Stirring device of shield tunneling machine

Abstract

The invention provides a stirring device of a shield tunneling machine, which solves the problem that accumulated materials are difficult to scrape off through a stirring blade and accumulate for a long time due to the fact that the stirring blade of the existing stirring machine adopts a fixed-length structure and a gap exists between the stirring blade and the inner wall of a tank body, and can be widely applied to the field of stirring machines. The device specifically comprises a tank body, a stirring shaft is arranged in the tank body, a plurality of fixing pipes are distributed on the surface of the stirring shaft, stirring blades are connected into the fixing pipes, first blades are arranged at the top ends of the stirring blades, second blades are connected to the surfaces of the first blades, the shapes of the second blades are matched with the inner wall of the tank body, and the ends of the second blades can be attached to the inner wall of the tank body; the ends of the second blades protrude out of the ends of the first blades, so that when the ends of the second blades are attached to the inner wall of the tank body, only the second blades make contact with the inner wall of the tank body.

Description

Technical Field

[0001] This invention belongs to the field of mixer technology, and specifically relates to a mixing device for a tunnel boring machine. Background Technology

[0002] As a large-scale construction machine integrating mechanical, electrical, hydraulic, and pneumatic technologies, the tunnel boring machine (TBM) involves a variety of complex fluid media, including hydraulic oil, lubricating oil, sealing grease, mortar, bentonite, compressed air, and foam, covering almost all media types in the field of fluid transmission. Among them, mortar, as the core medium for synchronous grouting, plays a crucial role. During the TBM's excavation, sufficient mortar needs to be rapidly injected behind the lining segments that have exited the shield tail to fill the annular structural gap between the shield tail and the segments. The solidification of the mortar stabilizes the segments, controls ground deformation, and ultimately reduces surface settlement. This places extremely high demands on the uniformity of mortar mixing and the timeliness of its supply. The TBM's mixing device, as the core equipment of the synchronous grouting system, directly determines the grouting quality and construction safety. However, in practical applications, the mixing devices of existing tunnel boring machines (TBMs) are limited by structural design flaws and struggle to adapt to the demands of complex construction scenarios. In particular, they suffer from significant shortcomings in internal cleaning convenience, becoming a key issue restricting construction efficiency and equipment lifespan. Specifically, the mixing blades of existing mixers mostly employ a fixed-length structure, creating a 3-5cm fixed gap with the inner wall of the tank. To ensure filling effectiveness, the mortar used for simultaneous grouting typically contains aggregates with a particle size of less than 2cm, and its viscosity needs to be dynamically adjusted according to geological conditions. High-viscosity mortar easily adheres to the inner wall of the tank and the surface of the mixing blades during mixing, forming a 5-8cm thick layer of accumulated material. Since the fixed gap cannot be eliminated, this accumulated material is difficult to remove by the mixing blades themselves, leading to the following problems over time: First, the cleaning process is cumbersome and time-consuming. Existing equipment requires stopping the machine during each construction break, and operators wearing protective gear must enter the mixing tank to manually clean the accumulated material using tools such as shovels and high-pressure water guns. If the accumulated material is dry and clumped together, the cleaning difficulty increases further, requiring the use of mechanical crushing tools, which may scratch the anti-corrosion layer on the inner wall of the tank and shorten the service life of the tank.

[0003] Secondly, incomplete cleaning leads to mortar contamination and equipment malfunction. Manual cleaning is insufficient to completely remove accumulated material from the corners of the tank's inner wall and the base of the mixing blades. Residual material mixes with fresh mortar, altering the mortar's mix ratio and viscosity, thus affecting the grouting effect. For example, residual dry material mixed with fresh mortar can form particles larger than 5cm in diameter, clogging the grouting pipes, causing a sudden increase in grouting pressure, potentially leading to pipe rupture or damage to the grouting pump. Simultaneously, residual material can enter the sensor detection area during the mixing process, distorting the detection data, affecting the accuracy of mixing control, and even causing equipment jamming.

[0004] Third, it cannot meet the needs of continuous construction. With the improvement of tunnel boring machine construction efficiency, some projects need to achieve 24-hour continuous tunneling. However, the frequent shutdown and cleaning of the existing mixers cause the synchronous grouting system to be unable to continuously supply mortar, requiring the additional configuration of a backup mixer, which increases the cost of equipment procurement and maintenance. Moreover, if the grouting interruption time exceeds 10 minutes during the switchover of backup equipment, it is easy to cause the gaps behind the tunnel segments to be filled in a timely manner, resulting in excessive ground settlement.

