Post-treatment method for multi-nozzle intelligent sand blasting at high temperature

Abstract

The invention discloses a post-processing method for multi-nozzle intelligent sand blasting at high temperature, and relates to the technical field of part post-processing. According to the post-processing method for multi-nozzle intelligent sand blasting at the high temperature, by means of multi-nozzle sand blasting operation at the specific high temperature of 670 + / -5 DEG C and in combination with intelligent path planning of aluminum oxide sand grains and the nozzles, the included angle between the nozzles and the normal of the surface of the part ranges from 45 degrees to 60 degrees, sand blasting impact force can be effectively decomposed and transmitted, the sand grains evenly and efficiently act on the surface of the part, and the surface of the part is protected. By means of the method, the stability and reliability of the treatment quality are ensured, the sand blasting aftertreatment efficiency can be improved, the surface roughness of the part can stably reach 10 microns, the reduction amplitude is more than 80% compared with that before treatment, the surface strength is remarkably improved to 380 MPa, the aftertreatment effect and efficiency are greatly improved, and the production cost is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of post-treatment of parts, and particularly to a post-treatment method for multi-nozzle intelligent sandblasting at high temperature. Background Art

[0002] Sandblasting technology plays an important role in the metal surface treatment process. It is a process that uses a high-speed sand flow to impact the metal surface. During sandblasting operation, the sand grains are driven by a powerful air flow or liquid flow, like a group of tiny but extremely powerful "impact drills", and violently impact the metal surface. This impact can effectively remove the oxide scale, rust and other impurities on the metal surface, making the metal surface brand new, and providing a clean, rough and highly active surface basis for subsequent processing or painting processes. Moreover, by reasonably controlling the sandblasting parameters, such as the particle size of the sand grains, the spraying speed and angle, etc., the surface roughness of the metal can be adjusted to a certain extent to meet different process requirements.

[0003] In the post-treatment of parts formed by traditional SLM printing, the improvement effect of surface roughness is not good and the surface strength is limited. For example, the Ra value of surface roughness is often greater than 50μm, and the surface strength is generally about 300MPa. The processing efficiency is also relatively low, and it is difficult to meet the requirements of industrial production for high-precision and high-performance parts. Traditional shot peening and other means cannot effectively solve these problems. Therefore, a post-treatment method for multi-nozzle intelligent sandblasting at high temperature is specifically proposed. Automatic intelligent sandblasting treatment with four nozzles is carried out in a high-temperature environment, which can improve the working efficiency and effectively ensure the surface treatment quality of parts, providing effective support for the improvement of the surface strength of parts. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a post-treatment method for multi-nozzle intelligent sandblasting at high temperature, which solves the problems that in the sandblasting post-treatment of parts formed by traditional SLM printing, the improvement effect of surface roughness is not good, the surface strength is limited, and the processing efficiency is relatively low, and it is difficult to meet the requirements of industrial production for high-precision and high-performance parts.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A post-treatment method for multi-nozzle intelligent sandblasting at high temperature specifically includes the following steps:

[0006] Step 1: Place the parts formed by SLM printing and with supports removed in a post-treatment device whose inner cylinder hot zone temperature is stably maintained at 670±5°C;

[0007] Step 2: Use a four-nozzle sandblasting device in the post-treatment device to perform sandblasting treatment on the parts, and the angle between the nozzle and the normal of the part surface is 45°-60°;

[0008] Step 3: Intelligently plan the movement path of the nozzle through Siemens NX-MCD software, and introduce the Q-learning algorithm to optimize the sandblasting operation.

[0009] The present invention is further configured as: The state space of the Q-learning algorithm is defined as:

[0010] The geometric shape, surface roughness, distribution of sandblasted and non-sandblasted areas of the object to be sandblasted, the nozzle position coordinates, and the nozzle sandblasting angle, and are represented using vectors.

[0011] The present invention is further configured as: The action set of the Q-learning algorithm includes:

[0012] Translation, rotation, and sandblasting operation control of the nozzle in three-dimensional space.

[0013] The present invention is further configured as: The reward of the Q-learning algorithm is designed as:

[0014] If the nozzle successfully sandblasts the non-sandblasted area evenly after one action, the coating thickness error is within ±0.05 mm and there is no collision or out-of-range situation, a positive reward of +10 is given; If the sandblasting is uneven, the surface roughness exceeds the range, a collision occurs, or the prohibited area is entered, a negative reward of -5 is given;

[0015] If there is no effective action for more than 5 s in a state, a negative reward of -1 is given.

