In summary, the research on the integrated cutter system has achieved certain results. However, there are still shortcomings, mainly reflected in the number of disassembly steps, locking capacity, and cutter system reliability. Based on the above research results, aiming at simplifying the cutter system structure and reducing the disassembly and assembly steps, this paper innovatively proposes an integrated new TBM Cutter system design scheme.
The study of the cutting cutter system in China started late. In 2018, Hongrun Construction Co., Ltd. [ 4 ] cooperated with universities to develop an integrated cutter system that uses an eccentric circular mechanism to lock the cutter. The cutter can be locked and unlocked by grasping the cutter counterclockwise, but it is difficult for the cutter-changing robot to grasp accurately. Meng [ 12 ] et al. of Dalian University of Technology proposed an integrated cutter system based on the topological structure theory of the planar mechanism, and the solution space is obtained for the six-bar single-DoF, which further improves the anti-loosening effect.
Universities and companies also try to improve the cutter system design scheme. In 2015, French companies BOUYGUES [ 9 ] combined with companies NFM proposed a connecting rod cutter system, which has simple disassembly actions, but it is easy to be stuck in the movement process and has a low reliability. In 2018, the Robot Innovation Center of the German Artificial Intelligence Research and Development Center [ 10 ] developed a slider cutter system, which has a simple structure and less disassembly actions, but it is hard to manufacture. Two French researchers, Derycke Jean-Noel and Rubrecht Sebastien [ 11 ], invented an integrated rotating-panel TBM cutter system, which has a simple structure and few disassembly actions, but has disadvantages such as excessive stress on the rotating panel and easy sticking of the rotating panel.
Factors such as the complex structure and multiple disassembly and assembly operation steps of the traditional cutter system restrict the cutter-changing actions, which makes the cutter-changing robot difficult to popularize and use. Some universities and companies first optimized the structure based on traditional industrial robots to achieve an automatic cutter replacement [ 4 5 ]. However, this scheme is difficult to apply due to the limitation of operational space and the load ratio of industrial robots in TBMs. Therefore, researchers try to design a new structure of cutter changing robots. Based on the KUKA’s 6-DOF industrial manipulator, the Bouygues Group [ 6 ] developed a cutter changing robot for the shield machine, designed a new cutter holder, and developed the corresponding end-effector. Huo [ 7 8 ] et al. of Dalian University of Technology proposed a 6-DOF cutter-changing robot and carried out a cutter-changing range analysis, joint trajectory planning, and motion control simulation. In addition, a hybrid curve trajectory planning method based on the trapezoidal velocity curve and quintic polynomial curve was proposed, and a variable structure PID controller based on dynamic model compensation was designed.
The disc cutter bears a great alternating load in the tunnel boring machine (TBM) rock-breaking process. The violent vibration causes the failure of the cutter system fastener and high consumption of the cutter, which requires monitoring and replacement frequently. It is challenging to improve the tunneling efficiency [ 1 2 ]. At the same time, the cutter replacement process is highly dependent on the manual operation; various safety hazards often threaten the life and safety of operators [ 3 ]. Therefore, it is imperative to use the robot to replace the manual cutter detection and replacement operation.
Traditional TBM cutter system mainly has three types: double wedge blocks cutter system, upper and lower wedge blocks cutter system, and rear-installed cutter system. The main structure defect of the above traditional cutter system is a complex structure and has numerous parts. The disc cutter disassembly steps are complicated, which will significantly reduce the disc cutter replacement efficiency.
This paper obtains inspiration from the automatic firearm’s mechanism and optimizes the traditional cutter system’s structure. We obtain a sliding block swing mechanism and evolve it into a fastening mechanism. Supplementing the cutter holder, high-strength bolts, disc cutter, wedge, shift fork, cutter side panels, and other components can get the complete design scheme of the wedge swing type new cutter system, as shown in Figure 1
The cutter system can be divided into the fastening module and supporting module according to the function. The main function of the fastening module is to fix the disc cutter tightly in the cutter system to ensure that it can reliably carry out a rock-breaking operation, and the main components are high-strength bolts, long shaft, shift fork, wedge, cushion block, fastening flange, etc. The main function of the support module is to limit the relative position between the cutter system’s two sides and provide a gripper for the operator or the cutter-changing robot when replacing the disc cutter. The main components are the grasping lever, side panels, cutter holder, cutter fixing bolts, etc.
