In the early 60s the revolution of design started changing form. From the use of simple means like pencils and rulers to the digital era. CAD has fundamentally changed the way design is done. Computer-aided manufacturing (CAM) is the software which used by the computer, in order to control the machine tools. This programs aims firstly to make the production much faster and reliable and at the same time to reduce the consumption of energy. In other words to be more efficiency .On the other hand Computer-aided design (CAD) is used to design items -objects which are either virtual or real.CAD used to design not only 2d objects but also 3d.CAD is widely used in industry ,in sectors such as automotive, shipbuilding, and aerospace. Nowadays it is also used to produce animations. In this assignment we are going to consider about the future of CAD/CAM in comparison with web-base technologies, artificial neural networks and genetic algorithms.CAD/CAM changed the way design it is done. We try to find if and how will contribute to the sectors which mentioned before.
the neurophysiologic knowledge of biological neurons and the networks of such biological neurons. But also differs from the proper computing machines that they have to replace, human brain computation enhanced without taking into account the organization of the computing elements and of their networking. In other words ANN (Artificial Neural Network), simulates a biological neural network and by using very simple mathematical operations manage to solve complex, mathematically problems, nonlinear problems or stochastic problems.
Principles of Artificial and Neural Networks
The basic principles of the artificial neural networks (ANNs) were first formulated by McCulloch and Pitts in 1943, in terms of five assumptions, as follows:
The activity of a neuron (ANN) is all-or-nothing.
A certain fixed number of synapses larger than 1 must be excited within a given interval of neural addition for a neuron to be excited.
The only significant delay within the neural system is the synaptic delay.
The activity of any inhibitory synapse absolutely prevents the excitation of the neuron at that time.
The structure of the interconnection network does not change over time.
ANN how it works
The way an artificial neuron works, is more or less the same as a real neuron works. For example a real neuron consists of 1) a cell body, 2) a set of fibers, called "dendrites" , 3)one special fiber leaving the soma , called "axon". The purpose of dendrites is to transmit trains of electrical pulses to the soma and the axon makes the reverse. If the transmission efficiency is positive the synapse called excitatory, if it is negative called inhibitory. The figure below simulates a simple model of artificial neuron.
McCulloch-Pitts model of a neuron
ANNs can be grouped in two categories if we consider the connection pattern
Feed-forward networks (graphs has no loops)
Recurrent networks (loops happened because of feedback connections)
ANN and the ability to learn
The ability to learn is one of the most basic characteristic of intelligent. Although, it is hard to give an exact definition of learning. The performance of the ANN is improved through updating the weights in the network. ANNs seems to learn underling rules, instead of following a set of rules specified by humans, from the collection of the representative examples. So can be easily understand that this is the biggest advantages of neural networks in comparison with traditional expert systems.
The future of CAD-CAM involved with Genetic Algorithms
Definition of Genetic Algorithms
A genetic algorithm (GA) is a search technique used in computing to find exact or approximate solutions to optimization and search problems. To be more accurate, algorithms encode a potential solution to a specific problem. Genetic algorithms are often viewed as function optimizers, although there is a wide range of problems to which genetic algorithms have been used. There are some specific examples of GAs uses:
* Aerospace engineering
* Astronomy and astrophysics
* Electrical engineering
* Financial markets
* Game playing
* Materials engineering
* Mathematics and algorithmics
* Military and law enforcement
* Molecular biology
* Pattern recognition and data mining
* Routing and scheduling
* Systems engineering
In this session it is visible that we will elaborate to the last use which is Systems engineering.
Genetic Algorithms in systems engineering
Benni and Toffolo in 2002 applied a genetic algorithm to the multi-objective task of designing wind turbines used to generate electric power.
Haas and Mills in 1997 used a multiobjective genetic algorithm to optimize the beam shape, orientation and intensity of X-ray emitters used in targeted radiotherapy to destroy cancerous tumors while sparing healthy tissue.
