Power System Planning Assessment for Optimizing Renewable Energy Integration in the Maluku Electricity System

Abstract

The planning of electrical power systems in remote island areas poses a few challenges, such as requiring many load centers, various energy sources, and certain geographical conditions, which leads to inefficiencies in energy production. For this reason, it is necessary to plan an electrical transmission system to efficiently transfer the power between load centers. Previous research has found that the completion of the most economical power system will be achieved in 2050 on the condition that the Ambon, Seram, Haruku, and Saparua (Ambon-SHS) systems are interconnected in 2025. Providing complementary support, this paper conducts a power system analysis to assess the proposed interconnection system in Maluku Island, which has several islands, small load centers, and local renewable energy resources. The power system analysis was performed using DIgSILENT PowerFactory and was carried out every 5 years of the system planning period until 2050. The results show that the design of the obtained transmission system meets the standard security criteria, which include thermal loading and voltage, being under both normal and N-1 contingency conditions, a short-circuit current, voltage stability, frequency stability, and transient stability. Finally, this paper demonstrates that the proposed plan is economically and technically feasible.

1. Introduction

Infrastructure for power systems is required for industrial and economic growth. Electricity consumption is a good indicator of a country’s economic growth, particularly in developing countries [1]. Among ASEAN countries, Indonesia was ranked sixth in 2014, with an average electricity consumption of 812 kWh per capita [2]. In comparison, Malaysia and Singapore have nearly six to nine times higher electricity consumption than Indonesia. One reason for the low electricity consumption is the fact that a considerable amount of electricity is consumed by households to meet their most basic needs, as presented in

Table 1. Energy sales per customer group [3].

Consumer Group	Annual Energy Sales (GWh)
Household	103,733
Industry	77,879
Business	46,901
Social	8622
Others	8383

. In 2019, household electricity sales totaled 103,733 GWh (42.25%), while industrial sales totaled only 77,879 GWh (31.72%) [3].

The different priorities among islands concerning electricity infrastructure planning also contribute to the low average electricity consumption. In addition, the difficulties in providing electricity to eastern Indonesia are related to social, geographical, and demographic conditions [4], as it consists of over 17,000 islands. The majority of the islands are in the eastern part of Indonesia and are divided into two major regions, Maluku and Papua, which are still underdeveloped when compared to other regions. Furthermore, the communal areas in those islands are far apart, which presents another challenge in power system planning.

These factors make it difficult to establish an electricity supply in Maluku and Papua. Currently, the electric power system in Maluku and Papua consists of 200 isolated systems with low electricity consumption. In 2019, the average electricity consumption of eastern Indonesia was only 390 kWh per capita [3]. This value is lower than Jakarta, Indonesia’s capital city, which has a value of 3236 kWh per capita [3]. Although it is still relatively low, the demand for electricity in Maluku and Papua, particularly in Maluku Province, has started to increase. Over the last 8 years, electricity sales in Maluku have increased by an average of 11%, from 337 GWh in 2011 to 599 GWh in 2018, as illustrated in Figure 1 [5]. In 2028, the demand for electrical energy in Maluku Province is expected to reach 1206 GWh [5]. However, due to the difficulty in transferring local energy resources, diesel generators continue to dominate the electricity production in Maluku, representing approximately 90.1% of the total generation in 2019 [3].

Considering the carbon dioxide emissions, generation expansion planning (GEP) must consider the emissions produced by generating units [6]. To avoid environmental problems and ensure energy security, the future energy mix should include a greater proportion of renewable energy [7]. Maluku has abundant potential for renewable energy sources, such as hydropower, wind, solar, and biomass. In addition, Maluku has substantial biomass energy resources, with a potential production forest area of 13,944.16 km² [4,8]. Some of the benefits of biomass energy include empowering the surrounding population to reduce poverty, reducing emissions using carbon-neutral technology, restoring underutilized land, and turning the land into a water retention area [9]. Based on these considerations, biomass-based power plants are a convincing option.

Seram Island has the greatest amount of biomass potential. However, the load center is on the island of Ambon (next to Seram). If the power plants are built near an energy source, as is the case with the resource-based approach, then a transmission system is required to transport energy from the location of generation to the location of the load center. Hence, transmission expansion planning is needed to determine where and when a new transmission line should be built to meet load demand over a given time horizon [10]. A transmission interconnection option for the neighboring system can be considered to optimize the utilization of local energy resources. In addition, the interconnection system’s reliability and security should be evaluated using power system analysis.

