Email Alert
Abstract
This paper addresses an optimal design of low-voltage (LV) distribution network for rural electrification considering photovoltaic (PV) and battery energy storage (BES). It aims at searching for an optimal topology of an LV distribution system as well as the siting and sizing of PV and storage over a time horizon of 30 years. Firstly, the shortest-path algorithm (SPA) and first-fit bin-packing algorithm (FFBPA) are used to search for the optimal radial topology that minimizes the total length of the distribution line and improves the load balancing. Then, the optimal siting of decentralized BES (DeBES) is determined using a genetic algorithm (GA) to eliminate the under-voltage constraints due to the load consumption. Two iterative techniques are elaborated to size the maximum peak power of PV and the minimum number of DeBES that can be connected to an LV network without violating the voltage and current constraints. Then, the sizing strategy of centralized BES (CeBES) is developed to avoid reverse power flows into the medium-voltage (MV) network. Finally, a Monte Carlo approach is used to study the impact of load profile uncertainties on the topology. A non-electrified village in Cambodia has been chosen as a case study.
Keywords
Battery energy storage (BES); low-voltage (LV) distribution network; Monte Carlo; photovoltaic (PV); electrification; planning
MORE than 80% of people are living in rural villages in Cambodia, and about 75% of the villages are currently non-electrified [
The LV distribution network is evolving towards a microgrid, i.e., a small-size LV network which is autonomous during some moments of the year and can be disconnected depending on the need of the main grid. The microgrid development for the electrification of territories has already been addressed in the literature. Their designs mainly include the optimization of the system components and operation strategies. To deal with the problem of the topology design, [
Regarding research papers on LV networks, [
The design of microgrid was studied as isolated systems in some works. The hybrid system of diesel-PV battery with an associated control strategy was examined in [
Numerous studies were conducted for the design of microgrids connected to the main grid by integrating PV and storage [
It is mandatory to study an alternative planning methodology with small-scale production integrated into the LV distribution networks compared to the traditional planning based on technical and economic considerations. This paper proposes the design and control of an LV distribution network integrating PV panels and battery storage to challenge the current electrification issues in Cambodia. An optimal radial topology will be proposed by applying the shortest-path algorithm (SPA) with the first-fit bin-packing algorithm (FFBPA) in order to minimize the costs and optimize the phase balancing. Then PVs and batteries will be sized and controlled to solve over-voltage and under-voltage problems preventing the DSO from over-sizing the network. The penetration of PV production will be maximized to avoid over-voltage problems, and decentralized storage is employed to remove under-voltage problems. Centralized battery storage will be installed at the MV/LV transformer in order to eliminate the reverse power flows that may occur and lead to potential over-loading of the MV network. This design of LV distribution network could be isolated from the main grid at some moments of the year, enabling a reduction of the investments in the MV network such as MV/LV substation and power loss reductions.
The novelty and contributions of this paper are as follows:
1) Provide a planning tool of distribution system integrating PV and storage for DSO, for countries similar to Cambodia.
2) Give a general methodology in order to build a microgrid in non-electrified countries, from the building of the lines to the siting and sizing of PV and storage units.
3) Develop simple operation rules of LV distribution network without a huge deployment of ICTs which integrate the uncertainties into loads and productions.
The characteristics of the developed distribution system will be the followings:
1) Maximize the autonomy of LV network to make it less dependent on the MV network with PV and storage.
2) Decrease the sizing of MV/LV equipment to reduce the network costs, e.g., MV/LV substation and power losses.
3) Increase the local production based on renewable energies while satisfying current and voltage constraints.
The remainder of this paper is organized as follows. The general methodology of the LV distribution network planning including the explanation of the algorithms is detailed in Section II. Section III describes a case study of a real non-electrified village in Cambodia including the simulation results. Section IV describes the comparison between traditional and novel planning methods. Section V presents the main conclusions and future perspectives.
