Optimization Techniques For Power System Problems

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 12

Optimization Techniques for

Power System Problems


Reporter: Ahsan Saleem ( 白杨 )
2021/06/18

School of Electronic Science & Engineering


Why do we need optimization in Power
System?
 Fuel Cost
 Improved operational reliability
 System security
 Increase of load demand
 Fossil fuel demand of thermal power plants
 Rising emissions into the environment.

School of Electronic Science & Engineering


Optimization Aim & Focus
 To minimize the cost of power generation in
regulated power systems.
 To maximize social welfare in deregulated power
systems, while satisfying various operating
constraints.

School of Electronic Science & Engineering


To Solve Power System Optimization
Problems
 Numerous methods are found
• Conventional
• Artificial intelligence
 Above methods are Constantly improved & Developed
• Deal with large size systems
• Interconnected systems
 Optimization problems
• Complex due to a large number of constraints.
Hence,
 Finding better solutions with shorter computation
times is the goal of these methods.

School of Electronic Science & Engineering


Optimization Issues in Power System
 Economic Dispatch (ED)
 Unit Commitment (UC)
 Hydrothermal scheduling
 Optimal power flow (OPF)
 Optimal Reactive power flow (ORDP)
 Voltage Stability
 Available Transfer Capability (ATC)
 FACTS Devices Placement
 Maintenance scheduling
 Distributed Generation
 Capacitor Placement in Radial Distribution Network(RDN)
 Phasor measurement unit (PMU) placement and etc..

School of Electronic Science & Engineering


Artificial Intelligence As A New Trend
In Optimization Problems
widely used for solving optimization problems.
ADVANTAGE
 Deal with complex problems that cannot be
solved by conventional methods.
 Easy to apply due to their simple mathematical
structure.
 Easy to combine with other methods to hybrid
systems adding the strengths of each single
method.
 Methods generally simulate natural phenomena
or the social behavior of humans or animals.

School of Electronic Science & Engineering


Artificial Neural Network
 It is Mathematical models by simulating the human biological
neural network for processing information.
 A Neural Network consists of some layers of Artificial Neurons
linked by weight connections.
 Various Neural Networks by their structure such as
• Feed Forward,
• Back Propagation,
• Radial Basis Function,
• Recurrent Networks, etc.

School of Electronic Science & Engineering


Artificial Neural Network
 Each type has some specific work after being trained.
 It is infer a function from observations which is particularly useful for
applications with the complex tasks faced in real life like function.
• Approximation
• Classification
• Data Processing, etc.
 Its primary advantage are capability to learn algorithms
• Online adaption of dynamic systems,
• Quick parallel computation
• Intelligent Interpolation of data.

School of Electronic Science & Engineering


Artificial Neural Network Optimization
 the objective function is mean square error function (loss/cost function).
 We have to find optimize values weights of neural network to minimize
the objective function.
 Gradient based search techniques such as backpropagation are currently
the most widely used optimization techniques for training neural
networks
 Global search techniques have been identified as a potential solution to
this problem. Two well-known global search techniques, Simulated
Annealing and the Genetic Algorithm can be used as well.

School of Electronic Science & Engineering


Back Propagation
 Backpropagation, an abbreviation for "backward propagation of errors",
is a common method of training artificial neural networks used in
conjunction with an optimization method such as gradient descent.
 The method calculates the gradient of a loss function with respect to all
the weights in the network.
 The gradient is fed to the optimization method which in turn uses it to
update the weights, in an attempt to minimize the loss function.
 Backpropagation requires a known, desired output for each input value
in order to calculate the loss function gradient. It is therefore usually
considered to be a supervised learning method.

School of Electronic Science & Engineering


Genetic Algorithm
 It is Search technique used to find the exact or approximately best
solution for optimization problems.
 The genetic algorithm belongs to evolutionary computation using the
techniques inspired by evolutionary biology such as
• Inheritance
• Mutation
• Selection
• Crossover

School of Electronic Science & Engineering


Genetic Algorithm
 The genetic algorithm was developed part by part from the 1950s onward and is
one of the most popular methods applied to various optimization problems in
• Bioinformatics
• Computer science
• Engineering
• Economics
• Chemistry
• Manufacturing,
• Mathematics,
• Physics and etc..
 This method can take long computational times to get the optimal solution.

School of Electronic Science & Engineering

You might also like