Journal of Power and Energy Engineering, 2014, 2, 86-97
Published Online September 2014 in SciRes. http://www.scirp.org/journal/jpee
http://dx.doi.org/10.4236/jpee.2014.29013
How to cite this paper: Chen, S., Wang, J. and Lie, T.T. (2014) Power Quality Regulation Using Free Market Oriented Ap-
proach. Journal of Power and Energy Engineering, 2, 86-97. http://dx. doi.org/10. 4236/jp ee . 2014.29013
Power Quality Regulation Using Free Market
Oriented Approach
Shiun Chen1, Jing Wang2, T. T. Lie3
1Serawak Energy, Kuching, Serawak, Malaysia
2Energy Market Authority, Singapore
3Department of Electrical and Electronic Engineering, Auckland University of Technology, Auckland,
New Zealand
Received May 2014
Abstract
Power quality is a compatibility issue between the supply systems and the connected loads. It is
often considered as a customer-side problem as its significance depends primarily on the sensitiv-
ity of the affected equipment and its function. Improvement solutions are implemented by or on
behalf of the customers only if the negative impact is causing great financial losses. However, as
the reliance on electronic devices continues to increase unabatedly, the percentage of equipment
being affected is expected to grow appreciably. In due course, the number of affected equipment
could be so large that it is only practical and economical to implement universal solutions instead
of individual solutions. This paper provides an early view of this scenario and explores the various
elements necessary for reining in the poor power quality, widely expected to deteriorate further
following deregulation of power industry. It attempts to use the free market oriented approach so
that solutions implemented are justified on economical basis.
Keywords
Power Quality, Economic Evaluation, Energy Market, Deregulation
1. Introduction
Power quality is an umbrella concept for a multitude of power system disturbances [1] [2]. In a broad sense, any
problem manifested in voltage, current or frequency deviation that results in failure or mis-operation of cus-
tomer equipment can be regarded as a power quality problem. Power quality regulation therefore refers to how
to provide a supply that is conducive to the connected loads. Technically, power supply system can only control
the quality of voltage; it has no control over the current that a particular load may draw. Therefore, power qual-
ity standards are generally devoted to maintaining the supply voltage within certain limits [3] [4]. Although the
many definitions of power quality in the literature and standards are not fully consistent with each other, but the
phenomena considered are generally the same. In IEC 61000-2-1 [1], a thorough description of various power
quality disturbances are categorized according to their typical duration, magnitude, and predominant frequency
as over-voltage transients, voltage variations, voltage fluctuations, harmonic distortions, frequency variations

S. Chen et al.
87
and voltage imbalance.
Maintaining power quality can be viewed as a financial consideration instead of a solely technical problem [5].
Unless the phenomenon and the subsequent disruptions on equipment or process result in significant monetary
losses, it may not be worthwhile to undertake any preventive or protective measures. This argument is further
enhanced by the fact that not every piece of equipment is affected by the same power quality disturbances. In
addition, similar equipment used for different purposes in different processes can have vastly different economic
impacts when disrupted. The nature of the disturbance, the condition of the supplying network, the equipment’s
sensitivity or tolerance, and the importance of the equipment to the customers, all combine to provide an indica-
tor for evaluating the severity of a power quality disturbance. Different customers are most likely to possess dif-
ferent importance levels and hence, under the spirit of deregulated market, they ought to be given the choice of
differentiated power quality at different prices.
Therefore, it appears logical to employ market mechanism that has been widely adopted in the restructuring of
power industry to regulate power quality. However, power quality encompasses many phenomena with distinct
characteristics, making it difficult for one to agree on a generic definition. This perhaps causes each phenome-
non to be treated separately and they are largely being managed in the same manner as before deregulation. This
paper evaluates the potential of using market oriented approach in regulating power quality. It takes a conceptual
view of the related issues, identifies the challenges that need to be overcome and studies the various issues that
define how such free-market idea can be employed.
2. Power Quality Regulation
The expectation of electricity consumers has evolved over the past years, and has risen partly due to the extra
publicities from the deregulation activities resulting in greater awareness, and better education [6]. This higher
demand is nevertheless logical because the modern industrial processes with new types of electrical equipment
are more adversely affected by poor power quality than the traditional equipment. In addition, integrated manu-
facturing supported by extensive automations and our today’s individual way of life are more dependent on
good quality of electricity than ever. All these can be narrowed to the ever-increasing use of semiconductor de-
vices that have poorer tolerances towards power quality disturbances comparing to traditional equipment.
