling state, the start point and stop point can be accurately found, so as to judge the complete tunneling cycle.

[0045] S2-1: Judge the start time point of the complete tunneling cycle: total thrust of propulsion > 0, propulsion state = 1 (indicating the start of tunneling), and the total thrust of propulsion at the previous time point = 0; if the time point meets the above conditions, then define this time point as the start time point of the complete tunneling cycle;

[0046] S2-2: Starting from the start time point of the complete tunneling cycle, judge the end time point of the complete tunneling cycle: total thrust of propulsion = 0, propulsion state = 0; if there is a subsequent time point that meets the conditions, then this time point is the end time point of the complete tunneling cycle;

[0047] S2-3: For the time between the start time point and the end time point of a complete tunneling cycle (including both time points), it is a complete tunneling cycle. Data of a complete tunneling cycle is as Figure 2 shown. The vertical coordinate in the figure is the total thrust of propulsion, and the horizontal coordinate is the serial number sorted by time.

[0048] In some embodiments of the present invention, as Figure 1 shown, in step S3, it specifically further includes: a descending section identification method. In a complete tunneling cycle, the stage from the time point when the first propulsion state = 0 (indicating tunneling stop) to the end of the complete tunneling cycle belongs to the descending section. Specifically, the TBM tunneling parameters include the propulsion state. If in a tunneling cycle, the first time point with the propulsion state = 0 appears, it means that the TBM stops tunneling, and the subsequent data is generated by the inertia of the TBM itself. The stage from the time point when the first propulsion state = 0 to the end of the tunneling cycle belongs to the descending section.

[0049] In some embodiments of the present invention, as Figure 1 and Figure 3 shown, in step S4, it specifically further includes: The Savitzky-Golay filtering algorithm is an algorithm for smoothing data based on the method of fitting polynomials. It fits polynomials within a sliding window to smooth the data, thereby removing noise while maximizing the retention of data details and peaks. Specifically, it selects a fixed-size sliding window on the signal and uses polynomials (usually low-order polynomials) to fit the data within each window. By calculating the value of the fitted polynomial at the center point of the window as the filtering result of this point. In this way, by moving the window and repeating the above process, the smoothing process of the entire signal can be realized. The key parameters include the window length and the polynomial order. The window length is a positive odd integer representing the number of adjacent points used for fitting; the polynomial order is a positive integer representing the order of the polynomial used for fitting. The higher the polynomial order, the higher the complexity of the fitting, which can better retain the details of the signal, but may also lead to overfitting and noise amplification.

[0050] In this embodiment, the parameter selection for Savitzky-Golay filtering processing is that the window length is 21 and the polynomial order is 3. Based on the above parameters, Savitzky-Golay filtering processing is performed to obtain new smoothed data F, as Figure 3 shown. The vertical coordinate in the figure is the total thrust of propulsion, and the horizontal coordinate is the serial number sorted by time.

[0051] By calculating the average value of all data after each point of the filtered data, the vibration of the data can be maximally reduced to ensure relatively stable data. Especially during the process from the rising section to the stable section, the data gradually grows more smoothly, which provides great help for distinguishing the rising section and the stable section.

[0052] At the demarcation point between the rising section and the stable section, the left side of the total thrust data shows an upward trend, and the right side shows a stable state. The derivative can represent the local upward or stable state. Therefore, the left and right derivatives are used to quantify the upward and stable states. In addition, relying solely on the upward or stable trend to judge the demarcation point between the rising section and the stable section has a drawback, that is, it cannot determine whether the specific value has reached the stable state. Therefore, by combining the logic of comparing the specific value of the current point with the subsequent average value, this drawback can be compensated, and thus the rising section and the stable section can be accurately distinguished.

[0053] In some embodiments of the present invention, as Figure 1 and Figure 4 shown, in step S5, it specifically further includes:

[0054] S5-1: Calculate the average value M of the data after each time point n . Assume that the current time point is the nth point and there are a total of m time points, then the average value M of the data after this time point n The calculation formula for the new data is:

[0055]

[0056] where F i is the data of the i-th point of the smoothed tunneling cycle data F obtained after Savitzky-Golay filtering processing.

[0057] S5-2: Calculate the left derivative L n of each new data point of the average value M n . Assume that the current new data point is the nth point, then the left derivative L n is calculated as:

[0058]

[0059] where M n is the average value after the nth point, and M n-9 is the average value after the (n - 9)th point.

