Coordinated Voltage Control in Electrical Power network

using predictive control model

(Case Study: On high Voltage level in Ethiopia Power Network)

M-Tech Thesis Project Proposal

Submitted in partial fulfillment of the requirement for the award of

degree of Master of Technology

In

Electrical Power Engineering

Specialization: Power system and Automation

By
Asamenew Niguesse
Reg.ID No. RPG/034/09
Supervisor: Dr. Man Mohan

Defense University
College Of Engineering

November, 2017
Bishoftu
 Abstract

The Ethiopian Electric Power (EEP) is, currently, undertaking a huge electric power generation
expansion. While plenty of electricity was available through the addition of the new power plant
to the national power grid, substations and power transmission lines were running out of
capacity, with frequent outages and shortages which resulted unexpected disturbances from time
to time when numbers of consumers lose power supply at a very expensive cost. Nowadays this
brings a race between available power generation capacities and the electrification and
availability of electricity. This huge expansion demands a complex power system network which
is more heavily loaded in which the voltage instability is becoming an increasingly serious
problem which leads to the control of voltage and reactive power as a major issue in power
system operation. The purpose of voltage control in electrical networks is to compensate for load
variations and events in the transmission system, such that customer supply voltages are kept
within certain bounds.
The basic concept of model predictive control (MPC) is to use a dynamic model to forecast
system behavior, and optimize the forecast to produce the best decision by providing a compact
and accessible overview of its essential element/s.
This thesis will use a general structure for coordinated system protection control scheme based
on Model Predictive Control to evolve strategies for the contributions of novel voltage control
schemes in electrical power network at the proposed voltage level to improve the effect of
disturbances on system operation and power system security. Algorithms which will function for
the desired control will be developed using MATLAB and Power System Analysis Toolbox
(PSAT).
Then the performance of the proposed system to be controlled will be investigated through
simulations to analyze voltage stability with/without any control mechanism in the proposed
power network system. Lastly, the simulation result will be established in MATLAB.

Keywords: electrical power network, coordinated voltage control, model predictive control.

ii
 I. Introduction

Voltage stability is the power system ability to maintain the voltage level in acceptable limits in
all the buses, under normal operating conditions as well as after perturbations. An electrical
power system enters a voltage instability state when a disturbance, load increase or change in the
system topology, causes a progressive and uncontrollable voltage level degradation in a node, in
an area or in the whole system.
The main cause of the voltage instability phenomenon is the voltage drops due to changes in the
power flows through the inductive elements of the transmission network following:
 load increase correlated with local or regional reactive power deficit;
 certain incidents that “weaken” either the local voltage control (tripping of some
generating units, exceeding some generators reactive power limits) or the transmission
network (tripping of some transmission lines, transformers or autotransformers, faults
occurring on substations bus bars) or cause heavily loading of the transmission network
(separation of network), and so on;
 malfunction of on-load tap changing transformers.

Although, in essence, the voltage instability is a local phenomenon, its consequences have a
major impact on the system operation, sometimes initiating a voltage collapse process.
The voltage “avalanche” or “collapse” phenomenon is characterized by a succession of
cascading events, associated with the instability phenomenon, that determine severe reduction of
the voltage level in an area or in the whole system and finally the loss of the system angle
stability.
Voltage collapse is typically associated with the reactive power demands of load not being met
because of limitations on the production and transmission of reactive power:
 limitations on the production of reactive power include generator and SVC reactive
power limits and the reduced reactive power produced by capacitors at low voltage;
 the primary limitations on the transmission of power are the high reactive power loss on
heavily loaded lines, as well as possible line outages that reduce transmission capacity;
 reactive power demands of loads increase with load increases, motor stalling, or changes
in load composition such as an increased proportion of compressor load [9].
Model Predictive Control
The model predictive control (MPC) method is a discrete-time form of control, with control
commands issued at certain periodic intervals [11][12]. This technique has been applied to both
voltage stability problems [13], and transient stability problems [15].Typically, a MPC model
can be depicted as shown in Figure.1.1. From Figure.1.1 it can be seen that it consists of two key
elements: the model of the system to be controlled and an optimizer which determines the
optimal control actions.
The model is used for prediction of the future behavior of the system when a specific control
action is applied. The prediction has two main components,

i. The free response, being the expected behavior of the output y0(t) assuming zero
future control actions.

ii. The forced response, being the additional component of the output dy(t) due to the
future control action u(t).

