An Overview and Countermeasure of Global Wave Energy Classification

Abstract

Wave energy will be an important support to deal with the energy crisis of human society. A scientific energy classification scheme is a prerequisite support for the macro-scale optimized layout, micro-scale accurate site selection and a blueprinting of development routes for wave energy. Based on the indicator considered, this study first divides the global wave energy classification into three stages: preliminary exploration stage, mid-term development stage and relatively mature stage, and then sorts out the main strengths and weaknesses of each stage. It is found that the current classification scheme has six typical bottlenecks such as inconsistency with physical mechanisms, inability to meet the needs of diverse tasks, inapplicability in some seasons/months, etc. To effectively address them, a dynamic adaptive wave energy classification scheme is proposed, which can consider all elements, is suitable for diverse tasks, is available at all times and is applicable to all regions. Based on this, the concepts of absolute and relative classes, a dynamic mapping of wave energy classification, and a future energy classification are proposed, with the expectation of promoting the industrialization and scaling of wave energy.

1. Current Status of Global Wave Energy Classification

With the rapid development of society, the crisis of resources and environment is becoming more and more severe. The developing countries and regions are generally weak in power supply capacity, and the developed countries in Europe and the United States are also facing the same energy crisis. Recently, the energy woes in Europe have led to soaring electricity prices that have broken through historical extremes, and some countries have even resumed coal-based power generation. Such advantages as regeneration and large reserves make wave energy an important support to break the energy dilemma. However, there are grand regional and seasonal differences in the distribution of wave energy resources. Therefore, an adequate knowledge of the resource characteristics is a prerequisite for the safe and efficient development of wave energy.

Iglesias and Carballo [1], Akpınar and Komürcü [2] and Liang et al. [3] carried out the analysis of the spatial and temporal characteristics of different waters. Roger [4], Reikard et al. [5] and Zheng and Song [6] proposed a short-term projection scheme for resources. Akpınar et al. [7] and Kamranzad and Lin [8] analyzed the long-term trend of wave power density (WPD) for climate. Zheng [9] created a dataset of global ocean wave energy resources. Previous work has made a great contribution to wave energy evaluation. However, there has never been a common wave energy classification scheme. Reasonable classification is the scientific and decision-making basis for the micro-scale accurate site selection, macro-scale optimized layout and technology roadmaps of a country’s ocean energy. Therefore, it is urgent to create a scientific and reasonable marine energy classification scheme. Based on the improvement of the indicators considered, there are three stages of global wave energy classification in this study.

1.1. Preliminary Exploration Stage

Wave energy classification has significantly lagged behind wind energy classification. The early prototype of global wave energy classification began in 2011 when Iglesias and Carballo [10] pioneered this work in the site selection of wave energy. In 2011, Shi and Wang [11] used the Analytic Hierarchy Process (AHP) to develop a siting scheme for offshore wave energy, with five main factors sea conditions, land traffic, construction conditions, floating water transport and environmental impacts. A case study was carried out in the waters around Qingdao. Through offshore tests, it was found that the wave power device operated well and the nearshore site selection was effective.

Referring to the wind energy classification scheme, Zheng et al. [12] created a wave energy classification scheme in 2012, which is based on significant wave height (SWH), WPD and effective wave height hours, as shown in

Table 1. Early wave energy classification scheme [12].

Class	Annual Average SWH (m)	Annual Average Wave Energy Density (kW·m−1)	Significant Interval (h)	Wave Energy Regionalized
1	0–0.5	<1	<2000	indigent area
2	0.5–1.5	1–6	2000–3000	available area
3	1.5–2.5	6–15	3000–5000	subrich area
4	>2.5	>15	>5000	rich area

. Based on this, the wave energy classification in China Seas is realized for the first time. It is the first time in China to use the numerical simulation method to study the wave energy of the entire China seas. In addition, the temporal and spatial distribution characteristics of a series of key indicators of wave energy in China Seas were analyzed. Combined with the scheme, the classification in the China Seas was carried out, as shown in Figure 1. The results show that the Bohai sea belongs to the poor area, most of the Yellow Sea belongs to the exploitable area and large areas of the East China Sea and South China Sea belong to the exploitable area and relatively rich area. The Beibu Gulf is mainly a poor area, and its southern sea area is the available area. The scheme was applied to a wave energy study in the Northwest Pacific Ocean by Wan et al. [13] to a wave energy resource analysis of an offshore wind farm in Guangdong by Zhou and Huang [14] of China Energy Engineering Group (CEG) and to a joint wind–wave energy assessment of Niger Delta Coasts by Ayodotun [15].

In 2012, Zhang et al. [16] used the annual average SWH as an indicator to carry out the wave energy classification of the Fujian coastal area, with H_1/10 above 1.3 m as class I, H_1/10 between 0.7 m and 1.3 m as class II, H_1/10 between 0.4 m and 0.7 m as class III, H_1/10 below 0.4 m as class IV. The results demonstrate that the central-eastern mountainous section of the area is class III while other sections are class I and class II, with sound development prospects.

The National Ocean Technology Center has made great efforts in the wave energy classification in China seas. Zhang et al. [17] leveraged one-year measured data from 50 observation stations and the Simulated Waves Nearshore (SWAN) hindcast wave data to get the distribution of WPD in China’s offshore waters and to develop a classification for its waters with WPD, with P above 4 kW/m as class I, P between 2 kW/m and 4 kW/m as class II, P between 1 kW/m and 2 kW/m as class III, P below 1 kW/m as class IV. The results show that southern Fujian, northern Guangdong, southwestern Hainan and most of the waters around Taiwan are class I areas (rich area), southern Zhejiang, northern Fujian and southwestern Guangdong are class II areas, Shanghai, northern Zhejiang and northern Hainan are class III areas, and Liaoning, Hebei, Tianjin, most of the waters around Shandong Peninsula and Jiangsu are class IV areas.

