Sie sind auf Seite 1von 4

International Journal of Technology & Engineering ISSN 2455-9121

Role of Artificial Intelligent in Mechanical Engineering


Honey
Assistant Professor, M.M.Modi College, Patiala, Punjab, India
Puneet_78620@yahoo.com

ABSTRACT
The paper presents an artificial intelligent application in mechanical engineering design model, which represents an
innovative knowledge of an artificial system for configuring technical systems. An artificial system contains functional
descriptions of building blocks of mechanical system. An artificial system can also synthesize functional structures which
are used then as shallow knowledge to configure technical systems with equivalent models of shape. Also a flexible
functional structure is used to manage models of shape properly.
Keyboard:
1. Introduction
Artificial Intelligence is a combination of computer science, physiology, and philosophy. Artificial Intelligence (AI)
is the area of computer science focusing on creating machines that can engage on behaviors that humans consider
intelligent. It has abilities to create intelligent machines has intrigued humans since ancient times and today with the
advent of the computer and 50 years of research into AI programming techniques, the dream of smart machines is
becoming a reality. On other hand Mechanical engineering is an innovative and novel discipline that uses the standards
of engineering, physics, and artificial science for the design, and evaluate the, industrialized and constructive feature
of mechanical systems. The central objective of mechanical engineering is to make new infrastructure where machine can
simulate the intelligent human behaviors.
In this work we have tried to make new mechanical engineering infrastructure with artificial intelligent. In this instance
most of the earlier techniques and tools of articial intelligence (AI), such as Artificial Neural Networks, Evolutionary
Computation, Fuzzy logic, Computational Swarm Intelligence and Artificial Immune Systems, have been developed for
decades. However, these applications are not provide much better solution in mechanical engineering field. In this work,
a survey on some applications of articial Intelligence to Mechanical Engineering will be presented. This review has been
based on statistical analysis of a large amount of data extracted from major bibliographic databases. This analysis
provides a quantitative map of the variety of CI methods and tools and their applications to some fields of Mechanical
Engineering.
This paper broadly divided into 4 sections.
2. Application of Artificial Intelligent in Mechanical engineering
There are many application
2.1 Neural Network
An artificial neural network (ANN) model was developed for the analysis and simulation of the correlation between the
friction stir welding (FSW) parameters of aluminum (Al) plates and mechanical properties. The input parameters of the
model consist of weld speed (Ws) and tool rotation speed (Rs). The outputs of the ANN model include property
parameters namely: tensile strength, yield strength and elongation. Good performance of the ANN model was achieved.
The model can be used to calculate mechanical properties of welded Al plates as functions of weld speed and Rs. The
combined influence of weld speed and Rs on the mechanical properties of welded Al plates was simulated. Simulated
annealing technique was used to prevent the network from getting stuck in local minima.

Vol 1 No. 1 (May 2016) IJTE 40


International Journal of Technology & Engineering ISSN 2455-9121

Figure 1: Schematic diagram of the FSW process


There are many research related to an applications of Artificial Intelligence (AI), especially Artificial Neural Networks
(ANN) for engineering fields such as manufacturing processes prediction, monitoring and controlling.
2.2 Deterministic Behavior
Deterministic behavior is exhibited by artificial-fact capable of achieving specified goals or sustaining desired behavior
only under predictable conditions. Data processing systems, conventional robots, production lines and computer
controlled machine tools are examples of such systems. Major strengths of this type of behavior are precision and
repeatability. The major weakness is its inability to cope with unexpected events. For many years automation was
synonymous with the economy of scale and mass production. It is now increasingly difficult and costly to construct and
maintain stable operating environments such as rigid production lines, required to implement automation. Therefore the
demand for machines with deterministic behavior is steadily declining. Under volatile market conditions an important
asset is flexibility which automated systems do not have.
2.3 ARTIFICIAL NEURAL NETWORKS IN MECHANICAL ENGINEERING
It follows from earlier discussion that for a problem where we have sufficient number of values sets for input and output
variables, we can develop a Neural Network, which can be then used to predict the unknown output values for the some
other input set.

AI
ME

AIME

Figure 2: High Level View of AIME Model


These features of AIME can be utilized to the benefit of mechanical engineers in the application areas such as Quality
Control, Production Planning, Job Shop Scheduling, Supply/Demand Forecasting, Mechanism design and analysis,
design optimization and several other areas. In the following section, we discuss a case problem, wherein AIME has been
implemented for determining natural frequencies of a cracked cantilever beam.
2.4 Ontology
Ontology is the explicit and abstract modeled representation of already defined finite sets of terms and concepts, involved
in knowledge engineering, knowledge management and intelligent information integration. Currently, Ontology is