In industrialized countries, the MV network is quite robust with a fast backup process in case of a fault, which is not the case in developing countries. The MV network might experience a power outage, which potentially leads to frequent blackouts in LV networks even if their system components have been properly sized. Thus, we propose to design LV microgrids that rarely depend on MV networks. The proposed strategy is that the DSO invests in the PV and storage systems.
Fig.1 Stages of novel planning method.
1) Build the optimal radial topology design to ensure the load balancing and to minimize the total length of lines.
2) Find the maximum penetration of PV peak power without violating current or voltage constraints.
3) Optimally sit and size the decentralized BES (DeBES).
4) Optimally size the centralized BES (CeBES).
The LV network in Cambodia comprises of single-phase and 3-phase LV feeders for a 3-phase MV/LV transformer to several single-phase poles where LV customers are connected. In this subsection, an LV topology is planned for a non-electrified area by minimizing the total length of the distribution line and optimizing the load balance. To achieve this objective, the SPA and FFBPA algorithms are used. Based on graph theory, the problem of SPA is to search the path between two vertices in a graph in order to minimize the sum of the length of its segments. The problem of FFBPA is filling all pieces into a well-defined number of bins, while minimizing the total weight difference for each bin. In this paper, the pieces are load demand, i.e., active and reactive power, and phases of the power system correspond to bins. Firstly, the SPA is started to search for the nearest pole to which the load is connected, and then the FFBPA is applied to balance the power of the connected loads for each phase. Also, this method has been chosen because LV networks are short and simple (houses along roads), so classical algorithms from graph theory are well suited.
For this primary study in Cambodia, the renewable sources considered are PV, which have the highest potential in Cambodia. In the microgrid, the DSO is able to choose the location of the PV. The operation cost of a distribution system depends on power flows. The closer the generation and consumption are, the smaller the line losses will be. Hence, it is assumed that PV installation is located near high-consumption loads. The objective of this step is to determine the maximum peak power of PV that could be connected to the distribution system without violating the voltage and current constraints.
Fig.2 Flowchart of maximum power of PV.
At the beginning, we input the network data such as the locations of MV/LV transformer and consumer, the line parameters, and the daily PV/load curves. Next, PVs are located at the customer side whose peak power is the highest. For every hour of the day, the PV power is incremented until an over-voltage, i.e., Vmax=1.06 p.u., and/or an over-current appear. The highest penetration of PVs in the network is the smallest of the highest value, which is computed every hour.
The problem of under-voltage almost always appears at the farthest household without/with the PV. Actually, the highest consumption generally happens when the production of PV is low, i.e., non-synchronized PV-load profile. Thus, the case of under-voltage constraint is not different without or with PV. The DeBES is placed in the network to eliminate the under-voltage constraint. Nevertheless, the integrated DeBES at the farthest household is not always the best place, especially if a producer is connected here. There are several techniques, e.g., particle swarm optimization (PSO), differential evolution (DE), GA, suitable for the siting and sizing problem of DeBES. The GA has been selected in this work because the computation time is shorter [
| (t) | (1) |
where (t) is the power injection of the battery for phase at the bus at t.
The objective function (1) is subject to the following constraints, i.e., voltage regulation of [10%, 6%] [
| (2) |
| (3) |
where Iline is the line current.
Moreover, the fitness function for a given node p is defined as the sum of the objective and penalties of over-voltages, under-voltage and over-current. The value of the fitness function for a given node at t is expressed as:
| (4) |
where ; is the penalty coefficient applied if the current is beyond the allowable line capacity ; , and is the voltage of the node; is the penalty coefficient if the voltage does not respect the permitted range; and are the coefficients of voltage and current constraints, respectively.
The penalty coefficients are used to measure the constrained violation. With numerous simulations, the adjustment parameters of the algorithm have been well-defined as a compromise between convergence and simulation time.
Once the location is obtained, the minimum DeBES capacity connected to the LV distribution network has to be computed to solve the problem of under-voltage. Firstly, as for PV in
| (5) |
The maximum injected power by this battery has been calculated at 7 p.m. when the worst under-voltage constraint occurs. This reference voltage is estimated using a load flow at 7 p.m. considering the minimum discharging power of the battery, with which a constraint of under-voltage will occur in the system.