2.1. Existing Approach
There are many voltage quality regulations aiming to keep disturbance levels under certain limits that would al-
low all the electrical equipment connected to the grid to function properly [7]-[11]. Comparisons of different
standards used in various countries were conducted for European countries [12] and Asia Pacific countries [13]
[14]. The various regulation techniques were studied according to the categories of the voltage imperfections.
For the surveys, it is clear that many countries have some form of voltage regulation, applied uniformly at the
national level. Voltage quality is usually part of the national regulation, and EN50160 is often used as the regu-
lation for low-voltage and medium-voltage levels in European countries. For higher voltage levels, only simple
criteria such as frequency and voltage magnitude as specified in EN50160 are adopted.
However, there is no explicit standard for the majority of voltage quality phenomena. In EN50160, exact le-
vels of compliance are stated for a few phenomena only. For most phenomena such as short duration variations,
only indicative values are given. It is left to the users to define their own compatibility levels. There are har-
monic limits defined in many countries but in some countries like Singapore, only frequency and long-duration
voltage variations are regulated. Despite having these standards and regulations, no country imposes penalty for
not meeting the regulations. Most utilities simply want to deliver a good product and therefore have committed
to that. This commitment appears to continue even after deregulation. Likewise, there is no reward for good
performance as there is no penalty for poor performance.
From the studies, it is observed that the regulation of power quality or voltage quality in most countries re-
mains the same as before the industry is deregulated. The regulation and emission limits are set out of concern
with system security and to achieve economical energy dispatch. The financial consequences of poor power
quality are still not considered.
2.2. Free-Market Approach
The first challenge in employing free-market approach in regulating power quality is how to quantify it as a sel-

S. Chen et al.
88
lable entity. This requires some forms of definition that can differentiate a quality level from another, allowing
different prices to be set to reflect its values. These definitions must allow the system quality level to be clearly
linked to the tolerances of sensitive equipment, and perhaps the following financial impacts too. The system
level is the basic or minimum level from which further improvement or correction measures can be implemented.
It needs to be carefully chosen as a too pessimistic value (high level of quality) would lead to undue measures,
higher costs and wastages since not all loads require such level. On the contrary, setting the value too low would
result in poor quality causing many disruptions and making it necessary for many loads to install improvement
or protection measures. In general, this value needs to be optimally chosen and set high enough to be deemed
adequate by most electrical equipment without causing any cost overrun.
Power quality is an umbrella concept encompassing many electromagnetic phenomena. Each of them has dis-
tinct characteristics, causing different problems and requiring different solution too. It therefore appears com-
plicated to produce a universal definition for grading power quality. One possible way of quantifying power
quality levels is to consider each of the negative symptoms individually. More specifically, as the main intent of
introducing free-market approach is to use economic signals in regulating power quality, this paper considers
how various phenomena can be controlled and the associated costs. The various phenomena are therefore di-
vided into two general groups of emission limit control and value added service mode of power quality man-
agement.
The former group accounts for phenomena that originate from customers and affect the quality of supply sys-
tems. It includes harmonic current emissions and randomly varying power demands from fluctuating loads like
the arc furnaces. The solution would be primarily corrective in nature for polluting equipment, and the regula-
tion would need to enforce certain limits on the customers. The latter group concerns the quality of supply volt-
age and accounts for disturbances like voltage dips, swells, interruptions, imbalances and distortions. These are
largely considered as inadequacies in the supply system and the solution would generally be protective in nature
for sensitive equipment. A minimum but acceptable quality level needs to be agreed upon between the custom-
ers and the power company.
In introducing free-market idea, monetary incentives or penalties are designed to deter emission in the former
group. Each user can be prescribed a set of limits according to the power demand and the supply voltage level.
