Shield tunneling machine tunneling speed prediction method

Abstract

The invention relates to the technical field of shield tunnels, and discloses a shield tunneling machine tunneling speed prediction method. According to the method, historical tunneling data of the shield tunneling machine are collected, the collected data are sorted and classified, abnormal data values are removed, the data are standardized, characteristics related to the tunneling speed are selected, and soil strength, soil humidity and dado pressure are calculated to serve as input factors for prediction. Deriving a prediction formula of the tunneling speed according to the processed data and the selected relevant characteristics of the tunneling speed, setting prediction parameters for formula calculation according to actual working conditions, and substituting the sorted data and the set parameters into the derived formula for speed calculation to obtain the predicted tunneling speed. And comparing the predicted tunneling speed with the actual tunneling speed, calculating a prediction error rate, evaluating prediction accuracy, and adjusting and optimizing a prediction formula and parameters according to a feedback result.

Description

Technical Field

[0001] The invention relates to the technical field of shield tunnels, and in particular to a method for predicting the excavation speed of a shield machine. Background Art

[0002] The tunneling speed of the shield machine is a key parameter in tunnel construction and has a significant impact on the entire construction process. The tunneling speed of the shield machine is directly related to the construction period of the tunnel and affects the overall construction period of the project. Increasing the tunneling speed will significantly shorten the tunnel construction time and reduce engineering costs. Appropriate tunneling speed can reduce safety accidents caused by geological changes, soil compression and other potential risks during construction. Reasonable control of the tunneling speed can effectively ensure the safety of workers. By predicting the tunneling speed, it is possible to effectively judge possible construction risks, such as soil mutations, equipment failures, etc., so as to carry out prevention and emergency preparations and reduce the possibility of unexpected shutdowns.

[0003] Currently, many methods for predicting tunneling speed often focus on the analysis of one or several single features, lack comprehensive consideration of multiple influencing factors, and are unable to fully reflect the various complex factors affecting the tunneling speed, thus reducing the accuracy and reliability of tunneling speed prediction. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a method for predicting the tunneling speed of a shield machine. The method comprises the following steps: collecting historical data of tunneling of the shield machine, including information such as soil type, stratum distribution, shield machine type, propulsion force, cutter head speed and tunneling speed, sorting and classifying the collected data, removing abnormal data values, and standardizing the data for subsequent analysis, selecting features related to the tunneling speed, calculating soil strength, soil moisture and retaining wall pressure as input factors for prediction, deriving a prediction formula for the tunneling speed based on the processed data and the selected features related to the tunneling speed, setting prediction parameters based on actual working conditions for formula calculation, substituting the sorted data and the set parameters into the derived formula for speed calculation to obtain a predicted tunneling speed, comparing the predicted tunneling speed with the actual tunneling speed, calculating the prediction error rate, evaluating the accuracy of the prediction, and adjusting and optimizing the prediction formula and parameters based on the feedback results. The speed prediction formula is derived by integrating multiple features that affect the prediction of the tunneling speed, comprehensively reflecting various complex factors that affect the tunneling speed, improving the accuracy and reliability of the tunneling speed prediction, and solving the above problems.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: a method for predicting the tunneling speed of a shield machine, comprising the following steps:

[0008] S1. Collect soil type data, stratum distribution data, propulsion data, cutter head speed data and initial speed data collected during the tunneling process of the shield machine;

[0009] S2. Arrange and classify all the data collected above, remove abnormal data values, and standardize the data for subsequent analysis;

[0010] S3, select the features related to the excavation speed, calculate the soil strength, soil moisture and retaining wall pressure as the input factors for prediction;

[0011] S4. Based on the processed data and the selected tunneling speed related features, a tunneling speed prediction formula is derived, prediction parameters are set according to the actual working conditions, used for formula calculation, and the sorted data and the set parameters are substituted into the derived formula for speed calculation to obtain the predicted tunneling speed;

[0012] S5. Compare the predicted excavation speed with the actual excavation speed, calculate the prediction error rate, evaluate the accuracy of the prediction, and return to S3 for recalculation if the prediction is inaccurate.

[0013] Preferably, S2 organizes and classifies the soil type, stratum distribution, shield machine type, propulsion force, cutter head rotation speed and excavation speed data as follows:

[0014] A1. Create a data table through SQL database, define the fields of the collected data, use classification labels to classify soil type data into soft soil, sand and clay, classify stratum depth data according to the stratum information recorded at depth, classify shield machine data according to TBM model, classify propulsion force data according to the numerical type kN, classify cutterhead speed data according to the numerical type rpm, and classify excavation speed data according to the numerical type m / h;

[0015] A2. Enter the historical data after the above classification into the data table one by one through batch processing. Batch processing is a computer data processing method used to process data sets or tasks in batches instead of one by one.