[0016]

[0017] The present invention is further configured as: The Q-table update formula of the Q-learning algorithm is:

[0018] Q(s,a) = Q(s,a) + α[r + γmax a ’Q(s',a') - Q(s,a)]

[0019] In the formula, Q(s, a) is the expected long-term cumulative reward for taking action a in state s, α is the learning rate, with a value range of 0.05 - 0.2, r is the reward obtained after executing action a and transferring from state s to the new state s', γ is the discount factor, with a value range of 0.8 - 0.9, Q(s', a') is the expected long-term cumulative reward for taking action a' in the new state s', max a 'Q(s', a'), represents the maximum value among the Q values of all possible actions in the new state s', and is used to calculate the Q value update of the current state-action pair.

[0020] ​The present invention is further configured such that: during the sandblasting treatment in step two, alumina sand grains are selected as the sand grains, wherein the particle size of the alumina sand grains is 20 - 50 mesh, and the Mohs hardness is greater than 9.

[0021] The present invention provides a post - treatment method for multi - nozzle intelligent sandblasting at high temperature. It has the following

[0022] Beneficial effects:

[0023] (1) By using multi - nozzle sandblasting operation at a specific high temperature of 670 ± 5 °C in the present invention, combined with the intelligent path planning of alumina sand grains and nozzles, the angle between the nozzle and the normal of the part surface is 45° - 60°, which can effectively decompose and transfer the sandblasting impact force, enabling the sand grains to act on the part surface evenly and efficiently, ensuring the stability and reliability of the treatment quality. It can not only improve the post - treatment efficiency of sandblasting, but also make the surface roughness of the part stably reach 10 μm, with a reduction amplitude of more than 80% compared with that before treatment, and the surface strength is significantly increased to 380 MPa, greatly improving the post - treatment effect and efficiency, and effectively reducing the production cost. Description of the Drawings

[0024] Figure 1 is a schematic flow chart of the present invention;

[0025] Figure 2 is a schematic structural diagram of the present invention. Detailed Embodiments

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention.

[0027] Please refer to Figure 1-2 , the embodiments of the present invention provide the following technical solutions: A post - treatment method for multi - nozzle intelligent sandblasting at high temperature, specifically including the following steps:

[0028] Step 1: Place the SLM - printed and support - removed part in a post - treatment device with the inner cylinder hot zone temperature stably maintained at 670 ± 5 °C. At this temperature, the crystal structure inside the metal is in a specific state, which is conducive to the optimization and recombination of the microstructure under the action of sandblasting impact, thereby improving the surface strength.

[0029] Step 2: Use a four - nozzle sandblasting device in the post - treatment device to perform sandblasting treatment on the part. The four - nozzle sandblasting device is as shown in the appendix Figure 2As shown, the nozzle is made of a special alloy material with a Rockwell hardness HRC greater than 60 and a temperature tolerance higher than 800 °C. When performing sandblasting, alumina sand grains are selected, where the particle size of the alumina sand grains is 20 - 50 mesh, and the Mohs hardness is greater than 9. The hardness and angular shape of the sand grains are screened and tested to be able to produce appropriate micro-cutting and impact strengthening effects on the surface of the part under the action of high temperature and impact force without damaging the part substrate. The nozzle forms an angle of 45° - 60° with the normal of the part surface, effectively decomposing and transmitting the sandblasting impact force, enabling the sand grains to act on the part surface evenly and efficiently, ensuring the stability and reliability of the treatment quality.

[0030] Step 3: Intelligently plan the movement path of the nozzle through Siemens NX - MCD software and introduce the Q - learning algorithm to optimize the sandblasting operation, specifically including:

[0031] State space definition: The geometric shape of the object to be sandblasted, surface roughness, distribution of sandblasted and non - sandblasted areas, nozzle position coordinates (accurate to the millimeter level), nozzle sandblasting angle (accurate to the degree), and is represented using vectors. For example, S = (object shape parameters, distribution of sandblasted areas, position of nozzle 1 (x1, y1, z1, angle of nozzle 1 θ1, position of nozzle 2 (x2, y2, z2, angle of nozzle 2 θ2,......);

[0032] Action definition: The action set for each nozzle includes translation in three - dimensional space (step size 0.5 cm), rotation (angle adjustment step size 5°), and sandblasting operation control (on or off, flow rate adjustment accuracy of ±0.05 L / min). For example, the action set A = {move forward 0.5 cm, move backward 0.5 cm, move left 0.5 cm, move right 0.5 cm, move up 0.5 cm, move down 0.5 cm, rotate the nozzle angle clockwise by 5°, rotate the nozzle angle counter - clockwise by 5°, turn on sandblasting (flow rate set to Q1), turn off sandblasting};