Lower limits are enforced at high voltage levels as such pollutions have more wide-ranging influence over the
entire network. The step-down transformers often act as a layer of separation between the system and disturbing
loads. The penalty can be set according to the cost of cleaning up the emissions. For the latter group, economic
signal is set to encourage investments in power quality improvement measures. However, there are two ways of
determining the prices for different quality levels. One obvious way is to consider the cost of improvement mea-
sures, which tends to increase with power ratings of the protected loads. The second way is to set the price at the
perceived “value” of the quality level, which depends on the economic consequence suffered by customers if no
action is taken. This economic impact may vary significantly between different classes of customers, depending
on how the affected equipment is used and its importance to the customer’s business function. This determina-
tion may be possible for individual customers but would be irrelevant for the masses. A computer is expected to
be more highly valued by a commercial customer using it for inventory control than a residential customer using
it for casual exchanges with friends.
Previously, power networks were planned taking into account a certain level of expected power quality. But
now, the base-rate regulation is being replaced by performance-based remuneration schemes in the liberalized
environment. Unfortunately, these new schemes encourage reducing operating costs and trimming infrastructure
investments. Many think that this will lead to deterioration of power quality. Hence, in such restructured envi-
ronment, an incentive-based mechanism is necessary for power quality regulation. Separate markets for energy
and reserve are already in operation and some are contemplating a reactive power market. Equally, a new market
can be conjured to manage power quality and to prevent it from further deterioration. The following sections
study the potential makeup of a power quality market and the associated pricing mechanism that is crucial for
the success of the market.
3. Power Quality Benefits and Costs
To establish a market for power quality as a product or service, it must be made quantifiable so that the benefits
and costs can be readily evaluated. The measurable quantity must enable decision to be made on whether or how

S. Chen et al.
89
much to invest in improvement solutions. Here, a quantity “Effectiveness” is introduced as measure of the
amount of savings achieved by power quality improvement services. It is defined as the percentage of distur-
bances that can be prevented through the particular service. For example, 50% Effectiveness means that with the
improvement service, half of the disturbances that originally cause disruptions can now be prevented from caus-
ing any disruption. Obviously, 100% is the absolute limit for this Effectiveness. The benefit gained by the cus-
tomers and the costs of improvement services are then defined according to the targeted level of Effectiveness.
This section shows the general shape and trend of the benefit and cost curves. These curves serve as founda-
tion for determining the basic power quality level based on monetary basis.
3.1. Customer Benefit Functions
The customer’s willingness to pay for improvement service depends on the potential benefits that the customer
can gain from the service. This value is purely decided by the importance of the load and tends to vary from in-
dividual to individual. This value determines the demand from customers for the service in term of quantity and
quality. There will always be some customers who value their loads more than the others. The customers with
important load shall benefit more from the same service and consequently may be more willing to pay more for
it. However, if a universal service is provided, customers with less important loads shall also benefit. The value
of benefit that realized by these less important loads is expected to be smaller. Thus, as the power ratings of the
improvement service increases, the unit benefit shall decrease with the total benefit increasing at a slower rate as
shown in Figure 1. The X1 and X2 denote services of different quality levels (or Effectiveness) where X1 is of a
lower quality than X2.
Translating this to unit benefit function to represent how much one unit of improvement service can bring, the
curves would be decreasing as shown in Figure 2. It is also piecewise as the different composition of loads en-
tails different value for the same service in accordance to their nature of operation. In general, higher valued
loads would be protected first. Hence, the curve starts with high values for this unit benefit and decreases sub-
sequently. As the improvement device reaches its power ratings limit, a different device would be needed re-
sulting in a discontinuity in the curve. It is only logical that the new device would be less effective (lower unit
benefit value) and the protected loads are valued less (continued trend of decreasing unit benefit). If the unit
benefit is defined as a monetary value, an equivalent price for the improvement service can be set on the curve.
For example, for a price of P1, only a quantity S1 of load can only derive financial benefit from this service. Ad-
ditional loads can only be protected if a lower cost service is available.
Assuming that each disruption causes a fixed amount of losses, the total benefit curve is then a linear function
of the Effectiveness as follows,
Power ratings (kVA)
X1
X2
Total benefits
Figure 1. Customer benefits w.r.t. ratings of
improvement solution.
X
2
P
1
X
1
S
1
S
2
kVA
Unit benefit
Figure 2. Customer unit benefit function.