[0016] Preferably, the formula for removing abnormal data values ​​from the classified and sorted data in S2 is as follows:

[0017]

[0018] In the formula, Kz i represents the ith data value after removing the anomaly, Xd i represents the value of data point i, Ybi represents the mean of data sample i, Bc i Represents the standard deviation of data sample i, where i represents the number of corresponding data.

[0019] Preferably, the formula for standardizing the classified and sorted data in S2 is as follows:

[0020]

[0021] In the formula, BH represents the standardized data value, Od represents the original data value, ZxOd represents the minimum original data value, and ZdOd represents the maximum original data value. This formula is used to convert each collected data value into the range of [0,1] for computer recognition and analysis.

[0022] Preferably, the formula for calculating soil strength in S3 is as follows:

[0023]

[0024] In the formula, Qu represents soil strength, Njlz represents cohesion, It represents the soil friction angle, which comes from the measurement of soil experiments. The 1 in the formula represents the benchmark value, which is regarded as the basic quantity of the system without other influencing factors.

[0025] Preferably, the formula for calculating soil moisture in S3 is as follows:

[0026]

[0027] In the formula, Ts represents soil moisture, Trsz represents the weight of water in the soil, and Tgdz represents the dry weight volume of the soil. The above values ​​are derived from instrument measurements.

[0028] Preferably, the formula for calculating the wall protection pressure in S3 is as follows:

[0029] Hy=γ*Sdzd+K*(P a -γ*Sdzd)

[0030] In the formula, Hy represents the retaining wall pressure, γ represents the unit weight of the soil, Sdzd represents the depth, K represents the horizontal soil pressure coefficient, which represents the static soil pressure coefficient and has a value range of 0.3-0.5, and P a Represents the effective stress of the surrounding environment, which is obtained by the stress sensor.

[0031] Preferably, the prediction formula for deriving the tunneling speed in S4 is:

[0032] After the above analysis, the characteristics related to the excavation speed include soil strength, soil moisture, arm pressure, cutter head speed and propulsion force, that is, the formula is as follows:

[0033] Jsdj=β0+β1*Qu+β2*Ts+β3*Hy+β4*Dpzs+β5*F+∈

[0034] In the formula, Jsdj represents the predicted excavation speed, β0 represents the intercept, β1, β2, β3, β4, and β5 represent the regression coefficients of the corresponding characteristics, and the regression coefficients are obtained by the least squares method. ∈ represents the random error term.

[0035] Preferably, the step S5 compares the predicted excavation speed with the actual excavation speed and calculates the prediction error rate, and the calculation formula is as follows:

[0036]

[0037] In the formula, Wcsj represents the prediction error rate, Sjds represents the actual excavation speed, and Jsdj represents the predicted excavation speed.

[0038] Preferably, S5 evaluates the accuracy of the prediction according to the prediction error rate, and the evaluation method is as follows:

[0039] When the predicted error rate Wcsj is calculated to be less than 10%, it means that the prediction of the tunneling speed is accurate and reliable, otherwise it means that the prediction of the tunneling speed is unreliable. At this time, return to S3 for a new round of feature selection.

[0040] Compared with the prior art, the present invention provides a method for predicting the tunneling speed of a shield machine, which has the following beneficial effects:

[0041] The present invention collects historical data of shield machine excavation, including information such as soil type, stratum distribution, shield machine type, propulsion force, cutter head rotation speed and excavation speed, organizes and classifies the collected data, removes abnormal data values, and standardizes the data for subsequent analysis, selects features related to the excavation speed, calculates soil strength, soil moisture and retaining wall pressure as prediction input factors, derives a prediction formula for the excavation speed based on the processed data and the selected excavation speed-related features, sets prediction parameters according to actual working conditions for formula calculation, substitutes the organized data and the set parameters into the derived formula for speed calculation to obtain a predicted excavation speed, compares the predicted excavation speed with the actual excavation speed, calculates the prediction error rate, evaluates the accuracy of the prediction, and adjusts and optimizes the prediction formula and parameters based on the feedback results, and derives the speed prediction formula by integrating multiple features that affect the excavation speed prediction, comprehensively reflects various complex factors that affect the excavation speed, and improves the accuracy and reliability of the excavation speed prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the steps of the method of the present invention. DETAILED DESCRIPTION

