The emergence of CNC machining is a landmark technological revolution in the machining industry. It connects the machining process with computers and other control equipment.
By using numerical control programming instead of manual control and adjustment, it realizes the automation of the machining process and improves the efficiency of the machinery manufacturing industry.
The core of CNC machining technology is CNC machine tools and CNC machining centers, which can be multi-axis linkage, and configured with automatic tool changer, so as to carry out complex configurations, multi-process automated machining.
CNC machining processing accuracy and processing efficiency are high, thus improving the efficiency of mechanical production. Compared with CNC machining, intelligent manufacturing is the third generation of technological revolution in the machinery manufacturing industry.
It introduces artificial intelligence into the field of mechanical manufacturing, and uses machine learning to carry out the task decomposition and task execution of the mechanical processing process, which further replaces the role of human beings in mechanical processing, and significantly improves the level of intellectualization of mechanical processing.
This paper combines intelligent manufacturing and CNC machining, using intelligent algorithms to amend the programming process in CNC machining, so as to achieve better machining results.
Intelligent manufacturing and hole group CNC machining task analysis
Machining has entered the era of intelligent manufacturing, which is gradually evolving from traditional processing, automated processing.
In the era of intelligent manufacturing, CNC machine tools, CNC machining centers as the core of the automated processing mode is still retained, but the CNC programming part can also be handed over to the intelligent manufacturing system, which does not need to be completed by a person.
Intelligent manufacturing system can analyze and understand the processing task by itself with the help of AI, intelligent algorithms, and then design its own CNC machining program, machining process and machining route.
Machining route selection, design and optimization in the entire machining process has a very important position, determining the efficiency of machining.
Reasonable machining path needs to meet the requirements of machining accuracy, as far as possible to reduce the total length of the machining path, thereby reducing the time used in machining.
On the basis of meeting the above requirements, minimize the loss of area to ensure the overall integrity of the remaining profile.
To improve the machining efficiency of multi-hole clusters, it is necessary to design better machining paths.
Hole group machining is the machining of multiple holes in a steel plate, which together constitute a hole group.
Since the location, internal diameter and depth of each hole are different, a reasonable machining path is required.
Analysis of hole group processing examples
A specific case of hole group machining is shown in Figure 1.
Vertical machining path
Figure 1, a total of 15 holes need to be machined on the steel plate group of holes, this paper sets a relatively simple case, 15 holes of the same diameter, the same depth, and uniformly arranged.
In the machining process, only one tool of the same type can be used to complete the machining.
Figure 1 gives the vertical direction of the main machining path, from the right side of the cut, first processing the first column of holes, and then horizontal steering, and then pick the head up, processing the second column of holes, and so on, to complete all 5 columns of holes processing.
However, in such a machining process, the tool needs to complete 8 times of steering, which will consume more machining time, so it needs to be changed to a horizontal tool path, as shown in Figure 2.
Horizontal machining paths.
In Fig. 2, the tool is mainly oriented horizontally, first cutting from the rightmost side, machining the first row of 5 holes in the horizontal direction, then downward machining the second row of 5 holes in the horizontal direction, and then downward machining the third row of 5 holes in the horizontal direction.
Comparing Figures 1 and 2, it can be seen that in the machining task for this group of holes, the horizontal direction of travel is the preferred machining path, and the tool only needs to complete 4 turns.
The machining tasks given in Figures 1 and 2 are relatively simple in form and are the most basic machining path optimization problems encountered in machining.
In a variety of complex machining tasks, the machining path will be selected as a starting point.
The selection of these 2 directions is shown in equation (1).
in Eq:
𝑖 is the 𝑖th machining point in the machining path of the
𝑛 the total number of machining points;
x𝑖 is the x-direction coordinate of the ith machining point;
x𝑖+1 is the X-direction coordinate of the i+1st machining point;
y𝑖 is the y-direction coordinate of the 𝑖th processing point;
y𝑖+1 is the Y-direction coordinate of the 𝑖+1th machining point.
Optimization of machining paths
In actual processing, hole group machining tasks need to be completed in a group of machining processes in order to complete the processing of a number of holes, the machining route shown in Figure 3.
As can be seen from Figure 3, this group of machining tasks, including multiple holes, from the machining route, not only continuously traverse all the holes, but also to the length of the machining route, which involves the optimization of the machining route.
In CNC machining, optimization is usually performed by manual programming.
The era of intelligent manufacturing enables intelligent algorithms to automatically handle machining path optimization and programming.
This paper applies a variety of intelligent algorithms to address machining challenges.
These include expert systems, fuzzy control, particle swarm optimization, ant colony optimization, genetic algorithms, and the simulated annealing method.
Additionally, techniques such as graphical cutting, support vector machines, fish swarm algorithms, decision trees, neural networks, and deep learning are also utilized.
Using these methods, the optimization process is automatically carried out under the constraints of the position, size and number of individual holes according to the machining task requirements of a multi-hole cluster, resulting in a more reasonable path, shorter machining time and lower machining cost for CNC machining.