S. Chen et al.
90
Benefit = k∙N∙X (1)
where
k is the aggregated loss of the customers due to one disruption;
X is the Effectiveness of the service, the percentage of the mitigated voltage sags that falls within the tolerance
curve of the loads;
N is the original number of voltage sags beyond the tolerances of the loads.
It is obvious that the customer benefit would depend on the system performance and the equipment’s Effec-
tiveness. The variable N representing the original sag performance and can be derived through measurements
[15] or estimations [16].
With the eventual power quality determined by the Effectiveness chosen, the customer benefit is therefore di-
rectly proportional to the quality too. The benefit functions will be straight lines, increasing with the Effective-
ness. An improvement service of higher power ratings would produce more benefit as it benefits more loads.
However, the rate of increase would be slower as the unit benefit is expected to decrease with the power ratings
as shown in Figure 3. In other word, the aggregated benefit k per disruption becomes smaller when a larger
volume of loads is being protected.
3.2. Cost Functions of Improvement Services
The cost of the service is dictated by the power conditioning equipment, the selection of which is subjected to
the targeted Effectiveness and the power ratings of the protected loads. At the initial stage when the Effectiveness
is low, only a modest cost is needed to mitigate the few minor disturbances. As the requirement rises, the cost
would increase at an increasing rate or exponentially as mitigating more severe disturbances is expected to be
more costly. Thus for a given power ratings, the total cost function shall exhibit an exponential-like characteris-
tic as shown in Figure 4. It also varies with the power ratings. Basically, the unit cost is small when the Effec-
tiveness requirement is low but increases gradually as the need becomes more stringent.
In addition, if considering the restriction from the technologies that are available to provide the improvement
service, the cost function would be made up of several discrete lines with each line representing a specific tech-
nology. Fig ure 5 shows such a scenario with each ladder step representing a different technology. For power
quality issues like voltage sags, a different technology may be needed to achieve a different Effectiveness of
protection. Likewise, different curves would need to be constructed for different power ratings.
The cost of the power conditioning equipment usually varies with its power ratings as well as its Effectiveness.
Naturally, it can be expected that for higher ratings, the cost would be higher as more energy needs to be stored
to provide the service. In addition, the unit cost is also expected to change with the ratings. Although no definite
100%
Effectiven ess
S
4
S
3
S
2
S
1
Total benefit
Figure 3. Customer benefit w.r.t. Effectiveness.
Total Cost
S2
S1
100%
Effectivenes s
Figure 4. Total service cost w.r.t. Effectiveness.

S. Chen et al.
91
Total Cost
Effectivenes s
Figure 5. Ladder like service cost curve.
pattern of relationship between the unit cost and power ratings is found for each protection device, very different
costs have been observed for some of them. From [5], the cost for each improvement device is given as unit cost
per kVA. Hence, the cost function shall be generalized as the following,
C = ai∙S (2)
where
i indicates different quality level in terms of type of equipment used or the Effectiveness of the solution
S is power ratings of the improvement service
At low ratings, the cost is dominated by some initial outlays related to the service installation. For a given
quality or Effectiveness, the total cost shall increase exponentially with the power ratings. In Figure 6, several
discrete lines represent the costs associated with different technologies (or different quality levels). The cost of
improvement service increases linearly until a different technology is needed to accommodate the next higher
power ratings.
3.3. Cost-Benefit Optimization
The main objective in carrying out cost-benefit analysis here is to determine a quality level that can satisfy both
the customers’ needs while keeping the cost in check. This requires determination of the service Effectiveness or
quality and the power ratings or quantity needed. This in turn leads to the determination of the price. The deci-
sion can be made through a process of maximizing the net benefit that can be achieved through the selection of
the right quality (Effectiveness) and quantity (power ratings) to be served.
For a given power ratings, the cost function exhibits an exponential-like curve. In Figure 7, it is drawn as a
continuous curve assuming that the improvement device can be smoothly enhanced to raise its Effectiveness. A
higher cost function is expected for higher power ratings. Assuming a fixed amount of losses per disruption and
the number of disruptions is inversely proportional to the Effectiveness, the benefit curve would be a straight and
increasing line. At higher power ratings, the benefit is expected to be higher but increasing at a lower rate with
Effectiveness since the additional loads that are being protected would not be valued as highly as the more im-
portant loads that are protected earlier.