In evaluating the criteria for system security, there are several technical parameters to consider. Depending on the characteristics of the system, the evaluation criteria can be determined based on the technical parameters. The common technical parameters used to evaluate the system are presented in Table 2. To evaluate them, static and dynamic analyses are required. The technical parameters based on static analysis include component loading, voltage profile, short-circuit currents, N-1 contingency, voltage stability, and small-signal stability [11,12]. On the other hand, the dynamic analysis includes dynamic voltage stability, frequency stability, and rotor angle stability, and is based on a time-domain simulation [12].

Many previous studies have been conducted to evaluate the security criteria of a power system. However, not every parameter is considered depending on the characteristics of the power system. Some researchers evaluate the power system plan according to the high penetration of renewable energy, such as solar [13,14], wind [15], or even both [16,17]. Those systems have intermittent characteristics, so general static and dynamic analyses are required. The intermittency of wind and solar power becomes an important aspect in that case. For that purpose, the prediction of intermittency is necessary to predict the wind and solar output as presented in [18,19]. Another study focuses on how major disturbances can affect the power system’s security and reliability as presented in [20], which focuses on dynamic analysis and system contingency.

The evaluation of long-term power system planning should consider some time horizons, and consists of some topology changes in the power system as presented in [21,22]. For that purpose, static and dynamic analyses should be performed to determine the power system’s component rating. HVAC and a hybrid HVAC–HVDC interconnection were decided on in [21], while two voltage levels for the proposed transmission system were selected in [22]. The power electronic technology, such as many inverter topologies for the energy conversion process, also plays an important role in the planning of the power system with a high penetration of renewable energy as presented in [23]. In some cases, the energy storage system technologies might be applied to increase the penetration of renewable energy, as presented in [24,25].

Table 2. The state-of-the-art technical parameters that are considered in system security evaluation.

Reference	Technical Parameters
	Component Loading	Voltage Profile	Short-Circuit Current	N-1 Contingency	Voltage Stability	Small-Signal Stability	Dynamic Voltage Stability	Frequency Stability	Rotor Angle Stability
[13]	✓	✓	-	-	✓	-	✓	✓	✓
[14]	-	✓	-	-	-	-	✓	-	-
[15]	✓	✓	-	-	✓	-	-	-	-
[16]	✓	✓	-	-	✓	-	✓	✓	✓
[17]	-	✓	-	-	-	-	✓	✓	✓
[21]	✓	✓	-	✓	-	-	-	-	-
[22]	✓	✓	✓	✓	✓	✓	-	-	-
[26]	-	-	✓	-	-	-	✓	-	-
[27]	-	✓	-	-	-	-		-	-
[28]	✓	✓	✓	✓	-	-	✓	✓	✓
[29]	-	-	-	-	-	✓	-	-	-

Table 2. The state-of-the-art technical parameters that are considered in system security evaluation.

Reference	Technical Parameters
	Component Loading	Voltage Profile	Short-Circuit Current	N-1 Contingency	Voltage Stability	Small-Signal Stability	Dynamic Voltage Stability	Frequency Stability	Rotor Angle Stability
[13]	✓	✓	-	-	✓	-	✓	✓	✓
[14]	-	✓	-	-	-	-	✓	-	-
[15]	✓	✓	-	-	✓	-	-	-	-
[16]	✓	✓	-	-	✓	-	✓	✓	✓
[17]	-	✓	-	-	-	-	✓	✓	✓
[21]	✓	✓	-	✓	-	-	-	-	-
[22]	✓	✓	✓	✓	✓	✓	-	-	-
[26]	-	-	✓	-	-	-	✓	-	-
[27]	-	✓	-	-	-	-		-	-
[28]	✓	✓	✓	✓	-	-	✓	✓	✓
[29]	-	-	-	-	-	✓	-	-	-

In a simple radial distribution system with distributed generation, it is necessary to conduct a short-circuit analysis and a dynamic voltage analysis, as presented in [26]. In the oil industry’s electricity network, which is also at the medium voltage level, the main analysis focuses on the voltage profile [27]. In this power system, an automatic tap changer and reactive power compensator are used to compensate for the system’s voltage profile. The planning of a power system in 2030 with high renewable energy penetration dominated by wind and solar generation was simulated in Korea [28]. The HVDC, FACTS, and TCSC are invested in to increase the system’s penetration level. In this proposed power system plan, static and dynamic analysis simulations are performed. The specific analysis of one aspect of power system planning can also be performed as presented in [29], which focuses on the small-signal stability analysis of wind-penetrated power systems.

The proposed Maluku power system plan was obtained from our previous work on the design of the Eastern Indonesia Masterplan, as presented in [4], and the generation and expansion planning of the Ambon-SHS system, including the decision to use interconnection, as presented in [8]. These papers show that the most economical plan will be achieved if the Ambon-SHS are connected through a submarine interconnection system in 2025. In supporting the proposed system planning process, a technical assessment based on power system analysis is required. The main contribution of this paper is to assess the proposed power system obtained from the long-term system planning of the Maluku system. The assessment considers the security of AC power flow, N-1 contingency, short circuits, voltage stability, frequency stability, and transient stability analysis based on applicable standards. The technical evaluation of a specific power system in the Ambon-SHS is this paper’s novel and original contribution. The assessment will determine whether the interconnection system should be built.