Lee and Zark used a genetic algorithm to evolve a set of rules to control an automotive anti-lock braking system. While the ability of antilock brake systems to reduce stopping distance and improve flexibility has saved many lives, the performance of an ABS depends on road surface conditions: as an example, an ABS controller that is optimized for dry asphalt will not work the same well on wet , mud or icy roads, and vice versa.
A GA also used in order to improved the efficiency of diesel machines by Dr. Peter Senecal of the University of Wisconsin. As a result the engine produced one-third as much nitric oxide exhaust and half as much soot. Moreover it consumes 15% less fuel than a normal diesel engine.
The use of genetic algorithms in CAD/CAM systems
It is apparent that design it is a human activity. Design it is not a purely sequential progress but it starts with existing constructions. The designers use the most desirable features and make through them new designs. Genetic algorithms constitute a category of methods of search are advisable particularly for the resolution of difficult problems of optimization.
They have been used successfully for the resolution one rapid developing number of complicated problems of mechanics at the last decade. Recently they have also it is considered as tools for creative planning of mechanics.
Before the 90's, CAD/CAM programs where developing under the philosophy of replacing the standard way of engineering design from paper based design to digital format. The evolution of these kind of programs was and still is remarkable. After the fast wide spread of the use of internet in our everyday life in middle of the 90's raised the need for evolution in design programs and further more for a web based CAD/CAM.
Manufacturing, design, production and services have evolved hardware and software wise. Due to the internet, engineering had to take a different approach and use the benefits of the internet. From sharing information on design within local area networks in companies offices to a wide area networks. This created the need for another level of engineering, which is called E-manufacturing. Web based CAD/CAM has advantages similar to the E-manufacturing like the following.
Each design itself is a pattern that consists of data (e.g. numerical, dimensions, programming code) packed in files. In order to manage all of the information concerning a product of a company there should be a system in place so as to assure that the data from each step of the lifecycle of a product, each department of the company or any data of the company are up to date and in a well stored way. This kind of system is called PDM that stands for product data management. Through PDM, web based CAD/CAM can use its functions (both user and utility) so as to retrieve, transport, translate data and not only. The translation of data between CAD/CAM is one of the vital functions which is used in web based collaboration because not of all the users use the same design CAD/CAM software. Most commonly used ISO that is used so as to cover the needs of exchange, implementation, sharing product database and archiving is ISO 10303. ISO 10303 is known as STEP (standard for the exchange of product model data).
Design in web based CAD/CAM has to be able to comply with the following key factors:
Tools for collaborative design
Design library modelling
Search of design knowledge
Web based product design via CAD/CAM is aiming at the feasibility and maximum outcome of the design based on the collaboration of the partners involved in the design process no matter where their location is. The tools for a collaborative design are the programming languages that are used so as to make possible and compatible the communication through the internet, like JAVA and VRML (virtual reality modelling language). Since the web based CAD/CAM is working through the internet, that raises the need for the use of servers that have databases of specifications, data relevant to the design process and libraries of materials, shapes, parts and in general whatever we need so as to complete the design process online. Through configuration design we can join different formats of CAD/CAM, computer languages together, an online program is used to do this and is CORBA (common object request broken architecture). In the design process engineers have to apply the proper knowledge so as to meet design requirements and specifications. G.Q.Huang developed in 2001 a special web based search engine called, web applications in the product introduction process (WAPIP) so as to provide design data and solutions to the engineers that would be accurate on a web based search. Many design applications have developed the past decade for online problem solving in special design.
Web based CAD/CAM came to life so as to cover the need for collaboration between designers no matter their location around the globe, for sharing knowledge in order to achieve maximum results in the design of the product life cycle. In designing a product online there are two major categories that describe the main aspects of web based CAD/CAM,
Collaborative design allows members of the design team of a company to make online conference so as to work on the same design. The integrative technology helps to improve the communication visual and data wise to support the product design by the interfaces that can provide to users the ability to evaluate at the same time the products designs and specifications. Sharing ideas and solutions the integration web tools is supporting the resolve of problems and team collaboration. Features of this kind of interfaces are the chat tools web cam tools so as to make the team more aware of the looks of the other members of the team rather not to be able to see with who they collaborate, in sort, to put a face at the voices and names of the rest team. This makes the collaboration more real life like, in real time except the fact that the members might be in different locations. This way of collaborating is more efficient and quicker than trying to collaborate through the phone or fax and mails. The meeting is encrypted so as to keep confidential the data and information that is exchanged between the members of the team. One of the programs that are used for conferencing is, spotlight meeting from Arel as shown above.
