Operation of an all-solar power system in Saudi Arabia

https://doi.org/10.1016/j.ijepes.2020.106466Get rights and content

Highlights

  • An all-solar electricity production mix makes sense in regions with high and stable solar irradiation, such as Saudi Arabia.

  • An all-solar power system heavily relies on storage to cover nighttime hours.

  • Simulation results for an all-solar power system in Saudi Arabia throughout one year are reported and discussed.

  • The best option is generally a combination of CSP plants, PV plants and storage facilities.

Abstract

This paper considers the operation of an all-solar electric energy system in Saudi Arabia, that is, a system which generation mix includes solely solar (PV and/or concentrated solar power (CSP)) and battery energy storage units (used to cover hours with no solar resource and hours beyond the CSP thermal storage capacity). We analyze the operation of this system using realistic data. These data pertain to the actual transmission system and the demand throughout the country in 2015. The Saudi Arabian power system exhibits a rather flat demand during both the summer and winter seasons (there are mostly two season in Saudi Arabia: winter and summer), being the peak demand for 2015 in these seasons 49 and 25 GW, respectively. Solar generation units and storage units are sized and sited using a fully-fledged capacity expansion model (not described in this paper). To comprehend the daily operation of such all-solar power system in Saudi Arabia throughout 2015, we analyze three cases in which the demand is served (i) using PV and storage units, (ii) using CSP and storage units, and (iii) using PV, CSP and storage units. We compare in detail the above three cases. We conclude that using an optimal combination of PV an CSP units results in minimum total (investment and operation) cost, minimum storage needs, and moderate solar energy spillage.

Introduction

Since solar resources are outstanding across the Arabian Peninsula [1], [2], [3], [4], it is reasonable to study the possibility of supplying Saudi Arabia electricity needs using solely solar resources and storage. Moreover, Saudi authorities are pursuing the transformation of the current electricity generation mix into a renewable dominated one [5]. It is important to note that the analyses carried out for Saudi Arabia are easily extrapolated to world regions of high and stable solar irradiation, such as northern Africa, southwestern USA, western China, northern Mexico, western and central Australia, and northern Chile.

Specifically, this paper considers the operation of an all-solar electric energy system in Saudi Arabia. That is, to supply the whole electric demand of Saudi Arabia using solely solar resources and battery energy storage systems (BESS). We assume that sufficient solar units and storage facilities are available throughout this system. Long-term planning models, beyond the scope of this paper, can be used to derive the solar/storage facilities needed. Specifically, we use the generation and transmission expansion planning model described in [6]. This is a green-field stochastic planning model that identifies the best generation and transmission capacity for a given target year. In addition to the future demand, the data required by this model include (i) all possible locations for solar plants of different technologies across the country and (ii) all available corridors for transmission expansion. Since the target year we analyze is 2015, we do not consider expanding the transmission network, and use this model to optimally replace the current thermal generation mix by solar plants and BESSs. The deterministic demand to be supplied is that of the year 2015.

We analyze the operation of the Saudi Arabia power system during one whole year using realistic data from 2015. These data pertain to the transmission system, the demand throughout the country, and solar generation (photovoltaic (PV) and/or concentrated solar power (CSP)) at locations identified by the generation expansion planning model [6] briefly described above. BESSs is used to cover the hours with no solar resource and hours beyond the CSP thermal storage capacity. We analyze three cases in which the demand is supplied:

  • using CSP and BESSs,

  • using PV and BESSs, and

  • using PV, CSP and BESSs.

We compare these three cases from the viewpoint of the daily operation.

A number of references regarding the operation of electric energy systems with high penetration of renewable power are available. References that are relevant to our work are described below. Reference [7] provide a multi-day stochastic operation model for power systems with CSP. Reference [8] analyzes the value of CSP for high renewable integration. It uses a case study based on the US southwest. Reference [9] explores the possibility of supplying large electrical demands in the US using renewable energy sources. Reference [10] analyzes the operation of CSP units in power systems with high renewable penetration and studies the benefits of these units. Reference [11] provide a methodology for the optimal operation of railway electric energy systems considering PV units, wind turbines, and storage systems. Reference [12] analyzes the benefits of using CSP and PV plants in the power system of Bangladesh. The result shows several advantages of CSP over PV. Reference [13] studies how to use storage to control the fluctuation of renewable energy. Reference [14] analyzes the performance of PV technology in Slovenia. Reference [15] discusses the key role that CSP technology would take in the Middle East due to high solar irradiation.

