NC machine tools and genetics.

Kovacic, M. ; Vaupotic, B. ; Balic, J. 等

Abstract: Manufacturing systems of the future foresee the use of intelligent machine tools, capable of autonomous and intelligent tool path planning, optimising and navigating through the production environment. Aims and scopes of manufacturing systems are usually defined. The paper proposes a new concept for programming of NC machines. The concept based on genetic algorithms assures evolutionary generation and optimization of NC programs on the basis of CAD production environment. Article presents a concept for NC machine tools path generating in production environment. Examples show universality and inventiveness of the approach. A brief discussion on potential applications is also provided.

Key words: NC machine tools, path planning, genetic algorithms

1. INTRODUCTION

The autonomous and intelligent NC machine tools control and programming will be of essential importance in the next-generation manufacturing systems (Mitchell, 1997). The aim of intelligent tool motion planning is to find the best part machining strategy. The systems for autonomous NC programs generation must have the capacity of self-learning and autonomous deciding in the dynamic production environment. Their behaviour must be based on the non-deterministic bases, since only in this way it is possible to ensure a high degree of adaptability to current requirements in the production environment (Kovacic, 2004).

This paper discusses about using genetic algorithms for intelligent NC machine tool path planning in CAD environment (Kovaeie & Balie, 2003). The concept imitates the natural evolution of living organisms, where in the struggle for natural resources the successful individuals gradually become more and more dominant, and adaptable to the environment in which they live, whereas the less successful ones are present in the next generations rarely. The structure, undergoing simulated evolution, is the population of NC programs. During the evolution the tool movement becomes more and more complex and intelligent solutions emerge gradually as a result of the low level interaction between the simple tool movements and the workpiece dynamic environment.

2. METHODS USED

Genetic algorithm (GA) (Holland, 1975) method was used in the research, which usually starts with creation of initial population of solutions. The solutions are called organisms or also chromosomes. Each generated organism represents a more or less accurate solution of the optimization problem. Then, the organisms are evaluated. Greater probability of cooperating in selection and variation operations is assigned to those organisms that solve the problem better (they are better adapted to the environment). The selection operation assures survival of more fit individuals of population and their advance in unchanged form into next iteration (generation). The variation operation has effect on one or more parental organisms and from them their offspring are created. After completion of selection and variation a new generation is obtained that is evaluated, too. The process is repeated until the termination criterion of the process is not fulfilled.

2.1 Coding of organisms

A chromosome represents a tool path as the sequence of points with the first point indicating the starting point followed by intermediate points, and the last point indicating the ending point. Each point can consists of the x, y and z coordinates.

A path (chromosome) can have a varied number of intermediate points. Each chromosome has a random number of intermediate points and randomly generated coordinates for each point.

2.2 Evaluation of the population

For solving the problem it is necessary to determine calculation of fitness. The fitness comprises three independent components:

* component taking into account the total path length (shortest path, lower value of fitness),

* component taking into account the collided volume of the machine tool and obstacles (more collided volume, higher value of fitness) and

* component taking into account the path smoothness (smoother path, lower value of fitness).

2.3 Changing of the population

Different genetic operations were used to evolve chromosomes into possibly better ones (Michalewicz, 1996). Operations change only the intermediate points of a chromosome.

3. EXAMPLE--GA LATHE

Figure 1 shows the contours of the simple blank and desired final shape of the product. The objective of turning is to assure relative motion of the tool with respect to the workpiece. The consequence of the relative motion is moving of the tool reference point and, consequently, forming of the desired shape of the product. The tool movement consists of the working and feeding motions. Machining of material takes place in several cuts whose optimum sequence is known in general only for very simple products. Each working motion takes off a certain quantity of material and thus it produces the cuttings.

[FIGURE 1 OMITTED]

The area of the possible tool motion is discretized into 11x11 squares. The tool of one square thickness can move discretely up, down, to left and to right, whereas it cuts only to right or downwards. Due to the discretization of the machining field the tool reference point is in the middle of the cutting edge.

3.1 GA parameters

In all runs of the system the same evolution parameters were used: population size of 50, probability of reproduction 0.2, probability of crossover 0.6 and probability of permutation 0.2. The tournament selection of organisms with tournament size 7 was used.

3.2 Results

Figure 2 shows the results of two independent runs. Only the best organisms are shown. At left of Figure 2 the system intelligently shortened the machining by a vertical cut at the right side of the workpiece. In this way, shorter tool path and reduced tool wear were assured.

[FIGURE 2 OMITTED]

4. EXAMPLE--GA MILLING

In the example machining of a simple product from relatively complex blank is shown (Fig. 3).

4.1 GA parameters

In all runs of the system the same evolution parameters were used: population size of 500, maximal number of generations 50, probability of reproduction 0.1, probability of crossover 0.6, probability of permutation 0.2 and probability of mutation 0.1. The tournament selection of organisms with tournament size 7 was used.

4.2 Results

For the comparison between conventional and genetic machining for a given case is shown in Figures 3 and 4.

In the conventional process the proposed machining strategy imposes the product machining in such away that the tool tries to move during the majority of time of machining in the plane parallel with the horizontal plane. By conventional machining type (Fig. 3) the tool made the product in 41 steps. The tool was in contact with the workpiece as much as 13 cuttings.

[FIGURE 3 OMITTED]

The best genetic based machining strategy of length 33 appeared already in generation 7 (out of 50) and is shown in Figure 4. The tool was in contact only with 7 cuttings. Conventional machining differs from the genetic machining for 24.24% with respect to the path length and for 85.71% with respect to the tool wear.

[FIGURE 4 OMITTED]

5. CONCLUSION

In the paper we presented the concept of automatic programming of NC machine tools by the method of genetic algorithms. Researches have shown that the proposed system of evolutionary searching for optimum NC program is efficient and universal, therefore it can be used also for programming of other NC machines.

First, the proposed intelligent system hierarchically divides the task of product machining into sub-tasks: fixing, cuts and cuttings. Then, it determines the sequences of fixing and intelligently prepares the NC programs for the individual blade and workpiece fixing. The system prepares the strategy of machining by means of simple rules for the tool movement. More and more optimal NC programs appear gradually, from generation to generation, as a consequence of interactions between the tool and the workpiece dynamic environment. Special emphasis is put on the evaluation of machining strategies.

The system is capable of autonomously planning the machining technology, detecting the machined and unashamed workpiece areas, planning and optimizing the tool working and feeding motions, detecting the collisions and verifying whether the product cannot be completely machined on a certain machine.

A disadvantage of shown systems is the discretization of the machining field, which considerably contributes to the time exactingness of searching for optimal solutions. Therefore, in future the researches will be oriented particularly towards conceiving an improved system with the possibility of use of different types of machining, different machining and tool parameters.

6. REFERENCES

Holland, J. (1975). Adaptation in Natural and Artificial Systems, University of Michigan Press, ISBN 0-262-58111-6, Ann Arbor

Kovaeie, M. & Balie, J. (2003). Evolutionary programming of a CNC cutting machine. International journal of advanced manufacturing technology, Vol. 22, No. 1/2, pp. 118-124. ISSN 0268-3768.

Kovaeie, M. (2004). Programming of numerical controlled machine tools with evolutionary computation methods, Doctoral thesis, Faculty for mechanical engineering, Maribor

Michalewicz Z. (1996). Genetic algorithms + data structures = evolution programs Springer-Verlag, ISBN 3540606769, New York

Mitchell, T. M. (1997). Machine Learning, The McGraw-Hill Companies, Inc., ISBN 0-07-042807-7, New York