اثر فرکانس های عامل غیر استاندارد در هزینه های اقتصادی شبکه های AC دور از ساحل

The effect of choosing a non-standard operating frequency on the equipment and infrastructure costs of an offshore AC network is investigated. The offshore AC network considered is similar in design to the European SuperGrid “SuperNode”. It is designed to connect several large wind arrays to multiple HVDC converters through which power may be transmitted to shore. As the offshore AC network is isolated from onshore networks by the use of HVDC links, it may be operated unsynchronised at any desired frequency. The cost associated with operating the network at a fixed frequency in the range 20–120 Hz is investigated, focusing on the frequency-cost scalings of electrical devices (such as cables, transformers and reactive compensation) and offshore infrastructures. A case study is presented based upon Tranche A area of Dogger Bank, UK, where a minimum point in the total cost of the offshore network is found at 93 Hz.

A European SuperGrid [1], [2], [3] and [4] will allow the connection of offshore renewable energy sources in remote locations to existing onshore networks. It has been recognised that offshore wind resources will play an important role in determining the design and placement of SuperGrid infrastructure. In order to provide large power transmission capacity over long distances it is proposed that the connection of offshore wind farms to onshore networks is performed by high-voltage direct-current (HVDC) links [5]. Multi-terminal HVDC (MT-HVDC) technology [6], [7] and [8] is one technical solution that allows the connection of multiple offshore wind farms to one or more onshore connection points. In order to provide adequate system reliability, an MT-HVDC system must be capable of blocking faults that occur in the DC grid [9] and [10]. Advances in hybrid DC breaker technology have recently been reported [11], however, the lack of a commercially proven DC circuit breakers may be a major barrier to the implementation of MT-HVDC systems. Some VSC topologies have been proposed that promise inherent fault blocking capability, however these remain to be commercially proven and suffer additional complexity and potentially higher losses [12] and [13]. An alternative or complementary option for the European SuperGrid is the “SuperNode concept” and is the focus of this paper. A SuperNode is an offshore network which allows the connection of multiple wind arrays and HVDC substations via an AC-hub arrangement and which does not require DC fault blocking capability [14]. Fig. 1 is a high-level representation of the SuperNode concept. The AC-hub arrangement eliminates the requirement for DC circuit breaking by employing only point-to-point HVDC links [15]; in this case AC circuit breakers located on the AC-hub and in the onshore AC network may be used to isolate faults occurring on any individual HVDC link whilst leaving other links operational. A further potential advantage of the AC-hub arrangement is the opportunity to use current-source converter (CSC) point-to-point technology in place of more costly voltage source converter (VSC) MT-HVDC technology. CSCs generally offer higher power transfer capability and greater efficiencies at a given cost point than VSCs [15] although they require significant reactive compensation and additional harmonic filtering which increases their overall space requirements. It should be noted that in an AC grid composed only of power electronic converters, at least one VSC or STATCOM will be required in order to provide a reference voltage source and enable black-start capability.As the SuperNode is connected to onshore AC networks only through point-to-point HVDC links, the offshore AC-hub may run unsynchronised with onshore networks. Indeed, in some cases synchronism with one onshore AC network (e.g., mainland Europe) may preclude synchronism with another (e.g. the UK). A further observation is that there is no requirement for the hub to operate at the same nominal frequency as any onshore network and that operation at non-standard frequencies (i.e., not 50 or 60 Hz) may confer technical and economic advantages. For instance, an operational frequency below 50 Hz can offer lower transmission losses and allow the use of longer cables, whilst an operational frequency above 50 Hz can reduce transformer sizes, in turn reducing offshore platform sizes. This paper considers operation of a SuperNode at non-standard fixed nominal operating frequencies in the range 20–120 Hz (dynamic variable frequency operation is not considered). An analysis of some of the technical and economic effects of operating at a non-standard frequency are examined. Tranche A of Dogger Bank [16] is used as a case study to highlight the potential economic advantages of operating at non-standard frequencies.

The SuperNode concept has been presented as an alternative to the use of MT-HVDC systems for the aggregation of geographically close wind power resources in a European SuperGrid, using Dogger Bank Tranche A as a case study. The effect of choosing non-standard operating frequencies on the costs of major electrical components (such as cables, transformers and reactive compensation) and in offshore infrastructure (platforms) has been investigated. The advantage of choosing lower operating frequencies are the large transmission distances which may be reached, however for the Dogger Bank case study presented here, large transmission distances are not required and the increased costs of other electrical components and infrastructure outweighs the benefits of the slight increase in transfer capacity per cable. This may in general be expected to be the case in SuperNode designs where geographically close wind farms are to be interconnected. The principal advantage of higher frequency operation (e.g., 100 Hz) is the substantial reduction in size of the major electrical elements and supporting platforms, the penalty being the reduced power transfer capacity per cable and therefore the need for a greater number of cables. However, as cable costs constitute the largest fraction of overall cost, even a relatively small increase in the required number of cables is seen to substantially balance out the savings made elsewhere, meaning that a cost minimum is found at some intermediate point. Whilst the economic analysis presented here is of a first-order nature only it serves to demonstrate the design tradeoffs that may be made when designing geographically small AC networks that may work unsynchronised from the wider grid. The conclusions of the present work are based on an analysis of component rating and associated capital costs and do not consider the operational challenges that may arise in such systems. Further detailed power flow and transient stability analyses will be required to capture the influence of the proposed non-standard operating frequencies on overall system performance. Not present in the analysis presented here is a consideration of installation costs. The decrease in transformer and substation size at high operating frequencies also implies a reduction of costs for their offshore installation; in other words, additional savings may be achieved beyond those considered here. For example, the size and/or the number of special purpose marine vessels used to transport and install offshore electrical infrastructure may be reduced since transformers and substations will be smaller at higher frequencies. Thus the cost savings associated with high operating frequencies may be expected to be somewhat greater than those already outlined here and in this case the cost minimum would be pushed towards a higher frequency. It can be concluded that higher operational frequencies, such as 100 Hz, could be economically advantageous for offshore AC networks such as the SuperNode. More detailed models of the cost dependence of electrical infrastructure on frequency as well as further analysis considering the costs of infrastructure transport and installation on a case-by-case basis will allow a more exact optimum frequency to be selected.