The structure of this paper is as follows: the existing system, system planning, and energy potential of Maluku region and the methods used to assess the power system’s planning are described in Section 2. Then, the data used and the simulation parameters will be described in Section 3. The result of the assessment is elaborated in Section 4. Finally, Section 5 will summarize the conclusions and implications of the research results.

2. Materials and Methods

2.1. Maluku Power System

The Maluku power system is made up of four different islands: Ambon, Seram, Haruku, and Saparua (Ambon-SHS). The existing electrical power system’s peak load (PL) demand, and generation installed capacity (IC) are shown in Figure 2. Most of the demand in Maluku is concentrated on the west side. The east side only has small isolated systems separated by great distances. Moreover, the east part of Seram Island is dominated by the forest. As a result, the western part of Maluku is more likely to be interconnected.

In Seram, there are four subsystems, including Taniwel, Piru, Kairatu, and Masohi. Based on the peak load demand, Ambon and Seram are classified as large systems, while Haruku and Saparua are classified as small systems. Ambon has the largest demand, but the area is relatively small. In addition, Ambon is the most populated island among the four islands because the capital of Maluku Province, Ambon City, is located there [30].

The current power system is still being developed to establish a transmission and distribution network for each subsystem. The existing network is only a medium-voltage distribution network in Ambon, which connects the Passo and Sirimau substations, along the 12.3 km circuit [5]. However, from 2019 to 2028, the government will establish a plan to construct 150 kV networks both in Ambon and Seram islands and upgrade the existing 70 kV line to 150 kV.

Maluku Island has abundant renewable energy potential. Some of the energy sources that can be found include hydro, wind, solar, and biomass. The potential for each energy source is shown in

Table 3. Renewable energy source potential in Maluku [4,8].

Renewable Energy Source	Amount of Potential (MW)
Hydro	109.6
Wind	404.8
Solar	496,587.9
Biomass	18,385.3

. The biomass potential is huge in eastern Indonesia since there is so much vegetation in the forest. More importantly, biomass energy is more controllable compared to hydro, wind, and solar energy, which have an intermittent nature. In addition, the reasonable generation cost makes biomass generating units a strong candidate for power system planning in the Maluku power system.

The biomass energy potential is mainly located on Seram Island. Figure 3 shows the map of the potential production forest area that can be utilized as a biomass supplier. The total potential area is 96,192.14 km² which is equal to 18,385.3 MW. According to the previous study [31], five types of biomass have the potential to be used in the Maluku region, one of which is Eucalyptus pellita. Because most of the electricity demand is located on Ambon Island, it is necessary to build inter-island transmission lines to maximize power transfer and the utilization of biomass potential. This strategy is used to achieve the best economic results while still adhering to technical requirements.

2.2. Transmission Expansion Planning Assessment Method

Transmission plays a role in distributing electrical energy from generation to load. Therefore, transmission expansion planning (TEP) is strongly influenced by the location of the generation. GEP is carried out to determine which power plant should be built, how much capacity it should have, and when it should be built [32]. On the other hand, the results of TEP show when the transmission candidate will be built. TEP optimization can be carried out simultaneously with GEP to produce the most economical total cost. The GEP and TEP of the Ambon-SHS system have been performed in previous studies [8]. This study will complete the technical analysis of this previous planning by performing the power system analysis to assess the technical security.

Transmission assessment includes power system analysis to determine the technical feasibility of the plan for the transmission line. It is also intended to anticipate several issues that may arise during the expansion of the generation system. The workflow of the assessment is shown in Figure 4. The data needed to assess the power system planning include the existing transmission system, the load forecasting result, the additional generating unit, and the proposed transmission line candidates, which are provided by the previous study [4,8]. After all of the data are collected, the power system is then modeled for a particular year in the commercial power system simulation package DIgSILENT PowerFactory [33], the user interface of which is presented in Figure 5. The proposed power system for the different years was built in the single line diagram for each respective year. Lines, generators, and load parameters were also adjusted, including the dynamic parameter. After the modeling is complete, simulations are carried out to assess the obtained model. The simulation consists of load flow analysis, N-1 load flow analysis, a short circuit calculation, a voltage stability analysis, frequency stability, and transient stability.

To validate transmission line candidates, several power system analyses are conducted. The power system analyses performed in this study can be divided into two studies, namely static and dynamic analyses. Static and dynamic analyses are both based on the same initial system operating conditions. The simulation is carried out during the annual peak load conditions to represent the most stressed operating conditions.