We propose a decentralized control strategy to manage the DeBES without expanding a large ICT network and devices. The DeBES will start to operate at 10 a.m. due to high PV generation and low consumption, as shown in
Fig.3 Normalized daily PV/load curves.
The CeBES is installed to store energy when reversed power goes from the LV network to the MV network, and to discharge when the network has to supply the LV consumers. It ensures that the SOC of the battery stays between SOCmin and SOCmax. The power is measured at the MV/LV transformer every hour. With a positive value, the battery storage will be discharged if the SOC of the battery storage remains inside the allowable boundaries. Otherwise, the battery storage will be charged. Such management aims to prioritize the use of the CeBES instead of the MV network.
The discount cost method is used in this paper to assess different proposed strategies over N planning years. Capital expenditure (CAPEX) and operation expenditure (OPEX) for energy losses of the LV network have been taken into account in this method. The discount cost [
| (6) |
where Cdiscountis the total discount cost; Cinvest (t) is the cost of investment for year t; Celect is the cost of electricity; ENloss (t) is the energy losses for year t; and i is the discount rate.
In addition, the DSO invests in PV/BES, so the investment of PV/BES has to be integrated into this discount cost. The reduction of the energy consumed due to the CeBES is also considered in the discount cost. Consequently, the energy losses in (6) have been replaced by the energy used by households and energy losses. Thus, the total discount cost of the system can be formulated as:
| (7) |
where ENuse(t) is the energy used for year t.
A non-electrified village in a rural area, located in Khum Sandek, Srok Batheay, Khet Kampong Cham, Cambodia, has been chosen as a case study in this paper, and PV is available in this area. There are 129 single-phase households located along the sides of the way. The consumers are powered by a 22 kV/0.4 kV substation. The active power Pactive of the network is about 43 kW with a power factor of 0.95 at the initial planning year. The maximum power of each consumer has been randomly generated following a normal distribution with a mean of 0.4 kW and a standard deviation of 0.05 kW because of no available information. For rural villages, typical conductor is 50 m
Some input parameters are listed in Table I. The planning study is 30 years, which is a typical time horizon. Over this duration, the uncertainties are too high to draw realistic conclusions. Three single-phase PV units per phase, i.e., 9 PV units in total, have been assumed because it represents about 10% of the total households of the network. It is the mean value of households equipped with a solar home system (SHS). Only one DeBES per phase has been considered to minimize the investment cost. Also, a high growing rate of 3% has been chosen since Cambodia is a developing country. Load profiles are built based on real measurements and considering uncertainties. Moreover, as real data of solar radiation are presently not available, the National Aeronautics and Space Administrations (NASA) data [
A normalized daily load curve is an average load profile taken from local measurements of three households. A PV curve taken from NASA is given in
We decide to place 3 PV units per phase where customers have the highest consumption to optimize the self-consumption. Typically, PV units are connected in single-phase up to 5 kW [
Furthermore, the parameters taken for the economic study are listed in Table II.
1) Optimal Radial Topology
An optimal radial topology of the LV network is obtained using the SPA/FFBPA algorithms. An optimal radial balanced topology is presented in
Fig.4 Optimal topology including integrated PV/DeBES/CeBES in Cambodia.
2) Maximum Peak Power of PV
The size of the nine PVs and the phase where they have been connected to the network are listed in Table III. Then, three PV inverters of 3.0 kW, 3.6 kW, and 4.0 kW are used with 90% [
3) Siting and Sizing of DeBES
By applying the algorithms described in Section II-C, the optimal siting of DeBES is found and the maximum power is set to the minimum power discharged at 7 p.m. to avoid the constraint of under-voltage. The parameters of the GA are as follows: the number of individuals is 150, and the number of generations is 300. This algorithm is integrated into a Monte Carlo simulation to estimate the average Vsetof each DeBES, as depicted in
Fig.5 Vset for DeBES of phase A at the end of planning period.
