Technical Writing and Research

Dick Arnie Q. Badang∗ , Abdul Aziz G. Mabaning†

Department of Electrical and Engineering Technology - College of Engineering and Technology
Mindanao State University - Iligan Institute of Technology
Iligan City, Lanao del Norte, Philippines
∗ [email protected], † [email protected]

Abstract—This paper presents a general overview for optimiza- short-term optimization process. Most of the numerical analy-
tion in power system including its application. The overall process sis performed are conventional and meta-heuristic optimization
of optimization is clearly presented along with the different power methods and algorithm. A number of uncertainties occur in
system optimization tools and methods. A number of applications
of optimization in different areas of science and engineering are power system problems because the network is large, complex
also discussed. Furthermore, expectations regarding the course and are widely distributed making it harder for mathematical
are emphasized along with the goals for this subject matter. The computation alone to solve the issues. During the analysis,
interesting research entitled ”Stochastic Energy Management of the solution derived are normally local optimum instead of
Microgrids During Unscheduled Islanding Period” is reviewed the more accurate global optimum. Because of this instances,
and supplementary ideas are also presented.
Index Terms—Optimization new methods were developed in order to determine the desired
global optimum [3].
I. I NTRODUCTION The optimization problem is composed of mathematical
model where its main objective is the minimization of un-
The advancement of technology leads to a rapid research on desirable quantities like cost, energy loss, errors and others or
the improvement of power system. The study of power system the maximization of beneficial quantities like profit, quality,
offers solution to the ever growing demand for energy use and efficiency and etc. while subject to constraints. The optimiza-
problems that arises with the modern electrical technology tion can be solve using algorithmic methods to obtain the
together with the complexities of the fast changing power optimal solution. This includes mathematical programming
system network [1]. One key component to solve this power technologies such as linear, quadratic and nonlinear program-
system issues is through the use of computer technology and ming, decomposition, mixed integer and dynamic program-
optimization technique. These optimization tools are relevant ming. Artifical intelligence is used as well together with fuzzy
for computationally calculating the optimal solution for the logic, evolutionary computation, genetic algorithm, simulated
given power system problems. The classical aspects of power annealing and other hybrid methods [7] [8].
system optimization includes the optimal power flow [2],
economic dispatch [3], unit commitment [4] [5] and power These developments of power system optimization drives
quality (PQ) optimization. The issues relating to siting, sizing the academic institutions to discuss applications relating to
and scheduling of distributed sources like renewable resources, this study. The goals of these is to educate researchers and
battery storage and other distributed generation are also cov- be involved on research relating to these fields. Personally,
ered through these techniques [6]. learning more advance research about microgrid looks promis-
The word optimization originated during the first industrial ing knowing its advantage for renewable energy generation
revolution in 1855 which means making the best of anything. and battery technologies. Wanting more real life or actual
In mathematical calculation, the best solution refers to the application together with theoretical assumption is also a
maximum or minimum of certain variables subject to different good objective which would boost understanding about these
constraints. The same is true on how humans deal with life, ongoing problems. It also preferable for this course to enhance
where striving for the best in all things and deriving better skills in programming and analysis where a lot of improvement
results with minimum effort, cost and time are at utmost could be done.
priority. In power systems, optimization is the study of solving The major aim of this paper is to present about the de-
problems and finding the best alternative with the given set of velopments of power system optimization and a review of a
criteria with less human effort, cost and time [7]. study entitled ”Stochastic Energy Management of Microgrids
Prior to the development of computer based technologies, During Unscheduled Islanding Period” along with the hiccups
manually calculating the problem through mathematical based and other possible improvements. The remainder of this paper
optimization were mostly used to derive the optimal solution. is organized as follows. Section II discusses the background
Presently, modern computers are now capable of solving of optimization and its components. Section III presents a re-