In 2013, Zhang et al. [18] proposed an energy classification scheme based on WPD, with P above 6 kW/m as class I, P between 4 kW/m and 6 kW/m as class II, P between 2 kW/m and 4 kW/m as class III, P below 2 kW/m as class IV. In 2014, based on their original work, Zhang et al. [19] supplemented it by considering annual effective wave hours, WPD and annual effective wave hours, with P above 6 kW/m and T_E above 5000 h as class I, P between 4 kW/m and 6 kW/m and T_E above 3000 h as class II, P between 2 kW/m and 4 kW/m and T_E above 1500 h as class III, P below 2 kW/m and T_E below 1500 h as class IV. With this scheme, the wave energy classification was carried out in the Bohai Sea, the northwestern and western Yellow Sea, the Yangtze River estuary and Hangzhou Bay waters. Four Class I areas, one Class II area and five Class III areas were found for wave energy development in these areas, which provides a better reference for development and site selection.

In 2013, Zheng et al. [20] used the simulated wave data obtained from the WW3 wave model to classify offshore China into several study areas: Bohai Sea, northern Yellow Sea, central Yellow Sea, southern Yellow Sea, East China Sea, Taiwan Strait, Beibu Gulf, Gulf of Thailand, northern South China Sea, central South China Sea and southern South China Sea. With the comprehensive consideration of a series of key factors such as wave power density, the effective time of resource development, the energy stability (coefficient of variation), the trend of SWH and wave power density, the total storage and effective storage of wave energy, the richness of wave energy in different areas were compared and analyzed and the concept of wave energy in selected advantaged areas was put forward, as shown in

Table 2. Comparison of wave energy classification in China Seas [20].

Region	Annual Mean WPD (kW/m)	Occurrence of Effective Time of Wave Energy (%)	Coefficient of Variation				Trend of Significant Wave Height (cm·a−1)	Trend of WPD (kW·m−1·a−1)	Energy Storage (104 kW·h·m−1)		Energy Class
			January	April	July	October			Total Storage	Effective Storage
Bohai Sea	<2	<10	2.0–3.0	1.5–3.0	2.0–3.0	2.0–3.0	1.0–2.0	<0.1	<2	<2	Poor
North of the Yellow Sea	<2	10–30	1.5–2.0	1.5–2.0	1.0–1.5	1.5–2.5	1.0–1.5	<0.1	<2	<2	Poor
Middle of the Yellow Sea	2–6	30–40	1.5–2.0	1.0–3.0	1.0–2.0	1.5–2.5	0.5–1.0	0.0–0.2	2–4	<4	Available
South of the Yellow Sea	6–10	40–60	1.0–2.0	1.0–2.0	1.5–2.0	1.0–1.5	1.0–2.0	0.1–0.3	4–8	4–6	Subrich
East Sea	10–16	60–70	<1.5	1.0–1.5	2.5–3.0	1.5–2.0	1.5–2.5	0.3–0.5	6–12	6–10	Rich
Taiwan Strait	8–10	40–50	0.5–1.5	1.5–2.0	2.0–2.5	1.0–1.5	2.0–3.0	0.1–0.2	4–6	4–6	Subrich
Beibu Gulf	<2	<30	1.5–3.0	1.5–2.5	1.0–1.5	1.5–2.5	1.5–2.5	0.0–0.2	<2	<2	Poor
Gulf of Thailand	<2	<30	1.0–3.0	2.0–3.0	1.0–1.5	2.0–3.0	1.0–1.5	<0.1	<2	<2	Poor
North of the South China Sea	10–22	60–80	0.5–1.0	1.0–2.0	1.5–2.5	1.0–1.5	1.5–2.5	0.2–0.5	6–16	4–14	Rich
Middle of the South China Sea	14–18	60–70	1.0–1.5	1.5–2.5	1.0–2.0	1.0–2.0	1.0–2.0	0.2–0.3	10–14	8–12	Subrich
South of the South China Sea	4–14	20–60	0.5–1.5	1.5–2.0	1.0–1.5	1.0–2.0	0.5–1.5	0.0–0.2	2–10	2–8	Available

. The results show that the northern South China Sea (from the east of Hainan Island to the Luzon Strait) and the ocean east of Taiwan and the waters around the Ryukyu Islands are the dominant areas, with resource abundance far exceeding the traditional estimates. This work is also the first time that the energy trend has been integrated into classification indicators.

Sun [21] adopted indicators such as energy level occurrence and availability proposed by Zheng and Li [12] to conduct a study in China seas. According to the two indicators of wave power density and coefficient of variation, the wave energy classification was carried out in China seas. The results show that the wave energy is abundant in the South China Sea and the East Sea of Taiwan island.

Advantages and disadvantages of the preliminary exploration stage: compared with wind energy classification, wave energy classification starts relatively late, but its starting point is relatively high. At this stage, SWH, wave power density and effective wave height hours can be considered in the scheme. Due to the lack of the definition of key indicators such as wave energy availability and energy level occurrence, the scheme fails to fully consider them. Therefore, the comprehensive consideration of energy characteristics, environmental risks and cost effectiveness has not yet been realized.