[0005] Furthermore, while some existing equipment attempts to reduce gaps using rigid scraping structures, the rigid material is in direct contact with the inner wall of the tank, easily scratching the inner wall. Moreover, it cannot accommodate the unavoidable roundness errors during tank manufacturing, resulting in over-scraping in some areas and gaps remaining in others, making it impossible to simultaneously achieve effective cleaning and equipment protection. Therefore, there is an urgent need for a mixing device for tunnel boring machines (TBMs) that can automatically remove accumulated material from the inner wall of the tank without frequent manual cleaning, to meet the continuous construction requirements of the TBM's synchronous grouting system and solve the problem of inconvenient cleaning in existing technologies. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the above-mentioned technology and provide a mixing device for a tunnel boring machine.

[0007] Therefore, the present invention provides a mixing device for a tunnel boring machine, including a tank, a mixing shaft disposed inside the tank, a plurality of fixed tubes distributed on the surface of the mixing shaft, a mixing blade connected inside the fixed tubes, a first blade disposed at the top of the mixing blade, a second blade connected to the surface of the first blade, the shape of the second blade being adapted to the inner wall of the tank, so that the end of the second blade can fit against the inner wall of the tank; the end of the second blade protrudes beyond the end of the first blade, so that when the end of the second blade fits against the inner wall of the tank, only the second blade contacts the inner wall of the tank.

[0008] Furthermore, the stirring blade also includes a sleeve, which is connected to a fixed tube; a telescopic rod is installed inside the sleeve, the top end of the telescopic rod is connected to the first blade, and the bottom end is connected to the sleeve via a spring. A control component is provided between the telescopic rod and the sleeve, which includes an electromagnet. The electromagnet provides a magnetic attraction force opposite to the force of the spring, so that the electromagnet can adjust the extension length of the telescopic rod by changing the magnitude of the magnetic force.

[0009] Furthermore, a pressure sensor is installed on the second blade to detect the contact pressure between the second blade and the inner wall of the tank, so as to adjust the extension length of the telescopic rod; a displacement sensor is installed on the first blade to detect the distance between the end face of the first blade and the inner wall of the tank, so as to assist in adjusting the extension length of the telescopic rod.

[0010] Furthermore, the control component also includes a controller, which has a storage module that stores a method for adjusting the length of the stirring blade, including the following steps: The received pressure sensor readings are evaluated, and the initial output of the electromagnet's coil current is determined. Based on the detection data from the displacement sensor, the initial output of the electromagnet's coil current is corrected. The modified coil current of the electromagnet is used as the final output to adjust the magnitude of the electromagnet's magnetic force.

[0011] Furthermore, the specific steps for judging the received pressure sensor readings and initially outputting the electromagnet's coil current are as follows: Using the pressure deviation as input and the increase in the coil current of the electromagnet as output, a PID control algorithm is established, and the pressure deviation e(k) of the k-th sample is denoted as: e(k) = P0 - P(k); Where P0 represents the target contact pressure between the second blade and the inner wall of the tank, and P(k) represents the actual pressure value of the kth sampling. The rate of change in pressure deviation is expressed as: Δe(k) = e(k) - e(k-1); Δe(k-1)=e(k-1)-e(k-2); Where Δe(k) represents the rate of change of pressure deviation between the k-th and (k-1)-th samples, Δe(k-1) represents the rate of change of pressure deviation between the (k-1)-th and (k-2)-th samples, e(k-1) represents the pressure deviation of the (k-1)-th sample, and e(k-2) represents the pressure deviation of the (k-2)-th sample. The current increment is expressed as: ΔI(k)=Kp×Δe(k)+Kn×e(k)+Kd×(Δe(k)-Δe(k-1)); Where ΔI(k) represents the coil current increment of the kth sampling, Kp represents the proportional coefficient, Kn represents the integral coefficient, and Kd represents the differential coefficient; The initial output current I(k) of the electromagnet coil is then expressed as: I(k) = I(k-1) + ΔI(k); Where I(k-1) represents the coil current sampled in the (k-1)th time.