complex system more faster and accurate with minimum view on power system research involving energy management
effort. Different areas of power system including the planning, of microgrid during unscheduled events. Lastly, Section IV
development, management and operation requires long-term or concludes the paper about the summary of observation.
 II. O PTIMIZATION M ODEL AND A LGORITHMS Most of the power system problems deals with on and off
status which can be modelled through mixed-integer program-
The optimization model can be represented by a set of
ming. Such problems will have a value of either ’1’ for on
unknowns x = x1 , x2 . . . xn required to be determined through
and ’0’ for off status. Mixed integer is commonly applied
the use of an objective function f (xi ), where i = 1, 2 . . . , n
for unit commitment problems [11], planning and scheduling
where its value is minimized or maximized while satisfying the
[7]. Artificial intelligence are also studied along with heuristic
given constraint function. These problem can be formulated
and metaheuristic methods. The behaviour of animals also
using the mathematical form:
inspired in developing nature based algorithm like the ant
Minimize : Y = f (x) (1) colony, elephant herding, grey wolf optimization and etc. The
genetic algorithm, particle swarm optimization, fuzzy logic
Subject to : zi (x) = 0, i = 1, . . . , N (2) and tabu search are also common in power system operation.
zi (x) ≤ 0, i = ne + 1, . . . , N (3) planning and multiobjective optimization.
xL
i ≤ xi ≤ xU
i (4)
III. R EVIEW OF THE S TUDY
where the x is the decision variable vector at quantity i up The study of power system optimization is broad and wide
to the total number of decision variable n. These decision covering many areas from load flow to scheduling and siting
variables may have continuous or discrete values. The main of energy sources. One of the interesting research that has
goal of the optimization process is to derive the values of been published by [4] deals with the energy management of
variable x which will result to a minimum or maximum microgrids during unscheduled islanding period. In their study,
objective function f (xi ). The objective function is represented they take into consideration the disturbances that happens in
by (1) measured as the effectiveness of a decision. The equa- the main grid causing islanding event in the microgrids with
tions (2) and (3) illustrates the given equality and inequality uncertainty over its duration. In order to mitigate the prob-
constraints, respectively, which imposed limits on the solution. lem, researchers proposed stochastic framework with optimal
Consequently, the lower xL and upper boundary xU which scheduling of microgrid resources all throughout the islanding
decide the continuity and range of variables. Although most event while taking into account the uncertainties of renewable
problems do not have constraints, it still require minimizing generation, load demand and duration. Risk management was
or maximizing the objective function. There are algorithm also imposed in order for the microgrid operator to have
which solves only unconstrained problems. For this instances, flexible decision making on its risk and cost.
the constrained problem is transformed to an unconstrained There are a lot of things which really picks my interest
optimization problem through the use of penalty function. One about this topic, one of which was on how they deal with
of which is shown in (5) unscheduled islanding event. Initially, I thought determining
" !#β the duration of outage is impossible, however, the idea of
m
X estimating it through the forecasted data, then managing the
W =Y 1+r zk (xi ) (5)
energy sources is good solution to mitigate possible issues
k=1
occurring after the disturbance in main grid. Nevertheless, the
where W refers to the unconstrained objective function while researchers methods and modelling are too ideal with lots of
r and β are our penalty coefficient. assumption in order to simplify the problem. Furthermore,
In the optimization process, the problem can be categorized there are inconsistencies with the author’s representation for
depending on its objective function and the nature of its the subsequent grid connected operation where battery storage
constraints with two types: linear and non-linear. In linear are discharged upon reconnection and was given a negative
problems, the decision variable in objective function, equality cost value in the objective function. I really admire how
and inequality are all in linear form while non-linear problems the researchers simplified the problem to make it easier to
are those where at least one of the objective function and solve using mixed integer linear programming based meth-
constrains is non-linear. The linear method is categorized to ods. Overall, the paper was really a great representation in
simplex and interior point —with the former having high modelling energy management during unscheduled islanding
computational efficiency but with exponentially increasing events.There are a lot of points to develop for the paper