1.2. Mid-Term Development Stage

In 2014, Zheng et al. [22] further improved the scheme created in 2012 by increasing the energy from four to six classes (as shown in

Table 3. Wave energy classification scheme [22].

Class	Annual Average Wave Height (m)	Annual Average Wave Energy Density (kW/m)	Significant Interval (h)	Wave Energy Regionalized
1	0–0.5	<1	<2000	indigent area
2	0.5–1.5	1–6	2000–3000	available area
3	1.5–2.5	6–15	3000–5000	subrich area
4	2.5–3.0	15–30	>5000	rich area
5	3.0–3.5	30–40
6	>3.5	>40

). Based on the ECMWF 45-year Re-Analysis (ERA-40), the first global wave energy classification was realized and its map was drawn, which properly reflects the overall energy class of global ocean. In this study, the ERA-40 wave reanalysis from ECMWF was used, which separated wind–sea and swell for 45 years (from September 1957 to August 2002). For the development of wave energy resource, two key indicators of available wave height occurrence and energy level occurrence are proposed. The spatial and temporal characteristics of the full set of key indicators of wave energy in global oceans are systematically studied by considering wave power density, energy level occurrence, effective wave height occurrence, energy stability and long-term trend, swell index and wave energy storage. On this basis, the wave energy classification of the global oceans was realized (Figure 2). The results show that the wave energy resource shows the characteristics of more in the south and less in the north. The rich areas are mainly distributed in the northern and southern hemisphere westerlies, such as most of the southern hemisphere westerlies, central and eastern waters of North Pacific Westerlies, and most of the waters in the north Atlantic westerlies. Most of the global oceans and China seas belong to the available areas. The poor areas are mainly distributed in some polar seas, Malaysia waters, the northern Caribbean Sea and Gulf of Mexico, Baffin Bay and Davis Strait, the Mediterranean Sea and Black Sea, etc. In the 1970s, judging from WPD, Tornkvist [23] pointed out that the rich areas are mainly concentrated in the northeast of the North Atlantic, the west coast of North America in the northeast of the Pacific Ocean, the southern coast of Australia, Chile in South America and the southwestern coast of South Africa. The extent of rich areas in Figure 2 is generally consistent with the conclusion of Tornkvist [23], but it is wider.

At the same time, a problem has also been identified: the regional differences of classification results based on this proposed scheme are not significant, and the wave energy is mainly class 3 at middle and low latitudes of global oceans. As a result, it is difficult to provide detailed guidance for site selection at middle and low latitudes. The north Pacific westerlies, the north Atlantic westerlies and the offshore China all face this problem. Therefore, there is an urgent need to establish a wave energy classification scheme that can fully reflect regional differences. Additionally, this scheme still fails to consider environmental risks and cost effectiveness.

Neary et al. [24] proposed the annual available energy (AAE) and divided it into four classes to describe the energy class. The AAE is calculated as follows.

$$\mathrm{AAE}\left(T_p\right)={\mathrm{T}}_{year}{\displaystyle \sum _{}^{}J(T_p)f(J,T_p)},$$

(1)

$$\mathrm{AAE}={\displaystyle \sum _{}^{}\mathrm{AAE}(T_p),}$$

(2)

where J is portioned wave power density and $T_p$ is peak period. The classification of AAE is that: class 0 sites (AAE < 10 MWh/m), representing sites with consistently low power that can only support very specialized energy projects, e.g., powering remote sensors, class 1 (10 < AAE < 50 MWh/m), with generally low power that may support specialized application, e.g., desalinization, and class 2 (50 < AAE < 200 MWh/m) and class 3 (200 < AAE MWh/m) sites with moderate to high power supporting utility scale projects. The index was used to classify wave energy resource offshore the U.S., showing that AAE in the West Coast, the south coast of Alaska, and Hawaii is up to 300 MWh/m, which belongs to class 3. AAE on the East Coast is 50–200 MWh/m and belongs to class 2. AAE in the Gulf Coast is less than 50 MWh/m and belongs to class 1.

Ahn et al. [25] presented a classification for the US coastal waters, which is based on two key indicators: WPD and dominant wind period band. Results show that the Energetic Total Power class I sites are predominant all along the West Coast, the northern and eastern shores of Hawaii, the southern coast of Alaska extending westward along the Aleutians and deep sites in the Bering Sea.

Shao et al. [26] established a classification scheme based on the literature review method and expert survey method, including resource and natural conditions, as well as economic and social factors. Additionally, they conducted case studies for Frauen Insel, Wheat Island, Zhucha Island and Sanping Island in Qingdao waters, finding that the order of resource development is Zhucha Island, Wheat Island, Frauen Insel, Sanping Island and Zhucha Island is the first choice for building wind–wave complementary power stations in Qingdao. The outcome provides a reference for the site selection of power stations in Qingdao waters. However, the scheme also does not consider environmental risks.

Feng [27] carried out a study on the site selection of wave power generation by using grey hierarchical analysis with multi-indicator effectiveness assessment, considering such elements as wave climate, site selection conditions of offshore power generation, onshore facility conditions, construction conditions and the surrounding environment. He conducted a case study of Donggang in Dandong and Longgang District in Huludao and found that the sea address of Donggang in Dandong is the best site selection option.

Yang [28] proposed a site selection scheme based on the improved Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) method, taking into account the conditions of wave climate, offshore power generation address, onshore facility, construction and the surrounding environment. It was found that through numerical experiments, the improved TOPSIS method is better than the traditional TOPSIS method in comparing the differences of schemes.

