Man-machine work efficiency quantitative decision-making method for sandstone compound stratum of shield tunneling machine

Abstract

The invention discloses a man-machine work efficiency quantitative decision-making method for a sandstone compound stratum of a shield tunneling machine. The method comprises the following steps: S1, defining a man-machine comprehensive work efficiency model based on a tunneling rate, a cutter loss rate and unit energy consumption; s2, screening independent variables; s3, establishing a multiple linear regression model by taking a man-machine comprehensive work efficiency model calculation value as a dependent variable and combining the screened independent variable; s4, performing variable standardization on the model and converting the model into a standard regression equation; s5, calculating the influence weight of each standardized variable; and S6, sorting the influence weights, and selecting a decision direction according to a result. According to the method, multi-dimensional parameters are integrated, the influence of the multi-dimensional parameters on the tunneling efficiency is quantified, a dynamic response scheme can be generated, the tunneling speed is increased, loss is reduced, and the continuous operation period of the shield tunneling machine is prolonged.

Description

Technical Field

[0001] This application relates to the field of shield tunneling scheme optimization technology, specifically to a quantitative decision-making method for human-machine efficiency in sandstone composite strata for shield tunneling machines. Background Technology

[0002] During tunnel excavation, tunnel boring machines (TBMs) face the problem of low human-machine efficiency when operating in sandstone composite strata. Specifically, the uneven rock hardness and frequent interface changes in sandstone composite strata often lead to abnormal wear of the cutting tools, requiring repeated adjustments to the excavation parameters, which greatly increases the complexity of human operation.

[0003] Current technologies primarily rely on experience to adjust equipment parameters and manpower allocation. However, experience sometimes struggles to reconcile the challenges of adjusting equipment parameters with the demands of manpower allocation, leading to either excessive manpower or reduced efficiency in adjusting equipment parameters. Essentially, experience-based decision-making isolates factors influencing work efficiency. It fails to systematically integrate geological parameters, equipment parameters, and personnel factors, and it lacks data support. When resource allocation depends on subjective experience, it becomes difficult to pinpoint key influencing factors. Summary of the Invention

[0004] This application provides a quantitative decision-making method for human-machine efficiency in sandstone composite formations for tunnel boring machines, and presents the collaborative quantitative results of human-machine-geology from a system perspective.

[0005] The technical solution of this application is as follows: A method for quantifying human-machine efficiency decision-making in sandstone composite strata for tunnel boring machines includes the following steps: S1. Define a human-machine integrated efficiency model based on tunneling rate, tool wear rate, and unit energy consumption; tunneling rate, tool wear rate, and unit energy consumption can all be obtained directly through monitoring.

[0006] S2. Select independent variables for the human-machine integrated ergonomics model from the initial factor pool; S3. Using the calculated values ​​of the human-machine integrated ergonomics model as the dependent variable and based on the independent variables selected in step S2, establish a multiple linear regression model. S4. Standardize the variables of the multiple linear regression model and transform it into a standard regression equation; S5. Calculate the influence weights of each standardized variable in the standard regression equation; S6. Sort the influence weights and select the decision direction based on the sorting results.

[0007] Furthermore, in step S2, the initial factor pool includes the coefficient of variation of sandstone compressive strength, density of soft and hard rock interface, gravel content, groundwater level, rock stratum dip angle, cutterhead torque fluctuation rate, propulsion pressure gradient, grouting pressure deviation, screw conveyor speed, articulated cylinder pressure difference, main drive temperature, slag temperature, team skill matching degree, abnormal response delay time, pre-shift meeting duration, operator fatigue, team communication frequency, and average age.

[0008] Furthermore, the independent variables selected in step S2 include: The independent variables of the tunneling rate include the coefficient of variation of sandstone compressive strength, density of the soft and hard rock interface, and gravel content; The independent variables of tool wear rate include cutterhead torque fluctuation rate, feed pressure gradient, and grouting pressure deviation. The independent variable for unit energy consumption includes the abnormal response delay time.

[0009] Furthermore, in step S3, a multiple linear regression model is established using the time-series data of the dependent and independent variables collected in real time.

[0010] Furthermore, in step S5, the standard regression equation is validated using leave-one-out cross-validation. The variance inflation factor is used to eliminate multicollinearity interference.

[0011] Furthermore, in step S6, standardized variables with an influence weight greater than 10% are defined as key variables, and decisions are made on all key variables.

