Offshore wind farm repowering optimization
Introduction
The history of offshore wind power can be traced back to 1991 when the first offshore wind farm, Vindeby, was installed in Denmark [1]. Compared with onshore wind power, it is still a novel energy technology and thus more attention has been paid to increasing energy production efficiency or improving installation technology while wind farm owners have seemed to disregard the significance of decommissioning [2]. Likewise, most existing research has concentrated on the development, construction, and operational stages of offshore wind farms [3]. In [4], [5], [6], [7], [8], the wind farm micro-siting optimization problem (WFMOP) was investigated. Due to the extremely nonlinear characteristic of the wake effect which is the dominant factor to be taken into account when solving the WFMOP, a heuristic algorithm was widely adopted [4], [5], [6], [7] while a recent work [8] also proposed a sequential convex optimization method to solve the WFMOP. Submarine cables are one of the important components in an offshore wind farm, in order to minimize the investment in cables, the cable connection layout is optimized in [9], [10]. In addition, some work has also been presented describing offshore wind farm control optimization [11], [12], [13], which can increase power production by tuning the pitch angle or rotor speed ratio. However, considering the increasing demand for decommissioning in the near future, decommissioning should be studied and planned at the very beginning of the project to prevent complications which may incur unexpected higher costs and environmental impact [14].
Decommissioning is considered to be the last step of the project. According to [15], decommissioning can be defined as the reverse of the installation phase; the objective of decommissioning is to return the site to its condition before project deployment as far as possible. The first offshore wind farm decommissioning on record (the Yttre Stengrud wind project) happened in 2016 [17]. This project only operated for 15 years [16]. However, due to the difficulty of finding spare parts and huge cost of repairs and upgrades, the wind farm owner decided to dismantle it [17]. Recently, several decommissioning plans were also announced at Vindeby and Lely. In addition, it is expected that offshore wind farm decommissioning will surge in the next decade since many offshore projects commercialized in the early 2000s. The information in Table 1 shows the operating offshore wind farms which have been in commission for more than 10 years, with the installed capacity for each wind farm (MW).
From Table 1, it can be easily seen that the decommissioning era is coming, and with great variety in the number of wind turbines (WT) and capacity of each wind farm. Taking into account the differences in foundation type, weather conditions, seabed conditions, etc., the decommissioning schemes are expected to be exclusive to and unique for each wind farm. In other words, it seems impossible to put in place a general method for offshore wind farm decommissioning [14]. In order to reduce the impact of the offshore wind farm on the local marine environment, the wind farm developer should follow legal obligations, as in UNCLOS (United Nations Convention on the Law of the Sea) [19], the Energy Act [20], and the Coast Protection Act [21]. All the obligations emphasize the responsibility of the wind farm owner to conduct a complete dismantling, which includes removing the foundation and cables (sometimes cables can be left in situ), to minimize the project’s impact on the marine ecosystem, despite findings from the oil and gas industry that keeping concrete foundations harms the ecosystem less than removing them [22]. As a reaction to the environmental impact of constructing and dismantling offshore WTs, research has been conducted on the increased environmental cost of decommissioning of offshore WTs when compared to the onshore counterparts [23], [24]. Further studies have even included strategies for decommissioning of foundations and cables in order to limit the impact on local marine life [25]. As an example, Northern America is expected to become a large offshore wind market, and there is already a study is estimating decommissioning costs and proposing strategies ensuring the decommissioning of WTs, foundations, and cables [26].