Also, the operation of the DeBES without the use of ICTs and only with a voltage sensor is shown in
Fig.6 Operation for DeBES of phase A at the end of planning year.
Thus, the size for each DeBES is 54.9 kWh, 48.8 kWh and 63 kWh, respectively. Their bi-directional inverters will be 18 kW, 18 kW and 20 kW with 90% efficiency [
4) Sizing of CeBES
According to the strategy described in Section II-D, the capacity of the CeBES is 162 kWh with a bi-directional inverter of 50 kW with 90% [
As a result, the network can be operated as an isolated microgrid with an autonomy of 91% for the initial year and 52% at the end of the planning studies due to the growth of load demand.
The traditional planning of developed countries aims to supply all the consumers at the lowest cost with high service quality. This method sizes the system so that it can supply the load at the end of the equipment life (about 30 years) to avoid multiple investments. And large section cables need to be used in the method.
Table IV provides the minimum voltage and maximum current in the network as a function of cables currently used in Cambodia at 7 p.m. when the peak load appears for the last planning year. It can be concluded that the section used in the network must be 150 m
As Vset for each DeBES has been statistically estimated, the values can be used as voltage references for the decentralized control considering the load curve uncertainty. Figures
Fig.7 Histogram of minimum voltage of network without DeBES for phase A at the end of planning year for 1000 samplings.
Fig.8 Histogram of minimum voltage of network with DeBES for phase A at the end of planning year for 1000 samplings.
Note that the problem of under-voltage is absolutely eliminated by connecting the DeBES according to a selection of Vset. Hence, it can be concluded that this decentralized control without ICT infrastructure is suitable to solve under-voltage problems.
In this subsection, over the planning period of 30 years, the cumulative discount costs of the two different planning methods are depicted in
Fig.9 Cumulative discount cost of two planning methods over 30 years.
The traditional planning solution is less expensive at the beginning because the CAPEX is lower than the novel planning solution. However, at year 12, the trend of the cumulative discount cost is reversed. The novel planning solution becomes more economical since the OPEX is always less expensive than the traditional planning solution. In the novel planning curve, we can see some discontinuities at years 5, 10, 15, 20 and 25 in comparison with the traditional planning curve. They are caused by additional investments for BES systems (every 5 years), inverters (every 15 years) and PV system (every 25 years). Thus, it can be concluded that the proposed solution is more economical than the traditional one.
The integration of PV/BES makes it possible to reduce the peak power flow coming from the MV network to supply loads at the end of the planning year from 116 kW to 94 kW, representing a reduction of about 19% as shown in Table V. Then, it is possible to size a smaller MV/LV transformer and reduce the section of the upstream MV distribution lines. The energy used per year is also reduced from 202 MWh to 146 MWh. Moreover, the autonomous operation of the system increases from 37 % to 52%, which corresponds to the situation when the consumption is equal to 0.
In this paper, a novel planning methodology for microgrid development has been studied for non-electrified areas of rural villages. The algorithm enables to find the optimal radial topology of LV distribution network as well as the siting and sizing of solar panels and batteries. The SPA and FFBPA have been applied to search for the optimal radial topology to ensure the load balancing. The maximum amount of installed PV has been determined considering the constraints of current and voltage. Besides, the siting and sizing of both decentralized and centralized BES systems have been carried out to resolve the problem of under-voltage and decrease the sizing of the MV/LV transformer. The integrated solar and battery storage into LV networks would be less expensive than that with traditional planning solutions for developing countries. Other benefits can be obtained such as the increasing autonomous operation of the network and the removal of reverse power flows. Additionally, the impact of the voltage reference values for the control of the decentralized BES has been investigated using Monte Carlo approach to consider the uncertainties of the load curve during the planning. A 5% slight variation in the case study does not have an undesirable impact on the system.