The quality and quantity to be provided shall be determined through a process of maximizing the net benefit
that can be achieved through the selection of the right quality to be served. From this figure, the optimization
process is to find the optimal power ratings (determining which pair of benefit and cost functions to use) and the
optimal quality level (deciding the targeted Effectiveness of improvement service) where the difference between
them is maximized. This would maximize the net benefit indicated as π in the figure. The price of this level of
power quality can then be determined from this Effectiveness level. At an aggregate load point, there are no
separate lines available to offer different customers different quality service, then only a uniform quality can be
delivered and the ratings of the load is fixed. Thus there is only one pair of cost and benefit curves, and the op-
timization is simplified to the determination of the optimal quality only.
In a “seller” market with limited option, the price would be P1 with all the net benefit goes to the service sup-
plier. On the other hand, in a “buyer” market where there are many service suppliers, a cost-based price of P2 is
expected to give all benefits to the customers. If the power quality market is centrally coordinated, the coordi-
nator can choose to apportion this benefit between the customers and suppliers by picking a price between these
two extremes such as P3. Through this mechanism, appropriate values can be derived as the differentiated prices
for different grades of power quality.

S. Chen et al.
92
KVA
X
1
Cost of service
Power ratings
X
2
Figure 6. Service cost w.r.t. power ratings.
Effectivenes s
S
2
π
S
1
$
Customer
benefit
curve s
Service
cost
curve s
P
1
P
2
P
3
Figure 7. Customer benefits and power quality im-
provement service costs.
4. Illustrative Examples
Using the information provided in various literature [5], [15] and [16], the following example show how the op-
timal quality and quantity of improvement service can be determined. Three types of the power conditioning
equipment are used here to denote the different levels of Effectiveness. They are the DVR (Dynamic Voltage
Restorer), flywheel ride-through devices and the UPS (Uninterruptible Power Supply) as a battery-based ride-
through method. Their costs for different sizes are given in Table 1 [5].
Here, the illustration is shown for loads that are between 2 - 10 MVA. The cost function is a piecewise ladder
curve as shown in Figure 8. For loads with other power ratings, similar procedure can be undertaken and the
resulted net benefits are compared across measures using improvement service of different power ratings.
The cost of disruption indirectly defines the benefit that can be achieved from carrying out the improvement
service. This cost can be derived from those defined for interruptions and is used to indicate the impacts of the
disruption on different customers, as shown in the Table 2.
The system sag performance from the EPRI survey [15] is then used as the basic level of quality that can be
expected from the distribution system as shown in Table 3. With this system performance, there are a significant
number of disruptions expected by the ITIC complaint loads. Therefore, improvement service can be contem-
plated and justified not only technically but also economically.
With this system performance and the loads are ITIC compliance, the number of disruptions due to voltage
sags beyond the load tolerance is 23.207 times. Using this information, the customer benefit can be calculated
since the cost a single disruption causes is known. The benefit shall be the avoided loss after the protection ser-
vice is implemented. The customer benefit shall be linear to the Effectiveness of the improvement equipment.
Figure 9 shows the benefit curves for different categories of customers and they are assessed against the costs of
the improvement options.
From the Figure 9 and Figure 10, the optimal quality of service can be determined by finding the optimal
quality level (Effectiveness) where net benefit is maximized. This is when the difference between the benefit
curve and the cost curve is at the greatest. For the commercial and industrial customers, the net benefit is maxi-
mized for selecting UPS as the protection measure. On the other hand, for the residential customers, flywheel
ride-through provides higher economical net benefit even though it is less effective than UPS. This is because
residential loads are not valued as much as those of commercial and industrial customers. Hence, it is not eco-
nomically worthwhile to apply the higher-cost UPS protection.

S. Chen et al.
93
Table 1. Cost of improvement services.
Cost ($/kVA-year) 2 - 10 MVA 10 - 300 kVA <5 kVA
DVR 45 30 150
Flywheel ride-through 75 85 -
UPS with battery 125 125 175
25%
DVR
95%
100%
Effectiveness
Flywheel
UPS
Cost ($/kva)
45
75
125
Figure 8. Cost function w.r.t. Effectiveness
within 2 - 10 MVA range.
Table 2. Customer benefits.