The cutter system installation process begins when the operator or the cutter-changing robot places the entire cutter system into the cutter holder by holding the grasping lever. Just rotate the high-strength bolt and drive the shift fork to move up. At this time, the wedge will swing in the direction of the cutter holder side panel around the long axis, and the wedge top will extend into the groove inside the cutter holder to limit the displacement.
Conversely, when the cutter system needs to be removed, just rotate the high-strength bolt in the opposite direction and drive the shift fork to move down. At this time, the wedge will swing in the direction of the disc cutter around the long axis and the wedge top will leave the groove inside the cutter holder. So that the operator or the cutter-changing robot can lift the entire cutter system from the cutter holder through the grasping lever.
The core advantage of the new TBM integrated cutter system is that it simplifies and integrates the structure of the cutter system, greatly reduces the operation steps, improves the disc cutter replacement efficiency, and provides the possibility for the cutter-changing robot popularization. The disassembly performance comparison between the new and the traditional cutter system is shown in Table 1
At the same time, the overall weight of the new TBM integrated cutter system is significantly reduced compared with the traditional cutter system because of the emphasis on the lightweight design at the beginning of the design process (the total weight of the new cutter system is only 354.4 kg, which is significantly less than that of the traditional cutter system). The smaller weight is helpful for the operation of the operator or the cutter-change robot, which can improve the construction efficiency.
The stress distribution rule of the cutter system under a nominal load can be obtained by using the finite element simulation calculation before the statics loading experiment. For the stress concentration part, the actual stress value can be measured by the statics loading experiment to verify the finite element calculation results.
In the finite element simulation process, a proper mesh size can form a balance between the computing efficiency, storage space, and calculation accuracy. Therefore, the finite element mesh convergence analysis of the cutter system should be carried out firstly to obtain the optimal mesh size.
After calculating and comparing the cutter system stress values under 5 groups of different mesh sizes (20 mm, 10 mm, 7 mm, 5 mm, and 3 mm), it is found that the stress values can obtain the highest stability and maintain a high precision under a 5 mm mesh size. Therefore, the finite element mesh size of the entire cutter system was set as 5 mm.
Take the real TBM working condition as an example, the disc cutter’s nominal load is a 315 KN vertical load, a 31.5 KN rolling load, and a 47.25 KN lateral load. However, the vertical load is obviously greater than the load in the other two directions (more than one order of magnitude). In the real rock-breaking process, the vertical load is the most important load which the disc cutter borne. Therefore, in order to simplify the calculation process, this paper ignored the rolling load and lateral load and instead just applied the vertical load (315 kN) to the model for the calculation of the finite element.
According to the assembly relationship between the TBM cutter head and cutter system, the holder is set as completely fixed in the finite element model, and the disc cutter and fastening devices are fixed as a whole part. Contact is set between the disc cutter shaft and the side panel circular groove surface, and the friction coefficient of this surface is set as 0.1.
Apply the nominal load to the disc cutter blade and start the finite element simulation. On this basis, we set other groups of different value loads and applied them to the finite element model in turn. The finite element model rationality can be verified by comparing the maximum stress position under different load values.
We calculate the cutter system’s stress by using finite element software. According to the results, the cutter system’s calculated maximum stress 131.92 MPa occurs at the apex of the shift fork, at point Max, shown in Figure 2
In equipment development and testing, prototypes are often scaled down to obtain similar models for various performance tests. The similarity model has the advantages of a low cost, short test cycle, and an ease to control, which helps form the empirical formula quickly [ 13 14 ]. We carried out statics and dynamic experiments on the new TBM cutter system based on the similarity theory in this paper.