Further works could investigate the use of wind turbines as they have high potential in mountainous and coastal regions of Cambodia. A comparison of several optimization techniques by considering the indicators such as the computation time and the accuracy will also be studied. The efficiency of the algorithms could be assessed by comparing the results obtained from other results of commercial software such as DIgSILENT PowerFactory. Finally, the development of models to tackle the intermittent behavior of both load demand and distributed generation will be investigated.
REFERENCES
C. Phok, “Contribution to electric energy generation for isolated-rural areas using 2nd life components and renewable energies: modelling and control of an induction generator,” Ph.D. dissertation, Department of Electrical Engineering, University of Toulouse III, Toulouse, France, 2016. [百度学术]
REF-EDC, “Program for the development of rural electrification of Electricité du Cambodge (EDC) through the department of Rural Electrification Fund (REF),” Department of Rural Electrification Fund, Electricité du Cambodge, Phnom Penh, Cambodia, 2017. [百度学术]
Electricity Authority of Cambodia. (2018). Report on power sector of the Kingdom of Cambodia. [Online]. Available: https://www.eac.gov.kh/site/annualreport?lang=en [百度学术]
V. Vai, E. Gladkikh, M.-C. Alvarez-Herault et al., “Study of low-voltage distribution system with integration of PV-battery energy storage for urban area in developing country,” in Proceedings of IEEE International Conference on Environment and Electrical Engineering and IEEE Industrial and Commercial Power Systems Europe, Milan, Italy, Jun. 2017, pp. 1-6. [百度学术]
V. Vai, E. Gladkikh, M.-C. Alvarez-Herault et al., “Planning of low-voltage distribution systems with uncertainty on load demand in urban areas,” in Proceedings of IEEE International Conference on Environment and Electrical Engineering and IEEE Industrial and Commercial Power Systems Europe, Milan, Italy, Jun. 2017, pp. 1-6. [百度学术]
V. Vai, E. Gladkikh, M.-C. Alvarez-Herault et al., “Low-voltage distribution system planning under load demand uncertainty: growth rate with connection of new loads,” in Proceedings ofInternational Electrical Engineering Congress, Pattaya, Thailand, Mar. 2017, pp. 1-4. [百度学术]
V. Kumar, R. Krishan, and Y. Sood, “Optimization of radial distribution networks using path search algorithm,” International Journal of Electronics and Electrical Engineering, vol. 1, no. 3, pp. 182-187, Sept. 2013. [百度学术]
N. G. Boulaxis, S. Member, and M. Papadopoulos, “Optimal feeder routing in distribution system planning using dynamic programming technique and GIS facilities,” IEEE Transactions on Power Delivery, vol. 17, no. 1, pp. 242-247, Jan. 2002. [百度学术]
V. Gouin, M.-C. Alvarez-Herault, and B. Raison, “Optimal planning of urban distribution network considering its topology,” in Proceedings ofInternational Conference on Electricity Distribution, Lyon, France, Jun. 2015, pp. 1-5. [百度学术]
M. G. Echeverri, R. G. Rendón, and J. M. L. Lezama, “Optimal phase balancing planning for loss reduction in distribution systems using specialized genetic algorithm,” Ingenieria y Ciencia, vol. 8, no. 15, pp. 121-140, Jun. 2012. [百度学术]
J. Zhu, M. Y. Chow, and F. Zhang, “Phase balancing using mixed-integer programming,” IEEE Transactions on Power Systems, vol. 13, no. 4, pp. 1487-1492, Nov. 1998. [百度学术]
H. M. Khodr, I. J. Zerpa, P. M. D. Jesus et al., “Optimal phase balancing in distribution system using mixed-integer linear programming,” in Proceedings of IEEE/PES Transmission & Distribution Conference and Exposition: Latin America, Caracas, Venezuela, Aug. 2006, pp. 1-6. [百度学术]