Benefit ($/kw) per disruption
Commercial 125
Industry 80
Residential 15
Table 3. Sag density table from EPRI survey.
Magnitude 0.00 - 0.02 s 0.02 - 0.05 s 0.05 - 0.20 s 0.20 - 0.40 s 0.40 - 0.60 s 0.60 - 0.80 s >0.8 s
90% - 100% 0 0 0 0 0 0 0
80% - 90% 4.22 6.33 12.683 6.4 1.367 0.666 0.067
70% - 80% 2.92 4.38 8.817 4.55 0.933 0.434 0.033
60% - 70% 0.439 0.657 1.388 0.875 0.116 0.167 0
50% - 60% 0.437 0.658 1.387 0.875 0.117 0.167 0
40% - 50% 0.439 0.6575 1.3875 0.875 0.117 0.166 0
30% - 40% 0.108 0.1625 0.7795 0.375 0 0.2 0
20% - 30% 0.109 0.162 0.679 0.375 0.1 0.2 0
10% - 20% 0.108 0.163 0.729 0.375 0.05 0.2 0
0% - 10% 0.52 0.78 0.525 0.675 0.5 3.4 1.7
For the RBTS system, disruption to loads sensitive to voltage sags beyond the ITIC tolerance is much smaller
at 2.322, with the system performance as summarized in Table 4. Therefore, the impact of the improvement
services is generally small. Only for some very important loads like those belonging to the commercial and in-
dustrial customers, the flywheel ride-through protection is good enough to achieve the maximum net benefit. For
residential customers, there is no need to apply any protection since the system performance is already adequate
according to the lower values of the loads. In fact, it is not worthwhile to apply UPS for any load since the ex-
isting system performance is already very good.
The above examples using EPRI survey data and RBTS simulation data are summarized in the following Ta-
ble 5, Table 6. The underlined figures indicate the solution with the highest net benefit.

S. Chen et al.
94
Effectivenes s
DVR
Flywheel
2900
1856
348
Benefit/cost
90%
100%
36%
UPS
Residential
Commercial
Industrial
Figure 9. Benefit-cost analysis using EPRI survey data.
Total Benefit
Effectivenes s
25%
95%
100%
DVR
UPS
35
Flywheel
Industrial
Residenti al
Commercial
290
185
Figure 10. Cost benefit analysis for RBTS system.
Table 4. Sag density table for RBTS system.
Magnitude 0.00 - 0.05 s 0.05 - 0.20 s 0.20 - 0.40 s 0.40 - 0.60 s 0.60 - 0.80 s >0.8 s
90% - 100% 0 0.471 0.073 0.088 0 0.192
80% - 90% 0 0.011 0 0 0 0.016
70% - 80% 0 0.035 0.007 0.008 0 0.003
60% - 70% 0 0.020 0.006 0.002 0 0
50% - 60% 0 0.187 0.017 0.009 0 0.019
40% - 50% 0 0.422 0.139 0.084 0 0.020
30% - 40% 0 0.318 0.018 0.120 0 0.095
20% - 30% 0 0.207 0 0.003 0 0.161
10% - 20% 0 0.008 0 0 0 0.101
0% - 10% 0 0.110 0.097 0 0 0.132
5. Pricing Strategy and Benefit Allocation
By maximizing the net benefit as described above, the optimal quality of service can be delivered. The price of
the service shall determine how this net benefit is allocated between the supplier and the customers. In the proc-
ess of determine the optimal service quality; it is assumed that the customers are willing to take up any service

S. Chen et al.
95
that can bring more benefit than the price they have to pay. For the benefit function defined with respect to
quantity, the unit benefit of the customers with respect to quantity can be obtained by simply taking the deriva-
tive of the total benefit function. And this function can be utilized as the demand function of the customers’
willingness to pay for the service.
As described earlier, customers can be categorized into three major groups of commercial, industrial and
residential, very much according to the sensitivity of their load equipment and importance. Consequently, the
customer demand function for improvement service would be expected to demonstrate different elasticity due to
the different composition of loads among them. However, to facilitate the study here, the demand function is
taken to be linear over certain range of quality and quantity of power quality improvement service. This assump-
tion enables the demand curve to be described as:
P = Pmax −
ε⋅ S (3)
where
S is power ratings of the customer load to be provided with power quality improvement service;
ε stands for the elasticity.