The static analyses include power flow, an N-1 contingency analysis, short circuit, and voltage stability. At the same time, the dynamic simulation carried out includes frequency stability and transient stability. If there are criteria that are not met, system modification is required. However, if nothing is violated, the simulation will be continued in the following years. Simulations are carried out for the years 2020 to 2050, with each step taking five years.

The power flow analysis is performed to determine the transmission lines as well as the transformer loading and voltage profile of the substation. At the same time, the contingency analysis is a power flow analysis performed when one of the transmission lines is faulted. The power flow analysis is carried out using the Newton–Raphson method. The active and reactive power injections on bus i are shown in Equations (1) and (2), respectively [11]. The power flow and contingency analysis results are then compared to the grid code standards, as shown in Table 4. The maximum allowable power flow loading under normal conditions is 80%. At the same time, the loading under the N-1 contingency condition must not exceed 100%. Furthermore, the permissible voltage values for both conditions are in the −10% to +5% range.

$$P_i={\displaystyle {{\displaystyle \sum }}_{j=1}^N}\left|V_i\right|\left|V_j\right|\left|Y_{ij}\right|\cos \left({\theta }_{ij}+{\delta }_j-{\delta }_i\right)$$

(1)

$$Q_i=-{\displaystyle {{\displaystyle \sum }}_{j=1}^N}\left|V_i\right|\left|V_j\right|\left|Y_{ij}\right|\sin \left({\theta }_{ij}+{\delta }_j-{\delta }_i\right)$$

(2)

where,

$P_i$: net active power injection at bus $i$

$Q_i$: net reactive power injection at bus $i$

$V_i,V_j$: voltage magnitude at bus $i,j$

${\delta }_i,{\delta }_j$: voltage angle at bus $i,j$

$\left|Y_{ij}\right|$: the magnitude of the Y-bus element between bus $i$ and $j$

${\theta }_{ij}$: the angle of the Y-bus element between bus $i$ and $j$

Table 4. Security criteria for the thermal loading and voltage [34,35].

Criteria	Allowable Value
	Normal Operation	N-1 Contingency
Voltage profile	≥0.9 p.u. and ≤1.05 p.u.	≥0.9 p.u. and ≤1.05 p.u.
Voltage angle difference between the substations	≤30°	≤30°
Loading of the transformer and transmission line	≤80%	≤100%

Table 4. Security criteria for the thermal loading and voltage [34,35].

Criteria	Allowable Value
	Normal Operation	N-1 Contingency
Voltage profile	≥0.9 p.u. and ≤1.05 p.u.	≥0.9 p.u. and ≤1.05 p.u.
Voltage angle difference between the substations	≤30°	≤30°
Loading of the transformer and transmission line	≤80%	≤100%

The short-circuit current is the flow of excess current caused by disturbances in system components. Short-circuit analysis is performed to determine the magnitude of the short-circuit current at each bus if a component, i.e., a transmission line, is faulted. Based on the standard, the short-circuit current has a certain limit at a certain voltage level. The maximum allowable short-circuit current for transmission systems with a voltage rating of 70 kV is 20 kA. At the same time, the maximum value for 150 kV is 30 kA, as shown in Table 5 [36]. The short-circuit current that is measured is the initial short-circuit current ($I_k^{″}$) in three-phase short-circuit conditions. The calculations for the initial short-circuit current and the short-circuit impedance for a symmetrical fault are shown in Equations (3) and (4), respectively [37]. For a rated voltage ($U_n$) of 1-550 kV, the voltage factor ($c$) used is 1.1 [37].

$$I_k^{″}=\frac{c·U_n}{\sqrt{3}\left|Z_k\right|}$$

(3)

$$Z_k=\sqrt{R_k{}^2+X_k{}^2}$$

(4)

where,

$I_k^{″}$: initial short-circuit current

$c$: voltage factor

$U_n$: rated voltage

$Z_k$: short-circuit impedance

$R_k$: total series resistance of one phase

$X_k$: total series reactance of one phase

Table 5. Short-circuit current limit at each voltage level [36].

Voltage Level (kV)	Short-Circuit Current (kA)
70	≤20
150	≤30

Table 5. Short-circuit current limit at each voltage level [36].

Voltage Level (kV)	Short-Circuit Current (kA)
70	≤20
150	≤30

The voltage stability is defined as the ability of the system to maintain a steady-state voltage value on all buses after a disturbance occurs or if there is an increase in the load [38]. The voltage stability can be evaluated from the relationship between the bus voltage and active power load that is represented by the P–V curve, as shown in Figure 6. If the bus’s active power load is increasing, the bus’s voltage will decrease as a result of the increased voltage drop. At a certain point, if the load continues to increase, a voltage collapse will occur. This point is known as the critical point. The value of the active power load that causes the voltage to reach a critical point is denoted as $P_{crit}$. Based on [12], voltage stability is achieved if the load is less than 90% of $P_{crit}$.

