Model based optimisation of distributed generation with respect to electric grid restrictions

Bernhard Wille-Haussmann*, Thomas Erge, Jochen Link, Christof Wittwer

Fraunhofer-Institut fur Solare Energiesysteme ISE, Heidenhofstrafle 2, 79110 Freiburg, Germany
Corresponding Author, bernhard. wille-haussmann@ise. fraunhofer. de


To integrate high volumes of renewable and decentralised energy resources (RES & DER) into the electricity grids it will be necessary to manage generation and loads much more than today. Within this paper we discuss operation strategies for cogeneration plants, which can contribute to active grid management with high volumes of RES. Typically cogeneration plants (CHP) are thermally driven, which is enforced through guarantied fixed feed-in tariffs (e. g. the KWK-G feed in tariff in Germany). With the growing share of CHP up to 25% of electricity production in 2020 (18% EU [1]), the rules of control devices increase and the feed-in tariffs are expected to change to a flexible structure. To manage cogeneration in the distribution grid we investigated additive variable local tariff components for the electricity production. The influences of different CHP operation strategies to the power quality are evaluated in a load flow simulation of an exemplary low voltage grid. The results show that with the local tariff based operation a bigger part of the electric load profile is covered by local generation and also the variation of voltage in the grid decreases.

Keywords: cogeneration, optimised operation, load flow analysis

1. Introduction

Driven by the liberalisation of the electricity markets and the support for distributed, climate friendly generation technologies, the production system of electricity develops more and more from a centralised to a decentralised structure. Especially lower voltage levels are faced with supply by distributed and/or fluctuating power plants (RES), such as photovoltaic, wind and cogeneration. This leads to new operation requirements in the grid. Management strategies for those grids with a high penetration of DER must be developed [2]. To integrate a great amount of these technologies, the grids will not be allowed to stay passive. They must become active and communication between the different participants is necessary. Those “smart grids” should be designed and managed by model based methods.

Cogeneration plants produce heat and electricity simultaneously. Due to the fact, that electric and thermal demand profiles do not match, a thermally driven CHP can not supply electricity peak demands when needed. With thermal storage systems it is possible to decouple the electric and thermal delivery, which offers a higher degree of freedom for the operation management. In contrast to most renewable generators, whose production fluctuates with the availability of sun or wind, cogeneration plants can be used quite flexibly, as long as the thermal restrictions are respected. This fact offers the possibility to use them in the grid to stabilise the power quality, e. g. the voltage range. With a high

penetration of RES the number of factors to be considered by grid management increases enormously. To manage those complex grids new optimisation and model reduction techniques are necessary to get rid with the complexity. Within the project NetMod [3], which is supported by the German “Bundesministerium ftir Bildung und Forschung”, together with several partners Fraunhofer ISE develops methods for such a management.

Подпись: Fig. 1. Grid and generation structure of a low voltage grid.

In this paper we discuss integration and management strategies for RES in a showcase of a low voltage grid. The structure of this grid is shown in Fig. 1. The grid consists of a long branch which feeds several houses in a rural area. Typically these structures are dimensioned weak, so that the high potential of renewable generation in this area is limited by the transport capacity of the local grid. For this showcase we assumed 4 houses with a nominal power of 10 kW each. At each house also one PV plant with a peak power of 25 kWp is installed. At the last house also a CHP plant with a nominal electric power of 25 kW (pel = 30 %, pth = 65 %) is installed, which supplies a local heat grid. To increase the degrees of freedom for each CHP a thermal storage with a capacity of 3000 l is installed.

In the following chapter we will discuss several possibilities to operate cogeneration plants. Based on this we introduce an optimisation algorithm for CHP operation, which is based on variable tariffs. For the showcase two price functions (global and local) are introduced. The operation of cogeneration only makes sense if the produced thermal energy can be used simultaneously or can be stored for a later use. There fore knowledge about the thermal load profile is necessary to generate optimal schedules. The prognosis of the thermal load will be subject of the following paragraph. Decentralised generation will influence the power quality in the grid. The following chapter evaluates the influence of different operation scenarios to the power quality by a load flow analysis and closes with suggestions for a grid optimised operation of decentralised power plants.

Updated: July 15, 2015 — 12:31 pm