Vol 1 No. 1 (May 2016) IJTE 41


International Journal of Technology & Engineering ISSN 2455-9121

divided into three categories .i.e., Natural Language Ontology (NLO), Domain Ontology (DO) and Ontology Instance
(OI). NLO is the relationship between generated lexical tokens of statements based on natural language, DO is the
knowledge of a particular domain and OI is the automatically generated web page behaves like an object. Ontology
development process starts with determining the scope by gathering and listing of terms based unstructured relevant
information. It includes information in the form of classes as nouns, properties as verbs containing values, constraints and
relationship with the other properties based on classes and their sub classes.
Terms are organized in a top down or bottom up class and sub class hierarchy called Taxonomy Hierarchy, works in a
way that the instance of the child class can behave as the instance of parent class and the property of the parent class can
also be applied to instance of child class. Size of Ontology varies with respect to the number of classes and instances; if
the number of classes and instances will increase then the size of ontology will increase.
2.5 Knowledge Management System (KMS)
an ontology-based KM system (KMS) is presented, which provides users with a product knowledge model (PKM) that
helps them to locate proper information, and supports reuse of product data among different application systems
throughout the product life cycle of mechanical engineering . The model is divided into information resource information
of engine, management activity of process cycle, product semantic ontology, as well as organization competence
ontology in order to realize semantic match for knowledge search, and enhance the performance of organization
capability and knowledge sharing.

Information
Management
Resource Product Semantic
Activities of
Informaion of Ontology
Process Cycle
Engine

Figure 3: Knowledge Management System


Furthermore, prototype system architecture is proposed, using the Semantic Web technology, for seamless sharing
domain-specific design knowledge among multidisciplinary organizations. Finally, the application of the prototype
system is presented.
2.6 TRAINING OF ARTIFICIAL NEURAL NETWORKS
The technique of setting the value of the weights enables the process of learning or training. The process of modifying
the weights in the connections between network layers with the objective of achieving the expected output is called
training of the network. The internal process that takes place when a network is trained is called learning. The most basic
method of training a neural network is trial and error. If the network isn't behaving the way it should, change the weights
of a random link by a random amount. If the accuracy of the network declines, undo the change and make a different one.
It takes time, but the trial and error method does produce results. The task is to mirror the status of the input row onto the
output row. The algorithm 1 Some of the popular training algorithms are-

 Back propagation Algorithm


 The simple neuron - the Single Layer Perceptron
 The Multilayer Perceptron
Of the above Algorithms the Back-Propagation Algorithm is widely used. The back-propagation algorithm uses
supervised learning. We provide the algorithm with examples of the inputs and outputs we want the network to compute,
and then the error (difference between actual and expected results) is calculated. The idea is to reduce this error, until the
ANN learns the training data. The training begins with random weights, and the goal is to adjust them so that the error
will be minimal. As the error depends on the weight
Conclusion

Vol 1 No. 1 (May 2016) IJTE 42


International Journal of Technology & Engineering ISSN 2455-9121

The advent and exponential developments in the field of computers and consequently in the area of Artificial
Intelligence has revolutionized an engineer‘s approach towards solving complex engineering problems. Engineers,
irrespective of their branch or field are starting to rely more and more on AI techniques. From the preceding discussion
and case study, it can be seen that ANN can help save time and efforts for complex problems, where analytical
techniques are very difficult and tedious to apply. Application of ANN in the case problem eliminates the need to
remodel the beam over and again for different crack parameters and re-run the FE Analysis.
References
[1] Rao S.S., Mechanical Vibrations, Pearson Education Inc, (2004)
[2] [6] Ismail Akbani, ―Crack Detection in Beams Using Vibration Analysis,M.Tech. Thesis, Visvesvaraya National
Institute of Technology, Nagpur, India (2010)
[3] [7] P.G.KIRMSHER 1944 Proceedings of the American Society of testing and materials 44,897-904 The effect of
discontinuities on the natural frequencies of beam.
[4] Pleune TT, Chopra OK. Using artificial neural networks to predict the fatigue life of carbon and low-alloy steels.
Nuclear Engineering and Design 2000;197:1–12.
[5] Velten K, Reinicke R, Friedrich K. Wear volume prediction with artificial neural networks. Tribology International
2000;33:731–6.
[6] Zhang Z, Friedrich K, Velten K. Prediction of tribological properties of short fiber composites using artificial
neural networks. Wear 2002;252(7–8):668–75.
[7] Heider D, Piovoso MJ, Gillespie Jr JW. Application of a neural network to improve an automated thermoplastic
tow-placement process. Journal of Process Control 2002;12:101–11.
[8] Rai N, Pitchumani R. Rapid cure simulation using artificial neural networks. Composites Part A: Applied Science
and Manufacturing 1997;28A:847–59.
[9] Heider D, Piovoso MJ, Gillespie Jr JW. Application of a neural network to improve an automated thermoplastic
tow-placement process. Journal of Process Control 2002;12:101–11.Rai N, Pitchumani R. Rapid cure simulation
using artificial neural networks. Composites Part A: Applied Science and Manufacturing 1997;28A:847–859.

Vol 1 No. 1 (May 2016) IJTE 43

Das könnte Ihnen auch gefallen