The establishment of similar structural parameters and quality attributes between the scaled cutter system and the prototype cutter system is a prerequisite for accurately simulating the real working conditions and loading of the TBM cutter system [ 15 ]. Therefore, select the similar parameters of the TBM prototype cutter system as follows:
Structural parameters: the side length
l
of a single cutter holder.Physical properties: material density
ρ
, Poisson’s ratioμ
, and elastic modulusE
.Based on the dimensional analysis method, and comprehensively considering the performance requirements and manufacturing costs of the scaled cutter system, as well as the structure and working range of the test bench, the similarity ratio of the cutter system’s structural parameters is formulated as 1:4. Since the materials used in the machining the scaled tool system are the same as those of a real TBM cutter system, the elastic modulus
E
similarity ratio in the material characteristic quantity isCE
= 1, Poisson’s ratio similarity ratio isCμ
= 1, and the density similarity ratio isCρ
= 1.Combining the Buckingham theorem, the similar constants of the related physical quantities of the scaled cutter system are shown in Table 2
F
and the side lengthl
of the cutter system. Therefore, the loading forceF
should be scaled according to the square of the scale ratio, that is, 1:16. Since the nominal load of the real size disc cutter is 315 KN, the equivalent load applied to the scaled cutter system in the loading test is 19.69 KN.Table 2 shows a square relationship between the loading forceand the side lengthof the cutter system. Therefore, the loading forceshould be scaled according to the square of the scale ratio, that is, 1:16. Since the nominal load of the real size disc cutter is 315 KN, the equivalent load applied to the scaled cutter system in the loading test is 19.69 KN.In this scaled loading experiment, arrange the strain sensors on the cutter system’s key positions and apply the equivalent load to the scaled cutter system, which can obtain the measured stress values of the cutter system. The accuracy of the finite element model can be verified by the error between the calculated and measured stress values.
According to the finite element stress calculation results, the stress will concentrate in two points when the nominal load is applied to the cutter system: one point is the wedge top that contacts the cutter holder inside the groove, another is the circular groove top on the side panel that contacts the cutter shaft. Therefore, this paper takes these two points as the key positions for the stress measurement.
To reduce the error and avoid the negative impact of the accuracy which is caused by the strain sensor quality, this experiment arranges the strain sensors symmetrically on the left and right sides of the cutter system. Strain sensors #1, #2, and #5 are used to measure the stress at two key positions on the cutter system’s left side, and strain sensors #3, #4, and #6 are on the right side. Finally, the stress data of the left and right sides are compared with each other to verify the accuracy.
Considering the arrangement operability, strain sensors #1 and #2 are arranged in the plane area around the wedge top of the cutter system’s left side to simultaneously measure the stress value and are used as the comparison verification to reduce the error. Strain sensors #3 and #4 are arranged in the same way. At the same time, strain sensors #5 and #6 are arranged, respectively, at circular grooves at the top of the left and right side of the cutter system’s panels to simultaneously measure the stress value. A detailed sensor arrange schematic diagram is shown as Figure 3 a.
The critical experimental devices in this statics performance experiment mainly include a test bench, wireless strain sensors, fixtures, and a hydraulic loader, shown in Figure 3 b. The fixtures are used to limit the cutter system’s displacement, which divides into upper, middle, and lower parts. Wireless strain sensors are used to collect the strain data. To maximize the accuracy, arrange the strain gauges on six key points, and set 4 sets of load values, specifically 5 kN, 10 kN, 15 kN, and 20 kN.
The critical experimental devices used in this experiment are as follows:
(1) A hydraulic servo actuator; (2) a control system; (3) two computers; (4) a wireless gateway; (5) four strain test nodes; (6) and testing software. Part of the experimental test instruments is shown in Figure 4
ε
on the measured key points of the cutter system, and the elastic modulusE
of the cutter holder material is known, so that stress valueσ
can be converted by Equation (1).σ = E ε
(1)
The statics performance experiment has obtained the strain valueon the measured key points of the cutter system, and the elastic modulusof the cutter holder material is known, so that stress valuecan be converted by Equation (1).
Strain data are collected 12 times, respectively. The average strain value
ε
is obtained after processing, and then we plugged it into Equation (1) to obtain the average stress valueσ
of the two parts, which can be compared with the finite element calculation results.We applied the load to the scaled cutter system separately according to 4 sets of the load values, collected the strain values of the six strain sensors 10 times, and the average values at each sensor in each loading. The maximum error between the measured and calculated stress is 8.35% through comparison and calculation. Thus, the accuracy of the finite element model and the static properties of the new cutter system is proved.