These relationships can be demonstrated in the following Figure 11. MR is the marginal revenue (or benefit)
while MC is marginal cost of the improvement service. P* and S* indicate the selling price and quantity where
the profit of the supplier can be maximized. The profit earned by the supplier is indicated by the area of
P*BAP**.
From the perspective of overall social economic benefits, the supply of power quality service should be pro-
vided as long as the price is higher than the marginal cost of the service. Then, the social welfare is indicated by
the area of DP**E. With the current supply quantity S*, the area of DBP* indicates the customer surplus or bene-
fit enjoyed by the customers. The area of P*BAP** indicates the consumer surplus transferred to the firm pro-
viding the service. The area of BEA indicates the potential consumer surplus that has not being tapped when
providing only S* amount of service. The existence of such loss shows that economic efficiency has not been
fully achieved. This loss should be minimized in order to maximize the overall benefit. In this case, the supply
Table 5. Summary of performance with EPRI survey data.
Total customer benefits ($) Maximum net benefit π ($)
DVR Flywheel UPS DVR Flywheel UPS
Commercial 1097.99 2616.63 2900.88 1052.99 2541.63 2775.875
Industrial 702.71 1674.64 1856.56 657.71 1599.64 1731.56
Residential 131.7585 313.995 348.105 86.7585 238.995 223.105
Table 6. Summary of performance with RBTS Simulation data.
Total customer benefits ($) Maximum net benefit π ($)
DVR Flywheel UPS DVR Flywheel UPS
Commercial 74.93 277.47 290.17 29.93 202.47 165.17
Industrial 47.96 177.58 185.71 2.96 102.58 60.71
Residential 8.99 33.30 34.82 −36.01 −41.70 −90.18
D
MR
P
*
B
A
E
P
**
Price ($)
S
*
S
**
Ratings (kVA)
MC
Demand curve
Figure 11. Marginal benefit and cost of PQ
improvement service.

S. Chen et al.
96
quantity should be traded at the S**. This can then lead to two possible outcomes. One is to set one uniform price
at the marginal cost P**, which is the situation with perfect competition, and the entire surplus goes to the cus-
tomer. The other is to practice price discrimination which is to set more than one price with the lowest one at P**
so that the selling quantity of S** can be realized.
5.1. Differentiated Pricing
As shown in Figure 12, by setting up a series of different prices and ratings to customers, the supplier can
achieve greater profit by selling more products at price above the average cost. Through this, more of the con-
sumer surplus is transferred to the firm. To implement differentiated pricing, it is necessary to find out the de-
mand curves for each category of the customers.
Under a monopolistic condition, the service provider is able to practice perfect price discrimination by charg-
ing different prices so that the entire consumer surplus goes to the firm. However, considering the practical con-
straints, it is impossible to implement perfect price discrimination over all customers. Partial price differentia-
tion may be possible since customers of different nature exhibit different sensitivity to the power quality distur-
bances. Different tariffs can be set for each category of customers, although essentially, they can all be protected
using the same type of power conditioning equipment. If the provider can offer different degrees of guarantee to
different customers, maximum benefits to all can be determined using the following formulation.
max
j jkkj
kj
MaxBenefitdr d ρ = −
∑ ∑∑
(4)
where
ρ
j is the tariff charge with PQ service for a consumer category;
rk is the average cost per power unit, which shall include fixed and variable costs, at different load points;
dkj is the consumption of power quality service by a category of consumer at each load point.
5.2. Influence of Government Activity
The above optimization is undertaken from the investor’s point of view. The power quality market regulator can
still exert influence by setting price caps. From the regulator’s viewpoint, the objective is to achieve the overall
social economic welfare. Especially in a public utility business such as the power utility, regulator often does not
allow the company to take all the benefits. There must be some sort of regulations to prevent a monopoly firm
from exploiting the customers.
As discussed above, the economic efficiency can be fully achieved when the price is equal to the marginal
cost. This is basically marginal pricing. To implement multiple tariffs, the low-price takers can be offered with
marginal price and at the efficiency that can best be achieved. However, there is still the possibility that the mo-
nopoly firm can still exploit those more affordable consumers through fixing of high prices.