The total future system response is the summation of the free response and the forced
response. This sum is calculated until the prediction horizon. A cost function is used to
evaluate the resulting total response of the system. Finally, an optimizer will determine the
best set of future control values u(t), for example, the control actions to minimize the cost
function. The choice of this function can be made according to the demands of the process
and is not dependent on the controller itself. Therefore it is possible to use linear or quadratic
cost functions and penalize the control error or the control effort with appropriate weighting
factors. Hence, a control engineer can design the controller in an optimal manner to fit the
requirement of the controlled system [10].

2
 Figure.1.1. Structure of a typical MPC model [10]

System protection scheme

A system protection scheme (SPS) is designed to sense abnormal system conditions and take
predetermined, corrective action (other than the separation of faulted elements) to reservation
system integrity and provide acceptable system performance [1]. System protection schemes
came up as a compromise between investments, operation cost and customer service quality.
They are smart mechanisms to improve system performance (stability, safety and security) due to
their low economic costs and environmental openness when compared to the alternatives such as
building new power plants and transmission lines. The most common SPS are probably triggered
by either under frequency, under voltage, or high rate of change of frequency. These have been
organized extensively in the Ethiopian national grid power system. However, these generally
have local measurement, local processing and local actions. It was not economical to construct a
new transmission line to alleviate this situation. Therefore, the design and implementation of a
system protection schemes was carried out to dismiss the instability and overload system
condition. Traditional system protection scheme is designed in off-line planning studies. It is
implemented as shown in the figure 1.2 when it is triggered by some specific system operation
condition [1].

3
 Figure 1.2 General Structure of a SPS

It consists of three main parts:

 Inputs (transmission line overload, status of circuits’ breakers, etc.)
 A decision-making system that initiates certain actions based on the inputs
 Actions (such as generator or load tripping, capacitor switching, etc.)
According to input variables, SPS can be categorized into two categories: response based SPS
and event-based SPS. Response-based SPS uses power system operation condition to initiate
control actions after the disturbance has caused the input variables significantly degraded. Under
voltage load shedding and under frequency load shedding are two examples of this type of SPS.
Event-based SPS is designed to activate control actions by the direct detection of a particular
combination of events. This type of SPS is rule-based. Rules are established from on-line
simulation. Examples of event-based SPS are load shedding or generator rejection by the tripping
of a transmission line. Currently, Ethiopia also has this kind of protection for high voltage (HV)
level in national grid of power system.

4
Model predictive control has its roots in optimal control. Thus the basic concept of MPC is to use
a dynamic model to forecast system behavior, and optimize the forecast to produce the best
decision to control more at the current time. Models are therefore central to every form of MPC.
Because the optimal control more depends on the initial state of the dynamic system, a second
basic concept in MPC is to use the past record of measurements to determine the most likely
initial state of the system. The state estimation problem is to examine the record of past data, and
reconcile these measurements with the model to determine the most likely value of the state at
the current time. Both the regulation problem in which a model forecast is used to produce the
optimal control action and the estimation problem in which the past record of measurements is
used to produce an optimal state estimate involve dynamic models and optimization.

