Elsevier

Applied Energy

Volume 226, 15 September 2018, Pages 957-966
Applied Energy

Optimal placement, sizing, and daily charge/discharge of battery energy storage in low voltage distribution network with high photovoltaic penetration

https://doi.org/10.1016/j.apenergy.2018.06.036Get rights and content

Highlights

  • Calculating optimal sitting, sizing, and scheduling of battery simultaneously.

  • Detailed cost/benefit analysis including all possible battery benefits and costs.

  • Mitigating impact of high photovoltaic penetration and increasing economic benefit.

  • Simulation was performed using real data and low voltage distribution network.

  • Reduction environmental emission is considered and converted to economic benefit.

Abstract

Proper installation of rooftop photovoltaic generation in distribution networks can improve voltage profile, reduce energy losses, and enhance the reliability. But, on the other hand, some problems regarding harmonic distortion, voltage magnitude, reverse power flow, and energy losses can arise when photovoltaic penetration is increased in low voltage distribution network. Local battery energy storage system can mitigate these disadvantages and as a result, improve the system operation. For this purpose, battery energy storage system is charged when production of photovoltaic is more than consumers’ demands and discharged when consumers’ demands are increased. Since the price of battery energy storage system is high, economic, environmental, and technical objectives should be considered together for its placement and sizing. In this paper, optimal placement, sizing, and daily (24 h) charge/discharge of battery energy storage system are performed based on a cost function that includes energy arbitrage, environmental emission, energy losses, transmission access fee, as well as capital and maintenance costs of battery energy storage system. All simulations are carried out in DIgSILENT and MATLAB linked together. Results show that by using the proposed approach, overvoltage and energy losses are decreased, reverse power flow is prevented, environmental emission is reduced, and economic profit is maximized.

Introduction

Recently, utilization of renewable energy sources (RES) in electrical networks is getting inevitable due to the global energy tension and environmental concerns of fossil-fuel-based electricity generation [1].

Photovoltaic (PV) generation is growing very fast while its cost is dropping rapidly [2]. Single phase rooftop PVs (<10 kW) owned by utility customers are being installed in low voltage (LV) distribution networks. The penetration of such PV systems is increased in many places throughout the world, including Iran, due to solar radiation, gradual elimination of energy subsidies, and government incentives.

Utilizing PV systems can help to reduce the dependence on conventional power plants, improve voltage profile, and decrease energy losses [3]. However, in the case of high PV penetration in LV distribution network, reverse power flow may occur when the PV production exceeds the consumers’ load [4]. This situation may lead to overvoltage, increase of total harmonic distortion (THD) and fault current, blinding of protection and false tripping, risk of islanding operation [5], and decrease reliability [6].

To reduce the negative impacts of high PV penetration, there are two main approaches including conventional (commercially available) and emerging mitigation methods [1]. Reconductoring and on-load tap changing (OLTC) are examples of conventional methods. Emerging methods include reactive power (VAR) control by PV inverters, distributed energy storage systems (DESSs), coordinated control between utility equipment and PV inverters, installation of devices such as dynamic voltage restorer (DVR) and distributed static compensator (DSTATCOM), etc.