J. Zhu, G. Bilbro, and M. Y. Chow, “Phase balancing using simulated annealing,” IEEE Transactions on Power Systems, vol. 14, no. 4, pp. 1508-1513, Nov. 1999. [百度学术]
L. N. An, T. Quoc-Tuan, B. Seddik et al., “Optimal design of an isolated photovoltaic-diesel-battery hybrid system by using an iterative algorithm,” in Proceedings of IEEE PES General Meeting, National Harbor, USA, Jul. 2014, pp. 1-6. [百度学术]
L. N. An, T. Quoc-Tuan, and B. Seddik, “Control strategies of a hybrid PV-diesel-battery system in different operation modes,” in Proceedings of IEEE PowerTech Conference, Grenoble, France, Jun. 2013, pp. 1-6. [百度学术]
A. Navarro-Espinosa and L. F. Ochoa, “Increasing the PV hosting capacity of LV networks: OLTC-fitted transformers vs. reinforcements,” in Proceedings of IEEE PES Innovative Smart Grid Technologies Conference, Washington DC, USA, Feb. 2015, pp. 1-5. [百度学术]
A. Gabash and P. Li, “Active-reactive optimal power flow for low-voltage networks with photovoltaic distributed generation,” in Proceedings of IEEE International Energy Conference, Florence, Italy, Sep. 2012, pp. 1-6. [百度学术]
B. Nasiri, A. Ahsan, D. M. Gonzalez et al., “Integration of smart grid technologies for voltage regulation in low voltage distribution grids,” in Proceedings of IEEE Innovative Smart Grid Technologies-Asia, Melbourne, Australia, Dec. 2016, pp. 954-959. [百度学术]
T. Degner, G. Arnold, T. Reimann et al., “Increasing the photovoltaic-system hosting capacity of low voltage distribution networks,” in Proceedings of International Conference on Electricity Distribution, Frankfurt, Germany, Jun. 2011, pp. 1-5. [百度学术]
A. A. E. M. Gado, “Impact the expansion of the production of generation of solar power on the low voltage network in Egypt,” in Proceedings of Saudi Arabia Smart Grid, Jeddah, Saudi Arabia, Dec. 2016, pp. 1-4. [百度学术]
S. Heslop, I. Macgill, J. Fletcher et al., “Method for determining a PV generation limit on low voltage feeders for evenly distributed PV and load,” Energy Procedia, vol. 57, pp. 207-216, Nov. 2014. [百度学术]
M. J. E. Alam, K. M. Muttaqi, and D. Sutanto, “Distributed energy storage for mitigation of voltage-rise impact caused by rooftop solar PV,” in Proceedigns of IEEE PES General Meeting, San Diego, USA, Jul. 2012, pp. 1-6. [百度学术]
M. J. E. Alam, K. M. Muttaqi, and D. Sutanto, “Mitigation of rooftop solar PV impacts and evening peak support by managing available capacity of distributed energy storage systems,” IEEE Transactions on Power Systems, vol. 28, no. 4, pp. 3874-3884, Nov. 2013. [百度学术]
F. Marra, G. Yang, C. Træholt et al., “A decentralized storage strategy for residential feeders with photovoltaics,” IEEE Transactions on Smart Grid, vol. 5, no. 2, pp. 974-981, Mar. 2014. [百度学术]
M. Zeraati, M. E. H. Golshan, and J. Guerrero, “Distributed control of battery energy storage systems for voltage regulation in distribution networks with high PV penetration,” IEEE Transactions on Smart Grid, vol. PP, no. 99, pp. 3582-3594, Jul. 2018. [百度学术]
I. Ranaweera and O. M. Midtgard, “Centralized control of energy storages for voltage support in low-voltage distribution grids,” in Proceedings of IEEE International Conference on Environment and Electrical Engineering, Florence, Italy, Jun. 2016, pp. 1-6. [百度学术]
Y. Wang, K. Tan, X. Peng et al., “Coordinated control of distributed energy storage systems for voltage regulation in distribution networks,” IEEE Transactions on Power Delivery, vol. 31, no. 1, pp. 1132-1141, Jun. 2016. [百度学术]