Advantages and disadvantages of the mid-term development stage: compared with the early stage, the wave energy classification scheme in the mid-term development stage has little great improvement in the considered factors. The progress is that the wave energy is increasing from four to six classes in this stage. Regardless of the stages, the scheme still consists of the following problems: (1) Insufficient consideration of resource factors. (2) Regional differences in energy classification are not obvious to guide micro-scale accurate site selection. (3) Comprehensive consideration of resource characteristics, environmental risks and cost effectiveness has not yet been realized.

1.3. Relatively Mature Stage

With the successive definition and continuous improvement of wave energy availability and energy level occurrence [29], the two key indicators are gradually applied to the classification. The availability of wave energy is defined as effective wave height occurrence (EWHO). The energy level occurrence is defined as the various levels of WPD, including available level occurrence (ALO, occurrence of WPD above 2 kW/m), moderate level occurrence (MLO, occurrence of WPD above 10 kW/m) and rich level occurrence (RLO, occurrence of WPD above 20 kW/m).

In 2014, Zheng et al. [22] continuously improved on the basis of the earlier scheme. Zheng and Li [29] created a scheme by the Delphi method, which considers resource characteristics, environmental risks, and cost effectiveness. In addition, a new concept of a dynamic adaptive energy classification to meet diverse demands was first proposed [29], which takes into account eight elements: WPD, EWHO, RLO, water depth (WD), distance to coast (DC), extreme wave height (EWH), coefficient of variation (Cv), and monthly variability index (Mv). Resource characteristic elements include WPD, EWHO, RLO, Cv and Mv, environmental risk elements include EWH, and cost effectiveness elements include WD and DC. A case study was carried out with the macro-scale classification of the Maritime Silk Road and the micro-scale classification of the Sri Lanka waters, as shown in Figure 3. This is the first time to realize the classification of the Maritime Silk Road at home and abroad. At the same time, the wave energy classification under the scenarios of commercial development, emergency power supply and tremendous power consumption was also carried out. This scheme has strong flexibility, which can adjust, add or delete the weight of relevant elements according to the actual demands. In addition, it can be widely used in classification and siting design of new energy under different demands.

In 2019, Zheng et al. [30] used the Coupled Model Intercomparison Project Phase 5 (CMIP5) wind data to drive the WW3 wave model to obtain the future wave data in China Seas, and then analyzed the future wave energy characteristics. Offshore China is divided into four study areas: north of the East China Sea, middle of the East China Sea, south of the East China Sea and the East China Sea. With the analysis of the WPD, EWHO, ALO and RLO, the advantages and disadvantages of the future resources in the above areas are analyzed and compared.

Fairley et al. [31] have conducted good work on the global coastal wave energy classification based on multivariate clustering. The elements considered in the classification include SWH, spectral peak period, coefficient of variation, extreme wave height with a 50-year return period, spectra, directionality, etc. The results show that the coasts with the highest energy are located in the areas influenced by large long-term swell and storm conditions and suggest that swell-dominated seas with low risk and variability are more suitable for developing resources. This work provides an important reference for wave energy development and exploitation of nearshore.

Kamranzad and Hadadpour [32] have conducted good work on local wave energy classification. They proposed a multi-criteria approach that combines climate change, exploitable storage, construction cost and availability to assist decision makers to plan for a pilot project. The results show that the individual criterion nominates different locations and wave energy converters (WEC) as the most suitable ones.

Referring to Zheng et al. [29], Ribeiro et al. [33] carried out the wave and wind energy classification of the NW Coast of the Iberian Peninsula through the Delphi method, with the comprehensive consideration of resource characteristics (availability and stability), environmental risks (extreme events) and cost effectiveness (WD and DC). The results show that most of the NW Iberian Peninsula presents good conditions for harvesting energy from wind and wave resources simultaneously. In particular, there are some optimal areas such as the areas near Cape Roca and the Galician coast.

Martinez and Iglesias [34] created a wave exploitability index (WEI) and carried out global wave energy classification. WEI is defined as divided $\overline{H_{rms}}$ by $\overline{H_{\max }}$. $\overline{H_{rms}}$ is the mean value of the root-mean-square wave height. $\overline{H_{\max }}$ is the maximum individual wave height over the period. The larger the WEI, the more conducive to resource development. Results show that: Class II and III mainly occur at low and low-middle latitudes, and Class IV and V mainly occur at middle and high latitudes. The greatest mean wave power values occur in the Southern Geoscience. Class V is specific to the Southern Ocean—an uninterrupted fetch that circumambulates the globe. This work has a great significance for the macro-scale site design.

Lavidas and Kamranzad [35] devised a Wave Energy Development Index (WEDI) to describe the exploitable degree of wave energy. The algorithm for the WEDI is $\mathrm{WEDI}={\overline{P}}_{wave}/J_{wave}$. ${\overline{P}}_{wave}$ is the annual average of WPD and $J_{wave}$ is the maximum of WPD. If ${\overline{P}}_{wave}$ is high but not very large in an area, it will indicate the area is rich in resources with low destructiveness, which is conducive to resource development. In other words, the higher the WEDI, the more favorable the resource exploitation and vice versa. They calculated the WEDI of global waters with two reanalysis which are ERA5 and ERA-Interim, respectively. The results show that the WEDI of mid-latitude waters is generally higher and that of the high-latitude waters is generally lower.