[0012] Furthermore, based on the detection data from the displacement sensor, the specific steps for correcting the initial output of the electromagnet's coil current are as follows: Let the displacement deviation Δx(k) be expressed as: Δx(k) = x0 - x(k); Where X0 represents the target displacement between the first blade and the inner wall of the tank, and x(k) represents the actual extension length of the telescopic rod after k samplings; ΔIx(k) = Kx × Δx(k); Where ΔIx(k) represents the current correction amount for the kth sampling, and Kx represents the displacement correction coefficient; After correction, the output current of the electromagnet's coil is: I final (k)=I(k)+ΔIx(k); Among them, I final (k) represents the corrected coil current of the kth sample, i.e., the corrected coil current of the electromagnet.

[0013] Furthermore, the method for adjusting the length of the stirring blade also includes a method for denoising the data detected by the displacement sensor, comprising the following steps: The five consecutive data points sampled each time are sorted, and the median value is taken as the preprocessing result. Let the original displacement data sequence of the k-th sampling be: x raw (k,1), x raw (k,2), x raw (k,3), x raw (k,4), x raw (k,5); Sort the sequence in ascending order: x sorted (k,1)≤x sorted (k,2)≤x sorted (k,3)≤x sorted (k,4)≤x sorted (k,5); Take the median as the preprocessed output: x mid (k)=x sorted (k,3); Where, x raw (k,n) represents the nth original displacement data from the kth sampling, where n=1,2,3,4,5; x sorted (k,n) represents the sorted displacement data; x mid (k) represents the displacement data after median filtering; Based on the discrete system, a Kalman filter model is established, and the state equation is: x(k)=A×x(k-1)+B×u(k-1)+w(k-1); Where x(k) represents the system state at time k, A represents the state transition matrix, B represents the control input matrix, u(k-1) represents the control input at time k-1, and w(k-1) represents the process noise, which follows a Gaussian distribution with mean 0 and variance Q. The observation equation is: z(k) = H × x(k) + v(k); Where z(k) represents the observed value at time k, H represents the observation matrix, and v(k) represents the observation noise, which follows a Gaussian distribution with mean 0 and variance R. According to the Kalman filter recursive formula, the prediction step is: x hat (k|k-1)=A×x hat (k-1|k-1)+B×u(k-1); P(k|k-1)=A×P(k-1|k-1)×A T +Q; Where, x hat (k|k-1) represents the prior state estimate at time k, x hat (k-1|k-1) represents the posterior state estimate at time k-1, P(k|k-1) represents the prior covariance matrix at time k, and P(k-1|k-1) represents the posterior covariance matrix at time k-1. T This represents the transpose of the state transition matrix A; The update steps are as follows: K(k) = P(k|k-1) × H T / (H×P(k|k-1)×H T +R); x hat (k|k)=x hat (k|k-1)+K(k)×(z(k)-H×x hat (k|k-1)); P(k|k)=(IK(k)×H)×P(k|k-1); Where K(k) represents the Kalman gain at time k, H T Let x represent the transpose of the observation matrix H. hat (k|k) represents the posterior state estimate at time k, I represents the identity matrix, and P(k|k) represents the posterior covariance matrix at time k.

[0014] This invention provides a mixing device for a tunnel boring machine, which has the following beneficial effects: This invention utilizes an independent electromagnet, sensor, and control circuit for each set of telescopic stirring blades, enabling individual control of each blade and avoiding overall deviations caused by adjusting all blades at once. Furthermore, through pressure-displacement dual closed-loop control, it can respond in real-time to changes in mortar conditions: when mortar viscosity increases and adhesion strengthens, the pressure sensor detects that the contact pressure exceeds the target bonding pressure, and the controller increases the coil current of the electromagnet, shortening the telescopic arm to prevent excessive pressure from over-compressing the second blade. When mortar viscosity decreases and adhesion weakens, the pressure sensor detects that the pressure is below the target bonding pressure, and the controller reduces the coil current of the electromagnet, extending the telescopic arm to ensure the second blade always fits tightly against the inner wall, scraping away residual mortar without manual cleaning. Simultaneously, the dual closed-loop control can adapt to changes in the stirring shaft speed. When the speed increases, the displacement sensor corrects the telescopic amount, preventing centrifugal force from causing the stirring blades to shift, ensuring wall scraping stability under different operating conditions, and significantly reducing equipment downtime for maintenance.