Another approach is through the rate of return regulation. This rate of return regulation is a traditional frame-
work to define the macro-economic performance of regulated companies. The regulator can devise the monop-
oly pricing schemes by setting a fair rate of return on investment. An appropriate rate of return can give the firm
the right incentives to minimize cost and the resulted price will give both the consumer and the firm equitable
benefits.
P
**
D
E
S
**
P
1
P
2
S
1
S
2
Price ($)
Ratings (kVA)
Figure 12. Differentiated pricing mechanism.

S. Chen et al.
97
6. Conclusion
Power quality problems are often regarded as financial considerations in addition to solving the technical issues.
A free-market approach where monetary returns take precedence would be a natural choice for regulating power
quality. Unless the benefits gained by the customers is higher than or equal to the price, there will not be a mar-
ket for such quality level. Similarly, unless suppliers can achieve a price higher than the cost of improvement
measures, there will not be any such services offered to the customers. To make use of market signals as incen-
tives for investments in power quality improvement, a power quality market needs to be established. Different
grades of power quality are to be defined and set at differentiated prices. Choosing the power quality grades and
setting of the corresponding prices need to compare the customer benefits against the costs of the improvement
services. Once the quality levels are chosen, the prices can be fixed by apportioning the marginal benefits be-
tween the customers and the service suppliers. Although the monetary aspect of power quality regulation can be
resolved by introducing a dedicated market with proper pricing mechanism, there remains many technical chal-
lenges such as how to deliver different grades of power quality over a common distribution system.
References
[1] IEC 61000-2-1 (1990) Electromagnetic Compatibility (EMC)—Part 2-1: Environment-Description of the Environ-
ment-Electromagnetic Environment for Low-Frequency Conducted Disturbances and Signalling in Public Power Sup-
ply Systems. 1st Edition.
[2] IEEE Std. 1159-1995 (1995) IEEE Recommended Practice for Monitoring Electric Power Quality.
[3] Dugan, R.C. (2003) Electrical Power Systems Quality . 2nd Edition, McGraw-Hill, New York.
[4] Arrillaga, J., Watson, N.R. and Ch en , S. (2000) Power System Quality Assessment . John Wiley & Sons, New York.
[5] McGranaghan, M. and Roett ge r, B. (2002) Economic Evaluation of Power Quality . IEEE Power Engineering Review,
22, 8-12.
[6] Shahidehpour, M. and Alomoush, M. (2002) Restructured Electrical Power Systems: Operation, Trading, and Volatil-
ity. Marcel Dekker, Inc., New York.
[7] BS EN 50160 (2000) Voltage Characteristics of Electricity Supplied by Public Distribution Systems.
[8] IEC 61000-2-2 (2002) Electromagnetic Compatibility (EMC)—Part 2-2: Environment-Compatibility Levels for Low-
Frequency Conducted Disturbances and Signaling in Public Low-Voltage Power Supply Systems.
[9] IEC 61000-2-4 (2002) Electromagnetic compatibility (EMC)—Part 2-4: Environment-Compatibility Levels in Indus-
trial Plants for Low-Frequency Conducted Disturbances.
[10] IEC 61000-3-2 ed. 2 (2000) Electromagnetic compatibility (EMC) —Part 3-2: Limits-Limits for Harmonic Current
Emission (Equipment Input Current up to and Including 16 A per Phase).
[11] IEEE Std. 519-2002 (2002) IEEE Standard Practices and Requirements for Harmonic Control in Electrical Power Sys-
tems.
[12] (2001) Quality of Electricity Supply: Initial Benchmarking on Actual Levels, Standards and Regulatory Strategies.
Council of European Energy Regulators.
[13] (2001) Electricity Distribution Code. Office of the Regulator-General, Victoria.
[14] (2002) Singapore Electricity Market Rules, Energy Market Authority.
[15] Gunther, E.W. and M eht a, H. (1995) A Survey of Distribution System Power Quality-Preliminary Results. IEEE
Transaction on Power Delivery, 10, 322-329.
[16] Wang, J., Chen, S. and Lie, T.T. (2007) A Systematic Approach for Evaluating Economic Impact of Voltage Dips.
Electric Power Systems Research, 77, 145-154.