Ahn et al. [36] presented an energy classification system for global coastal wave sites based on three available wave power zonings with five parameters: zoning 1 based only on the total omnidirectional (frequency and directionally unconstrained) wave power, zoning 2 based on the frequency constrained wave power and its dominant peak period band containing this power and zoning 3 based on the frequency-directionally constrained wave power and its dominant peak period band containing this power. Combining all three zonings results in a five-parameter zoning matrix that delineates distinct global wave energy resource classes. This is a positive contribution to the coastal wave energy development.

In 2021, with the dynamic adaptive wave energy classification scheme created by [29], Zheng [37] carried out a global wave energy classification, comprehensively considering resource characteristics, environmental risks and cost effectiveness. The classification index of global oceans is shown in Figure 4 (weighting of the expert assessment without special needs). The classification under three scenarios of availability, cost effectiveness and resource output is shown in Figure 5. The results show that: (1) The global wave energy is generally operable and most of the sea areas belong to resource-rich areas (energy class is above 4). (2) The spatial distribution of classification shows a significant difference under diverse demands. In the scenario of commercial development (focusing on cost), the extent of poor areas is much smaller than that in the other two scenarios (huge electricity consumption and emergency power supply). In the scenario of huge electricity consumption (focusing on resource enrichment), it is much greater than that in the other two scenarios (commercial development and emergency power supply). In the scenario of emergency power supply (focusing on resource availability), the extent of dominant areas above class 6 is the largest. In the scenario of commercial development (focusing on cost), the extent in areas above class 6 is the smallest. (3) Common points of the classification under different demands: the energy class of the southern hemisphere is higher than that of the northern hemisphere, and the energy class of the eastern ocean is better than that of the western ocean. The relatively inferior areas are always distributed in the Arabian Sea, Bay of Bengal, the sea area west of the line connecting Kamchatka Peninsula-Kuril Islands-Ogasawara Islands and New Guinea Islands, the western North Atlantic Ocean at the middle and low latitudes and the polar regions.

In 2021, Zheng [37] launched the first comprehensive wave energy resource dataset (WERD) and took the Maritime Silk Road as an example. Dynamic adaptive energy classification is the key component of this dataset. This dataset extracts a series of key indicators closely related to the wave energy development from raw big wave data including the temporal-spatial distribution characteristics and the classification, historical variation, long-term projection, swell characteristics, short-term projections, climate features and key points of wave energy characteristics.

Currently, the marine big data widely used in the world mainly belongs to raw data with fatal weaknesses of big storage and low information density, which leads to long cycles and low efficiency of the marine construction. Extracting key information from massive raw data and establishing an application of big data is the key point of industrialization and scaling of marine renewable energy and is also an urgent challenge for developed countries to overcome. The WERD in global oceans launched by Zheng [37] is a typical application of big data of marine renewable energy. It can provide comprehensive data support for the evaluation and development of resources, which is conducive to promoting the industrialization and scaling of energy projects such as generation.

Advantages and disadvantages of relative maturity stages: The energy classification at this stage can realize a more comprehensive consideration of elements (including WPD, EWHO, RLO, Cv and Mv) and resource characteristics, environmental risks and cost effectiveness and can also better reflect the regional differences. However, there are still two shortcomings: (1) Insufficient prediction of future energy classification to support long-term planning layouts. (2) Energy classification for each month has not been carried out, which makes it difficult to meet the needs of all-time development.

2. Bottlenecks in Wave Energy Classification

Many active attempts have been made and much progress has been obtained in wave energy classification. However, up to now, the research has been still extremely scarce, which is the basis for decision making on the macro-scale optimized layout, micro-scale accurate site selection and the roadmap design of wave energy. There are six major bottlenecks in the current classification: (1) Inconsistency with physical mechanisms results in inaccurate site selection. (2) Failure to comprehensively consider resources, risks and costs makes it difficult to support the macro-scale optimized layout. (3) Inability to meet the needs of diverse tasks results in a lack of universality. (4) Inapplicability in some seasons/months makes it difficult to meet the needs of all-time development. (5) Without notable regional differences in wave energy class, it is difficult to guide micro-scale accurate site selection. (6) Ignoring the long-term projection of wave energy class makes it difficult to support long-term planning and layout. In response to the above bottlenecks, this study put forward corresponding countermeasures.

2.1. Inconsistent with Physical Mechanisms

The wave energy classification in the mid-term development stage mainly refers to wind energy classification [22], and the results of this stage show that most of the low-latitude waters of the Indian Ocean and the mid-low latitude waters in the Pacific Ocean belong to class 3, with no significant difference between the eastern and western waters. However, the results in the relatively mature stage [37] show that at the low latitudes of the Indian Ocean, the wave energy class in the east of the ocean is significantly higher than that in the west and at the mid-low latitudes of the Pacific Ocean, the class in the east is significantly higher than that in the west. Which scheme is more scientific? Which one is more consistent with physical mechanisms? In the Indian Ocean, strong swells excited by the south Indian westerlies usually propagate along a northeasterly path to the Bay of Bengal waters, bringing abundant wave energy to the central–eastern part of the Indian Ocean. In the Pacific Ocean, strong swells excited by the south Pacific westerlies usually propagate to the Hawaiian waters and the Cologne Islands (equatorial eastern Pacific Ocean), bringing rich wave energy to the central–eastern part of the mid-low latitudes of Pacific Ocean. Apparently, the classification scheme is more consistent with the physical mechanisms in the relatively mature stage, while it underestimates the resources in the eastern part of the low latitude of the Indian Ocean and the east–central part of the mid-low latitude of the Pacific Ocean in the mid-term development stage.

