Uploaded by Okec Steven

soft computing solutions pdf

advertisement
SOFT COMPUTING
Question Six (20 marks)
(a) With reference to Artificial Neural Networks;
I.
Define an Epoch
➢ An epoch is when all of the data in the training set is presented to the neural network once.
II.
Describe the training process of set of pairs of input patterns with corresponding desired
output patterns. (5 marks)
➢ Training set is a set of pairs of input patterns with corresponding desired output patterns. Each
pair represents how the network is supposed to respond to a particular input. The network is
trained to respond correctly to each input pattern from the training set.
➢ Training algorithms that use training sets are called supervised learning algorithms. We may
think of a supervised learning as learning with a teacher, and the training set as a set of
examples. During training the network, when presented with input patterns, gives ‘wrong’
answers (not desired output).
➢ The error is used to adjust the weights in the network so that next time the error was smaller.
This procedure is repeated using many examples (pairs of inputs and desired outputs) from the
training set until the error (E total) becomes sufficiently small.
(b) Write short notes on the following. (2 marks @
. Supervised learning
-
Every input pattern is used to train the network.
-
Learning process is based on comparison, between network's computed output and the correct
expected output, generating "error".
-
The "error" generated is used to change network parameters that result improved
performance.
2. Unsupervised Learning
-
The expected or desired output is not presented to the network.
-
The system learns of it own by discovering and adapting to the structural features in the input
patterns.
3. Reinforced learning
-
The expected or desired output is not presented but only indicated if the computed output is
correct or incorrect.
-
The information provided helps the network in its learning process.
-
A reward is given for correct answer computed and a penalty for a wrong answer
(b) Describe the four classical applications of artificial neural networks. (8 marks)
1. Clustering:
➢ A clustering algorithm explores the similarity between patterns and places similar patterns in
a cluster. Best known applications include data compression and data mining.
2. Classification/Pattern recognition:
➢ The task of pattern recognition is to assign an input pattern (like handwritten symbol) to one
of many classes. This category includes algorithmic implementations such as associative
memory.
3. Function approximation :
➢ The tasks of function approximation is to find an estimate of the unknown function subject to
noise. Various engineering and scientific disciplines require function approximation.
4. Prediction Systems:
➢ The task is to forecast some future values of a time-sequenced data. Prediction has a
significant impact on decision support systems. Prediction differs from function
approximation by considering time factor. System may be dynamic and may produce different
results for the same input data based on system state (time).
Explain why hybrid systems need to be designed.
➢ A Hybrid system is an intelligent system that is framed by combining at least two
intelligent technologies like Fuzzy Logic, Neural networks and genetic algorithm
➢ Hybrid systems are designed in order for machines to achieve autonomy. These
systems are capable of reasoning and learning in an uncertain and imprecise
environment.
Explain the three main classes of hybrid systems.
➢ The Neuro-fuzzy system is based on fuzzy system which is trained on the basis of the
working of neural network theory.
➢ A Neuro Genetic hybrid system is a system that combines Neural networks and
a Genetic algorithm.
➢ A Fuzzy Genetic Hybrid System is developed to use fuzzy logic-based techniques for
improving and modeling Genetic algorithms and vice-versa
Describe the concept of Adaptive Neural Fuzzy Inference System (ANFIS).
➢ Adaptive Neural Fuzzy Inference System (ANFIS) is one of the intelligent principles
that based on integrating fuzzy logic and neural networks layers. It is a fuzzy
inference system implemented in the framework of adaptive networks.
➢ ANFIS=Fuzzy Logic +Neural Networks layers
Question Seven (20 marks)
Write short notes on the following hybrid systems with illustrations where possible;
(i)
•
Genetic Algorithm based Back Propagation Network (Neuro-Genetic system)
Genetic Algorithm based Back Propagation Network
Back propagation is a supervised learning algorithm used for training multilayer artificial neural
networks
(ii)
Fussy Associative Memory/FAM (Neuro-Fuzzy system)
➢ Associative Memory is a type of memory with a generalized addressing method. The address is
not the same as the data location, as in traditional memory. An associative memory system
stores mappings of specific input representations to specific output representations.