[0043] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Many current methods for predicting tunneling speed often focus on the analysis of one or several single features, lack comprehensive consideration of multiple influencing factors, and cannot fully reflect the various complex factors that affect tunneling speed, resulting in reduced accuracy and reliability of tunneling speed prediction. Therefore, a method for predicting tunneling speed of shield machine is proposed. Figure 1 , the method comprises the following steps:

[0045] S1. Collect soil type data, stratum distribution data, propulsion data, cutter head speed data and initial speed data collected during the tunneling process of the shield machine;

[0046] In order to effectively collect and analyze the historical data of shield machine excavation, various data related to construction are recorded in real time by using high-precision sensors and monitoring equipment, including soil type, stratum distribution, shield machine model and its working parameters such as propulsion force and cutterhead speed. These sensors can be directly integrated on the shield machine to ensure the accuracy and real-time nature of data collection. Geological exploration technologies such as geological radar and drilling sampling can be used to analyze the stratum characteristics and soil properties of the construction site in detail. These technologies can provide detailed information on soil density, particle composition and moisture conditions, thereby providing data support for subsequent cave excavation strategies;

[0047] S2. Arrange and classify all the data collected above, remove abnormal data values, and standardize the data for subsequent analysis;

[0048] The data was collated and categorized as follows:

[0049] A1. Create a data table through SQL database, define the fields of the collected data, use classification labels to classify soil type data into soft soil, sand and clay, classify stratum depth data according to the stratum information recorded at depth, classify shield machine data according to TBM model, classify propulsion force data according to the numerical type kN, classify cutterhead speed data according to the numerical type rpm, and classify excavation speed data according to the numerical type m / h;

[0050] A2. Enter the classified historical data into the data table one by one through batch processing. Batch processing is a computer data processing method used to process data sets or tasks in batches instead of one by one.

[0051] The formula for removing outliers from the data is as follows:

[0052]

[0053] Outliers are usually caused by measurement errors, data entry errors or extreme environmental factors. Removing these values ​​can significantly improve the accuracy and reliability of the data and ensure that subsequent analysis is based on real and valid data. In the formula, Kz i represents the ith data value after removing the anomaly, Xd i represents the value of data point i, Yb i represents the mean of data sample i, Bc i It represents the standard deviation of data sample i, where i represents the number of corresponding data. When predicting the tunneling speed, abnormal values ​​may lead to parameter deviation and thus cause errors in the prediction results.

[0054] S3, select the features related to the excavation speed, calculate the soil strength, soil moisture and retaining wall pressure as the input factors for prediction;

[0055] The soil strength calculation formula is as follows:

[0056]

[0057] Soil strength is one of the important factors affecting the tunneling speed of shield machines. Calculating soil strength can help engineers understand the bearing capacity of the soil layer, thereby evaluating the feasibility of shield operations and formulating reasonable construction plans. In the formula, Qu represents soil strength, Njlz represents cohesion, It represents the soil friction angle, which comes from the measurement of soil experiments. The 1 in the formula represents the reference value, which is regarded as the basic quantity of the system without other influencing factors. By calculating the soil strength, it can provide a basis for optimizing the operating parameters such as the thrust and cutterhead speed of the shield machine, thereby improving the excavation efficiency and ensuring safety.

[0058] The soil moisture calculation formula is as follows:

[0059]

[0060] Calculating soil moisture helps optimize the proportion and injection volume of mud, ensure the fluidity and effectiveness of mud during construction, thereby improving the soil's retention capacity and avoiding adverse effects on excavation progress. In the formula, Ts represents soil moisture, Trsz represents the weight of water in the soil, and Tgdz represents the dry weight volume of the soil. The above values ​​are measured by instruments. Humidity data can be analyzed in conjunction with other key parameters such as soil strength to comprehensively judge the impact of various factors on excavation speed, thereby formulating a more scientific advancement strategy;

[0061] The formula for calculating the retaining wall pressure is as follows:

[0062] Hy=γ*Sdzd+K*(P a -γ*Sdzd)

[0063] The calculation of arm pressure is an important indicator for evaluating the stability of the stratum during shield construction. Reasonable arm pressure can prevent the damage, collapse or instability of the shield wall and ensure the safety of the construction process. In the formula, Hy represents the shield pressure, γ represents the unit weight of the soil, Sdzd represents the depth, K represents the horizontal soil pressure coefficient, which represents the static soil pressure coefficient and has a value range of 0.3-0.5, P a Represents the effective stress of the surrounding environment. This value is obtained through the stress sensor. The arm pressure plays a vital role in the prediction of the shield machine's excavation speed. It not only improves the safety and efficiency of construction by optimizing construction parameters and strategies, but also provides a scientific basis for the smooth implementation of the project, ensuring the efficiency, economy and safety of tunnel construction.