[0012] Due to the adoption of the above technical solution, the beneficial effects of this application are as follows: 1. This application integrates geological parameters, equipment parameters, and personnel parameters, which can quantify the impact of multi-dimensional parameters on tunneling efficiency, and select dimensions with greater influence from the quantification results for targeted improvement.

[0013] 2. This application is based on real-time calculation of parameters collected in real time, which can generate dynamic response parameter schemes, improve the tunneling team's understanding of complex strata, facilitate real-time adjustments, and accumulate experience for subsequent tunneling work.

[0014] 3. The technical solution of this application can improve tunneling speed, reduce tool wear, and reduce unplanned downtime in long-term practice, thereby extending the continuous operation cycle of the tunnel boring machine in sandstone strata. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application.

[0016] Figure 1 A flowchart of a human-machine efficiency quantification decision-making method for tunnel boring machines in sandstone composite strata, provided in this application. Detailed Implementation

[0017] Based on the background technology described above, please refer to the appendix. Figure 1 This application provides a quantitative decision-making method for human-machine efficiency in sandstone composite strata for tunnel boring machines, including the following steps: S1. Define a human-machine integrated efficiency model based on tunneling rate, tool wear rate, and unit energy consumption; The human-machine integrated ergonomics model is as follows: ; In the formula, This represents the calculated value of overall human-machine efficiency. Indicates the tunneling rate. Indicates tool wear rate. Indicates unit energy consumption. The weighting coefficient is the industry standard; tunneling rate, tool wear rate, and unit energy consumption can all be obtained directly through monitoring.

[0018] The entropy weighting method determines the weight of each indicator based on its dispersion within the historical dataset. Indicators with greater dispersion are considered to contain more information and thus have a higher weight. The steps of the entropy weighting method are as follows: (1) Collect the original data of m shield tunneling sections to form an m×3 dimensional data matrix; (2) Standardize the raw data to eliminate the influence of dimensions; (3) Calculate the first j The first item under the indicator i The proportion of each sample value ; (4) Calculate the first j Entropy value of the item index ; (5) Calculate the coefficient of variation for each indicator. ; (6) Normalize the difference coefficients to obtain the weights of each indicator.

[0019] S2. Select independent variables for the human-machine integrated ergonomics model from the initial factor pool; The tunneling rate is related to geology, so its independent variable is classified as geology; the tool wear rate is classified as equipment; and the unit energy consumption is classified as personnel.

[0020] Independent variables are sorted by their ordinal numbers: Geological independent variables Arguments of the device class The independent variable of the personnel class ; In this embodiment, the geological independent variables include the coefficient of variation of sandstone compressive strength, density at the interface between soft and hard rock, and gravel content; the equipment independent variables include cutterhead torque fluctuation rate, propulsion pressure gradient, and grouting pressure deviation; and the personnel independent variables include the team skill matching degree and abnormal response delay time. Among these independent variables, the geological and equipment independent variables can be obtained through monitoring. The statistical methods for the personnel independent variables are: personnel operation certificate level rating and monitoring system timescale comparison.

[0021] The independent variables for geology include: The coefficient of variation of sandstone compressive strength, density at the interface between soft and hard rocks, gravel content, groundwater level, and rock strata dip angle.

[0022] The coefficient of variation of sandstone compressive strength represents the ratio of the standard deviation to the mean of the uniaxial compressive strength of sandstone within a segment, reflecting the lithological uniformity. In this embodiment, it was selected as an independent variable because it is significantly related to tool wear and feed speed fluctuations.

[0023] The density of the soft and hard rock interface represents the number of interfaces between soft and hard rock per unit tunneling length. In this embodiment, due to the frequent changes in the interface, the parameters are frequently adjusted, which is a core characteristic of the composite strata and is therefore selected as an independent variable.

[0024] Gravel content represents the percentage of particles larger than 5 mm in the soil. In this embodiment, it was selected as an independent variable because it directly affects the cutter head torque and the impact wear of the cutter.

[0025] The groundwater level elevation represents the water level height at the center of the tunnel section. In earth pressure shield tunneling, groundwater can be effectively controlled through soil improvement. Its correlation with work efficiency is weak, so it is not selected.

