Page 276 - Decision Making Applications in Modern Power Systems

Distributed generation in deregulated Chapter | 9 239

constraints detailed in Section 9.5.2. Accordingly, PF correction capacitors
were considered to enhance the base system PF to reach 0.9229 using four
capacitor banks located at buses 9, 18, 21, and 25, respectively, and each of
them has a rating of 1.2 MVAr as per the optimal results achieved in Ref.
[28]. HC enhancement of the EDS is investigated as a constrained optimiza-
tion problem using particle swarm optimization (PSO) technique in both
deterministic and probabilistic assessments. The widely used PSO algorithm
[29] is used to solve the optimization problem presented in this work. PSO
method is a heuristic-based method, which is applied effectively in various
electrical engineering problems. The parameters of PSO are set to a popula-
tion size of 10 individuals, cognitive and social acceleration coefficients c1
and c2 set to 2, minimum and maximum inertia weights set to 0.1 and 1.1,
respectively [23], and the maximum number of iterations is set to 100.

9.6.1 Deterministic hosting capacity results

In this chapter, two DG integration schemes have been examined; three DG
and four DG schemes. Other arrangements could be explored and the optimal
configuration should consider technical, economic, and network operator
requirements. Results are reported at two selective durations, at the 14th
hour (100% loading) and at the 5th hour (60% loading), according to the
daily load profile in Fig. 9.6. The deterministic HC approach simulations are
performed under two main conditions:
1. Reverse power flow in network branches is not allowed (no back-feed).
2. Back-feed is allowed, under operators’ control and supervision, while
updating the network protection settings and configurations to allow for
these bidirectional power flows.

In addition, simulations were developed at two loading times, namely,
full loading (at the 14th hour) and light loading (at the 5th hour) as shown in
the daily load profile in Fig. 9.6. The obtained results of the deterministic
HC approach for both conditions (1) and (2) are presented in Tables 9.5 and
9.6, respectively.
From the achieved results in Tables 9.5 and 9.6, it can be noticed that the
probabilistic HC levels are higher than the deterministic HC results. In other
words, ignoring the uncertainty of electrical parameters results optimistically
that cause a noticeable underestimation to the HC levels achieved from prob-
abilistic assessments. Besides, it can be noticed that the HC of a network is
not a fixed value along the day. However, as emphasized by the results of
Tables 9.5 and 9.6, it is concluded that the network’s HC level at full loading
(H 5 14) duration is higher than the HC level during light loading durations
(H 5 5). In addition, it is found that higher HC levels can be obtained by the
proper utilization of both active and reactive powers injected by the intelli-
gent DG units, such as PV units controlled by smart inverters. As concluded