2.2. Failure to Comprehensively Consider the Resources, Risks and Costs

For a long time, the traditional wave energy classification scheme has played an important role in promoting projects such as wave power generation and seawater desalination. However, it only considers some resource characteristics (mainly WPD, SWH and availability). The lack of a definition of the energy level occurrence leads to this scheme failing to take into account the abundance of the resource. Additionally, resource stability related to the efficiency of energy collection is also not considered. In addition, the scheme does not take into account a series of key indicators that are closely related to power generation costs and environmental risks such as WD, DC, EWH, etc. In short, it mainly considers some resource characteristics but fails to achieve a comprehensive consideration of resource characteristics, environmental risks and cost effectiveness. In addition, it does not consider the energy crisis brought by extreme weather and regional conflicts, which is unfavorable for leveraging new energy to improve the ability to solve them.

2.3. Inability to Meet the Needs of Diverse Tasks

In the actual wave energy development, if the specific needs are different, the focus on each element is also different in the classification. It is very misleading to use fixed results to guide site selection for different development needs. The following are three typical scenarios: (1) Commercial development. WD and DC, which are closely related to the difficulty of network connection and investment costs, tend to receive much attention. Therefore, the weight of them should be appropriately increased. (2) Emergency power supply. In the resource development, much attention is paid to the resource availability to enhance the power self-sufficiency as much as possible under the demand of emergency power supply of islands and reefs. Therefore, its weight should be increased appropriately. (3) Huge power consumption. During the construction of some major deep-sea projects, there is often a great demand for electricity. Additionally, much special attention is paid to the richness of resources, so it is necessary to appropriately increase its weight. However, the traditional classification scheme with fixed results can hardly meet the needs of diverse developments, which is prone to mislead the site selection design by referring to it step by step.

2.4. Inapplicability in Some Seasons/Months

The traditional wave energy classification is mainly a year-round classification scheme, the results of which can provide a good reference and guidance for wave energy devices. However, there are some tasks that only need to be carried out in some months or specific months, such as unmanned vehicles for short-term missions. In this case, the results of the classification usually are usually not instructive which creates an urgent need for monthly energy classification.

2.5. Without Notable Regional Differences in Wave Energy Class

After the macro-scale site selection for wave energy, the final resource development needs to be implemented in specific micro-scale sea areas, which requires a micro-scale wave energy classification in a small, concerned area to provide decision support for micro-scale accurate site selection. However, the existing energy classification is mostly macro-scale, whose results are difficult to guide micro-scale accurate site selection. The wave energy classification results based on traditional schemes have insignificant regional differences [22], and wave energy at mid-low latitudes is mainly class 3, which is difficult to provide fine guidance for its site selection. As the wave energy in the north Pacific Ocean westerlies mainly belongs to class 3–4 and in the north Atlantic Ocean westerlies mainly belongs to class 4–5, it is also difficult to provide detailed guidance for the accurate site selection of wave energy in the global westerlies. In addition, the wave energy in a large area of China’s offshore waters is classified as class 2, which also makes it difficult to provide detailed guidance. Therefore, there is an urgent need to create a classification scheme that fully reflects the regional differences in energy classes.

2.6. Ignoring the Long-Term Projection of Wave Energy Class

The existing classification is mainly carried out for historical resources, and the results can provide some references for the layout and site selection of resource development. However, the development will be carried out in the future, and it is more practical to carry out the future wave energy classification, so there is an urgent need to carry it out. Zheng et al. [30] and Costoya et al. [38] successively put forward an idea of the future wind energy classification and carried out the future offshore energy classification by using CMIP5 and Coupled Model Intercomparison Project Phase 6 (CMIP6) data to provide the technical approach. In 2021, Ribeiro et al. [39] carried out the wave energy classification of Iberian Peninsula waters by using WW3 and SWAN.

3. Countermeasures

To effectively deal with the six typical bottlenecks, a dynamic adaptive wave energy classification scheme is proposed, which can consider all elements, be suitable for diverse tasks, be available at all times and be suitable in all regions. On this basis, the concepts of absolute and relative class, the dynamic mapping of classification and future energy classification are proposed, with the expectation of promoting the industrialization and scaling of wave energy.

3.1. To Match with Physical Mechanisms

The scheme designed by Zheng [9] is suggested to solve the problem of inconsistency between classification results and mechanisms. In addition, this part of the analysis also reveals that the swell has an important influence on wave energy classification and that it is necessary to use swell as an important indicator for classification in the future.

3.2. To Integrate the Resources, Risks and Costs

In 2018, Zheng et al. [29] proposed a classification concept that covers all elements and established a scheme that can integrate three aspects: resource characteristics, environmental risks and cost effectiveness (as shown in Figure 6), specifically including eight elements (resource characteristic includes WPD, EWHO, RLO, Cv and Mv, cost effectiveness includes WD and DC and environmental risk includes EWH). Through this scheme and the Delphi method, the wave energy classification has been carried out in the global oceans, the Maritime Silk Road waters and the Sri Lankan waters. In the future, the Delphi method will be used in the classification, but not limited to this. Other methods such as hierarchical analysis will also be used.

3.3. To Be Suitable for Diverse Tasks

Zheng and Li [29] took advance in proposing the concept of dynamic adaptive energy classification (as shown in Figure 7), which is to reasonably and flexibly add or delete relevant elements according to the actual demands and to reasonably adjust the weights of them, so as to develop classification under diverse demands. This effectively solves the difficulty that traditional classification with fixed results can hardly meet the needs of diverse tasks. In addition, it is found in the analysis in Section 3.1 that the swell plays a significant role in the classification. At the same time, it has the characteristics of high energy and sound stability, which are conducive to the collection and conversion of wave energy, so it can be regarded as a resource indicator.