[0015] This invention also specifically addresses two types of noise interference through a combination of median filtering and Kalman filtering algorithms: median filtering can eliminate pulse noise caused by mortar splashing, such as the instantaneous error caused by stones hitting the sensor probe. By sorting the five data points of a single sampling and taking the median, the pulse error can be reduced from ±0.5mm to below ±0.1mm; Kalman filtering can eliminate high-frequency noise caused by electromagnetic interference, such as electromagnetic field interference generated when the drive motor or electromagnet is working. By correcting the detection data in real time through state equations and observation equations, the displacement detection fluctuation amplitude is reduced to less than 0.02mm. Attached Figure Description

[0016] Figure 1 This is an external structural diagram of the present invention; Figure 2 This is a structural diagram of the stirring shaft of the present invention; Figure 3 This is a side view of the stirring shaft of the present invention; Figure 4 This is a diagram of the internal structure of the stirring blade of the present invention.

[0017] The markings in the diagram are: 1. Tank body; 2. Feed pipe; 3. Fixed platform; 4. Drive motor; 5. Inspection window; 6. Discharge port; 7. Stirring shaft; 8. Stirring blade; 9. First blade; 10. Second blade; 11. Fixed pipe; 12. Sleeve; 13. Telescopic rod; 14. Control room; 15. Electromagnet; 16. Spring; 17. Circuit board; 18. Battery; 19. Pressure sensor; 20. Displacement sensor. Detailed Implementation

[0018] The present invention will be further described below with reference to the accompanying drawings and specific embodiments to aid in understanding its content. Unless otherwise specified, the methods used in this invention are conventional methods; the raw materials and apparatus used, unless otherwise specified, are conventional commercially available products.

[0019] like Figure 1 As shown, this invention provides a mixing device for a tunnel boring machine, including a tank 1. The upper part of the tank 1 is square, forming a cuboid, and the bottom is semi-circular, forming half of a cylinder. The bottom surface of the cuboid matches the cross-section of the cylinder. Multiple discharge ports 6 are located at the bottom of the semi-circular portion. The semi-circular shape allows the mixed mortar to flow out through the discharge ports 6 under its own weight within the semi-circular portion. A fixed platform 3 is installed on the side wall of the tank 1, and a feed pipe 2 is installed on the other side. A drive motor 4 is installed on the fixed platform 3. The output shaft of the drive motor 4 is fixedly connected to a mixing shaft 7 located on the inner wall of the tank 1, and the drive motor 4 drives the mixing shaft 7 to rotate synchronously. An inspection window 5 is also provided on the side wall of the tank 1 for replacing or repairing internal parts of the tank 1.

[0020] like Figure 2 and 3 As shown, the surface of the stirring shaft 7 is spirally distributed with multiple fixed tubes 11, each of which is equipped with a stirring blade 8. The stirring blade 8 includes a sleeve 12, the surface of which is welded inside the fixed tube 11. A telescopic rod 13 is fitted inside the sleeve 12, and a first blade 9 is fixedly connected to the top of the telescopic rod 13. The first blade 9 is arc-shaped, with the same arc as the semicircular portion of the tank body 1. A second blade 10 of the same shape as the first blade 9 is also fixedly connected to the first blade 9. The outer arc edge of the second blade 10 protrudes beyond the outer arc edge of the first blade 9, allowing the outer arc edge of the second blade 10 to adhere tightly to the inner wall of the semicircular portion of the tank body 1, scraping away the mortar adhering to the inner wall. To ensure the durability of the stirring blade 8, the first blade 9 is made of rigid metal, such as Q355B steel; the second blade 10 is made of hard rubber, such as Shore A polyurethane rubber greater than 80. The hard rubber also prevents the second blade 10 from being too hard and scratching the inner wall of the tank body 1.

[0021] In actual operation, during the tunnel boring machine excavation process, the mortar is transported to the feed pipe 2 through the synchronous grouting pump. The mixing shaft 7 rotates continuously under the drive of the drive motor 4. The second blade 10 contacts the inner wall of the tank 1. When the mixing shaft 7 rotates, the mixing blade 8 sticks tightly to the inner wall of the tank 1, so that the mortar cannot adhere to the inner wall of the tank 1, and there is no need to clean the inside of the mixer afterward.

