1 Introduction

During the almost three hundred years of industrialized process, a huge number of fossil energy has been exploited and utilized, which gave a profound promotion to development of the global economy and human society.However, the problems are more and more prominent that the nonrenewable fossil energy is nearly exhausted, along with the phenomenon of environmental pollution and climate change. Therefore, it is obvious that human society can hardly survive with the traditional method of energy utilization, and the energy revolution is extremely urgent[1-4].

One important part of the energy revolution is the widely using of renewable energy resources, for example,wind energy and solar energy. The renewable energy resources have unexhausted reserves all over the world and are friendly to the environment, which could satisfy the increasing energy demand of human society. Nevertheless,the allocation of renewable energy resources is pretty unbalance among countries and regions, for example,there are abundant wind energy resources in the Arctic,which has few energy demands. Moreover, such a natural distribution is unchangeable, which cause more challenges to the optimal utilization of renewable energy resources [5].

Based on the background above, the concept of “Global Energy Interconnection (GEI)” has been proposed, which has aroused more and more attention around the world.The GEI aims at developing a global power grid, which can transmit the electrical power converted from renewable energy resources and other forms of energy all over the world, become the vital hub of the energy allocation optimization, and promote the smooth progress of energy revolution. The significant features of GEI mainly include[6]:

1) Extensively interconnected. This is the most fundamental feature of GEI, which has plentiful meaning.From microgrids to regional distribution networks,intracontinental power grids to intercontinental power grids, every part no matter huge or small is connected closely and develops coordinately, making a complete system of GEI.

2) Strong and Smart. The power grids are the most complex dynamical system, and GEI is the highest form.The flexible and stable operation, as well as the robust adaptation and self-healing capability of GEI, both need the most advanced techniques to support and make the GEI strong and smart enough.

Fig. 1 shows the diagram of economic dispatch for global energy interconnection. In each country, there are thermal units, renewable energy generation and load,which are connected as a national power system through transmission lines. Between the countries, there are the tielines to transmit electrical power and exchange dispatching information, which realize the interconnection of national power systems. The optimization problem of DED for GEI is formulated as follows.

检测条件：洗脱液为柠檬酸缓冲溶液（pH值3.3～4.9），显色液为茚三酮∶乙二醇甲醚∶乙酸钠缓冲液=2∶75∶25，除羟脯氨酸在440 nm检测外，其他氨基酸均在570 nm检测。

Among the studies of GEI, economic dispatch (ED) is a basic and vital content. Generally speaking, the ED is an optimization problem aims to minimize the operation cost of power system while satisfying various constraints. From the view of time scale, the ED problem can be divided into two kinds: static economic dispatch (SED) [7], [8] and dynamic economic dispatch (DED) [9-11]. The SED only considers the ED problem with one single time period,i.e. the connections among different time periods are omitted. Oppositely, the DED fully considers the coupling relationship of different time periods, for example, the ramp limits of units. Despite that the DED is more complex, it can better reflect the reality of system operation, and get more practical results [12]. In the GEI, the penetration of renewable energy generation is large, which bring more volatility to the system and make a closer connection of time periods. If SED is applied to the GEI, probably the dispatching results will be infeasible. Thus, DED is a better choice for GEI and used in this paper.

由训练误差曲线可以看出，两个CNN框架在训练迭代15 000次后误差已基本达到收敛，其中CNN-1收敛速度较慢，可能的原因是多尺度模块的引入，导致梯度在反向传播时计算量增加。由测试正确率曲线可以看出，在迭代约24 000次后，CNN-1的测试正确率已超过CNN-2，且正确率大小整体趋于稳定。

1317 BICD cargo adaptor 1 gene as a biomarker of grade progress of glioma

As it is mentioned above, the ED is an optimization problem which has its objective function. In the traditional ED, the objective function mainly consists of the fuel costs of thermal units. However, this cannot meet the various requirements of GEI, for instance, the sustainability.Thus, multi-objective function [13], [14] should be introduced into the ED of GEI. One common method to handle the objective functions with different dimension is distributing certain weight coefficients to them and adding components into one single objective function. In this paper, the curtailment of renewable energy generation is put into the objective function to balance the economy and sustainability.

Assume the penalty coefficients areρ w =ρs = 50, and the capacities of transmission lines and tie lines are large enough, i.e. there is no line blocking problem. Solve the isolated model without tie-lines and the connected model with tie-lines respectively, and the dispatching results are shown in Table 1. Obviously, the connected mode has great superiority compared with the isolated mode, which both decreases the renewable energy generation curtailment rates and saves lots of operation costs. Under the isolated operation mode, for the renewable energy bases like Mongolia, the permitted wind power and photovoltaic power are far larger than the demand, and these clean energies can only be curtailed without tie-lines; for the load center like Japan, the outputs of thermal units have to increase to meet the load requirement. Thus, it is verified that the GEI is highly beneficial to the economy and sustainability of human society.