➢ Associative memory allows a fuzzy rule base to be stored. The inputs are the degrees of
membership, and the outputs are the fuzzy system’s output. Fuzzy Associative Memory (FAM)
consists of a single-layer feed-forward fuzzy neural network that stores fuzzy rules.
(iii)
Fuzzy Back Propagation Network (Neuro-Fuzzy system)
➢ The back-propagation (BP) neural network displays a strong learning ability using nonlinear
models with a high fault tolerance. It can overcome the deficiencies of traditional medical
models and is suitable for pattern recognition and disease diagnosis
(iv)
Fuzzy logic controlled Genetic Algorithm (Fuzzy-Genetic system)
➢ Genetic-Fuzzy Hybrid System is a combination of fuzzy system and system based on genetic
algorithm. The knowledge base used in this system is a combination of qualitatively different
components, and is not just a homogeneous structure.
Question Five (20 marks)
(a) Explain the term optimization problem.
(2 marks)
(b)Explain the three factors that are considered in analysis of an optimization problem. (3 marks
(c) Describe the basic structure of genetic algorithm (GA). (3 marks)
(d) State the four types of GA encoding for chromosomes. (2 marks)
(e) Using genetic algorithm of one generation and initial population of four chromosomes,
maximize the function f (x) x2 for 0 x 15. (10 marks)
Question Four (20 marks)
(a)Using a clear illustration and equation, describe the components of McColloch-Pitts (M-P) Neuron.
(b)State the main difference between M-P neuron and a perceptron. (2 marks)
A perceptron An arrangement
of one input layer of McCulloch-Pitts neurons, that is
feeding forward to one output layer of McCulloch-Pitts neurons .while
for the M-P neuron; One neuron can not do much on its own. Usually have many neurons labeled
by indices k, i, j and activation flows between them via synapses with strengths wki, wij
(c) Using network diagrams differentiate between single layer perceptron and a multiple Layer
Perceptron.
(3 marks)
➢ A neural network consisting of a layer of nodes or perceptrons between the input and the
output is called a single layer perceptron.
➢ A network consisting of several layers of single layer perceptron stacked on
top
of other,
between input and output , is called a multi-layer perceptron
(d) The basic structure of a neural network with two inputs, two hidden neurons, two output neurons
is shown in the Figure 1 below. Additionally, the hidden and output neurons will include a bias.
(i)
Assuming that the network followed a sigmoid activation function for hidden and
output layers, determine the network outputs.
(ii)
Calculate the total error of the network.
Question Three (20 marks)
(a) Clearly differentiate between a fuzzy set and a classical set.
(2 marks)
➢ Fuzzy set A defined in the universal space X is a function defined in X which assumes values in
the range [0, 1]. while a classical set is any well-defined collection of objects.
(b) With an illustration, state the condition under which a membership function distribution is said
to be convex. (3 marks)
➢ Convex fuzzy set is described by a membership function whose membership values are strictly
Monotonically Increasing or Monotonically Decreasing or Initially Monotonically Increasing
then Monotonically Decreasing with the increase in the values of the elements of that particular
fuzzy set.
(c) With
the aid of an illustration, briefly describe the Fuzzy Inference System (FIS) operation
sequence.
(5 marks)
The input data are fuzzified in order to obtain membership degrees to each of the terms of
the input fuzzy variables; then the inference machine applies the aggregation rules, using the
knowledge base and thus membership degrees to the terms of output variables are
calculated and finally, after defuzzification, the output result is obtained
(d)
In a steam power plant, a steam engine performs mechanical work using steam as its
working fluid. Mass flow rate, inlet temperature and pressure of the working fluid are the
parameters that affect the output power. However, the inlet temperature (x) and pressure O')
are more significant thus need to be controlled for the intended output mechanical power (z).
The steam engine thus employs fuzzy control system based on temperature and pressure to
monitor the output power. The fuzzy system follows a two input-single output Sugeno (TSK)
model with 4 rules as; given below.
Rule 1: IF x is low ANDY is low, THENz = —x +120y+ 13
Rule 2: IF x is low ORY is high, THEN z = -12y+ 152
Rule 3: IF x is high OR y is low, THEN z = —0.6x + 397
Rule 4: IF x is high ANDY is high, THEN z =
+ 70
power (W) of the steam engine at the inlet temperature of 4200C and pressure of 5.3 bars.