[0022] like Figure 4As shown, the bottom end of the telescopic rod 13 is connected to the inner bottom wall of the sleeve 12 via a spring 16. The spring 16 is a compression spring, providing elastic force in the direction of extension of the telescopic rod 13. A control chamber 14 is formed in the space between the telescopic rod 13 and the sleeve 12. A circuit board 17 is installed on the bottom wall of the control chamber 14. The circuit board 17 is equipped with a controller, an electromagnet 15, and a battery 18. The electromagnet 15 provides magnetic attraction in the direction of retraction of the telescopic rod 13, and the battery 18 provides an independent power supply for the control chamber 14. The controller is an S7-1200 model and is equipped with a digital output module, a wireless receiving module, and a storage module.

[0023] like Figure 3 As shown, near the outer arc edge of the second blade 10, multiple pressure sensors 19 are installed inside the second blade 10 to detect the pressure between the second blade 10 and the inner wall of the tank 1. Multiple displacement sensors 20 are installed on the outer arc edge end face of the first blade 9 to detect the distance between the first blade 9 and the tank 1. The pressure sensors 19 and displacement sensors 20 transmit the detected data to the controller via a 2.4G radio frequency module. The controller, using the length adjustment method of the stirring blade 8 stored in the storage module, controls the extension length of the telescopic rod 13, ensuring that the second blade 10 is tightly pressed against the inner wall of the tank 1.

[0024] The process of establishing the method for adjusting the length of the stirring blade 8 includes the following steps: The magnetic force F of electromagnet 15 can be expressed as: F=K i ×I×μ0×(N²×S) / (2δ²); Among them, K i δ represents the current coefficient; I represents the coil current of electromagnet 15; μ0 represents the free permeability; N represents the number of coil turns; S represents the cross-sectional area of ​​the iron core; δ represents the air gap between electromagnet 15 and the armature disk.

[0025] The force balance equation for telescopic rod 13 is: F = Fs + Fp + Ff; Where Fs represents the spring force of the return spring; Fp represents the reaction force corresponding to the contact pressure; and Ff represents the sliding friction force of the telescopic rod 13.

[0026] Fs = k × x; Where k represents the spring constant of spring 16; x represents the extension length of telescopic rod 13.

[0027] Fp = P × A; Where P represents the contact pressure between the rubber block and the inner wall of tank 1; A represents the contact area between the rubber block and the inner wall of tank 1.

[0028] Ff = μ × (m × g + Fn); Where μ represents the coefficient of sliding friction; m represents the mass of the stirring blade 8; g represents the acceleration due to gravity; and Fn represents the normal force of the telescopic rod 13.

[0029] The relationship between the extension length of the telescopic rod 13 and the air gap is as follows: δ = δ0 - x; Where δ0 represents the initial air gap when the telescopic rod 13 is fully retracted; x represents the extension length of the telescopic rod 13.

[0030] Based on the above relationship equation, an incremental PID algorithm is used to adjust the coil current of electromagnet 15. The pressure deviation is taken as input, and the increment of the coil current of electromagnet 15 is taken as output. The control PID algorithm is established, including the following steps: e(k) = P0 - P(k); Where e(k) represents the pressure deviation of the kth sampling; P0 represents the target contact pressure between the second blade 10 and the inner wall of the tank 1, which is preset to 0.5MPa; P(k) represents the actual pressure value of the kth sampling.

[0031] The rate of change deviation is then expressed as: Δe(k) = e(k) - e(k-1); Δe(k-1)=e(k-1)-e(k-2); Where Δe(k) represents the rate of change of pressure deviation between the k-th and (k-1)-th samples; Δe(k-1) represents the rate of change of pressure deviation between the (k-1)-th and (k-2)-th samples; e(k-1) represents the pressure deviation of the (k-1)-th sample; and e(k-2) represents the pressure deviation of the (k-2)-th sample.

[0032] The current increment is expressed as: ΔI(k)=Kp×Δe(k)+Kn×e(k)+Kd×(Δe(k)-Δe(k-1)); Where ΔI(k) represents the coil current increment in the k-th sampling, Kp represents the proportional coefficient, Kn represents the integral coefficient, and Kd represents the derivative coefficient. Through simulation optimization, the values ​​are Kp=0.8, Kn=0.05, and Kd=0.1.

[0033] Finally, the coil current of electromagnet 15 is expressed as: I(k) = I(k-1) + ΔI(k); I min ≤I(k)≤I max ; Where I(k) represents the coil current sampled at the kth time; I(k-1) represents the coil current sampled at the (k-1)th time; I min Indicates the minimum operating current of electromagnet 15; I maxThis indicates the maximum operating current of electromagnet 15.