The voltage stability simulation is performed by gradually increasing the load on the bus and then executing the load flow’s simulation. The simulation is repeated until the load flow’s solution cannot be obtained due to a nonconvergent solution, i.e., when the voltage collapse occurs. Observations on the voltage stability are carried out one by one on each substation, and then the resulting P–V curve is observed.

Frequency stability is defined as the ability of a power system to retain a constant frequency after a severe disturbance occurs that results in a serious imbalance between the generation and load [39]. The response of a generator to an imbalance between the generation and load is expressed in the swing equation shown in Equation (5) [12].

$$\frac{df}{dt}=\frac{f_0}{2H}(P_m-P_e)$$

(5)

where,

f: system frequency

f₀: factor initial frequency

H: system inertia

P_m: total mechanical power

P_e: total electrical power

Frequency stability analysis is carried out to determine the lowest point of frequency (nadir) when a disturbance occurs in the system. Based on the Indonesian grid code [39], the frequency nadir is allowed a deviation of ± 0.5 Hz from the nominal value of 50 Hz. If the frequency drops below 49.5 Hz and the available generation reserves are not sufficient to return the frequency to normal, manual load shedding (LS) will be carried out. Furthermore, if the frequency drops below 49 Hz, the under-frequency load shedding (UFLS) relay will be triggered. There is a manual load shedding stage and seven automatic load shedding stages in the defense scheme’s mechanism against the frequency drop, as shown in Figure 7. If the frequency continues to fall under the last load shedding stage, the islanding operation will occur.

The transient stability is one branch of rotor angle stability, namely, the ability of the synchronous generator to remain synchronized with the system after a disturbance occurs [12]. When a disturbance occurs in a power system, the generator rotor angle will increase. If the rotor angle continues to rise and reaches a critical angle, then the generator will experience a loss of synchronism. Critical clearing time (CCT) is the maximum duration for the generator to remain synchronized after a disturbance [40]. To prevent the loss of synchronism, the disturbance must be cleared before the CCT. The CCT is calculated based on the generator’s critical angle (${\delta }_c$), as shown in Equation (6) [11].

$$t_{cct}=\sqrt{\frac{2H\left({\delta }_c-{\delta }_0\right)}{\pi f_0{P}_m}}$$

(6)

where,

t_cct: system frequency’s critical clearing time

H: system inertia

δ_c: total mechanical power’s critical rotor angle

δ₀: total electrical power’s initial rotor angle

f₀: initial frequency

P_m: total mechanical power

A circuit breaker (CB) operates to clear faults with a certain duration. Based on [39], the fault clearing time of a CB for each voltage level is limited, as shown in

Table 6. Fault clearing time limit at each voltage level [39].

Voltage Level (kV)	Critical Clearing Time (ms)
66	150
150	120
275	100

. If the CB clearing time is shorter than the CCT, the generator will not experience a loss of synchronism. In other words, the longer the CCT, the better. In this analysis, the CCT is observed at all transmission lines and the fault location is specified in the middle of the line.

3. Simulation’s Set-Up

A typical electrical power planning system is made up of three parts: load forecasting, generation expansion planning, and transmission expansion planning. Each step is completed in the proper sequence. Transmission expansion planning, load forecasting, and generation expansion planning were already carried out in the previous step. Peak load and energy demand during the planning period are included in load forecasting data. These data have an impact on the transmission line’s ability to transfer electrical power from the generation to the demand side. When a transmission line becomes overloaded, it is necessary to expand the transmission by adding new lines. At the same time, generation expansion planning entails the installation and retirement of power plants during the planning period. These data are required to determine whether additional lines for connecting new power plants to the electrical power system are required.

3.1. Load Forecasting Projection

The results of load forecasting are very influential on the results of system planning. If the results obtained are too low, there will be an overinvestment. Moreover, if the result is too low, there will be reliability issues due to the insufficient system capacity [41]. For this reason, a proven forecasting method is needed. The load forecasting result that is used in this study refers to a study performed in [42]. The load forecasting method employs multivariate regression analysis that incorporates various variables. The model is widely used by electrical power companies around the world, with the hope that the resulting output will be more precise and accountable. Figure 8 depicts the load forecasting, with an average yearly peak load growth of 4.78%.

3.2. Generating Unit Candidates

Several types of power plant that are considered as options include gas-machine, steam turbine, geothermal, biomass, hydro, wind turbine, and solar.

Table 7. The total capacity of each generating unit type in MW [4,8].