[0026] The dip angle of a rock stratum represents the angle between the rock stratum interface and the horizontal plane. If the dip angle of the bottom layer in a region does not vary much, the VIF value will be too high, indicating multicollinearity between the density of the soft and hard rock interfaces. In this embodiment, the dip angle of the rock stratum is not selected as an independent variable.

[0027] The independent variables for equipment include: cutterhead torque fluctuation rate, propulsion pressure gradient, grouting pressure deviation, screw conveyor speed, articulated cylinder pressure difference, main drive temperature, and slag temperature.

[0028] The cutterhead torque fluctuation rate represents the ratio of the standard deviation of the torque to the mean, reflecting the stability of the equipment load. In this embodiment, it is selected as an independent variable because it reflects the frequency and magnitude of the operator's adjustments to changes in the formation.

[0029] The grouting pressure deviation value represents the degree of deviation between the actual grouting pressure and the set pressure range. In this embodiment, this indicator will affect the stability of the tunnel segments. If the grouting pressure deviation is too large, it will easily lead to the displacement of the tunnel segments and require additional correction work, which will occupy the effective tunneling time and have a direct impact on the overall efficiency of manpower and machinery. Therefore, it was selected as an independent variable.

[0030] The rotational speed of the screw conveyor is a key parameter for controlling the amount of soil discharged. In this embodiment, it is highly linearly correlated with the "propulsion speed", which will cause serious collinearity. In order to avoid model redundancy and ensure calculation accuracy, and since the "propulsion speed" can already reflect the impact of soil discharge on work efficiency, it was not selected as an independent variable.

[0031] The pressure difference of the articulated hydraulic cylinder is used to control the pressure difference value of the shield machine attitude. In this embodiment, although it has a certain effect on the adjustment of the shield machine attitude, its influence is indirectly reflected in the "propulsion pressure gradient" and "tunneling trajectory deviation". If it is included in the model, it will cause the variable to be calculated repeatedly and interfere with the quantitative analysis of the efficiency influencing factors. Therefore, it was not selected as an independent variable.

[0032] The main drive temperature refers to the temperature of the main bearing. In this embodiment, the main drive temperature is always kept within a stable threshold range under normal operating conditions, and the data variance is extremely small. It is impossible to reflect the impact on the overall human-machine efficiency through temperature changes. The temperature will only fluctuate abnormally when the equipment fails. However, the fault data has been removed during the data acquisition stage, so it was not selected as an independent variable.

[0033] Slag temperature refers to the slag temperature at the screw conveyor outlet. In this embodiment, the analysis found that it is related to the wear of the cutting tools. However, the amount of slag temperature data accumulated in the current project is insufficient to meet the data sample size requirements of the multiple linear regression model, making it difficult to accurately quantify its impact on the overall efficiency of human and machine operations. Therefore, it was not selected as an independent variable.

[0034] The independent variables for personnel include: abnormal response delay time, pre-shift meeting duration, operator fatigue, team communication frequency, and average age.

[0035] The team skill matching degree is a comprehensive score calculated based on the team members' qualifications, years of service, and historical performance. In this embodiment, all team members have many years of work experience, and the differences between the members are not significant, so they were not selected as independent variables.

[0036] The abnormal response delay time refers to the average time from system alarm to manual intervention and adjustment. In this embodiment, this indicator directly reflects the judgment, experience level and emergency handling ability of personnel. The shorter the response delay, the more the tunneling interruption time caused by abnormal situations can be reduced, which has a significant impact on the overall efficiency of human and machine work. Therefore, it was selected as an independent variable.

[0037] The duration of the pre-shift meeting refers to the duration of the daily pre-shift technical briefing and safety education. In this embodiment, it has no significant linear relationship with the overall efficiency of human and machine operations. The core value of the pre-shift meeting is to ensure construction safety rather than directly improve tunneling efficiency, and it cannot have a substantial impact on efficiency. Therefore, it was not selected as an independent variable.

[0038] Operator fatigue is estimated based on working hours and physiological monitoring (such as eye trackers). In this embodiment, on the one hand, the measurement cost is high, requiring additional investment in physiological monitoring equipment; on the other hand, it involves operator privacy and is difficult to form a unified, standardized quantitative indicator, resulting in poor engineering measurability. Even when "shift time" is used as a proxy variable, statistics show that its correlation with work efficiency is still not significant, so it was not selected as an independent variable.

