Impact of P2P trading on distributed generation curtailment in constrained distribution networks

Highlights

• Paper illustrates the effects that P2P trading can have on the network operation.

• Presents OPF based tool to integrate P2P into distribution network operation.

• Investigates influences of P2P on DGs under Last-In-First-Out principle of access.

• Understanding unintended consequences of P2P trading on network and DGs is key.

Abstract

The increasing uptake of Distributed Energy Resources and advancement in blockchain technology has led to the growth in interest in the peer-to-peer (P2P) based energy trading. However, there has been a lack of consideration how these trades may affect the network operation, in particular in networks that apply Active Network Management using Last-in-First-out (LIFO) access rules, such as in the UK. This paper presents a novel application of Optimal Power Flow that manages the P2P trading whilst optimising legacy DGs under a LIFO access agreement. Analysis show that under such arrangements, and in combination with autonomous P2P trading, lower priority DGs are vulnerable to excessive curtailment levels.

Introduction

There is a significant increasing number of distributed energy resources (DERs) such as renewable energy sources, storage and domestic ‘smart’ appliances [1] becoming integrated into the power system. These technologies are having profound impacts on the way the power system is operated, particularly on the lower voltage distribution networks. As a result, Distribution Network Operators (DNOs) are modifying their traditional passive operating philosophy and adopting a more active role. Moreover, as penetration levels of these stochastic and intermittent DERs increase, the challenges become more prominent and complex. Inadequate management of DER integration can lead to both thermal and voltage violations. These can cause significant underutilisation of network capacity for future (and current) renewable generation [2], as well as blocking further connections of renewable resources.

Distributed Generation (DG) capacity was, in the UK, initially granted on a case-by-case basis through a so-called ‘fit and forget’ approach, where the DG was granted a firm allocation of capacity based on conservative network planning. However, low voltage network capacity became more limited, and, as a result, DNOs struggled to facilitate DGs with firm connections [2]. This posed a significant problem to DNOs who searched for innovative solutions to enable additional DG connections while avoiding costly network reinforcements. The main approach tested in the UK, which has since become the Business as Usual (BaU), is Active Network Management (ANM). As described in [3], ANM includes a monitoring and control system that, in real time, enables management of DGs’ outputs in the case where network limitations are reached.

For DGs integrated into ANM schemes, it is common to have a non-firm arrangement that is subject to curtailment and managed in accordance with a prearranged Principles of Access (PoA) agreement [4] and [5]. The most commonly used and easily implemented PoA for ANM schemes is the Last in First Out (LIFO) arrangement, under which the last DG to connect to the ANM scheme is the first to be curtailed, with initial connections having priority over network capacity. This arrangement has been embedded for a long time, under which many DGs have made investment decisions and to unbundling/changing this arrangement would be difficult process.

Furthermore, recent developments have seen a trend towards decentralised energy trading at the distribution level. This has been investigated both in research work and through a number of pilot projects [6]. For example, the use of Peer-to-Peer (P2P) energy transactions in local markets to facilitate energy exchanges between small-scale energy producers and consumers has been investigated in [7], [8], [9], [10]. In P2P electricity trading, ‘peers’ have the ability to directly and autonomously negotiate energy exchanges and prices with any other on the network, regardless of magnitude or location. However, debates on the specific coordination, architecture and management of this trading structure are still ongoing.

The authors in [11] have identified three prominent market platform designs that may influence the development in decentralised energy trading 1) Fully Peer-to-Peer Market, 2) Community-based Market and 3) Hybrid P2P market. The distinguishing components between the three deigns are the topology of allowable trades and the degree of which there is hierarchical oversight/control. The work presented in [12,13] are examples of Fully Peer-to-Peer Markets, that present market structures based on a bilateral economic despatch formulation that allows for individual consumer preferences. An Example of a Community-based Market is presented in [14], which illustrates P2P trading within a μ-grid where an energy sharing structure is proposed that integrates the peers into an energy sharing zone. In this zone, the peers are equal and in agreement on a unified price with intention that a unified μ-grid is overall more economically beneficial (when compared to using a structure of individual self-interested participants). Lastly, [15] and [16] have proposed work that uses a Hybrid P2P market approach. In [15] a higher level of control is introduced that expands on the work from [14], where the μ-grid itself acts as one ‘peer’. Trading between the μ-grid peers is detailed alongside a localized fully P2P market within the μ-grid itself. In addition, [16] expands work from [15], with trading between peers (on the same network) solved through a relaxed form of Optimal Power Flow (OPF).

In practice, participants in P2P markets will have restrictions imposed by network limits as well as technical constraints, and the absence of such consideration may lead to operational issues. Despite this, only a few studies have assessed the impact that fully P2P trading mechanisms may have on the network [17], [18], [19]. In [18], unique network fees were imposed on certain transactions depending on the location and distance of the negotiated trade. These fees were informed through electrical distance methods such as Power Transfer Distance (PTD) and analysed using a DC power flow. A novel methodology based on a linearized power flow was presented in [19], which used partial derivatives of the Jacobian matrix to extract voltage and power flow sensitivities that are then used in an iterative process to solve the power flow.

It is the aim of this paper to illustrate the effects that P2P trading can have on the network operation, particularly when considering legacy DGs under a PoA agreement, i.e. LIFO. In order to do that, a full representation of network constraints in an AC OPF solution is required. Extending the work of [20], we introduced P2P trading that, similarly to arrangements at the transmission level [21], are allowed without conventional intermediaries. We analyse the impact such trading can have on the levels of curtailment under various LIFO arrangements and in doing so demonstrate the need for further understanding when considering the rules of governance for these markets. Following this, our novel technique based on the OPF formulation is applied, which relaxes the inherit strictness of the non-technical PoAs e.g. strict LIFO scheme, while at the same time to control the change in the curtailment levels caused by autonomous P2P trading. In particular, the changes caused by P2P that can have negative effect on system operation and overall DG curtailment.

The paper progresses as follows. Section II describes the methodology for modelling the extended formulation of the OPF tool. Section III summaries how the bilateral contracts are modelled for the P2P trades. Section IV presents the model for the case studies and discussion of the results and lastly, Section V presents the conclusion.