[0060] S5-3: Calculate the right derivative R n of each new data point of the average value M n . Assume that the current new data point is the nth point, then the right derivative R n is calculated as:

[0061]

[0062] Among them, M n is the average value after the nth point, and M n+9 is the average value after the (n + 9)th point.

[0063] S5-4: Method for distinguishing the driving ascending section and the stable section: If the nth point simultaneously satisfies the following conditions, then it can be determined that n is the demarcation point between the driving ascending section and the stable section.

[0064] |L n - 1| ≤ 0.05,

[0065] |R n - 1| ≤ 0.05,

[0066] |F n - M n | < M n * 0.03

[0067] Among them, L n is the left derivative of the new data of the average value M n , R n is the right derivative of the new data of the average value M n , and F n is the data of the nth point of the smoothed driving cycle data F obtained after Savitzky-Golay filtering processing.

[0068] According to the distinguishing algorithm for the ascending section and the stable section, the found demarcation point, as Figure 4 shown, the ordinate in the figure is the total driving thrust, and the abscissa is the serial number sorted by time.

[0069] Up to this point, those skilled in the art should recognize that although multiple exemplary embodiments of the present invention have been shown and described in detail herein, still, without departing from the spirit and scope of the present invention, many other variations or modifications that conform to the principles of the present invention can be directly determined or derived based on the content disclosed in the present invention. Therefore, the scope of the present invention should be understood and determined to cover all these other variations or modifications.

Claims

1. A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering, characterized in that, It includes the following steps: Based on the tunneling data collected by sensors on the TBM; Adopt the complete tunneling cycle judgment method to judge the complete tunneling cycle of the tunneling data; Adopt the identification method of the descending section in the complete tunneling cycle to judge the descending section data and remove the descending section data from the tunneling cycle data; Perform Savitzky-Golay filtering on the tunneling cycle data after removing the descending section data to obtain the smoothed tunneling cycle data F; According to the smoothed tunneling cycle data F obtained after filtering processing, first calculate the average value M of the data after each time point n ; According to the obtained average value M n , calculate the left derivative L n of the new data point for each average value M n and the right derivative R n , and then traverse all points to find the first demarcation point that meets the conditions of the tunneling rising section and stable section discrimination method, and this point is the point that separates the tunneling rising section and the stable section.

2. The method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to claim 1, wherein, The performing Savitzky-Golay filtering on the tunneling cycle data after removing the descending section data to obtain the smoothed tunneling cycle data specifically further includes: adopting the Savitzky-Golay filtering parameter selection, with the window length of 21 and the polynomial order of 3.

3. A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to claim 1, characterized in that Based on the smoothed tunneling cycle data F obtained after filtering processing, first calculate the average value M of the data after each time point. n ; Based on the obtained average value M n , calculate the left derivative L n of each new data point of the average value M n and the right derivative R n , then traverse all points to find the first demarcation point that meets the conditions of the tunneling rising section and stable section discrimination method, and this point is the point that separates the tunneling rising section and the stable section.

4. A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to claim 1, characterized in that, The adopting the complete tunneling cycle judgment method to judge the complete tunneling cycle of the tunneling data specifically further includes: A complete tunneling cycle refers to the time period from the start state to the stop state of the TBM; Judge the start time point of the complete tunneling cycle: The total thrust of propulsion > 0, the propulsion state = 1, and the total thrust of propulsion at the previous time point = 0; if the time point meets the above conditions, then define this time point as the start time point of the complete tunneling cycle; Starting from the start time point of the complete tunneling cycle, judge the end time point of the complete tunneling cycle: The total thrust of propulsion = 0, the propulsion state = 0; if there is a subsequent time point that meets the conditions, then this time point is the end time point of the complete tunneling cycle; For the time between the start time point and the end time point of the complete tunneling cycle, including both end time points, it is a complete tunneling cycle.

5. A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to claim 1, characterized in that, The adopting the identification method of the descending section in the complete tunneling cycle specifically further includes: Descending section identification method: In a complete tunneling cycle, the stage from the first time point when the propulsion state = 0 (indicating tunneling stop) to the end of the tunneling cycle belongs to the descending section.

6. A method for distinguishing the ascending section and the stable section of an open TBM based on Savitzky-Golay filtering according to claim 1, characterized in that, The tunneling data includes: the total thrust of propulsion, the propulsion state, and time.

Images