[0033] Reward design: If after a single action of the nozzle, it successfully sandblasts the non - sandblasted area evenly, with the coating thickness error within ±0.05 mm and no collisions or out - of - range situations, a positive reward of +10 is given; 2 If the sandblasting is uneven, the surface roughness exceeds the range, a collision occurs, or it enters the prohibited area, a negative reward of - 5 is given;

[0034] If there is no effective action in a state for more than 5 s, a negative reward of - 1 is given;

[0035] Q - table update and learning: The Q - table update formula of the Q - learning algorithm is:

[0036]

[0037] ​Q(s,a) = Q(s,a) + α[r + γ maxa’Q(s',a') - Q(s,a)]

[0038] Wherein, Q(s,a) is the expected long-term cumulative reward for taking action a in state s, α is the learning rate, with a value range of 0.05 - 0.2, r is the reward obtained when transferring from state s to a new state s' after executing action a, γ is the discount factor, with a value range of 0.8 - 0.9, Q(s',a') is the expected long-term cumulative reward for taking action a' in the new state s', and max a 'Q(s',a'), represents the maximum value among the Q-values of all possible actions in the new state s', and is used to calculate the Q-value update of the current state-action pair.

[0039] The present invention realizes that the surface roughness of the parts can stably reach below 10μm, with a reduction amplitude of over 80% compared to before treatment, the surface strength is significantly increased to 380MPa, and the sandblasting treatment efficiency is increased by 3 - 4 times compared to the traditional single-nozzle sandblasting process, effectively solving many drawbacks of the traditional post-treatment process and providing an efficient solution for the post-treatment of additive manufacturing.

[0040] Furthermore, during the sandblasting process of the present invention, the supply flow rate of the sand grains can be monitored in real time, and its fluctuation range is controlled within ±5%, the sandblasting pressure is stably maintained at 0.4 - 0.6MPa, and the sandblasting time is accurately set between 10 - 30min according to the part size and complexity. Through these precise control measures, the consistency and reliability of the treatment effect are ensured, the resource utilization efficiency is improved, and unnecessary losses are reduced.

[0041] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0042] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A post-processing method for multi-nozzle intelligent sandblasting at high temperature, characterized in that: The specific steps include: Step 1: Place the SLM-printed and support-removed parts in a post-processing device where the temperature of the inner cylinder hot zone is stably maintained at 670±5℃; Step 2: Use a four-nozzle sandblasting device in the post-processing device to sandblast the parts, with the nozzles forming an angle of 45°-60° with the normal line of the part surface; Step 3: Use Siemens NX-MCD software to intelligently plan the nozzle motion path, and introduce Q-learning algorithm to optimize the sandblasting operation.

2. A post-processing method for multi-nozzle intelligent sandblasting at high temperature according to claim 1, characterized in that: The state space of the Q-learning algorithm is defined as: The geometric shape of the object to be blasted, the surface roughness, the distribution of blasted and unblasted areas, the nozzle position coordinates and the nozzle blasting angle are represented by vectors.

3. The post-processing method of multi-nozzle intelligent sandblasting at high temperature according to claim 1, characterized in that: The set of actions for the Q-learning algorithm includes: Control of the translation, rotation and blasting operations of the nozzle in three-dimensional space.

4. The post-processing method of multi-nozzle intelligent sandblasting at high temperature according to claim 1, characterized in that: The reward design of the Q-learning algorithm is: If the nozzle successfully moves 10cm after one operation 2 If the unblasted area is sandblasted evenly, the coating thickness error is within ±0.05mm and there is no collision or out-of-range situation, a positive reward of +10 will be given; If the sandblasting is uneven, the surface roughness exceeds the range, a collision occurs, or the prohibited area is entered, a negative reward of -5 will be given; If there is no effective action in a state for more than 5 seconds, a negative reward of -1 will be given.

5. The post-processing method of multi-nozzle intelligent sandblasting at high temperature according to claim 1, characterized in that: The Q-table update formula of the Q-learning algorithm is: Q(s,a)=Q(s,a)+α[r+γmax a ’Q(s',a')-Q(s,a)] Where Q(s, a) is the expected long-term cumulative reward for taking action a in state s, α is the learning rate, which is 0.05-0.2, r is the reward for transferring from state s to the new state s' after executing action a, γ is the discount factor, which is 0.8-0.9, Q(s', a') is the expected long-term cumulative reward for taking action a' in the new state s', max a 'Q(s', a'), which represents the maximum value of the Q-values ​​of all possible actions in the new state s', is used to calculate the Q-value update of the current state-action pair.

6. The post-processing method of multi-nozzle intelligent sandblasting at high temperature according to claim 1, characterized in that: In the step 2, the sand particles used for the sand blasting are aluminum oxide sand particles, wherein the particle size of the aluminum oxide sand particles is 20-50 meshes and the Mohs hardness is greater than 9.

Images