3.4. To Be Available in Each Seasons/Months

In 2017, Zheng et al. [40] proposed a concept of all-time energy classification (i.e., monthly energy classification) and used the ERA-interim wind data from ECMWF to achieve the first all-time global offshore wind energy classification (monthly energy classification). The results show that there are significant differences in energy classification between different months and also between monthly and annual classification. Therefore, it can be extremely misleading to use the annual classification results to guide wave energy development in a certain month. In the future, based on a dynamic adaptive energy classification scheme, the wave energy classification in global concerned oceans can be carried out for the whole year, seasons and months with reference to the concept of Zheng et al. [40], so as to achieve an all-time classification and solve the problem that the traditional classification scheme is not applicable in some seasons/months.

3.5. To Show Regional Differences in Wave Energy Class

With a scheme that covers all elements, Zheng and Li [29] and Zheng [9] have developed a macro-scale wave energy classification for global oceans (as shown in Figure 8), a mesoscale classification for the Maritime Silk Road and a micro-scale classification for the Sri Lankan waters. This study further proposes the concept of an absolute and relative class of resources: the result of the global wave energy classification is the absolute class and of the regional wave energy classification is the relative class. Repeat the steps in Section 3.2, and continuously focus on narrowing the area and revealing the relative resource class of the concerned sea area, until the micro-scale classification can meet the demand to provide decision support for micro-accurate site selection.

3.6. To Carry Out Long-Term Projection of Wave Energy Class

It is recommended to refer to the concepts of Zheng et al. [30] and Costoya et al. [38] to carry out the future wave energy classification under different demands, and to provide decision support for the long-term planning and layout of the resource development based on a dynamic adaptive energy classification scheme. The study of wave energy classification under different scenarios in the future should be systematically carried out, and the effect of climate changes on the class distribution (especially the distribution and mitigation of dominant areas) also can be revealed to effectively deal with the energy crisis caused by future extreme weathers. From the spatial perspective, it is necessary to cover the macro-scale classification of global oceans and the micro-scale classification of focus waters. From the temporal perspective, it is necessary to cover the annual and monthly classification in the future. For example, a wave energy classification of the global oceans over the future 40 years could be developed. The CMIP6 data can be used to drive the current international advanced wave models WW3 and SWAN, and the wave data for a long time series covering the global oceans can be obtained through simulation (as shown in Figure 9), and then the classification in the future can be carried out including the annual and monthly classification of the global ocean at the average state for the next 40 years (2025–2064), as well as the first decade (2025–2034), the second decade (2035–2044), the third decade (2045–2054) and the fourth decade (2055–2064), which are shown in

Table 4. Key elements of the future wave energy classification.

Decades	Seasons	Contents
The first decade (2025–2034)	Annual energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
	January–December energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
Second Decade (2035–2044)	Annual energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
	January–December energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
Third decade (2045–2054)	Annual energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
	January–December energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
Fourth decade (2055–2064)	Annual energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
	January–December energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
2025–2064	Annual energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters
	January–December energy classification	Global macro-scale classification
		Micro-scale classification of the focus waters

. It is also necessary to calculate and analyze the characteristics of interannual and interdecadal changes of resource class in the future and compare the distribution and changes of dominant areas.

4. Case Study of Wave Energy Classification

For easily understanding, the wave energy classification of the China seas was carried out as a case study by using the dynamic adaptive wave energy classification scheme proposed in this study. The detail process is:

Step 1, elements selection. Three aspects were considered including the resource characteristic, environmental risk and cost. The resource characteristic elements include WPD, EWHO, RLO, Cv and Mv. The environmental risk element is EWH. The cost elements include WD and DC.

Step 2, Standardization of each element. The ERA-interim data, global self-consistent, hierarchical, high-resolution shoreline database (GSHHS) coastline data and gtopo30 water depth data were used to calculate the WPD, EWHO, RLO, Cv, Mv, EWH, DC and WD. The min-max normalization method is used to standardize the above eight elements.

Step 3, weight assessment of each element. Well-known experts in the field of wave energy development are invited to evaluate eight elements of energy classification. The weights given by k experts are arranged into a matrix. Then, the weighted average is taken for the weights.

Step 4, calculation of energy classification index. With the weight set of each element and the standardized values of each element, the wave energy classification index at each grid point is calculated.

Step 5, calculation of relative class and classification. The classification result for global oceans is absolute class and that for local waters is relative class. The absolute class of wave energy in the China seas is presented in Figure 4 and Figure 5. Here, the relative class of wave energy in the China seas will be calculated. Based on the wave energy classification index obtained in step 4, the relative class was calculated by using min-max normalization method. At last, the relative class can also be divided into seven classes by using

Table 5. The absolute and absolute classification scheme of wave energy.

Class	Expectation Value (y)	Resource Potential	Grade Division
1	y ≤ 0.4	Poor	Indigent area
2	0.4 < y ≤ 0.5	Marginal	Available area
3	0.5 < y ≤ 0.6	Fair	Subrich area
4	0.6 < y ≤ 0.7	Good	Rich area
5	0.7 < y ≤ 0.8	Excellent
6	0.8 < y ≤ 0.9	Outstanding
7	y > 0.9	Superb

. Finally, the wave energy classification index y is divided into seven classes (as shown in Table 5), thus realizing the energy classification of the China seas, as shown in Figure 10.