Additionally, reference [1] analyzes the possibility of integrating renewable energy in the Saudi power system. Reference [2] analyzes the relative properties of solar energy production for sites in Saudi Arabia. In [3], the National Renewable Energy Laboratory’s (NREL) and King Abdulaziz City for Science and Technology (KACST) describes a project to estimate the solar resource capability of Saudi Arabia. Reference [4] analyzes the benefit of solar power at the residential scale in Riyadh city. Reference [16] studies the performance of PV and CSP technologies in three different sites in Saudi Arabia. They conclude that CSP plant has better electricity generation performance. Reference [17] analyzes economically the integration of solar PV in the Saudi power system. Reference [18] studies solar radiation resources in Saudi Arabia and they concludes that PV technology would perform well at any location. Reference [19] presents an analysis of one-year solar data at 44 locations across Saudi Arabia. Overall, the study finds that coastal areas have a lower amount of global horizontal irradiance as compared to inland areas, and that the northern province of Tabuk is the most suitable region for solar PV plants.

Considering the above references, the contributions of this paper are threefold:

  • 1.

    Developing an operation model including CSP, PV and storage for the yearly operation of an all-solar power system.

  • 2.

    Analyzing all-solar operation outcomes in Saudi Arabian throughout one year.

  • 3.

    Comparing yearly operation outcomes considering (i) PV and BESS, (ii) CSP and BESS and (iii) PV, CSP and BESS.

The rest of this paper is organized as follows. Section 2 describes the proposed operation model. Section 3 discusses an illustrative example. Section 4 provides and analyzes outcomes from a case study based on the Saudi Arabian power system. Finally, Section 5 provides conclusions.

Section snippets

Optimization model formulation

By efficiently using storage facilities, the proposed model addresses the need of minimizing operation cost while eliminating or minimizing unserved energy, and has the form:minimizet=1TcΩCCcCFctCPcCmax+pΩPCpPFptPPpPmax+dΩDCdUpdtUsubject to:es,t+1=est+pstCηs-pstDs,t0pstDPsDmaxs,t0pstCPsCmaxs,t0estEsmaxs,t0pdtUPdtDd,tpltL=Blθslt-θrltl,t-plmaxpltLplmaxl,tcΩnCFctC.PcCmax+pΩnPFptP.PpPmax-cΩnCpctCspill-pΩnPpptPspill+sΩnSpstD-l|sl=npltL+l|rl=npltL=dΩnDPdtD

Illustrative example

The simple example bellow illustrates the functioning of an all-solar system with a stable demand profile (the case of Saudi Arabia) for a given day.

The power system in this example has a load of 50 GW throughout the study horizon (similar to the summer peak demand of the Saudi Arabian system).

The daytime is assumed to be 10 h and the nighttime 14 h. This assumption is solely valid in this illustrative example. In the case study, we use the actual value of solar irradiation each hour of the

Case study

In this section we apply model (1)-(11) to the Saudi Arabian power system. We consider year 2015 because we have detailed and accurate data for that year. However, we note that the proposed methodology can be applied to a future year provided and accurate forecasts regarding the demand and the generation and transmission systems are available.

Conclusions

This paper reports studies pertaining to the operation of an all-solar power system in Saudi Arabia during 2015. The Saudi system exhibits a relatively flat demand of about 50 GW during the summer and of about 25 GW during the winter. The generation mix considered in this study consists of: (i) 12 CSP units (108.5 GW total) and 17 battery energy storage facilities (482.1 GWh total), (ii) 12 PV units (202.0 GW total) and 17 battery energy storage facilities (584.2 GWh total) and (iii) 8 CSP

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (29)

  • H. Jin et al.

    Power system multi-day stochastic scheduling considering the uncertainty of CSP/wind plants

    Proc IEEE Power Energy Soc Gen Meet

    (2016)
  • Paul D, Mehos M. Enabling greater penetration of solar power via the use of CSP with thermal energy storage. Book...
  • Hand M, Baldwin S, DeMeo E, Reilly J, Mai T, Arent D, et al. Renewable Electricity Futures Study. National Renewable...
  • E. Du

    Operation of a high renewable penetrated power system with CSP plants: a look-ahead stochastic unit commitment model

    IEEE Trans Power Syst

    (2019)
  • Cited by (7)

    • A new monitoring technique for fault detection and classification in PV systems based on rate of change of voltage-current trajectory

      2021, International Journal of Electrical Power and Energy Systems
      Citation Excerpt :

      Photovoltaic (PV) power resources are free-pollution resources of electrical power energy because they do not generate any exhausted contamination [1].

    • Evaluating properties of Arabian desert sands for use in solar thermal technologies

      2021, Solar Energy Materials and Solar Cells
      Citation Excerpt :

      Saudi Arabia has a growing interest in generating electricity from solar technologies including 25 GW of CSP systems by 2030 [38]. Several studies have considered using CSP for power generation in Saudi Arabia [39], taking into account the particularities of different locations, including Jeddah [40], Riyadh [41–43], Dhahran [44], Neom [45], and AlKhobar [46,47]. However, few studies have investigated the use of sand as TES material and direct solar absorber and have mainly focused on Riyadh or fracking sands [15,42,48].

    View all citing articles on Scopus
    View full text