II. Literature survey

In literature, many algorithms have been developed to maintain a satisfactory voltage profile. A
detailed study of on-line voltage/var control performed on the Ontario Hydro system is presented
in [2]. The objective function is to minimize the transmission loss as well as the amount of
controls.
An optimal power flow (OPF) formulation is adopted to schedule generator voltages and
transformer tap positions in the bulk transmission system. Furthermore, the frequency of the
generator voltages and transformer tap changers to be optimized is studied to achieve most of the
possible transmission loss savings. An optimal power flow based real time voltage control
method is proposed in [3]. The primary goal is to eliminate voltage violations. However, under
conditions where violations cannot be removed, the objective is to minimize violations. Under
the normal operating condition where all the voltages are satisfactory, the objective is to
minimize system losses. The controls include transformer tap changers and generator voltage set-
points. Besides the OPF based voltage control, Artificial Intelligence especially expert system
has also been applied in the study.
[8] Develops an expert system to assist the decision-making of power system in presence of a
voltage violation problem. Empirical rules are generated to mitigate voltage problem using tap
changers, reactive injections, and generator voltage setting points. The proposed approach is
composed of four steps: identification of the knowledge required to solve a selected problem,
justification of the identified empirical rules, development production rules and testing and

5
modification. The sensitivity of voltage magnitude change with respect to various controllers is
used to evaluate the maximum voltage increase with the available control.
A rule-based approach for decentralized voltage control is presented in [6]. A network
decomposition technique is used to alleviate a bus voltage limit violation. Only the information
local to the bus with a voltage violation is adopted in control decision making. Network voltage
sensitivity with respect to the available voltage controllers is calculated and a rule-based
approach is developed to select the optimal set of control actions for alleviating voltage
violations. The controls include generator voltage set points, reactor switching and tap changers.
Distributed expert systems are developed in [5] for voltage control. The controls are composed
of load shedding, shunt capacitors, tap changers, static var compensator as well as load flow
control. Each var compensating device is controlled by a dedicated computer. An expert system
is designed in each computer. Communication exists among those computers. Contribution
degree is defined to evaluate the ability of a control in restoring voltages.
In [7] two rule-based techniques in a voltage control expert system are introduced. Reactive path
concept is adopted to determine control regions of each reactive power compensator and efficient
controllers for each observed bus. The decision is based on two sets of rule-based techniques. All
the above work is based on static analysis, in which only the real and reactive power balance in
each bus of a power system is considered.
The power system is assumed to have a stable operating point. The voltage performance criteria
could be met only if the system reaches a post-contingency stable operating point. However, if
disturbances are severing, the power system may lose stability. Under this situation, the control
strategy to restore the stable equilibrium point requires a dynamic analysis. The dynamic
behavior of system components such as dynamic load characteristics, dynamic behavior of load
tap changers need to be taken into account. In this thesis project proposal, computation of the
optimal strategies considering the dynamic behavior of a power system based on model
predictive control (MPC) is proposed. One approach to study the coordinated voltage control is
based on system response. In other words, direct telemetry of voltage at power system pivot
nodes is used as feedback to design voltage controller.

6
 III. Statement of the problem

The main motivation for this research proposal is drawbacks of traditional system protection
scheme which is observed in the Ethiopian high voltage power network that are severe for the
network planners and operators due to voltage instability which may result in the form of a
progressive fall or rise of voltage of some buses. That is, this traditional or response-based
system protection scheme uses power system operation condition to initiate control actions after
the disturbance has caused the input variables (the electrical parameter, voltage) significantly
degraded. Thus this system protection schemes cannot carry out the control actions to mitigate
the effects of potential instability or a safety/security degradation of a power network noticed by
an online dynamic security assessment program and it cannot maintain reliability of the network.

IV. Objectives

General objective
The general objective of this thesis is:
 To apply the model predictive control technique in Ethiopia high voltage electric power
network system to improve the voltage stability performance.

Specific objectives
The specific objectives of the thesis are:
 To study mathematical model of proposed power network system for voltage stability
control design.
 To analyze voltage stability performance with and/or without coordinated control
mechanism in the proposed power network system.
 To design model predictive control for voltage stability controls in the proposed power
network with coordinated control mechanism and evaluate performance of the propose
control technique by using PSAT (power system analysis tool box) and MATLAB
software.