[0064] S4. Based on the processed data and the selected tunneling speed related features, a tunneling speed prediction formula is derived, prediction parameters are set according to the actual working conditions, used for formula calculation, and the sorted data and the set parameters are substituted into the derived formula for speed calculation to obtain the predicted tunneling speed;

[0065] Among them, after the above analysis, the features related to the excavation speed include soil strength, soil moisture, arm pressure, cutter head speed and propulsion force. That is, the formula for predicting the excavation speed is as follows:

[0066] Jsdj=β0+β1*Qu+β2*Ts+β3*Hy+β4*Dpzs+β5*F+∈

[0067] By real-time calculation and monitoring of the tunneling speed, abnormal conditions in the construction process can be discovered in time to prevent equipment failure or safety accidents caused by excessive tunneling speed. In the formula, Jsdj represents the predicted tunneling speed, β0 represents the intercept, β1, β2, β3, β4, and β5 represent the regression coefficients of the corresponding characteristics, and the regression coefficients are obtained by the least squares method. ∈ represents the random error term. The mutual influence between different parameters can better reflect the complexity of the actual construction process, so that the prediction formula can more accurately calculate the predicted tunneling speed, providing more scientific planning and decision-making for tunneling equipment before construction.

[0068] The specific calculation formula of the regression coefficient is as follows:

[0069] y=X*β+∈

[0070] In the formula, y is an m*1 column vector containing m observed parameter eigenvalues, X is an m*(n+1) matrix, where the first column is all 1s, used to estimate the intercept term β0, and the remaining columns are eigenvalues. β represents a (n+1)*1 column vector containing the regression coefficients we want to estimate;

[0071] By minimizing the residual sum of squares, the estimated values ​​of the regression coefficients are obtained, where:

[0072]

[0073] In the formula, X represents the feature matrix, where each row represents an observation sample and each column corresponds to a feature. T represents the rotation matrix of the feature matrix, (X T *X) -1 The inverse matrix represents the result of transposing the feature matrix and multiplying the feature matrix. This method can be used to obtain the regression coefficients β1, β2, β3, β4 and β5 of the corresponding features.

[0074] S5, comparing the predicted excavation speed with the actual excavation speed, calculating the prediction error rate, evaluating the accuracy of the prediction, and returning to S3 for recalculation if the prediction is inaccurate;

[0075] The prediction error rate is calculated as follows:

[0076]

[0077] The prediction error rate is one of the key indicators to measure the performance of the prediction model. By calculating the error between the predicted value and the actual value, the accuracy and reliability of the prediction formula can be directly evaluated. If the error rate is high, it means that the formula may have defects and needs to be adjusted or optimized. In the formula, Wcsj represents the prediction error rate, Sjds represents the actual tunneling speed, and Jsdj represents the predicted tunneling speed. During the tunneling process of the shield machine, the construction conditions and parameters may change. By continuously monitoring and calculating the error rate, the model can be iterated and improved in the process to make it more adaptable and cope with new environments and new challenges.

[0078] When the prediction error rate of the tunnel boring machine's excavation speed is less than 10%, this result shows that the formula has excellent prediction ability, and the difference between the actual excavation speed and the predicted result is very small, which makes the prediction result highly reliable. This accuracy not only enhances the validity of the data and provides a stable foundation for the construction team, but also enables engineers to more confidently adjust construction strategies, such as propulsion force and cutter head speed, thereby improving overall construction efficiency and reducing resource waste. In addition, an error rate of less than 10% provides an effective feedback system for the construction team, enabling it to quickly adjust various parameters, enhancing the flexibility and safety of the construction process, and improving the decision-making confidence of project management. With the accumulation of historical data, accurate predictions can guide the planning of future projects and promote the intelligent development and innovation of shield machine technology. Therefore, a prediction error rate of less than 10% is not only a proof of the performance of the formula, but also an optimization and guarantee of the entire tunnel engineering construction process.

[0079] By integrating multiple characteristics that affect the prediction of tunneling speed to derive the speed prediction formula, the various complex factors that affect the tunneling speed are fully reflected, thereby improving the accuracy and reliability of the tunneling speed prediction.