Year	Total Capacity of Power Plant (MW)
	Gas-Machine	Steam	Geothermal	Biomass	Hydro	Wind	Solar
2020	129	0	0	0	0	0	0
2025	125	50	7	6	18.6	0	0
2030	125	80	7	6	72.6	0	0
2035	125	80	7	6	72.6	0	0
2040	125	80	7	6	72.6	37.7	3
2045	122.5	80	7	49	72.6	50.6	21.4

and

Table 8. The total energy production of each generating unit type in GWh [4,8].

Year	Total Energy of Power Plant (GWh)
	Gas-Machine	Steam	Geothermal	Biomass	Hydro	Wind	Solar
2020	439.9	0.0	0.0	0.0	0.0	0.0	0.0
2025	241.0	306.6	36.8	36.8	105.9	0.0	0.0
2030	117.3	490.5	35.7	36.8	247.8	0.0	0.0
2035	282.0	491.2	36.8	36.9	272.3	0.0	0.0
2040	339.0	515.7	39.1	39.2	319.5	66.7	3.2
2045	327.0	502.3	37.8	312.5	301.9	58.0	18.3

display the installed capacity of the generating unit facilities, i.e., the total capacity and energy generation, of the Maluku power system. Based on the proposed planning result, hydropower and biomass are projected to have a substantial contribution to the overall energy generation.

3.3. Transmission Line and Interconnection Candidates

The construction of the transmission line between subsystems follows two plans, namely the plan based on Indonesia’s 2019–2028 Electricity Procurement Plan [5] and the power system planning obtained in previous studies [4,8]. Figure 9 shows the transmission line candidates for the Maluku system from 2020 to 2050 and the single line diagram is shown in Figure 10. The total length of the transmission line addition is 680 circuit km, with a single voltage level of 150 kV, as detailed in

Table 9. Transmission line candidates [4,5,8].

From Bus	To Bus	Year Built	Length (Circuit km)
Piru	Kairatu	2020	110
Masohi	Kairatu	2020	210
Piru	Taniwel	2022	60
Passo	Sirimau	2025	12
Waai	Haruku	2025	30
Haruku	Saparua	2025	48
Kairatu	Saparua	2025	40

. Transmission line candidates that will connect the islands of Seram, Saparua, Haruku, and Ambon are proposed to transfer the biomass energy in Seram to the neighboring islands, the generating unit facilities and peak loads of which are presented in

Table 10. The installed generating unit capacity of the Ambon and SHS power system [4,8].

Year	Installed Capacity (MW)
	Waai	Passo	Sirimau	Wayame	Haruku	Saparua	Kairatu	Piru	Taniwel	Masohi
2020	60	27	0	0	0.3	3	0	4.5	0	0
2025	10	27	0	3	1.3	5.3	0	19	0	0
2030	15	27	0	3	1.3	5.5	0	19	1.5	0
2035	15	27	0	3	1.3	5.5	0	19	1.5	0
2040	15	27	74	3	4.3	26.4	0	30	1.5	0
2045	55.4	439.1	187	153	20.3	33.9	140	25.5	5.4	95.2

and

Table 11. The peak load of the Ambon and SHS power system [4,8].

Year	Installed Capacity (MW)
	Waai	Passo	Sirimau	Wayame	Haruku	Saparua	Kairatu	Piru	Taniwel	Masohi
2020	0	22.1	34.3	13.7	0	3.1	4.1	3.1	0	7.5
2025	0	29.4	44.2	21.5	1.6	4.4	5.7	4.3	1.3	10.4
2030	0	36.9	54.2	30.5	2.1	6	8.3	5.6	1.7	13.4
2035	0	44.6	65.3	40.7	2.5	8.1	12.9	7.1	2.1	15.8
2040	0	52.9	77.1	53.1	2.9	10.8	19.1	8.9	2.4	18.2
2045	0	61.7	85	67.8	3.1	14	27.3	11	2.7	20.6

, respectively. The optimal planning will result in the inter-area interconnection being built by 2025.

4. Results and Discussion

4.1. Effect of the Interconnection

The construction of inter-island transmission lines changes the pattern of power flow and generator loading. Figure 11a shows the total generation and peak load demand of the Ambon system. In this case, Seram, Haruku, and Saparua (SHS) are assumed to be in a group, and their load and generation are shown in Figure 11b. In both groups of regions, the load demand always increases over the planning period. However, different things happen in the power generation trend of the two regions. In Ambon, the generation began to decrease in 2040 and only increased slightly in the following years. The situation in which the generation is smaller than the load indicates that the system is importing power from another system. On the other hand, the generation in SHS continues to increase and get higher than the load demand. This shows that the power generated in SHS is not entirely absorbed by the load in that region, but is exported to another area, namely Ambon.