From the absolute class of wave energy in the global oceans (Figure 4 and Figure 5), the regional difference of energy class in the China seas is not obvious. It is difficult to guide accurate micro-scale site selection in this region. From the relative class of wave energy (Figure 10), the regional difference in energy class is obvious, which can provide decision support for micro-accurate site selection in the China seas. In addition, it is found that there is an obvious difference in spatial distribution characteristics of energy classes under different focuses, which can meet the needs of diverse tasks.

5. Conclusions and Prospects

Reasonable energy classification is the scientific basis for the micro-scale accurate site selection of marine energy development, as well as the decision-making basis for the macro-scale optimized layout and the blueprint of development routes of a country’s marine energy. However, there has long been a lack of scientific energy classification schemes. The study found that several countries that are leaders in ocean energy development have not yet developed ocean energy technology roadmaps [41,42]. A large amount of work and contribution has been made in wave energy classification, but it is not enough to meet the practical needs of macro-scale optimized layout, micro-scale accurate site selection and blueprint of development routes. Based on the improvement of the indicators considered, this study firstly divides the global wave energy classification into three stages: the preliminary exploration stage, the mid-term development stage and the relatively mature stage and compares the main achievements, advantages and shortcomings of each stage. It is found that there are still typical bottlenecks in the current classification such as inconsistency with physical mechanisms, inability to meet the needs of diverse tasks, inapplicability in some seasons/months, etc.

In response to the difficulty that traditional scheme fails to comprehensively consider resources, risks and costs, this study proposes a classification concept that covers all elements to support the macro-scale optimized layout. The scheme in this study considers three aspects of resources, risks and costs, specifically including eight elements such as WPD, EWHO, RLO, Cv, Mv, EWH, WD and DC. In the future, the relevant elements can be adapted, added or deleted according to the specific needs of different tasks.

In response to the difficulty that the traditional scheme is unable to meet the needs of diverse tasks and lacks universality, this study proposes a dynamic adaptive energy classification concept, which can reasonably and flexibly add or delete relevant elements according to actual needs and adjust the weights, thus developing classification under diverse demands and effectively solving the problem.

In response to the challenge of no significant regional differences in wave energy class, which makes it difficult to guide micro-scale accurate site selection, this study proposes the concept of absolute and relative class of resources, with the classification result for global oceans as absolute class and that for local waters as relative class. The macro-scale classification in global oceans, the mesoscale classification in local waters and the micro-scale classification in focus waters are carried out. Constant attention is paid to the shrinking areas to reveal the relative resource class of the focus areas until the micro-scale accurate site selection can be met, which effectively solves the problem.

In response to the traditional scheme focusing only on annual energy classification, which is not applicable in some seasons/months and can hardly meet the needs of all-time development, this paper proposes the concept of all-time energy classification to carry out classification in all time based on the dynamic adaptive energy classification scheme.

In response to the problem that the long-term projection of wave energy class is not considered enough to support long-term planning and layout, it is suggested to systematically carry out wave energy classification under different scenarios in the future to reveal the effect of climate changes on the energy class distribution and effectively deal with the energy crisis caused by future extreme weathers. Additionally, the interannual and interdecadal variation characteristics of future energy classes should be further calculated and analyzed. The distribution and changes of dominant areas need to be compared to support the long-term energy planning and layout as well as the energy roadmaps.

Overall, the dynamic adaptive wave energy classification scheme proposed in this study, which can consider all elements, be suitable for diverse tasks, be available at all times and be suitable in all regions, is more consistent with physical mechanisms than the traditional schemes. In the future, it is necessary to further develop classification standards for regulation and to further develop a dynamic mapping covering global and key waters, at all times (all year round and all months, historical states, different future emission scenarios), and macro-scale and micro-scale classification of marine renewable energy under different demands, so as to accumulate first-hand basic information and theoretical reserves for the energy development and actively contribute to the “double carbon” goal and sustainable development.

In the actual wave energy classification, the following steps are suggested: the first step is to target the task. Once it is defined, the weights of the relevant elements are reasonably adjusted according to the different tasks, and then an all-element, all-time and dynamic adaptive energy classification scheme can be carried out. The second step is to lock the time. If the wave energy device is deployed all year round, the annual energy classification will be developed, and if it is deployed in some seasons or months, the classification for the corresponding seasons or months will be developed. The third step is to target the sea areas. Firstly, global wave energy classification is developed, and the result is the absolute class to obtain the absolute class of wave energy in the focus areas. Then, the wave mesoscale classification of the local sea area is further developed to obtain the relative energy classes within the sea area. If the regional differences in resource class are sufficient to meet the requirements of site selection, the next step can be taken. If they are still not significant, the sea area continues to be reduced and a micro-scale classification of small areas is carried out to obtain the relative wave energy class until the requirements can be met.

With the rapid development of science and technology, wave energy converters have also developed rapidly [43,44,45]. In the future, it is also necessary to utilize some advanced devices and place them in areas with different energy classes to fully test the reliability of the wave energy classification scheme proposed in this study.

Finally, it is hoped that this study could provide a scientific basis and auxiliary decision-making for the macro-scale optimized layout, micro-scale accurate site selection and development roadmaps to further promote the industrialization and scaling of wave energy development and provide references for other marine energy development, in the hope of contributing to the sustainable development of mankind and the “double carbon” goal.