[0034] To avoid the limitations of single-point detection by pressure sensor 19, displacement sensor 20 is introduced for auxiliary correction. The correction method is as follows: Since the second blade 10 is in close contact with the inner wall of the tank 1, and the relative distance between them is 0, the target displacement between the first blade 9 and the inner wall of the tank 1 is expressed as: X0=Rrtw; Where: R represents the inner wall radius of the tank, r represents the radius of the stirring shaft 7, t represents the preset compression amount of the second blade 10, which is preset to 0.002m; w represents the width of the second blade 10 protruding from the first blade 9.

[0035] Displacement deviation calculation: Δx(k) = x0 - x(k); Where x(k) represents the actual extension length of the telescopic rod 13 after k samplings.

[0036] ΔIx(k) = Kx × Δx(k); Where ΔIx(k) represents the current correction amount for the kth sampling; Kx represents the displacement correction coefficient, which is preset to 0.2A / m.

[0037] After correction, the output current of the coil of electromagnet 15 is: I final (k)=I(k)+ΔIx(k); Among them, I final (k) represents the corrected coil current of the kth sampling. Displacement-assisted correction can avoid deviations in the extension and retraction of the telescopic rod 13 caused by pressure sensor 19 failure or local adhesion, thus improving the stability of the fit control.

[0038] Because dust and slurry splashes generated during mortar mixing can cause high-frequency noise in the displacement sensor data, a combined Kalman filter and median filter algorithm is used to reduce the noise in the displacement sensor data. The noise reduction steps are as follows: The five consecutive data points sampled each time are sorted, and the median value is taken as the preprocessing result.

[0039] Let the original displacement data sequence of the k-th sampling be: x raw (k,1), x raw (k,2), x raw (k,3), x raw (k,4), x raw (k,5); Sort the sequence in ascending order: xsorted (k,1)≤x sorted (k,2)≤x sorted (k,3)≤x sorted (k,4)≤x sorted (k,5); Take the median as the preprocessed output: x mid (k)=x sorted (k,3); Where, x raw (k,n) represents the nth original displacement data from the kth sampling, where n=1,2,3,4,5; x sorted (k,n) represents the sorted displacement data; x mid (k) represents the displacement data after median filtering.

[0040] Based on the Kalman filter model for discrete systems, further noise reduction is performed on the median-filtered data. The model is established as follows: The state equation is: x(k)=A×x(k-1)+B×u(k-1)+w(k-1); Where x(k) represents the system state at time k, i.e., the actual displacement value detected by displacement sensor 20; A represents the state transition matrix, with a preset value of 1; B represents the control input matrix, with a preset value of 0; u(k-1) represents the control input at time k-1, with a preset value of 0; and w(k-1) represents the process noise, which follows a Gaussian distribution with a mean of 0 and a variance of Q, where Q = 1 × 10⁻⁶. -6 mm 2 .

[0041] The observation equation is: z(k) = H × x(k) + v(k); Where z(k) represents the observed value at time k, i.e., the displacement data after median filtering. mid (k); H represents the observation matrix, with a default value of 1; v(k) represents the observation noise, which follows a Gaussian distribution with a mean of 0 and a variance of R, where R = 5 × 10⁻⁶. -6 mm².

[0042] According to the Kalman filter recursive formula, the prediction step is: x hat (k|k-1)=A×x hat (k-1|k-1)+B×u(k-1); P(k|k-1)=A×P(k-1|k-1)×A T +Q; Where, x hat (k|k-1) represents the prior state estimate at time k; xhat (k-1|k-1) represents the posterior state estimate at time k-1; P(k|k-1) represents the prior covariance matrix at time k; P(k-1|k-1) represents the posterior covariance matrix at time k-1; A T This represents the transpose of the state transition matrix A.

[0043] The update steps are as follows: K(k) = P(k|k-1) × H T / (H×P(k|k-1)×H T +R); x hat (k|k)=x hat (k|k-1)+K(k)×(z(k)-H×x hat (k|k-1)); P(k|k)=(IK(k)×H)×P(k|k-1); Where K(k) represents the Kalman gain at time k; H T x represents the transpose of the observation matrix H; hat (k|k) represents the posterior state estimate at time k, which is the displacement data of displacement sensor 20 in the final filtered output; I represents the identity matrix; P(k|k) represents the posterior covariance matrix at time k.