[0080] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the tunneling speed of a shield machine, characterized in that: include: S1. Collect soil type data, stratum distribution data, propulsion data, cutterhead speed data and initial speed data during the tunneling process of the shield machine; S2. Arrange and classify all the data collected above, remove abnormal data values, and standardize the data for subsequent analysis; S3, select the features related to the excavation speed, calculate the soil strength, soil moisture and retaining wall pressure as the input factors for prediction; S4. Based on the processed data and the selected tunneling speed related features, a tunneling speed prediction formula is derived, prediction parameters are set according to the actual working conditions, used for formula calculation, and the sorted data and the set parameters are substituted into the derived formula for speed calculation to obtain the predicted tunneling speed; S5. Compare the predicted excavation speed with the actual excavation speed, calculate the prediction error rate, evaluate the accuracy of the prediction, and return to S3 for recalculation if the prediction is inaccurate.

2. A shield machine excavation speed prediction method according to claim 1, characterized in that: The S2 organizes and classifies the soil type, stratum distribution, shield machine type, propulsion force, cutter head speed and excavation speed data as follows: A1. Create a data table through SQL database, define the fields of the collected data, use classification labels to classify soil type data into soft soil, sand and clay, classify stratum depth data according to the stratum information recorded at depth, classify shield machine data according to TBM model, classify propulsion force data according to the numerical type kN, classify cutterhead speed data according to the numerical type rpm, and classify excavation speed data according to the numerical type m / h; A2. Enter the historical data after the above classification into the data table one by one through batch processing. Batch processing is a computer data processing method used to process data sets or tasks in batches instead of one by one.

3. A shield machine excavation speed prediction method according to claim 2, characterized in that: The formula for removing abnormal data values ​​from the classified and sorted data in S2 is as follows: In the formula, Kz i represents the ith data value after removing the anomaly, Xd i represents the value of data point i, Yb i represents the mean of data sample i, Bc i Represents the standard deviation of data sample i, where i represents the number of corresponding data.

4. A shield machine excavation speed prediction method according to claim 3, characterized in that: The formula for standardizing the classified and sorted data in S2 is as follows: In the formula, BH represents the standardized data value, Od represents the original data value, ZxOd represents the minimum original data value, and ZdOd represents the maximum original data value. This formula is used to convert each collected data value into the range of [0,1] for computer recognition and analysis.

5. A shield machine excavation speed prediction method according to claim 4, characterized in that: The S3 formula for calculating soil strength is as follows: In the formula, Qu represents soil strength, Njlz represents cohesion, It represents the soil friction angle, which comes from the measurement of soil experiments. The 1 in the formula represents the benchmark value, which is regarded as the basic quantity of the system without other influencing factors.

6. A shield machine excavation speed prediction method according to claim 5, characterized in that: The formula for calculating soil moisture in S3 is as follows: In the formula, Ts represents soil moisture, Trsz represents the weight of water in the soil, and Tgdz represents the dry weight volume of the soil. The above values ​​are derived from instrument measurements.

7. A shield machine excavation speed prediction method according to claim 6, characterized in that: The formula for calculating the wall pressure of S3 is as follows: Hy=γ*Jdzd+K*(P a -γ*Judge) In the formula, Hy represents the retaining wall pressure, γ represents the unit weight of the soil, Sdzd represents the depth, K represents the horizontal soil pressure coefficient, which represents the static soil pressure coefficient and has a value range of 0.3-0.5, and P a Represents the effective stress of the surrounding environment, which is obtained by the stress sensor.

8. A shield machine excavation speed prediction method according to claim 7, characterized in that: The prediction formula for the tunneling speed derived from S4 is: After the above analysis, the characteristics related to the excavation speed include soil strength, soil moisture, arm pressure, cutter head speed and propulsion force, that is, the formula is as follows: Jsdj=β0+β1*Qu+β2*Ts+β3*Hy+β4*Dpzs+β5*F+∈ In the formula, Jsdj represents the predicted excavation speed, β0 represents the intercept, β1, β2, β3, β4, and β5 represent the regression coefficients of the corresponding characteristics, and the regression coefficients are obtained by the least squares method. ∈ represents the random error term.

9. A shield machine excavation speed prediction method according to claim 8, characterized in that: The S5 compares the predicted excavation speed with the actual excavation speed and calculates the prediction error rate. The calculation formula is as follows: In the formula, Wcsj represents the prediction error rate, Sjds represents the actual excavation speed, and Jsdj represents the predicted excavation speed.

10. A shield machine excavation speed prediction method according to claim 9, characterized in that: The S5 evaluates the accuracy of the prediction according to the prediction error rate, and the evaluation method is as follows: When the predicted error rate Wcsj is calculated to be less than 10%, it means that the prediction of the tunneling speed is accurate and reliable, otherwise it means that the prediction of the tunneling speed is unreliable. At this time, return to S3 for a new round of feature selection.

Images