A method for monitoring mud cake on a cutter head of a shield machine

Abstract

The application discloses a kind of on-line monitoring method of shield machine cutterhead mud cake, comprising: arranging multiple groups of temperature sensors on the surface of shield machine cutterhead;Establish mud cake tunneling parameter database and temperature database under different strata, establish temperature database under different strata according to the temperature data of the surface of shield machine cutterhead under different strata;Obtain temperature variation prediction curve set;Judge.This application combines temperature sensor and BP neural network algorithm, after obtaining the corresponding temperature variation prediction curve subset of different strata, real-time acquisition of the temperature of shield machine cutterhead and the internal residue soil of soil bin after excavation, selects the corresponding temperature variation prediction curve subset of different strata, compares and analyzes the corresponding temperature variation prediction curve in temperature variation prediction curve subset with the collected temperature data, determines whether cutterhead surface is mud cake, can adapt to the tunneling environment of different strata, and can accurately and effectively reflect cutterhead mud cake phenomenon, guarantee the normal tunneling of shield machine.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield construction, in particular to a method for monitoring mud cake of a cutter head of a shield machine. BACKGROUND

[0002] With the acceleration of urbanization, the demand for underground transportation construction is increasing. As a common underground tunnel construction equipment, the problem of mud cake of the cutter head of the shield machine is gradually highlighted. The mud cake of the cutter head of the shield machine can cause the cutter head to be stuck and damaged, and even cause serious accidents. Therefore, it is necessary to develop a monitoring system that can prevent the mud cake of the cutter head of the shield machine.

[0003] The formation of the mud cake of the cutter head of the shield machine is due to the insufficient friction between the cutter head and the soil, which causes the soil to be squeezed into the cutter head to form mud cake. These mud cakes will hinder the normal operation of the cutter head, greatly affecting the working efficiency and safety of the shield machine.

[0004] At present, there are two kinds of mud cake detection methods commonly used in China:

[0005] One is direct detection, such as published patents: CN104747197A, CN206208418U, which detects by drilling a hole in the soil pressure shield soil bin wall and then inserting a drill rod into the soil bin for detection. This direct detection method requires a hole to be drilled in the soil bin wall in advance and sealed, which not only complicates the installation of the device, but also makes it difficult to maintain a sealed state during operation, is prone to danger, and is difficult to apply to mud water shield for mud cake detection.

[0006] The other is indirect detection, such as published patents: CN107355227A, CN106885642A, CN108131150A, which detects mud cake by monitoring the change of the cutter head temperature. Specifically, when the cutter head temperature rises to a certain temperature, it is determined that mud cake has formed. For example, in patent CN106885642A, the cutter head temperature is below 50℃, the mud cake is initially formed when the temperature is 70-80℃, and the mud cake is formed in large area when the temperature is higher than 100℃. However, due to different strata and different cutter head structures of the shield machine, the temperature reached when the mud cake is formed is not the same. At the same time, when the strata in front of the cutter head becomes hard during excavation, the increase of the thrust torque will also cause the temperature of the cutter head to rise. Therefore, when a single temperature threshold is used to judge the mud cake, the final result is not accurate.

[0007] In addition, in the existing research on the mud cake of the cutter head of the shield machine, the cutter head and the foam agent are usually designed and added during or before construction to prevent the mud cake of the cutter head of the shield machine and reduce the impact of the mud cake of the cutter head on the construction. These traditional methods often lack timeliness and cannot reflect the state of the mud cake of the cutter head in time.