[0044] Simulation verification was performed using Matlab. The original displacement signal was set as x(t) = 20 + 2sin(2πt), and random noise with an amplitude of 0.5 mm was superimposed to simulate mortar interference. Single median filtering, single Kalman filtering, and the combined filtering algorithm of this invention were used for processing. The filtering results are as follows:

[0045] Simulation results show that the combined filtering algorithm can effectively eliminate noise caused by mortar interference, improve the accuracy and stability of the data detected by displacement sensor 20, and provide a reliable input signal for the control of telescopic rod 13.

[0046] It should be noted that, since there are multiple displacement sensors 20 and pressure sensors 19, the final output results need to be averaged.

[0047] The controller's control flow is as follows: After the mixer starts, the controller initializes the control parameters; Electromagnet 15 is energized, and the initial coil current is set to I. max The telescopic rod 13 extends slowly under the action of the electromagnet 15; The displacement sensor 20 collects the extension length data of the telescopic rod 13 in real time, and transmits it to the controller after processing by a combined filtering algorithm. When the protruding second blade 10 contacts the inner wall of the tank 1, the pressure sensor 19 detects the pressure signal, and the controller starts PID regulation based on the pressure deviation and displacement deviation. During the stirring process, the controller continuously collects pressure and displacement data, dynamically adjusts the coil current of the electromagnet 15, and maintains the contact pressure of the second blade 10 within the target range.

[0048] In the description of this invention, it should be understood that the terms "left", "right", "up", "down", "top", "bottom", "front", "back", "inner", "outer", "back", "middle", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0049] However, the above description is merely a specific embodiment of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any substitution of equivalent components or equivalent changes and modifications made in accordance with the scope of protection of the present invention should still fall within the scope of the claims of the present invention.

Claims

1. A mixing device for a tunnel boring machine, comprising a tank, a mixing shaft disposed within the tank, a plurality of fixed pipes distributed on the surface of the mixing shaft, and mixing blades connected within the fixed pipes, characterized in that, The top of the stirring blade is provided with a first blade, and a second blade is connected to the surface of the first blade. The shape of the second blade is adapted to the inner wall of the tank so that the end of the second blade can fit against the inner wall of the tank. The end of the second blade protrudes beyond the end of the first blade, so that when the end of the second blade is in contact with the inner wall of the tank, only the second blade is in contact with the inner wall of the tank.

2. The mixing device for a tunnel boring machine according to claim 1, characterized in that, The stirring blade also includes a sleeve, which is connected to a fixed tube. A telescopic rod is installed inside the sleeve. The top end of the telescopic rod is connected to the first blade, and the bottom end is connected to the sleeve via a spring. A control component is installed between the telescopic rod and the sleeve. The control component includes an electromagnet, which provides a magnetic attraction force opposite to the force of the spring, so that the electromagnet can adjust the extension length of the telescopic rod by changing the magnitude of the magnetic force.

3. The mixing device for a tunnel boring machine according to claim 2, characterized in that, A pressure sensor is installed on the second blade to detect the contact pressure between the second blade and the inner wall of the tank, so as to adjust the extension length of the telescopic rod; a displacement sensor is installed on the first blade to detect the distance between the end face of the first blade and the inner wall of the tank, so as to assist in adjusting the extension length of the telescopic rod.

4. The mixing device for a tunnel boring machine according to claim 3, characterized in that, The control component also includes a controller, which contains a storage module that stores a method for adjusting the length of the stirring blade, including the following steps: The received pressure sensor readings are evaluated, and the coil current of the electromagnet is initially output. Based on the detection data from the displacement sensor, the coil current of the electromagnet initially output is corrected; The modified coil current of the electromagnet is used as the final output to adjust the magnitude of the electromagnet's magnetic force.