The optimization process is completed automatically without human intervention, which is a typical intelligent optimization process and the core work in intelligent manufacturing.
Comparing various intelligent algorithm theories, this paper finally selects the multi-objective optimization framework, and carries out the intelligent algorithm construction and design.
Intelligent manufacturing algorithm based on multi-objective optimization
The machining route optimization of multi-hole hole group is a typical multi-objective optimization problem, both to meet the continuity of processing, but also to meet the shortest machining route, but also to take into account the cost of machining factors, so this paper adopts intelligent manufacturing algorithms based on multi-objective optimization for the optimization of machining routes.
Optimizing machining time
The first optimization objective, i.e., the total machining time is shown in Equation (2).
In Eq:
t𝓌 is the total time to complete all the machining tasks in the multi-hole hole group;
t𝓂 is the time consumed to complete the cutting tasks in the machining of the multi-hole hole cluster;
tℴ is the auxiliary time for completing all machining tasks in the machining of the porous hole group;
t𝒸 is the time for one tool change in completing the machining of a cluster of holes;
T is the time for completing the tool reliability
in multi-hole group machining.
t𝓌 is the number of tool changes needed to complete the total task of hole group machining.
Optimizing total machining cost
The second optimization objective, i.e., the total machining cost, is shown in Equation (3).
In Eq:
C is the total cost of completing all machining tasks in the multi-hole cluster.
M is the unit time cost of completing the machining in the multi-hole cluster;
Ct is the tool cost in completing the machining of a multi-hole group.
After obtaining the above 2 optimization objectives, the multi-objective optimization function of multi-hole hole group machining can be constructed, as shown in Equation (4).
In Eq:
φ is the multi-objective optimization function of multi-hole hole group machining;
k1 is the weight size of the first optimization objective;
k2 is the weight size of the second optimization objective;
t𝓃 is the theoretical total time to complete all the machining tasks in the multi-hole hole group;
C𝓃 is the theoretical total cost of completing all machining tasks in a cluster of multi-hole holes.
Hole group CNC machining of intelligent algorithm path optimization test
The above analysis of the relationship between intelligent manufacturing, CNC machining and the possible entry point of the combination of the two, the multi-hole hole group machining as the object of study, in the hole machining in the horizontal machining routes, vertical machining routes and multiple folding combinations of the significance of the path optimization was analyzed.
The multi-objective optimization algorithm is selected in the field of intelligent manufacturing, and the time optimization objective, cost optimization objective and total optimization function are constructed respectively.
In order to verify the efficiency of the proposed intelligent algorithm in the optimization of the machining paths of multiple hole groups, a pilot study is conducted in this paper.
Processing time optimization test
During the experimental process, the first set of experiments verified the machining time of multiple groups of porous hole groups, and the comparison between the intelligent algorithm before and after optimization is shown in Table 1.
Optimization Solution Analysis
In Table 1, taking the 1st group of hole clusters as an example, the intelligent algorithm plans a total of 2 machining schemes.
According to Eq. (4), set the weight of the first optimization objective k1 = 0.6, the weight of the second optimization objective k2 = 0.4, the t𝒲 /t𝓃= 1.2, C/C𝓃 = 1.3 of scheme one, and derive the objective optimization function φ = 0.6×1.2+0.4×1.3 = 0.72+0.52 = 1.24;
The t𝒲 /t𝓃 = 1.1 and C/C𝓃 = 1.2 for scheme II, and the objective optimization function φ = 0.6 × 1.1 + 0.4 × 1.2 = 0.66 + 0.48 = 1.14 is derived.
Therefore, option 2 is chosen and the total processing time obtained is 7.2 min.
The calculations for each set of data were carried out as described above.
Visualization of optimization effects
Further the data in Table 1 was plotted as a bar chart in order to form a visual comparison as shown in Figure 4.
From the graph, it can be seen that the second group of multi-hole hole cluster machining tasks took the longest time and the machining time was reduced after optimization.
The most obvious optimization effect is seen in the fourth and sixth groups.
After processing with the intelligent algorithm based on the multi-objective optimization model, a clear improvement is observed. The machining time for the porous hole groups in these two groups is significantly reduced.
The multi-hole group machining in this paper’s test is carried out on a flat steel plate profile.
After processing with the multi-objective optimization intelligent algorithm, the steel plate profile shows improved material-saving efficiency. This efficiency is illustrated in Figure 5.
Savings efficiency analysis
From Fig. 5, it can be clearly seen that the multi-objective optimization intelligent algorithm has a positive effect. After processing, each group of multi-hole group machining saves the steel plate profile to a certain extent.
This indicates that the method presented in this paper can not only obtain a more reasonable machining path, but also save more material.
Conclusion
Machining manufacturing has experienced the transformation from traditional processing to CNC machining methods. It is now stepping into the era of intelligent manufacturing.
CNC machining not only realizes the automation of the machining process. In the foreseeable future, artificial intelligence will also be introduced into the field of machining and manufacturing.
Machine learning will further replace human decision-making and programming.
This shift will bring a comprehensive change to the field of machining and manufacturing.