In the above description, decommissioning has been defined as the process of dismantling the entire wind farm including removal of the foundations, removal of the WTs and cables, etc. However, some components of the wind farm usually have a longer lifetime. For instance, the foundations can last over 100 years (for gravity based foundations) [27] and the internal array and transmission cables can be used for more than 40 years [28]. In addition, a two-year period of monitoring and remediation is required to ensure that the site returns to the state before wind farm construction [29]. Hence, some wind farm owners may decide to repower the offshore wind farms to continue to use the majority of the original electrical system (and/or foundations) to install bigger WTs or change some components, such as drive trains or electronic devices which can improve the production efficiency. In [30], three end-of-life scenarios for offshore wind turbines were summarized as life extension, repowering and decommissioning, and it was pointed out that failure mode identification throughout the service life of an offshore wind farm is necessary for the end-of-life decision. In [14], two repowering strategies: partial repowering (refurbishment) and full repowering were introduced. Partial repowering is the process of installing minor components within the wind farm such as rotors, blades, gearboxes, etc., which is similar to life extension as described in [30]. Full repowering involves replacing the old turbines with newer, bigger ones to obtain higher energy efficiency. Repowering is considered as one end-of-life decision for an offshore wind farm in [2], [3], [14], [19], [25], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36]; it is sustainable and there is potential value in recycling or reusing the dismantled spares. It has become an increasing common practice for Germany and Denmark [31]. Different repowering options considering the Spanish regulatory framework were analyzed in [32], [33] for two existing wind farms - Bustelo and San Xoan. An economic analysis of wind project repowering decisions in California was conducted using the common evaluation index LCoE in [34]. In [37], the profitability for full and partial repowering was analyzed using net present value (NPV) as the evaluation index. It concluded that full repowering would be attractive after 20–25 years of operation, while before this time the benefits of repowering are insignificant. Moreover, partial repowering shows only about 10% cost savings compared with full repowering, so it is not preferable unless advanced technology can be applied to promote generation efficiency or minimize operating costs. Nevertheless, not all WTs will be decommissioned at the exact same time as assumed in [37]. Several studies have been conducted on the decommissioning of oil and gas rigs with a special focus on avoiding damage to the local marine environment [38], [39]. However, since oil and gas is a limited resource, reuse-in-place of facilities and potential optimization of this was not touched upon in these studies. One lesson learned, however, is the fact that the least damage is done to the marine environment by keeping the concrete structures under the sea [22], instead of removing them, which only strengthens the proposals of this research paper. A number of attributes are relevant in particular to repowering, such as the cost of infrastructure, the environmental aspects, the regulatory framework, the logistics, insurance, etc. However, in this paper, we focus on the economic analysis of the full repowering option using an optimization method. The wake effect is the dominant factor considered and the relation between energy production and additional investment is investigated. The replacement of merely one WT within a wind farm may cause changes in the wind conditions observed for the other WTs, due to changes in wakes. It is therefore worth considering which WTs to remove and which types of WTs to install to maximize the energy output of the whole wind farm. In this research, both bigger WTs and smaller ones are considered for full repowering. However, different WTs would have a different hub height and blade diameter compared with the original one. The wake losses in such a mixed hub height wind farm should be estimated so that the profitability of the repowering decision can be properly analyzed.
The contributions of this research can be summarized as follows:
- 1.
The repowering option for an offshore wind farm was formulated as a non-convex optimization problem and solved by heuristic optimization algorithm. Though the heuristic algorithm has been widely used in wind farm layout design, to our knowledge, this is the first paper which quantitatively analyzes the profitability of the repowering strategy using this optimization method.
- 2.
The existing full repowering option is to replace the existing WTs with a fewer number of bigger capacity WT of the same type [32], [33], while this paper considers the possibility of constructing a wind farm with mixed types of WTs (bigger or smaller WTs) and regards it as one solution for improving the efficiency of the original wind farm.
- 3.
Mathematical derivations for estimation of wake losses in a wind farm with mixed hub height WTs are given for the proposed optimized repowering method, which can help the wind farm owner to create their own repowering strategy quantitatively.
Section snippets
Mathematical models
In this section, the wake model and energy model for energy production estimation are introduced.
Methodology and problem formulation
This section first introduces the methods and materials applied to examine and optimize a strategy for repowering an offshore wind farm. Then the objective function is presented. The assumptions are stated at the end.
Case study
In this section, the input database and two scenarios are presented. Afterwards, the optimized layout and results are given.
Conclusions
This research introduces a novel strategy for repowering WTs in offshore wind farms, in order to maximize the value of a wind farm investment. The results are obtained using the data from a real offshore wind farm, Horns Rev 1, and by simulating various optimization scenarios in order to reveal the proper strategy. The APSO algorithm revealed the best solution after 113th iterations, where the minimum LCoE, 10.43% smaller than in the “new for old” strategy, was achieved. The proposed method can
Acknowledgments
The authors would like to thank the Norwegian Centre for Offshore Wind Energy (NORCOWE) under grant 193821/S60 from the Research Council of Norway (RCN).
References (53)
- et al.
Offshore wind farm development in Europe and its comparison with onshore counterpart
Renew Sustain Energy Rev
(2011) - et al.
Offshore wind farms’ decommissioning: a semi quantitative Multi-Criteria Decision Aid framework
Sustain Energy Technol Assess
(2016) - et al.
Combined optimization for offshore wind turbine micro siting
Appl Energy
(2017) - et al.
Optimization of wind turbine layout position in a wind farm using a newly-developed two-dimensional wake model
Appl Energy
(2016) - et al.
Towards realistic design of wind dams: An innovative approach to enhance wind potential
Appl Energy
(2016) - et al.
Layout optimization for maximizing wind farm power production using sequential convex programming
Appl Energy
(2015) - et al.
A new method for simultaneous optimizing of wind farm’s network layout and cable cross-sections by MILP optimization
Appl Energy
(2016) - et al.
Sustainable decommissioning of an offshore wind farm
Renew Energy
(2017) - et al.
Decommissioning of offshore oil and gas facilities: a comparative assessment of different scenarios
J Environ Manage
(June 2006) - et al.
Life cycle assessment of onshore and offshore wind energy-from theory to application
Appl Energy
(2016)