[0039] Team communication frequency refers to the number of wireless calls within the team per hour. In this embodiment, the actual collected data has a large dispersion, and the correlation test found that its relationship with "abnormal response delay time" and human-machine comprehensive work efficiency is unclear. Therefore, it is not selected as an independent variable because the impact on work efficiency cannot be accurately quantified through communication frequency.

[0040] Average age refers to the average age of team members. In this embodiment, this indicator is ethically sensitive, and the "team skill matching degree" already covers key information such as members' experience and operational ability. Age itself is not a good indicator of the relationship between personnel ability and work efficiency, and it cannot effectively quantify its impact on human-machine comprehensive work efficiency. Therefore, it was actively excluded and not selected as an independent variable.

[0041] S3. Using the calculated values ​​of the human-machine integrated ergonomics model as the dependent variable and based on the independent variables selected in step S2, establish a multiple linear regression model.

[0042] The basic equation of the multiple linear regression model is: ; In the formula, Represent the fundamental equation, The intercept is... These are partial regression coefficients. This is the error term; A multiple linear regression model is established using real-time collected time-series data of the dependent and independent variables. In this embodiment, in N Real-time data collection was performed on sections of a sandstone composite stratum. The data is time-series data. The data acquisition process requires removing outliers, such as data generated during periods of equipment failure, and normalizing the dimensions of the acquired data.

[0043] S4. Standardize the variables of the multiple linear regression model and transform it into a standard regression equation.

[0044] The formula for standardizing variables is: ; In the formula, This represents the mean, and its subscript indicates the parameter it belongs to; σ The standard deviation is indicated by its subscript, which represents the parameter it pertains to. and These represent the standardized independent and dependent variables, respectively.

[0045] The standardized regression equation is: ; In the formula, Represents the standard regression coefficient; S5. Calculate the influence weights of each standardized variable in the standard regression equation; ; In the formula, Indicates the influence weight. These are the terms of the standard regression coefficients in the above regression equation.

[0046] After the calculation was completed, the standard regression equation was validated using leave-one-out cross-validation. The variance inflation factor is used to eliminate multicollinearity interference.

[0047] S6. Sort the influence weights and select the decision direction based on the sorting results.

[0048] In practical implementation, in addition to ranking, key variables should not be ignored. In this embodiment, standardized variables with an influence weight greater than 10% are defined as key variables, and decisions are made based on all key variables.

[0049] For example, when the weight of geological factors is relatively high, the decision-making direction is to adjust the cutter head speed and thrust; When the weight of equipment factors is high, the decision-making direction tends to favor the real-time torque control system; When the weight of human factors is high, it is necessary to conduct skills enhancement training for personnel.

[0050] In this application, the result is calculated in real time through step S1. Then, by fitting the data to multiple linear regression... To fit.

[0051] Example: Taking a subway tunnel crossing a river as an example: The geological features consist of a sandstone composite stratum comprising quartz sandstone (compressive strength 85-120 MPa, accounting for 65%) and silty mudstone interlayers (compressive strength 20-35 MPa, accounting for 35%), with a dip angle of 15°–28° and a gravel content of 8%–12%. A Herrenknecht earth pressure balance tunnel boring machine (6.7m in diameter) is used. The personnel configuration is a 3-shift rotation, with 6 people per shift (including the main operator, maintenance personnel, and monitoring staff).

[0052] First, calculate the dependent variable. In practice, tunneling rate, tool wear rate, and unit energy consumption can all be directly obtained through monitoring. The values ​​are 0.4, 0.3, and 0.3 respectively.

[0053] The selected geological independent variable is: sandstone strength variation coefficient. Density at the interface between soft and hard rocks , Obtained through core laboratory testing. Obtained through ground-penetrating radar scanning.

[0054] The independent variable for the selected equipment is: cutter head torque fluctuation rate. and propulsion pressure gradient , Record in real time through the PLC system Monitoring is conducted using hydraulic sensors.

[0055] The independent variable for personnel selection is: skill matching degree within the work group. and abnormal response delay The skill matching degree of the work team is rated based on the operator's operation certificate level. In practice, experts can also evaluate and score the performance. Obtained by comparing time stamps with the monitoring system.