The rest of this paper is organized as follows. Section 2 gives the model of DED for GEI, which including the parts of renewable energy generation. Section 3 presents the case studies of the application on the transnational energy interconnection of Northeast Asia, and the conclusion is followed in Section 4.

那么，学科思维有什么特性呢？首先，学科思维具有较高层次的抽象性；其次，学科思维的获得过程具有长期性，不可能一蹴而就，必须经历长时间、系统而复杂的学习活动和心理过程才能获得；再次，学科思维还具有社会性[2].从中我们可以得到启示，学科思维的培养具有长期性、深刻性，它需要有“长度”的教学，更需要有“深度”的教学.

2 Problem formulations

绩效评价应当涵盖农业科研项目的全过程，即活动的投入、过程、产出以及活动的效果，将绩效评价视为连续、系统的过程，注重项目全过程考核。所谓投入类指标是反映农业科研投入的人力、物力和财力的变量，如农业科研经费数、专项主持人情况等。过程类指标是反映农业项目研究过程中对预算执行质量的控制，如该项目是否按进度开展，财政资金的使用有无浪费等。

Fig. 1 Diagram of economic dispatch for global energy interconnection

2.1 Objective function

5) Maximum transmission capacity of lines and tie-lines

In fact, the weight coefficient ωp can be fixed asω p=1,and the other two weight coefficients change according to proportion, which can reduce the number of parameters and guarantee obtaining the same results. The objective function is rewritten as:

where ρw andρs are the proportionally changed weight coefficients of wind power curtailment and photovoltaic power curtailment. The two terms withρwand ρshave clear meanings, which can be considered as penalty terms to balance choice tendency between economy and sustainability.

2.2 Constraints

1) Power balance equation

There might be tie-lines faults between two countries for various reasons, such as political factors and natural disasters. Table 2 gives the dispatching results when the tie-lines faults occur between some two countries. When the tie-lines faults occur between China and Mongolia,the renewable energy generation curtailment rate rise up rapidly, and the total cost also increases. When the tie-lines faults occur between South Korea and Japan, the total cost rises up greatly. However, because there is a loop network among China, North Korea, and South Korea, as for other tie-lines faults, the dispatching results remain unchanged.Therefore, more loop networks in the GEI can improve its reliability and security.

3) Output bounds of wind farms and photovoltaic plants

4) Ramp limits of thermal units

The objective function of DED for GEI is divided into two parts. The first part is the same as traditional ED, which is composed of the operating costs of thermal units. The fundamental requirement of power systems, i.e.the economic efficiency, is reflected in the first part. The second part mainly considers the sustainable character,which consists of the renewable energy curtailment. The objective function is given as:

The outputs of wind farms and photovoltaic plants are considered with enough flexibility, which can be arbitrarily controlled within the maximum range. Thus, there is no ramp limit of wind farms and photovoltaic plants. In addition, the permitted transmission capacity of tie-lines not only depends on the electrical characteristics, but also the agreement between the countries.

裁判的判罚会偶尔失误，判罚错误或误判，这种失误有时会影响运动员比赛的心情，更严重时候，如在关键分时候，有可能会改变比赛的结果。

3 Case studies

The development of GEI is planned as three steps:intracontinental interconnection, intercontinental interconnection, and global interconnection. Nowadays, the GEI is in the early stage, and the transnational energy interconnections are in the ascendant, among which the Northeast Asia plays an essential role.

3) Open and interactive. The well cooperation of countries is of great necessity for the GEI. Thus, the construction and operation of GEI must be open and equal for each country. In addition, the load side should also participate actively in the GEI to realize the two-way interaction between the grid side and load side.

Fig. 2 gives the simulation framework of the transnational energy interconnection in Northeast Asia. As shown in the figure, there are mainly five countries and regions included in the Northeast Asia: Mongolia, North and Northeast China, North Korea, South Korea and Japan.Broadly speaking, Mongolia and North and Northeast China are major wind power and solar power bases, and North and Northeast China, South Korea, and Japan are main load centers. In this case, 80 nodes are constructed according to the partition of small region power grids in China and geographical position relationships in other countries, with 9 tie-lines combining these countries. The time periods studied totally is 168 hours in a week. In addition, the countries are located in different time zones,and a time standard (Beijing time in this case) need be chosen for time conversion.