computational burden with the increasing number of iteration and areas needed to make it more realistic and applicable
while the latter stay within the interior of the constrained in real life scenario. Studying this field will greatly help for
region and progressively moving towards the optimal point. future microgrids especially now where the earth temperature
Interior point algorithm has been applied in a variety of areas is rising and global warming are more prevalent.
of power system problems like power flow, economic dis- The flowchart for the overall process of [4] is presented
patch and reactive power optimization. In contrast, non-linear in Figure 1. It can be seen that the RESs sources and load
problem are solved through various methods like Newton and are forecasted, then an estimation for the islanding duration
Quasi Newton and are distinguished with their definition and was predicted trough historical data, Monte Carlo simulation
step length. Its application in power system includes optimal and other means. Then, set of scenarios are established which
power flow [9] and hydrothermal scheduling [10]. represents the uncertainties of load and RES sources over the
 of the probability method of [4] which was used for uncertain-
Start ties in scenario and duration. The authors also focuses more on
the price of distributed generation which maximizes the overall
profit of the microgrid operator. Te advantages of method in
Load and RESs economical and reliability aspect was also emphasized and
Power Forecasts compared with other existing methods.
IV. C ONCLUSION
Estimate the
This paper provided a brief summary of power system
Islanding Duration
optimization along with its model, algorithm and application.
A review was also presented on the research paper by [4]
and [5]. Its overall process and its idea was discussed and the
Prepare the Set of Scenario
author’s interests on the topic were explained. The observed
improvement for the work was also established which would
greatly help in understanding the behaviour of microgrids
Determine Commitment during unscheduled islanding events.
of Dispatchable DERs
R EFERENCES
[1] A. Zobaa, S. Abdel Aleem, and A. Abdelaziz, Classical and Recent
Determine Microgrid Aspects of Power System Optimization 1st Edition, 06 2018.
Schedule for Different [2] F. Feng and P. Zhang, “Enhanced microgrid power flow incorporating
hierarchical control,” IEEE Transactions on Power Systems, vol. 35,
Load and RES Scenarios no. 3, pp. 2463–2466, 2020.
based on DGs commitment [3] D. P. Kothari and J. S. Dhillon, Power system optimization. PHI
Learning Private Ltd., 2012.
[4] H. Farzin, M. Fotuhi-Firuzabad, and M. Moeini-Aghtaie, “Stochastic
energy management of microgrids during unscheduled islanding period,”
Evaluate Expected IEEE Transactions on Industrial Informatics, vol. 13, no. 3, pp. 1079–
Cost and Risk Measure 1087, 2017.
[5] M. Vahedipour-Dahraie, H. Rashidizadeh-Kermani, A. Anvari-
Moghaddam, and P. Siano, “Flexible stochastic scheduling of
microgrids with islanding operation complemented by optimal offering
Implement the strategies,” CSEE Journal of Power and Energy Systems, pp. 1–10,
Optimal Schedule 2020.
[6] H. Chen, Y. Zhang, and H. Ngan, Power system optimization: large-scale
complex systems approaches. Wiley Blackwell, 2016.
[7] R. C. Bansal, “Optimization methods for electric power systems:
An overview,” International Journal of Emerging Electric
Power Systems, vol. 2, no. 1, 12 Mar. 2005. [Online].
Available: https://www.degruyter.com/view/journals/ijeeps/2/1/article-
no End of Islanding? ijeeps.2005.2.1.1021.xml.xml
[8] J. A. Momoh, Electric power system applications of optimization. CRC
Press, 2019.
[9] H. R. Baghaee, M. Mirsalim, G. B. Gharehpetian, and H. A. Talebi,
“Three-phase ac/dc power-flow for balanced/unbalanced microgrids
including wind/solar, droop-controlled and electronically-coupled dis-
yes tributed energy resources using radial basis function neural networks,”
IET Power Electronics, vol. 10, no. 3, pp. 313–328, 2017.
[10] N. Pathak, T. S. Bhatti, and A. Verma, “Discrete data agc of hydrother-
End mal systems under varying turbine time constants along with the power
system loading conditions,” IEEE Transactions on Industry Applications,
vol. 53, no. 5, pp. 4998–5013, 2017.
[11] L. S. M. Guedes, P. de Mendonça Maia, A. C. Lisboa, D. A. G. Vieira,
Fig. 1. Energy Management Framework in [4] and R. R. Saldanha, “A unit commitment algorithm and a compact
milp model for short-term hydro-power generation scheduling,” IEEE
Transactions on Power Systems, vol. 32, no. 5, pp. 3381–3390, 2017.

estimated islanding period. The scenario referred are sequence

of load and RES with varying timeslots. Next, microgid sched-
ule is determined in order to solve the stochastic optimization
problem through the objective function. In this paper, five
different scenario were considered with outputs having low
risk but high cost and some have high risk but low cost. The
output implies how these two are balanced to derive the most
feasible solution.
A research was also published by [5] where an improvement