[0008] Therefore, it is of great significance to develop a monitoring method that can effectively monitor whether there is sludge cake formation on the cutterhead. Summary of the Invention

[0009] To address the current problem of ineffective and timely monitoring of cutterhead mud cake formation during tunnel boring machine (TBM) construction, this invention provides an online monitoring method for TBM cutterhead mud cake formation, specifically including the following steps:

[0010] A method for online monitoring of mud cake buildup on the cutterhead of a tunnel boring machine includes the following steps:

[0011] Step 1: Arrange multiple sets of temperature sensors on the surface of the tunnel boring machine cutterhead;

[0012] Step 2: Establish a database of mud cake excavation parameters and a database of temperatures in different formations. Specifically:

[0013] A mud cake tunneling parameter database is established by combining key tunneling parameters, geological data obtained from geological surveys, and shield tunneling methods. The key tunneling parameters include grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, and vibration signals.

[0014] A temperature database for different geological formations was established based on temperature data of the cutterhead surface of tunnel boring machines in different geological strata.

[0015] Step 3: Obtain the temperature change prediction curve set. Specifically, the temperature change prediction curve set includes subsets of temperature change prediction curves corresponding to different formations. The method for obtaining the subset of temperature change prediction curves corresponding to a single formation is as follows: the data in the mud cake tunneling parameter database and the temperature database of the corresponding formation from Step 2 are divided into training set and test set; a neural network algorithm is used to predict the temperature change range of the cutterhead, and the temperature change prediction curves of each set of temperature sensors corresponding to that formation are obtained, thus obtaining the subset of temperature change prediction curves corresponding to that formation.

[0016] Step four involves making a judgment, specifically: real-time collection of the temperature of the tunnel boring machine cutterhead and the excavated soil inside the soil chamber; based on the excavated soil, determining the physical properties of the strata to which the current excavation area belongs and obtaining the geological parameters of the strata; selecting a subset of temperature change prediction curves corresponding to different strata based on the geological parameters; comparing and analyzing the temperature data collected by each set of temperature sensors with the corresponding temperature change prediction curves in the subset of temperature change prediction curves to determine whether mud cake has formed on the surface of the cutterhead.

[0017] Preferably, in step one: multiple sets of temperature sensors are arranged at intervals with the center of the tunnel boring machine cutterhead as the center point; the temperature sensors are arranged adjacent to the blades; and the temperature sensors are set through prefabricated through holes on the cutterhead.

[0018] Preferably, multiple temperature sensors are arranged at equal intervals.

[0019] Preferably, in step two, the temperature sensor acquires temperature data at a rate of 5-10 seconds per acquisition.

[0020] Preferably, in step three:

[0021] The ratio of training set to test set data size is 7:3;

[0022] The structure diagram of the neural network algorithm is as follows: The input layer contains key tunneling parameters, geological data, and shield tunneling method; the hidden layer contains grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation; the output layer contains temperature data and temperature change prediction curves.

[0023] The temperature variation range at each temperature sensor location in the cutter head is denoted as C. n C n For {t min t max Each temperature sensor performs 900-1200 temperature predictions, where: t min The lowest temperature value, t max This is the highest temperature value.

[0024] Preferably, the temperature data of the cutter head is calculated using the following formula:

[0025] t=f(w1*a1+w2*a2+w3*a3+......w n *a n );

[0026] Where: a i This represents the i-th influencing indicator, which includes grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation; i = 1, 2, 3, ..., N; w i This represents the weight factor corresponding to the i-th influencing indicator.

[0027] Preferably, the determination in step four specifically involves:

[0028] If the actual measured temperature T of the cutter head n ∈C n If so, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head;

[0029] If the actual measured temperature of the cutter head Then calculate K n To make a judgment, specifically:

[0030] If K n If the value is greater than 0.05, it is determined that a mud cake has formed in the area where the temperature sensor is installed in the cutter head;

[0031] If K n If K is less than or equal to 0.05, then continue calculating K. n+1 To make a judgment, specifically including: ①K n+1 If the value is greater than 0.05, it is determined that a cake of mud has formed in the area where the temperature sensor is installed in the cutter head; ②K n+1 If the value is less than or equal to 0.05, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head, where: K n =(T n -T n-1 ) / T n .