7
 V. Scope of the Study
In the thesis, a coordinated control of static var compensators (SVCs) will be designed to
improve voltage performance. Given the locations and capabilities of SVCs the control design
problem will determine the control sequences and the control amounts to satisfy voltage
performance requirements.

VI. Methodology
Generally the following methodology will applied in conducting this research work
 exhaustive survey of literature will be made based on a clear definition of the scope and
focus of the work
 theory and modeling of the concerned power system including its method of solution
will be discovered
 survey of data and information sources will be conducted for the selected electrical
power network
 algorithms of the controller for part of the focus area will developed based on the
method of solution
 The developed algorithms will be made functional with MATLAB and Power System
Analysis Toolbox (PSAT)
 salient observations on the results will be obtained
 summary of salient observations and trends will be made to show how the research will
fill some gaps in the literature, scope and desirability of further work on the problem,
applications and potential areas

VII. Expected Outcome of the Project

Since the proper coordination of rule-based schemes control is highly dependent on the system
loading and state, algorithmic approaches that employ detailed network models to account for the
contemporary power flow in the system and thereby the effect of controls will be better
coordinated to minimize the usage of this traditional or rule-based schemes and to improve
voltage stability and increase reliability of the network.

8
 VIII. Importance of the Proposed Project
The importance of this work is to determine an optimal coordinated control strategy consisting of
continuous and discrete power system controls to improve voltage performance and prevent
voltage instability.

9
 IX. Work plan
Table.1: shows the activities to be accomplished within the scheduled time for phase-I.

S.N Months of the first semester

o.
Activities Sep.2017 Oct.2017 Nov.2017 Dec.2017 Jan .2017

Weeks Weeks Weeks Weeks Weeks

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 Exhaustive survey of
literature

2 Discussion with advisor

3 Project Proposal preparation

and submission.
4 Presentation of project
Proposal.
5 Submission of Formal
written proposal
6 Review of related literature
and survey of data and
information sources

7 Studying related software to

the desired control
mechanism
8 study mathematical model of
proposed power network
system
9 Organizing the data

10 analyze voltage stability for

the proposed power network

11 Thesis Progress- I report

Preparation and submission

12 Thesis Progress- I report

presentation

10
 Table.2: shows the activities to be accomplished within the scheduled time for phase-II.

S.N Months of the first semester

o.
Activities Feb .2017 March .2017 April .2017 May .2017 June .2017

Weeks Weeks Weeks Weeks Weeks

1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4

1 Analyzing and make

correction for the feedback
given from phase-I
presentation
2 Developing functional
algorithms for the controller
3 Discussion with advisor

4 penultimate draft paper

preparation
5 Thesis Progress- II report
Preparation and submission

6 Thesis Progress- II report

presentation

7 Correcting errors

8 Work on final thesis report

9 Hand in final thesis

document

10 Presentation

11
 X. Materials and Financial Requirements

Table.3. Stationery

S.No. Item Unit Quantity Unit price Total price

1 WR-CD Number 10 25 250

2 Flash memory ,, 1 400 400

3 Printing paper N/A 4 200*4 800

4 Photocopy paper ,, 3 150*3 450

SUB-TOTAL 1900

Table.4. Per-diem
S.No Person Number of days Daily rate Total

1 Student 5 160 800

2 Advisor 2 240 480

Sub Total 1280

Table.5. Transportation

S.No Person Departure Destination Number of Cost of Total cost

trips trip

1 Student Bishoftu Addis Ababa 6 80 480

Sub-total 480

12
Table.6. Miscellaneous

S.No Item/services Price

1 Printing the draft copy of the thesis and progress reports to the advisor 800

2 Data collection 400

3 Book qualifying final binding 1000

4 Photocopy and telephone 500

Sub total 2700

Table.7. Budget summary

No Description Sub-total(Birr)