The power transfer between the regions of Ambon and SHS is possible because submarine lines are connecting them. The construction of submarine interconnection between Ambon and SHS altered the power flow between the systems, as illustrated in Figure 12. After the interconnection is built, the new power flow starts from Kairatu to Saparua, Haruku, and ends in Ambon. The power flow generally flows from Seram to Ambon, where the load center is located.

Figure 13 shows the magnitude of the power transfer between the regions of SHS and Ambon. A positive value indicates the power flow from SHS to Ambon. Because the transmission line is expected to be completed in 2025, the power flow between regions will begin in that year. A significant rise will begin to occur in 2035 and it will continue to experience a similar increase until 2050. This is due to the rapidly increasing number of generations in the SHS region, especially renewable energy sourced from biomass, hydropower, wind, and the sun.

Figure 14 shows the overall energy mix in the Ambon-SHS interconnection system. For 2025, the energy mix is shown in Figure 14a. The additional power plants are hydro, coal, biomass, and geothermal power plants. The contribution of SHS areas is 260 GWh, at 33%. In 2035, as seen in Figure 14b, the SHS areas will have begun to contribute more to the overall energy mix, with a total contribution of 386.4 GWh, at 39%. In the final year of planning, the total generation of the system will be dominated by generations from the SHS region, as shown in Figure 14c. The generation system in SHS is responsible for 1055.1 GWh, at 57% of the total generation. Biomass-powered generation dominates the generation from the SHS region, with a contribution of 26% to the overall energy mix. This shows that the construction of an inter-area transmission system can maximize the potential of biomass energy in the SHS region by transmitting the energy to Ambon.

4.2. Evaluation of The Proposed Power System Planning

4.2.1. Power Flow Analysis

Figure 15 depicts the voltage profile of the Ambon-SHS interconnection system’s 150 kV substation during the planning period from 2020 to 2050 with a 5-year interval. The voltages in all substations are still within the permissible range, ranging from 0.9 to 1.05 p.u. The voltage on the new substation in Sirimau appears in 2045 and 2050 because the substation is projected to start operating in 2045. At the end of the planning year, the Wayame substation is expected to have the lowest voltage, which is at 0.953 p.u. This is because the Wayame substation is one of the furthest substations. At the same time, the Taniwel substation is predicted to have the highest voltage magnitude, at 1.01 p.u. A slightly over-voltage on the Taniwel substation is caused by the line capacitance connecting Piru and Taniwel, as well as the reactor’s configuration.

Figure 16 depicts the loading of several transmission lines with high loading occurring throughout the planning period. All line loadings remain within safe limits, at less than 80% under normal conditions. At the end of the planning year, the Waai–Passo line is expected to have the highest line loading, at 58.86%. This is because most of the power plants on Ambon Island are located in the Waai area. The power flow resulted in 3.14 MW of losses, or approximately 0.88% of the total generation.

Figure 17 depicts the difference in the voltage angle between two substations connected by the aforementioned lines. The difference in the voltage angle is still within safe limits, which are less than or equal to 30°. The largest voltage difference at the end of the planning year is 0.53°, which is located on the Waai–Passo line.

4.2.2. Contingency Analysis

Contingency simulation is used to determine the impact of an outage of a single component on the substation voltage and line loading. This simulation’s contingency event is a trip on the most heavily loaded line. Figure 18 depicts the voltage profile on each substation following a contingency. Throughout the planning period, the overall magnitude of the substation’s voltage remains within permissible limits.

Figure 18 and Figure 19 show the contingency simulation results for the line loading and the voltage difference, respectively. In 2045 and 2050, the line loading of Waai–Passo is not recorded because Waai–Passo is the faulted line. This happens as it has the highest loading during those years. Even though all the line loadings are increasing, they are still below the standard limit of 100%. Furthermore, the difference in voltage between the substations remains below the specified standard limit of 30°, as presented in Figure 20.

4.2.3. Short Circuit Analysis

Figure 21 depicts the magnitude of the symmetrical short-circuit current at the Ambon-SHS interconnection system substation. It can be seen from all of the short-circuit current values that the overall short-circuit current value is still less than the allowable limit of 30 kA for a voltage level of 150 kV. As a result, the existing power system network can withstand short-circuit events. The largest short-circuit current at the end of the planning year is 6.35 kA, which is located at the Saparua substation.

4.2.4. Voltage Stability Analysis

Figure 22 shows the relative value of the load demand to the critical point ($P_{crit}$) at each substation. A load demand above 90% of the $P_{crit}$ is considered to cause voltage instability. From the simulation, it is found that during the planning period, voltage stability is obtained in the substation. This is indicated by the load demands that are still below 90% of $P_{crit}$. The highest value is recorded on the Sirimau bus in 2045. Figure 23 shows the P–V curve of the Sirimau substation in that year. During that period, the load on the Sirimau bus is 87 MW, while the $P_{crit}$ was recorded at 182.5 MW. In other words, the existing operating point is located at 48% of the $P_{crit}$, meaning that the voltage on the bus is still stable.