[0032] Preferably, it also includes an early warning processing step, specifically: after determining the mud cake in step four, release mud cake information, and at the same time, provide feedback on the location of the temperature sensor with abnormal temperature change and the temperature rise rate of the temperature sensor; inject water into the soil chamber and / or increase the injection of foam aqueous solution to cool down the soil chamber and cutter head.

[0033] The effect of applying the technical solution of this invention is:

[0034] 1. This invention combines temperature sensors and a BP neural network algorithm. After acquiring subsets of temperature change prediction curves corresponding to different strata, it collects the temperature of the tunnel boring machine cutterhead and the excavated soil inside the soil chamber in real time. Based on the excavated soil, it determines the physical properties of the strata to which the current excavation area belongs and obtains the geological parameters of the strata. Based on the geological parameters of the strata, it selects subsets of temperature change prediction curves corresponding to different strata. It compares and analyzes the temperature data collected by each set of temperature sensors with the corresponding temperature change prediction curves in the subset of temperature change prediction curves to determine whether mud cake has formed on the cutterhead surface. It can adapt to the tunneling environment of different strata and can accurately and effectively reflect the phenomenon of mud cake formation on the cutterhead, ensuring the normal tunneling of the tunnel boring machine.

[0035] 2. This invention uses the center of the tunnel boring machine cutterhead as a pivot point, arranging multiple sets of temperature sensors at intervals; the temperature sensors are positioned adjacent to the cutterhead blades; and the temperature sensors are installed through pre-fabricated through holes on the cutterhead. The temperature sensor design is reasonable, objectively and accurately reflecting the temperature of the cutterhead surface while avoiding damage to the temperature sensors due to friction and collision between the cutterhead and the soil layer.

[0036] 3. This invention employs a neural network algorithm to predict mud cake formation on the cutterhead surface of a tunnel boring machine (TBM). By monitoring and analyzing various data such as tunneling thrust, cutterhead rotation speed, and vibration signals, the likelihood of mud cake formation can be predicted in advance. The TBM's tunneling parameters can then be adjusted promptly, including grouting pressure, TBM thrust, tunneling speed, cutterhead rotation speed, torque, and grouting volume, thereby preventing mud cake formation. This improves the safety of the TBM, reduces equipment damage rates, increases TBM efficiency, shortens construction time, and reduces construction costs.

[0037] 4. This invention employs a unique judgment rule, namely, if the actual measured temperature T of the cutter head... n ∈C n If the temperature sensor installation area in the cutter head is not caked, it is determined that no mud cake has formed; if the actual measured temperature of the cutter head is... Then calculate K n To make a judgment, specifically: if K n If the value is greater than 0.05, it is determined that a cake of mud has formed in the area where the temperature sensor is installed in the cutter head; if K n If K is less than or equal to 0.05, then continue calculating K. n+1 To make a judgment, specifically including: ①K n+1 If the value is greater than 0.05, it is determined that a cake of mud has formed in the area where the temperature sensor is installed in the cutter head; ②K n+1 If the value is less than or equal to 0.05, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head, where: K n =(T n -T n-1 ) / T n It can effectively monitor the formation of mud cake in its early stages, mitigate further deterioration of mud cake, and allow for timely and effective measures to reduce the cutterhead surface temperature to a normal range, ensuring the normal tunneling of the tunnel boring machine. Attached Figure Description

[0038] Figure 1 This is a schematic diagram showing the arrangement of the surface temperature sensors on the cutterhead of the tunnel boring machine in an embodiment of the present invention;

[0039] Figure 2 yes Figure 1 Schematic diagram of medium temperature sensor installation method;

[0040] Figure 3 This is a neural network structure diagram in an embodiment of the present invention. Detailed Implementation

[0041] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0042] Example:

[0043] A method for online monitoring of mud cake buildup on the cutterhead of a tunnel boring machine includes the following steps:

[0044] Step 1: Arrange multiple sets of temperature sensors on the surface of the tunnel boring machine cutterhead. Specifically:

[0045] Taking the common earth pressure balance tunnel boring machine as an example, see details. Figure 1 The diameter of the tunnel boring machine (TBM) cutterhead 1 is approximately 8m. Temperature sensors 3 (see the pentagram markings) are placed at initial intervals of 1m, with the center of the cutterhead as the focal point. These sensors are positioned approximately 1m, 1.5m, 2m, 2.5m, 3m, and 3.5m from the cutterhead center. Since mud cake formation commonly occurs near the cutter blades 2, the temperature sensors are placed adjacent to the blades (i.e., areas prone to mud cake formation). Additionally, the temperature sensors can be placed in pre-drilled holes at the center of the cutterhead to prevent damage from friction and collision between the cutterhead and the soil, which could cause them to malfunction.

[0046] An integrated temperature sensor passes through a pre-drilled hole in the center of the cutter. The signal transmitter converts the signal collected by the temperature sensor into a digital signal via an A / D signal conversion module, which then transmits it wirelessly, enabling data exchange with the signal receiving and display modules. The transmitter box contains the A / D signal conversion module, the signal transmitter, and a power supply. Installing a power isolation component effectively isolates measurement errors caused by power supply noise; the transmitter box is sealed with adhesive to ensure dustproof, waterproof, and temperature-resistant performance. Figure 2 Temperature detection: The temperature sensor is placed in a pre-made through hole in the middle of the cutter and is connected to the A / D signal conversion template and the transmitter box through a transmission line.

[0047] Step Two: Establish a database of mud cake tunneling parameters and a temperature database for different geological strata. Specifically, this involves combining key tunneling parameters with geological data obtained from geological surveys and the shield tunneling method to establish a mud cake tunneling parameter database. The key tunneling parameters include grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck removal speed, muck removal volume, cutterhead diameter, cutterhead coating material, and vibration signals. The mud cake tunneling parameter database also contains data information on the installation of detection elements on the cutterhead panel and the corresponding positions of the cutters. Utilizing the built-in function of the timer in the temperature sensor, data is transmitted to the computer processing system every five seconds, and each set of data is statistically represented as T. n Each dataset consists of a set of seven monitoring points at different distances {D1, D2, D3, D4, D5, D6, D7}. Since six temperature sensors are set at each distance, the dataset D1 collected by the first set of temperature sensors (1m away from the cutter head) includes {t... 11 t 12 t 13 t 14 t 15 t 16}, t ij This represents the temperature data collected by the j-th temperature sensor in the i-th group of temperature sensors. The statistics are presented in a six-row, seven-column matrix format.

[0048]

[0049] A temperature database for different strata is established based on the temperature data of the cutterhead surface of the tunnel boring machine (TBM) under different strata. Specifically, the excavated soil in the TBM's soil chamber is collected and the geological parameters of the current excavation stratum are analyzed. When the TBM passes through different strata, the temperature of the cutterhead surface will also change accordingly. The temperature data under different strata are summarized and organized to establish a temperature database for different strata.

[0050] Step 3: Obtain the temperature change prediction curve set. Specifically, the temperature change prediction curve set includes subsets of temperature change prediction curves corresponding to different formations. The method for obtaining the subset of temperature change prediction curves corresponding to a single formation is as follows: divide the data into a training set and a test set; use a neural network algorithm to predict the temperature change range of the cutterhead, and obtain the temperature change prediction curve for each set of temperature sensors corresponding to that formation, thus obtaining the subset of temperature change prediction curves corresponding to that formation. Specifically:

[0051] The established database of mud cake tunneling parameters and the corresponding stratum temperature database were divided into training and testing sets, with a data volume ratio of 7:3. The training set included data on grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, and data related to open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation. This data was used to adjust the network weights during the training phase. The testing set was used to test the network's classification performance on data not present in the training set. Based on the network's performance on the testing set, the network structure may need adjustment, or the number of training iterations may need to be increased. A neural network algorithm was used to calculate the weights of key tunneling parameters, geological data, and shield method on the temperature impact of the cutterhead. A neural network structure diagram was drawn by selecting some of the more heavily weighted influencing indicators, as shown in the figure. Figure 3 As shown. The cutterhead temperature data is predicted based on weighting factors. The expression for calculating the cutterhead temperature data is as follows:

[0052] t=f(w1*a1+w2*a2+w3*a3+......w n *a n );

[0053] Where: a i This represents the i-th influencing indicator, which includes grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation; i = 1, 2, 3, ..., N; w i This represents the weight factor corresponding to the i-th influencing indicator.

[0054] The specific calculation process is as follows: A random number generator program is used to generate a set of random numbers ranging from -0.5 to +0.5, which are used as the initial weights {w1, w2, w3...} of the network. The training rate is determined empirically; the larger the training rate, the greater the weight changes and the faster the convergence. However, an excessively large training rate can cause system oscillations. Therefore, the training rate should be as large as possible without causing oscillations, with a minimum training rate of 0.9. The allowable error is 0.0001, and the number of iterations is 1000. The input training set sample values ​​are normalized. The normalization of the positive index can be performed using the formula V = (XX...) min ) / (X max -X minThe normalization of the inverse index can be performed according to V = (X). max -X) / (X max -X min X represents the input training set sample values. max X min These represent the maximum and minimum values ​​in the sample, respectively, and V is the normalized sample value. A positive indicator indicates a positive correlation, and a negative indicator indicates a negative correlation.

[0055] The final result is calculated by combining the initially set weight coefficients. The error between the actual output and the ideal output is backpropagated to the hidden layer. The weights are adjusted and iterative calculations are performed until the error between the actual output and the ideal output of the training set is less than the allowable error. The final weight assignment result is then output.

[0056] Based on the final weight assignment results, the predicted cutterhead temperature data is calculated, presenting the temperature change prediction curves for each set of temperature sensors. The temperature data collected during the tunnel boring machine's operation is used as a validation set to continuously correct weight errors after the network is determined, enabling better testing and measurement of the network's performance. One minute after the tunnel boring machine begins excavation, a neural network algorithm is employed, based on the previously collected dataset T1-T... n-1 The data is used as a validation set for correction. Based on subsequent tunneling parameters and soil physical properties, the next set of data is predicted, namely the temperature change range C at the location of each temperature sensor on the cutterhead. n C n The temperature change of each temperature sensor is predicted one thousand times, and the predicted temperature data is organized into a temperature change prediction interval for the location detected by that temperature sensor, which is {t}. min , t max}, where: t min The lowest temperature value, t max This is the highest temperature value.

[0057] Step four involves making a judgment, specifically: Real-time data collection of the temperature of the tunnel boring machine cutterhead and the excavated soil inside the soil chamber; determining the physical properties of the strata in the current excavation area based on the excavated soil to obtain geological parameters; selecting subsets of temperature change prediction curves corresponding to different strata based on these geological parameters; and comparing the temperature data collected by each set of temperature sensors with the corresponding temperature change prediction curves in the subsets to determine whether mud cake has formed on the cutterhead surface. The specific judgment is as follows:

[0058] If the actual measured temperature T of the cutter head n ∈C n If so, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head;

[0059] If the actual measured temperature of the cutter head Then calculate K n To make a judgment, specifically:

[0060] If K n If the value is greater than 0.05, it is determined that a mud cake has formed in the area where the temperature sensor is installed in the cutter head;

[0061] If K n If K is less than or equal to 0.05, then continue calculating K. n+1 To make a judgment, specifically including: ①K n+1 If the value is greater than 0.05, it is determined that a cake of mud has formed in the area where the temperature sensor is installed in the cutter head; ②K n+1 If the value is less than or equal to 0.05, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head, where: K n =(T n -T n-1 ) / T n .