Negative impacts of high PV penetration such as increased voltage magnitude, reverse power flow, and energy losses can be mitigated by optimal placement, sizing and/or charge/discharge scheduling of battery energy storage system (BESS). In this regard, many researchers have studied proper installation of energy storage in distribution networks with high PV penetration. In [7], optimal daily energy profiles of storage systems co-located with PV generation are calculated and it is shown that significant control abilities in peak shaving, voltage stability, and reducing distribution losses can be achieved. Optimal sizing of battery energy storage co-located with PV is evaluated in [8] for the goals such as voltage regulation. In another study, a coordinated hierarchical control scheme is presented for static synchronous compensators (STATCOM) and BESS in order to mitigate the overvoltage problem, but, cost/benefit analysis is not performed for the BESS [9]. Cost/benefit analysis is performed in [10] to determine the optimal location and size (without optimal operation) of community energy storage (CES) by considering energy arbitrage, peak power generation, energy loss reduction, upgrade deferral of transmission and distribution (T & D) systems, CO2 emission reduction, and reactive power support. BESS is applied in [11] for peak shaving and smoothing the distribution load profile. To achieve these goals, a real time control is developed which performs smoothing and peak shaving, simultaneously. In [11], the economic purpose (price arbitrage) is not considered, therefore, BESS charge/discharge is only calculated for peak shaving. Authors of [12] proposed an algorithm that is capable of integrating sizing, placement, and operational strategies of BESS taking into account energy losses, but, without considering environmental emission. The minimum energy storage required to be installed in LV grid to prevent the overvoltage is calculated in [13]; optimal sizing and placement of BESS is calculated, but, daily charge/discharge is not considered. Authors of [14] proposed optimal sizing (without sitting) of BESS in the residential LV distribution network for peak shaving, valley filling, load balancing, and management of distributed RES. In [15], sizing energy storage based on Open Distribution Simulator (OpenDSS) is proposed, but, optimal sizing, sitting, and charge/discharge are not done simultaneously. Authors of [16] proposed a new framework to integrate CES units in an existing residential community system with rooftop PV units. In [16], the location, sizing, and operational characteristics of CES are calculated to minimize the annual energy loss, enhance load following control, and improve the voltage profile, respectively. In [17], a coordinated control of distributed BESS with traditional voltage regulators including the OLTC and step voltage regulators (SVR) is proposed, but, environmental effects are not analyzed. Authors of [18] discussed optimal sizing and operation of BESS to contribute to local distribution network operation through peak shaving, voltage control, and levelling out power production from RES. The work in [19], optimizes the size of BESS based on a cost/benefit analysis when BESS is applied for voltage regulation and peak load shaving, but, optimal charge/discharge is not taken into account. Optimal planning and operation of energy storage is performed in [20] for peak shaving, reducing reverse power flow, and energy price arbitrage in distribution network with high penetration of RES, but, voltage regulation is not taken into account. In [21], the storage is utilized to compensate long-term and short-term voltage variations originated from sudden change of PV output. The strategy of charge/discharge is presented without any optimization. Authors of [22] determined the soft open point (SOP) of distribution network with the aim of optimal operation of energy storage to mitigate overvoltage arising from high RES penetration. A method is proposed in [23] to optimize the location and size of the DESS. The optimization function is based on best economical investment without considering energy price arbitrage. In [24], by considering high RES penetration, optimal sizing and operation of BESS is proposed to maximize the house independence from the grid and minimize the power flow peaks from and to the grid. An optimization method is developed in [25] for allocation of BESS in distribution system considering capital, land-of-use, and installation costs without taking into account the benefit of energy price arbitrage. Authors of [26] proposed an optimal planning approach for DESS to achieve better economic solution considering total power losses, but, without analyzing environmental effects. In [27], an optimization model is presented to minimize the net present value (NPV) of BESS and energy losses while reduction of environmental emission is not considered. Optimal location, capacity, and power rating of batteries are calculated in [28] to determine the economic technology by considering high RES penetration. Authors of [29] presented a strategy for optimal integration of BESSs by considering voltage regulation and loss reduction without taking into account the benefit of energy price arbitrage. An approach for proper utilization of the energy storage system to mitigate the effects of intermittent nature of PV has been presented in [30], but, optimal BESS planning is not included.

In the present work, it is assumed that distribution system operator (DSO) has got the ownership of BESS. Optimal placement, sizing, and operation of BESS are taken into account in LV distribution network considering high PV penetration. Optimal planning and operation of BESS is performed based on a cost function in order to make the BESS installation economical. In addition, sizing and sitting are done simultaneously with daily charge/discharge. Also, the objectives including energy price arbitrage, transmission access fee, energy losses, and environmental emission are taken into account simultaneously. The objective (cost) function consists of these objectives, and capital and maintenance costs of BESS. In this objective function, loss reduction and environmental benefits are converted to economic benefits. Other technical goals including reverse power flow and voltage regulation are considered as constraints.

Benefits of energy price arbitrage, environmental emission, and transmission access fee are maximized when BESS is charged in low energy price, emission rate, and transmission access fee and discharged while these rates are high. On the other hand, overvoltages that occur due to high penetration of PV are decreased by charging the BESS when PV systems produce maximum energy. Therefore, the optimal charge/discharge of BESS is complicated. In this paper, an auxiliary objective function is defined for increasing energy price arbitrage, reducing transmission access fee and environmental emission, and mitigating undesired impacts of high PV penetration by considering BESS constraints.