4.2.5. Frequency Stability Analysis

The frequency nadirs resulting from the worst contingency in each year are shown in Figure 24. It can be seen that, during the planning period, the frequency nadirs are above the limit of manual LS at 49.5 Hz. This means that the system will not lose any load demand due to load shedding, even though a major disturbance occurs. For example, in 2050, the frequency nadir is at 49.69 Hz due to the outage of a coal power plant in Ambon, as shown in Figure 25.

4.2.6. Transient Stability Analysis

The CCT that is evaluated in the analysis is the most critical, i.e., the shortest. The shortest CCT in each simulation year is shown in Figure 26. It can be seen that all of the CCTs are above the CB clearing time standard for 150 kV, which is 120 ms. The fault locations that caused the lowest CCT for each year are MVPP to Waai (2020), Gas-PP AP to Incomer Waai–Passo (2025–2035), PV Haruku to Haruku (2040 and 2050), and Hydro PP Wai Tala to Incomer Masohi–Kairatu (2045). The simulation results show that all the generators in the system can remain in sync despite disturbances.

4.2.7. Assessment Summary

Table 12. Summary of the power system’s security evaluation results.

Analysis	Parameter	Planning Period
		2020	2025	2030	2035	2040	2045	2050
Load flow under normal conditions	Voltage profile	Meets the limit of 0.9–1.05 p.u.
	Difference in the voltage angle between substations	Less than 30°
	Loading of the transformer and transmission line	Less than 80%
Load flow under a Contingency of the N-1 line	Voltage profile	Meets the limit of 0.9–1.05 p.u.
	Difference in the voltage angle between the substations	Less than 30°
	Loading of the transformer and transmission line	Less than 100%
Short circuit	Symmetrical short-circuit current	Less than 30 kA
Voltage stability	Operating point	Load demand of less than 90% of $P_{crit}$
Frequency stability	Frequency nadir	Above 49.5 Hz
Transient stability	CCT	Longer than 120 ms

depicts a recapitulation of the simulation results. The simulation was conducted every 5 years during the planning period under certain operating conditions, especially during the peak load condition. To assess the system’s security, the power flow analysis, short-circuit analysis, and dynamic analysis were simulated. The power flow analysis demonstrates that the system remains in a normal state of voltage profile and component loading, both under normal conditions and when one of the lines is faulted. This condition is important for providing the power system’s component sizing when the power system would expand. In addition, the short-circuit current caused by the fault is also within the limits specified, i.e., 30 kA. The short-circuit analysis would more likely correlate to the system’s protection.

For the dynamic analysis, the voltage, frequency, and transient stability were simulated. The voltage stability analysis reveals the system loading margin during certain operating conditions. The results of the voltage stability analysis show that the system’s substations were stable throughout the planning period and still have a significant margin for additional demand. The frequency stability was assessed by deteriorating the system with the active power loss in the system. For that purpose, the generator outage was simulated. In the case of the most severe generator outage, the system can still respond well by keeping the frequency nadir above 49.5 Hz. This means that other generators can respond to the generation imbalance rapidly. Finally, the great disturbance was simulated to assess the CCT of each generator. The CCT of all generators remains above 120 ms, indicating that the CB has enough time to clear the disturbances so that all generators do not experience a loss of synchronism. Both static and dynamic simulation results demonstrate that all the parameters assessed meet the allowable limits, indicating that the development design for the proposed power system is technically feasible.

5. Conclusions

A power system assessment of the proposed power system in Maluku Island was conducted in this research. The proposed power system would interconnect the Ambon-SHS system, which currently operates as an isolated system. The power system consists of power flow analysis, N-1 contingency, short-circuit currents, frequency stability, and transient stability. The assessment was executed under different operating conditions within the planning horizon period. From the simulation result, the proposed power system plan is technically safe and capable of meeting electricity demands until 2050.

The proposed power system plan would optimize the utilization of biomass energy on Seram Island and transmit the electric power to the load center on Ambon Island. Under this scheme, the generation from biomass energy is projected to reach 33% of the total generation, exceeding the coal energy mix of 27%. This target is achievable because the power generated in the SHS area can be transmitted to Ambon Island, at which point the demand will increase. In 2050, power transfer from SHS to Ambon is expected to reach 89.8 MW, which is projected to serve 35% of Ambon’s peak demand. Because the plan for the power system can meet the economic and technical criteria, it can be realized in the Maluku power systems.