1 Stationery 1900

2 Per-diem 1280

3 Transport 480

4 Miscellaneous 2700

Total 6360 ETB

13
 References

[1] Karlsson.D.and Waymel.X.,”System protection schemes in power networks”, Technical

report, CIGRE Task Force, 2001.
[2] Khalil, H. K.,” Nonlinear System (Third Edition)”, Prentice Hall, New Jeasey, 2002.
[3] El-Kady, M. A., Bell, B. D., Carvalho, V. F., Burchett, R., Happ, H. H., and Vierath, D. R.,”
Assessment of real - time optimal voltage control”, IEEE Transactions on Power Systems,
PWRS-1(2):,pp.98-105,1986.
[4] Vu, H., Pruvot, P., Launay, C., and Harmand, Y. (1996),”An improved voltage control on
large-scale power system”, IEEE Transactions on Power Systems, 11(3):pp.1295-1303, 1991.
[5] Matsuda, S., Ogi, H., Nishimura, K., Okataku, Y., and Tamura, S.,” Power system voltage
control by distributed expert systems”, IEEE Transactions on Power Systems, 37(3):pp.236-240,
1990.
[6] Wagner, W. R., Keyhani, A., Hao, S., and Wong, T. C.,” A rule based approach to
decentralized voltage control”, IEEE Transactions on Power Systems, 5(2):pp.643-651,1990.
[7] Godart, T. F. and Puttgen, H. B.,” A reactive path concept applied within a voltage control
expert system”, IEEE Transactions on Power Systems, 6(2):pp.787-793.
[8] Liu, C. C. and Tomsovic, K.,”An expert system assisting decision-making”, IEEE
Transactions on Power Systems, PWRS-1(3): pp.195-201, 1986.
[9] Mohammad Shahidehpour “Handbook of Electrical power System dynamics Modeling,
stability, and control”, wiley-IEEE press(2013)
[10] Guanji Hou ,’Trajectory Sensitivity Based Power System Dynamic Security Assessment “
Arizona state university May 2012

[11] J. Rawlings, “Tutorial overview of model predictive control,” IEEE Con-trol Systems
Magazine, v. 20, No. 3, pp. 38 – 52, June 2000
[12] R. Findeisen, L. Imsland, F. Allgower, B. Foss, “State and output feed-back nonlinear
model predictive control: an overview,” European Journal of Control, v. 9, pp. 190 – 206, 2003
[13] R. Findeisen, M. Diehl, Z. Nagy, F. Allgower, H. G. Bock, J. P. Schloder, “Computational
feasibility and performance of nonlinear model predictive control schemes,” in European
Control Conference, Porto, Sept. 2001
[14] M. Larsson, D. Karlsson, “Coordinated system protection scheme against voltage collapse
using heuristic search and predictive control,” IEEE Trans. Power Syst., v. 18, No. 3, pp. 1001 –
1006, 2003

14
[15] Y. Li, W. Liu, J. Liu, “The study on real-time transient stability emergen-cy control in
power system,” Proc. IEEE Canadian Conf. Electrical and Computer Engineering, v. 1, pp. 138
– 143, Aug. 2002

15
Name of the student: Asamenew Niguesse Reg.ID No. RPG/034/09 Signature: ___________

Name of project Advisor: Dr. Man Mohan Designation: PhD Signature: ________

Date of submission: November 23, 2017

16
 Defence University
College of engineering
Department of electrical power engineering
M.Tech. Project/Thesis Proposal Approval

I. Project/Thesis title:
Coordinated Voltage Control in Electrical Power network using
predictive control model
(Case Study: On high Voltage level in Ethiopia Power Network)
II. Student’s Name: Asamenew Niguesse Reg ID.No: RPG/034/09
Specialization: Power System and Automation
III. Name of Advisor: Dr. Man Mohan
IV. Reviewed and acceptance:
The M.Tech. Project/Thesis proposal has been reviewed and accepted by department
graduate committee (DGC) after the candidate defends his project/thesis title.
V. Approval by committee members

Name Signature Date

Subject expert
Subject expert
Advisor
HoD
Dean,PGP&R/PGPC

17
18