Question Two (20 marks)
(a) State the basic goal and aim of soft computing.
(4 marks)
➢ The main goal of SC is to develop intelligent machines to provide solutions to real world
problems, which can not be modeled or complex to model mathematically.
➢ Its aim is to exploit the tolerance for Approximation, Uncertainty, Imprecision, and Partial
Truth in order to achieve close resemblance with human like decision making. Therefore, the
role model for soft computing is the human mind
(b) State five differences between soft computing and hard computing.(10 marks)
➢ Hard computing is best for solving the mathematical problems which don’t solve the problems
of the real world while Soft computing is better used in solving real-world problems as it is
stochastic in nature i.e., it is a randomly defined process that can be analyzed statistically but
not with precision.
➢ Hard computing relies on binary logic and predefined instructions like a numerical analysis and
brisk software and uses two-valued logic whereas Soft computing is based on the model of the
human mind where it has probabilistic reasoning, fuzzy logic, and uses multivalued logic.
➢ Hard computing needs exact input of the data and is sequential; on the other hand, Soft
computing can handle an abundance of data and handles multiple computations which might
not be exact in a parallel way.
➢ Hard computing takes a lot of time to complete tasks and is costly while soft computing
tolerance of uncertainty and imprecision is estimated to achieve Machine Intelligence
Quotient (MIQ) and lower cost. It also provides better communication.
➢ Hard computing is best suited for solving mathematical problems which give some precise
answers while Soft computing resolves the nonlinear issues that involve uncertainty and
impreciseness as it has human-like intelligence that can resolve the real-life issue
➢ Hard computing takes a lot of time in computing as it requires the stated analytical model
while the model for soft computing is based on is that of human intelligence.
✓ Soft Computing also known as Computational Intelligence is defined as the fusion of
methodologies designed to model and enable solutions to real world problems, which
can not be modeled or complex to model mathematically
(c) Outline the three broad classes of Soft Computing. (3 marks)
(d) State any three applications of soft computing. (3mark)
➢ Approximation : the model features are similar to the real ones, but not
the same.
➢ Uncertainty :
we are not sure that the features of the model are
the same as that of the entity (belief).
➢ Imprecision :
the model features (quantities) are not the same as
that of the real ones, but close to them.
Input Vector (Crisp input) : X = [x1 , x2, . . . xn ] are crisp values, which are
transformed into fuzzy sets in the fuzzification block.
Output Vector (Crisp output) : Y = [y1 , y2, . . . ym ] comes out from the
defuzzification block, which transforms an output fuzzy set back to a crisp value.
Fuzzification: a process of transforming crisp values into grades of membership for
linguistic terms i.e. "far", "near", “very far" of an input variable “distance” of fuzzy
set.
Fuzzy Rule base: a collection of propositions containing linguistic variables; the rules
are expressed in the form: If (x is A ) AND (y is B ) . . . . . . THEN (z is C)
Fuzzy Inferencing: combines the facts obtained from the Fuzzification with the rule
base and conducts the Fuzzy reasoning process.
Defuzzification: The process of converting a fuzzified output into a single crisp value.
Question One (Compulsory)
Humidification in a spinning mill is mandatory as it maintains appropriate moisture content thus
increasing yarn tensile strength and reducing waste. The humidification system at Southern Range
Nyanza Limited (SRNL) essentially has three parameters that is; temperature, humidity and compressor
speed. Monitoring the compressor speed to give the required humidity at SRNL is a challenge. The
company seeks to establish a control system that processes the temperature and humidity in the ranges
of OOC to 500C and 0% to 100 % respectively and outputs an appropriate compressor speed in the range
of 0% to 100%. As an industrial engineer of the company, design a fuzzy controller with a set of rules
that determines the compressor speed of the humidification system provided that each of the inputs
has three linguistic variables which follow triangular membership ftnctions while the output has five
linguistic variables which also follow triangular membership functions. Use the designed fuzzy controller
to determine the compressor speed at room temperature and normal humidity. (20 marks)
Download