[0062] In addition, the system includes early warning and handling steps, specifically: after mud cake formation is identified in step four, mud cake information is released, and feedback is provided on abnormal temperature changes from temperature sensors and the rate of temperature rise; water is injected into the soil chamber and / or the injection of foam solution is increased to cool the soil chamber and cutterhead. Simultaneously, grouting pressure, tunnel boring machine thrust, tunneling speed, cutterhead rotation speed, torque, and grouting volume are adjusted to control soil chamber pressure, muck discharge speed, and muck discharge volume, reducing the cutterhead surface temperature to the normal range until tunneling is completed.

[0063] Taking a certain subway tunnel as an example, this subway tunnel mainly traverses silty clay layers, gravelly silty clay layers, completely weathered mudstone, argillaceous sandstone layers, strongly weathered mudstone, moderately weathered mudstone, argillaceous sandstone (fractured) layers, moderately weathered gravelly sandstone, sandstone and conglomerate layers, and sandstone and conglomerate (fractured) layers. Regarding the geological conditions of this section, the tunnel boring machines used are two articulated earth pressure balance composite tunnel boring machines from China Railway Equipment, with an excavation diameter of 6470mm and a composite cutterhead, suitable for tunneling through the complex strata of this section; the maximum rated torque of the tunnel boring machines is 7757KN.m, the escape torque is 9309KN.m, and the maximum thrust is 40860kN.

[0064] During tunnel boring machine (TBM) excavation, this invention's method compares and analyzes measured temperature data with corresponding temperature change prediction curves within a subset of temperature change prediction curves to determine whether mud cake has formed on the cutterhead surface. Furthermore, it determines the optimal excavation parameters for different soil layers to reduce mud cake formation.

[0065] The relevant tunneling parameters were adjusted as follows during the shield tunneling process of the subway tunnel:

[0066] ① When constructing in a cross-section of silty clay soil, adjust the relevant tunneling parameters as shown in Table 1:

[0067] Table 1. Parameter statistics during construction in slurry clay sections of the entire cross section.

[0068]

[0069] ② When constructing in soft upper and hard lower strata, adjust the tunneling parameters as shown in Table 2:

[0070] Table 2 Parameter Statistics Table for Construction in Soft Upper and Hard Lower Strata

[0071]

[0072] ③ When constructing in full-section mudstone and sandstone, adjust the tunneling parameters as shown in Table 3:

[0073] Table 3. Parameter statistics during full-section mudstone and sandstone construction.

[0074]

[0075]

[0076] ④ When excavating the full-section sandstone and conglomerate on the right line, adjust the excavation parameters as shown in Table 4:

[0077] Table 4. Parameter statistics during full-section sandstone and conglomerate excavation on the right line.

[0078]

[0079] During the shield tunneling construction in this section, the cutterhead temperature was close to the tunnel temperature, with temperature variation controlled within 3℃. Opening inspections were carried out at the 310th, 560th, 725th, and 840th rings, and the cutterhead wear and mud cake formation on the cutterhead surface were found to be good.

[0080] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for online monitoring of mud cake buildup on the cutterhead of a tunnel boring machine, characterized in that, Includes the following steps: Step 1: Arrange multiple sets of temperature sensors on the surface of the tunnel boring machine cutterhead; Step 2: Establish a database of mud cake excavation parameters and a database of temperatures in different formations. Specifically: A mud cake tunneling parameter database is established by combining key tunneling parameters, geological data obtained from geological surveys, and shield tunneling methods. The key tunneling parameters include grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, and vibration signals. A temperature database for different geological formations was established based on temperature data of the cutterhead surface of tunnel boring machines in different geological strata. Step 3: Obtain the temperature change prediction curve set. Specifically, the temperature change prediction curve set includes subsets of temperature change prediction curves corresponding to different formations. The method for obtaining the subset of temperature change prediction curves corresponding to a single formation is as follows: the data in the mud cake tunneling parameter database and the temperature database of the corresponding formation from Step 2 are divided into training set and test set; a neural network algorithm is used to predict the temperature change range of the cutterhead, and the temperature change prediction curves of each set of temperature sensors corresponding to that formation are obtained, thus obtaining the subset of temperature change prediction curves corresponding to that formation. Step four involves making a judgment, specifically: real-time collection of the temperature of the tunnel boring machine cutterhead and the excavated soil inside the soil chamber; based on the excavated soil, determining the physical properties of the strata to which the current excavation area belongs and obtaining the geological parameters of the strata; selecting a subset of temperature change prediction curves corresponding to different strata based on the geological parameters; comparing and analyzing the temperature data collected by each set of temperature sensors with the corresponding temperature change prediction curves in the subset of temperature change prediction curves to determine whether mud cake has formed on the surface of the cutterhead.

2. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 1, characterized in that, In step one: multiple sets of temperature sensors are arranged at intervals with the center of the tunnel boring machine cutterhead as the center point; the temperature sensors are arranged adjacent to the blades; and the temperature sensors are set through prefabricated through holes on the cutterhead.

3. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 2, characterized in that, Multiple temperature sensors are arranged at equal intervals.

4. The method for online monitoring of mud cake buildup on the cutterhead of a tunnel boring machine according to claim 1, characterized in that, In step two, the temperature sensor collects temperature data at a rate of 5-10 seconds per acquisition.

5. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 1, characterized in that, In step three: The ratio of training set to test set data size is 7:3; The structure diagram of the neural network algorithm is as follows: The input layer contains key tunneling parameters, geological data, and shield tunneling method; the hidden layer contains grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation; the output layer contains temperature data and temperature change prediction curves. The temperature variation range at each temperature sensor location in the cutter head is denoted as C. n C n For {t min t max Each temperature sensor performs 900-1200 temperature predictions, where: t min The lowest temperature value, t max This is the highest temperature value.

6. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 5, characterized in that, The temperature data of the cutter head is calculated using the following formula: t=f(w1*a1+w2*a2+w3*a3+......w n *a n ); Where: a i This represents the i-th influencing indicator, which includes grouting pressure, shield machine thrust, tunneling speed, cutterhead rotation speed, torque, grouting volume, soil chamber pressure, muck discharge speed, muck discharge volume, cutterhead diameter, cutterhead coating material, vibration signal, soil moisture content, soil internal friction coefficient, soil unit weight, compression modulus, deformation modulus, elastic modulus, soil coarse ore, fine ore, clay content, open shield excavation, mechanical cutting excavation, compression excavation, and grid excavation; i = 1, 2, 3, ..., N; w i This represents the weight factor corresponding to the i-th influencing indicator.

7. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 4, characterized in that, The determination in step four specifically involves: If the actual measured temperature T of the cutter head n ∈C n If so, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head; If the actual measured temperature of the cutter head Then calculate K n To make a judgment, specifically: If K n If the value is greater than 0.05, it is determined that a sludge cake has formed in the area where the temperature sensor is installed in the cutter head; If K n If K is less than or equal to 0.05, then continue calculating K. n+1 To make a judgment, specifically including: ①K n+1 If the value is greater than 0.05, it is determined that a cake of mud has formed in the area where the temperature sensor is installed in the cutter head; ②K n+1 If the value is less than or equal to 0.05, it is determined that no mud cake has formed in the area where the temperature sensor is installed in the cutter head, where: K n =(T n -T n-1 ) / T n .

8. The online monitoring method for mud cake formation on the cutterhead of a tunnel boring machine according to claim 7, characterized in that, It also includes an early warning processing step, specifically: after determining the mud cake in step four, release mud cake information, and at the same time, provide feedback on the location of the temperature sensor with abnormal temperature changes and the temperature rise rate of the temperature sensor; inject water into the soil chamber and / or increase the injection of foam aqueous solution to cool down the soil chamber and cutterhead.

Images