[0056] After collecting the data, the standardized regression equation was established as follows: ; The weighting of each influence is ranked as follows: Cutter head torque fluctuation The standardized regression coefficient is 0.52, the influence weight is 22.3%, and the ranking is 1. Team skill matching The standardized regression coefficient is 0.47, the influence weight is 20.2%, and the ranking is 2. Sandstone strength variation coefficient The standardized regression coefficient is 0.41, the influence weight is 17.6%, and the ranking is 3. Abnormal response delay The standardized regression coefficient is -0.31, the influence weight is 13.3%, and the ranking is 4. Density at the interface between soft and hard rocks The standardized regression coefficient is -0.28, the influence weight is 12.0%, and the ranking is 5. Promoting the pressure gradient The standard regression coefficient is -0.19, the influence weight is 8.1%, and the ranking is 6.

[0057] Model validation was performed, and the model fit was R² = 0.87, which is highly significant.

[0058] Cross-validation error: RMSE=0.13, which is less than 0.15 and meets the requirements.

[0059] Multicollinearity detection: Maximum VIF = 3.8 < 5, no severe multicollinearity.

[0060] In this embodiment, there are many key factors exceeding 10%, so we select independent variables exceeding 15% for optimization. The optimization measures are as follows: Cutter head torque fluctuation The installation of a torque adaptive control system automatically reduces the rotational speed by 5%-8% when the geological radar detects a sudden change in sandstone strength, reducing the measured torque fluctuation rate from ±25% to ±12%.

[0061] Team skill matching Develop dynamic skills training programs: When sandstone accounts for more than 60% of the total rock mass: Strengthen training on tunneling operations in high-hardness rock formations (100% pass rate in simulator assessment); For the soft and hard rock interface section, geological map reading and parameter switching drills will be added (the transition time for each shift will be shortened by 3.2 minutes).

[0062] Sandstone strength variation coefficient The pre-crushing grouting process was adopted: chemical softeners were injected into the high-pressure sandstone section through drilling, which reduced the strength variation coefficient from 35% to 22%.

[0063] For any parts not mentioned in this application, existing technologies may be used or referenced.

[0064] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A quantitative decision-making method for human-machine efficiency in sandstone composite strata for tunnel boring machines, characterized in that, Includes the following steps: S1. Define a human-machine integrated efficiency model based on tunneling rate, tool wear rate, and unit energy consumption; S2. Select independent variables for the human-machine integrated ergonomics model from the initial factor pool; S3. Using the calculated values ​​of the human-machine integrated ergonomics model as the dependent variable and based on the independent variables selected in step S2, establish a multiple linear regression model. S4. Standardize the variables of the multiple linear regression model and transform it into a standard regression equation; S5. Calculate the influence weights of each standardized variable in the standard regression equation; S6. Sort the influence weights and select the decision direction based on the sorting results.

2. The method for quantitative decision-making on human-machine efficiency in sandstone composite strata for tunnel boring machines according to claim 1, characterized in that, In step S2, the initial factor pool includes the coefficient of variation of sandstone compressive strength, density of soft and hard rock interface, gravel content, groundwater level, rock stratum dip angle, cutterhead torque fluctuation rate, propulsion pressure gradient, grouting pressure deviation, screw conveyor speed, articulated cylinder pressure difference, main drive temperature, slag temperature, abnormal response delay time, pre-shift meeting duration, operator fatigue, team communication frequency, and average age.

3. The method for quantitative decision-making on human-machine efficiency in sandstone composite strata for tunnel boring machines according to claim 2, characterized in that, The independent variables selected in step S2 include: The independent variables of the tunneling rate include the coefficient of variation of sandstone compressive strength, density of the soft and hard rock interface, and gravel content; The independent variables of tool wear rate include cutterhead torque fluctuation rate, feed pressure gradient, and grouting pressure deviation. The independent variable for unit energy consumption includes the abnormal response delay time.

4. The method for quantitative decision-making on human-machine efficiency in sandstone composite strata for tunnel boring machines according to claim 3, characterized in that, In step S3, a multiple linear regression model is established using time-series data of the dependent and independent variables collected in real time.

5. The method for quantitative decision-making on human-machine efficiency in sandstone composite strata for tunnel boring machines according to claim 4, characterized in that, In step S5, the standard regression equation is validated using leave-one-out cross-validation. The variance inflation factor is used to eliminate multicollinearity interference.

6. The method for quantitative decision-making on human-machine efficiency in sandstone composite strata for tunnel boring machines according to claim 5, characterized in that, In step S6, standardized variables with an influence weight greater than 10% are defined as key variables, and decisions are made on all key variables.

Images