S. Hashemi and J. Ostergaard, “Methods and strategies for overvoltage prevention in low voltage distribution systems with PV,” IET Renewable Power Generation, vol. 11, no. 2, pp. 205-214, Apr. 2017. [百度学术]
S. Qi, L. Chen, H. Li et al., “Assessment of voltage control techniques for low voltage networks,” in Proceedings of IET International Conference on Developments in Power System Protection, Copenhagen, Denmark, Mar. 2014, pp. 1-6. [百度学术]
S. Hashemi, J. Ostergaard, T. Degner et al., “Efficient control of active transformers for increasing the PV hosting capacity of LV grids,” IEEE Transactions on Industrial Informatics, vol. 13, no. 1, pp. 270-277, Feb. 2017. [百度学术]
V. Vai, M.-C. Alvarez-Herault, B. Raison et al., “Study of low-voltage distribution system with decentralized PV-BES and centralized BES for rural village in developing country,” in Proceedings of International Electrical Engineering Congress, Pattaya, Thailand, Mar. 2017, pp. 1-4. [百度学术]
A. Ouarda and M. Bouamar, “A comparison of evolutionary algorithms: PSO, DE and GA for fuzzy C-Partition,” International Journal of Computer Applications, vol. 91, no. 10, pp. 32-38, Apr. 2014. [百度学术]
F. Jia and D. Lichti, “A comparison of simulated annealing, genetic algorithm and particle swarm optimization in optimal first-order design of indoor TLS networks,” in Proceedings ofISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, Wuhan, China, Sept. 2017, pp. 75-82. [百度学术]
Electricity Authority of Cambodia. (2004, Jul.). General Requirements of electric power technical standards of the Kingdom of Cambodia. [Online]. Available: https://eac.gov.kh/site/standards?lang=en [百度学术]
M. Mao, P. Jin, L. Chang et al., “Economic analysis and optimal design on microgrids with SS-PVs for industries,” IEEE Transactions on Sustainable Energy, vol. 5, no. 4, pp. 1328-1336, Oct. 2014. [百度学术]
M.-C. Alvarez-Herault, D. Picault, R. Caire etal., “A novel hybrid network architecture to increase DG insertion in electrical distribution systems,” IEEE Transactions on Power Systems, vol. 26, no. 2, pp. 905-914, May. 2011. [百度学术]
V. Vai, M.-C. Alvarez-Herault, L. Bun et al., “Design of LVAC distribution system with PV and centralized battery energy storage integration-a case study of Cambodia,” ASEAN Engineering Journal, vol. 9, no. 2, pp. 1-16, Jul. 2019. [百度学术]
T. Aziz and N. Ketjoy, “PV penetration limits in low voltage networks and voltage variations,” IEEE Access, vol. 5, pp. 16784-16792, Aug. 2017. [百度学术]
J. Karppanen, A. Lana, and T. Lähdeaho, “Comparison of LVDC distribution network alternatives: full-DC vs Link-type solutions,” in Proceedings ofInternational Conference on Electricity Distribution, Glasgow, U.K., Jun. 2017, pp. 1-5. [百度学术]
C. Lao and S. Chungpaibulpatana, “Techno-economic analysis of hybrid system for rural electrification in Cambodia,” Energy Procedia, vol. 138, pp. 524-529, Oct. 2017. [百度学术]
Surrette 6CS25P. (2018, Jun.). Rolls battery engineering-6CS25P. [Online]. Available: http://www.rollsbattery.com/catalog/ [百度学术]
A. Ahmadian, M. Sedghi, B. Mohammadi-Ivatloo et al., “Cost-benefit analysis of V2G implementation in distribution networks considering PEVs battery degradation,” IEEE Transactions on Sustainable Energy, vol. 9, no. 2, pp. 961-970, Apr. 2018. [百度学术]
P. Peerapong and B. Limmeechokchai, “Optimal electricity development by increasing solar resources in diesel-based micro grid of island society in Thailand,” Energy Reports, vol. 3, pp. 1-13, Nov. 2017. [百度学术]