Concurrent design is the integration of the product's life cycle development into the design process. Is based on Design for Manufacturing (DFM) and Design for Assembly (DFA). The design can be in 2D or 3D depending on the type of the web based CAD that a team will work on. Concurrent design involves the development of a design by a team with combined skills and expertise from sectors as materials, assembly, simulating/ testing, etc. The members of a team can make modifications on the same design at the same time and evaluate the progress of the design process until they are going to reach the wanted result by the client. In concurrent design except the design we have processes like the development of the product, analysis and troubleshooting. Concurrent design is aiming to bring together a design team that will be able to complete the life cycle of the product from conception to manufacture.
Virtual reality in design
Virtual reality (VR) is the simulation of a real like environment that is represented in 3D and can provide an accurate reciprocation of a process based on the rules of real world. The language that can support VR is called Virtual Reality Modelling Language (VRML). VRML designs comprise a set of nodes that can contain many types of data like 3D geometry, JPEG images and more. The language includes primitive shapes like tubes, cylinders and in general all the basic shapes. For animation VRML can move objects using standard transformations of computer graphics. The growth of the use of VRML environments is a sub product of the demand in our days for the delivery of high quality products in low cost in shortest time. Through CAM using VRML we can simulate real life reaction of a machine or test the material or run the milling process in a virtual multi axis CNC machine (i.e. CyberCut application).
Discussion and conclusion
Genetic algorithms used more and more in the engineering. Gas are "tools" for optimization and parameter tuning but also used for design. Genetic algorithms conduct a search through the space of potential solutions to the problem. They try to balance between "exploration and "exploitationÂ». They explore the search space and at the same time exploit the good features of already found solutions. Gases performs an independent sampling on a population of design solutions, then selects members of the population and creates new designs by crossover and by mutation.There are a lot factors to consider if an algorithm is strength full. They are flexible and can be
adapted to different problems. They are flexible and we can use them so as to solve different design problems. Realization of GA procedures results in robust and stable algorithms and computer codes. They had the ability to work with many parameters in a search space of a complicated structure.
It is essential in design applications to be a built-in tendency to find global optima. GAs can also be used beyond parameter optimization, for creative design.
As far as ANNs is concerned my conclusion is summarized as follows:
ANN has a intense interesting in manufacturing
In some manufacturing problems paid more attention than some others
The majority of the applications are simulated in computers instead of implemented in hardware
Last but not least, there is a trend to integrate neural network system techniques in order to, manufacturing problems come to an end.
ANNS techniques are very new and the development of them is very rapid. I assume that for the future ANN will not going to replace conventional computing but they will be integrated with conventional computing techniques, so as to develop intelligent manufacturing systems.
The future of web based CAD/CAM is clearly going towards a virtual reality based environment that it will offer the real time collaboration of a design team to interact on the design, at the same time with an interface that will offer the conference features and the design tools so as to make the interaction stronger and most of all to have all the means to produce the optimum result. Complex assemblies and drafts should be easy and fast to handle. The environment in VR can have options for testing, different kind of machines from libraries that will be stored in servers that will offer efficiency and accuracy so as to conduct tests with real life like results. Simulation in real like conditions of designs of parts that will be fitted in and run so as to record reactions and offer useful conclusions for improvement or adjustments that need to be done. In order to support the VR environment the application of the 3D technology via glasses would give an other view to the interaction, just imagine working together in the virtual workshop using all the dimensions for a more exciting experience in design.