5. The mixing device for a tunnel boring machine according to claim 4, characterized in that, The specific steps for judging the received pressure sensor readings and initially outputting the electromagnet's coil current are as follows: Using the pressure deviation as input and the increase in the coil current of the electromagnet as output, a PID control algorithm is established, and the pressure deviation e(k) of the k-th sample is denoted as: e(k) = P0 - P(k); Where P0 represents the target contact pressure between the second blade and the inner wall of the tank, and P(k) represents the actual pressure value of the kth sampling. The rate of change in pressure deviation is expressed as: Δe(k) = e(k) - e(k-1); Δe(k-1)=e(k-1)-e(k-2); Where Δe(k) represents the rate of change of pressure deviation between the k-th and (k-1)-th samples, Δe(k-1) represents the rate of change of pressure deviation between the (k-1)-th and (k-2)-th samples, e(k-1) represents the pressure deviation of the (k-1)-th sample, and e(k-2) represents the pressure deviation of the (k-2)-th sample. The current increment is expressed as: ΔI(k)=Kp×Δe(k)+Kn×e(k)+Kd×(Δe(k)-Δe(k-1)); Where ΔI(k) represents the coil current increment of the kth sampling, Kp represents the proportional coefficient, Kn represents the integral coefficient, and Kd represents the differential coefficient; The initial output current I(k) of the electromagnet coil is then expressed as: I(k) = I(k-1) + ΔI(k); Where I(k-1) represents the coil current sampled in the (k-1)th time.

6. The mixing device for a tunnel boring machine according to claim 5, characterized in that, Based on the detection data from the displacement sensor, the specific steps for correcting the initial output coil current of the electromagnet are as follows: Let the displacement deviation Δx(k) be expressed as: Δx(k) = x0 - x(k); Where X0 represents the target displacement between the first blade and the inner wall of the tank, and x(k) represents the actual extension length of the telescopic rod after k samplings; ΔIx(k) = Kx × Δx(k); Where ΔIx(k) represents the current correction amount for the kth sampling, and Kx represents the displacement correction coefficient; After correction, the output current of the electromagnet's coil is: I final (k)=I(k)+ΔIx(k); Among them, I final (k) represents the corrected coil current of the kth sample, i.e., the corrected coil current of the electromagnet.

7. The mixing device for a tunnel boring machine according to claim 4, characterized in that, The method for adjusting the length of the stirring blade also includes a method for reducing noise in the displacement sensor detection data, comprising the following steps: The five consecutive data points sampled each time are sorted, and the median value is taken as the preprocessing result. Let the original displacement data sequence of the k-th sampling be: x raw (k,1), x raw (k,2), x raw (k,3), x raw (k,4), x raw (k,5); Sort the sequence in ascending order: x sorted (k,1)≤x sorted (k,2)≤x sorted (k,3)≤x sorted (k,4)≤x sorted (k,5); Take the median as the preprocessed output: x mid (k)=x sorted (k,3); Where, x raw (k,n) represents the nth original displacement data from the kth sampling, where n=1,2,3,4,5; x sorted (k,n) represents the sorted displacement data; x mid (k) represents the displacement data after median filtering; Based on the discrete system, a Kalman filter model is established, and the state equation is: x(k)=A×x(k-1)+B×u(k-1)+w(k-1); Where x(k) represents the system state at time k, A represents the state transition matrix, B represents the control input matrix, u(k-1) represents the control input at time k-1, and w(k-1) represents the process noise, which follows a Gaussian distribution with mean 0 and variance Q. The observation equation is: z(k) = H × x(k) + v(k); Where z(k) represents the observed value at time k, H represents the observation matrix, and v(k) represents the observation noise, which follows a Gaussian distribution with mean 0 and variance R. According to the Kalman filter recursive formula, the prediction step is: x hat (k|k-1)=A×x hat (k-1|k-1)+B×u(k-1); P(k|k-1)=A×P(k-1|k-1)×A T +Q; Where, x hat (k|k-1) represents the prior state estimate at time k, x hat (k-1|k-1) represents the posterior state estimate at time k-1, P(k|k-1) represents the prior covariance matrix at time k, and P(k-1|k-1) represents the posterior covariance matrix at time k-1. T This represents the transpose of the state transition matrix A; The update steps are as follows: K(k)=P(k|k-1)×H T / (H×P(k|k-1)×H T +R); x hat (k|k)=x hat (k|k-1)+K(k)×(z(k)-H×x hat (k|k-1)); P(k|k)=(IK(k)×H)×P(k|k-1); Where K(k) represents the Kalman gain at time k, H T Let x represent the transpose of the observation matrix H. hat (k|k) represents the posterior state estimate at time k, I represents the identity matrix, and P(k|k) represents the posterior covariance matrix at time k.

8. The mixing device for a tunnel boring machine according to claim 1, characterized in that, The tank consists of a square section and a semi-circular section. The bottom of the semi-circular section has multiple discharge ports. The semi-circular section is half the size of a cylinder, which allows the material to be discharged quickly by its own weight.

Images