DIgSILENT and MATLAB are linked together because modeling of network equipment such as transformer, feeder, load, and power flow study are more accurate and realistic in DIgSILENT while MATLAB provides more powerful optimization tools.

Section snippets

BESS modeling

In the case of high PV penetration in LV distribution network, reverse power flow may occur when the PV production exceeds the consumers’ load. This situation may lead to overvoltage and increase energy losses [4] (Fig. 1).

BESS can mitigate these disadvantages. Recently, thanks to the technological developments, the price of BESS is decreased, but still is high. As a result, economic, environmental, and technical objectives should be considered for planning and operation of BESS, in order to

Optimization function and constraints

As mentioned before, when PV penetration in LV distribution network increases, some problems may occur. Local BESS can improve these disadvantages. On the other hand, in order to ensure affordable BESS utilization, this paper introduces a cost function to increase benefit and mitigate the disadvantages. This cost function is expressed as (5) based on NPV:CF=n=1NBARB+BENV+BLOSS×365+BTRANS×12-CM&O×1+ir1+drn-CBESSwhere BARB is energy price arbitrage benefit, BENV, BLOSS, and BTRANS are the profit

Optimal management approach

A solution method that combines the genetic algorithm with linear programing method (GALP) is proposed in this paper to find the optimal solution for number, placement, sizing, and scheduling of BESS [35]. In [35], GA is used to transform the cost function to an LP that can be solved by Simplex Method. Also, GA and LP are run only in MATLAB. Modeling of network equipment such as transformer, feeder, load, and power flow study are more accurate and realistic in DIgSILENT than MATLAB. In

Simulation results

Fig. 5 portrays an unbalanced LV distribution system located in Yazd province, Iran. This system is connected to a medium voltage system through a 20 kV/0.4 kV transformer feeding 137 residential loads. In this network, 2 single phase PV systems each with the capacity of 5 kW are connected between phase a and neutral and located at the end of feeders. In the real distribution network, there are two PV systems named PV1 and PV2, however, PV3 and PV4 each with the capacity of 5 kW are also

Conclusion

This paper proposed an optimal method for simultaneous placement, sizing, and daily charge/discharge of battery energy storage system which improved the performance of the distribution network to mitigate disadvantages of high photovoltaic penetration. Technical and environmental benefits were converted to economic benefit and thus, problem was expressed as a cost function. The optimization includes this cost function, an auxiliary objective function, and constraints of battery energy storage

References (47)

  • Ch.J. Bennett et al.

    Development of a three-phase battery energy storage scheduling and operation system for low voltage distribution networks

    Appl Energy

    (2015)
  • J. Sardi et al.

    Strategic allocation of community energy storage in a residential system with rooftop PV units

    Appl Energy

    (2017)
  • P.F. Lyons et al.

    Design and analysis of electrical energy storage demonstration projects on UK distribution networks

    Appl Energy

    (2015)
  • Kh. Mahani et al.

    Network-aware approach for energy storage planning and control in the network with high penetration of renewables

    Appl Energy

    (2017)
  • M. Chen et al.

    Optimal allocation method on distributed energy storage system in active distribution network

    Energy Procedia

    (2017)
  • J.M. Santos et al.

    Technical and economic impact of residential electricity storage at local and grid level for Portugal

    Appl Energy

    (2014)
  • O. Babacan et al.

    Siting and sizing of distributed energy storage to mitigate voltage impact by solar PV in distribution systems

    Solar Energy

    (2017)
  • L. Bai et al.

    Distributed energy storage planning in soft open point based active distribution networks incorporating network reconfiguration and DG reactive power capability

    Appl Energy

    (2018)
  • M. Daghi et al.

    Factor analysis based optimal storage planning in active distribution network considering different battery technologies

    Appl Energy

    (2016)
  • M. Aneke et al.

    Energy storage technologies and real life applications – a state of the art review

    Appl Energy

    (2016)
  • M. Beaudin et al.

    Energy storage for mitigating the variability of renewable electricity sources: an updated review

    Energy Sustain Dev

    (2010)
  • H. Chen et al.

    Progress in electrical energy storage system: a critical review

    Prog Nat Sci

    (2009)
  • R.C. Leou

    An economic analysis model for the energy storage system applied to a distribution substation

    Int J Electr Power Energy Syst

    (2012)
  • Cited by (159)

    View all citing articles on Scopus
    View full text