Fig. 2 Transnational energy interconnection of Northeast Asia

Table 1 Dispatching results of isolated and connected operation

Operation mode Total curtailment rate Total cost(108¥)Wind Photovoltaic Isolated 15.33% 24.23% 4.3489 Connected 0.00% 0.00% 2.6580

某高速公路通车以来交通量逐年增加，部分沥青路面先后出现各种病害，尤其在连续降雨的影响下，路面病害的扩展速度显著加快，经研究决定，采用冷补沥青修补料对该高速公路发生严重坑槽病害的路段进行修补。

新型浸梗机中烟梗由刮板式输送机构强制阻挡，从而前进速度缓慢可调，使除去杂质后的烟梗在40℃～80℃循环水中强制浸泡20～120s(可调)。浸泡后烟梗温度与水温能达到较好的平衡，同时烟梗含水率达到36.02%，能满足后续加工要求，增温增湿能力较强。表1为传统的水洗梗和新型的浸梗机应用后梗丝质量的对比分析。

Fig. 3 Total energy transmission in Northeast Asia (TWh)

Furthermore, the total energy transmission results of the connected mode are presented in Fig. 3. It can be concluded that the power flow direction is from the renewable energy bases to the load centers, which realizes the energy complementary and optimal allocation. Moreover, the forecasted maximum electrical energy from wind and photovoltaic resources in Mongolia is calculated as 3.58 TWh, which is just a little larger than the sending-out electrical energy (3.40 TWh). That is to say, the renewable energy resources in Mongolia are abundant but the loads are light, which accords with the characters of power system in Mongolia.

Table 2 Dispatching results with tie-line faults

C=China, M=Mongolia, N=North Korea, S=South Korea, J=Japan

Tie-lines faults location Total curtailment rate Total cost(108¥)Wind Photovoltaic None 0.00% 0.00% 2.6580 C-M 14.81% 24.74% 2.8758 C-N 0.00% 0.00% 2.6580 C-S 0.00% 0.00% 2.6580 N-S 0.00% 0.00% 2.6580 S-J 0.00% 0.00% 3.2094

2) Output bounds of thermal units

Generally speaking, the renewable energy resources and load demands have obvious seasonal characteristics.For the wind energy, because of the monsoon climate in the Northeast Asia, the average wind speed is larger in spring and winter than that in summer and autumn. For the photovoltaic energy, which is mainly influenced by the solar light intensity, it rises up to the maximum in summer and falls down to the minimum in winter, which gets a medium value in spring and autumn. For the load demand,the peaks appear in summer and winter with the reason of cooling and heating supplies. Base on the analysis above,the cases of four different seasons are constructed and the results are shown in Table 3.

Table 3 Dispatching results of renewable energy generation in different seasons

W=Wind, P=Photovoltaic, S=Summary

Season Generation energy (TWh) Generation percentage W P S W P S Spring 10.17 8.04 18.21 14.26% 11.28% 25.54%Summer 7.82 12.06 19.88 7.84% 12.08% 19.92%Autumn 7.82 8.04 15.86 10.97% 11.28% 22.25%Winter 11.74 4.02 15.76 13.71% 4.70% 18.41%

As it is presented in Table 3, the total generation energy of wind power and photovoltaic power is larger in spring and summer, which is smaller in autumn and winter. However, the generation percentage shows different feature, which is larger in spring and autumn,while smaller in summer and winter. This is a result from the seasonal fluctuation of load demand. Thus, these two statistical standards of renewable energy generation should both be considered in practice to obtain a more reasonable evaluation. Moreover, it can be observed that the wind power and photovoltaic power have complementarity to some degree, which is beneficial for the stable operation of power systems. For example, from autumn to winter, the wind power rises up yet the photovoltaic power goes down,but their summary almost maintains unchanged. Therefore,it is important to develop the wind farms and photovoltaic plants coordinately in the GEI.

Table 4 Dispatching results with different wind power penalty coefficients

Wind power penalty coefficient pw Total cost(108¥)0 12.11% 3.0304 50 11.27% 3.0377 100 10.89% 3.0489 500 10.49% 3.0805 1000 10.46% 3.0901 Wind power curtailment rate

Table 4 gives the dispatching results with different wind power penalty coefficient in the objective function. In order to highlight the influence of pw , the outputs of photovoltaic plants are assumed as zero, the forecasted outputs of wind farms are reconstructed with greater volatility, and the maximum ramp rate of thermal units are decreased to 15%of the capacity per hour. It can be observed that when pw increases, the wind power curtailment rate is reduced but the total cost of the system gets larger. This is because that for the better accommodation of wind power, the thermal units have to change the operation mode to obtain more total ramping ability, which goes against to the economy.In conclusion, these results validate that the economic efficiency and sustainable capability are often opposite, and pw reflects the selection preference between them. Similar analysis can be put onto ps and the conclusion goes the same.

4 Conclusions

This paper mainly proposes a dynamic economic dispatch model for the GEI, and the renewable energy generations of wind power and photovoltaic power are consider in it. The case studies which apply the model to the transnational energy interconnection of Northeast Asia validate the huge benefits of GEI. Further conclusions from the case studies are:

坚持用习近平新时代中国特色社会主义思想武装干部头脑，教育引导党员干部牢固树立“四个意识”，坚定“四个自信”。坚决维护习近平总书记的核心地位，坚决维护党中央权威和集中统一领导，在急难险重任务中加强政治训练，确保系统内干部政治上绝对可靠、对党绝对忠诚，确保各项工作始终沿着正确政治方向前进。

1) More loop networks among countries in the GEI can improve its reliability and security, especially when the tielines faults occur;

2) It is beneficial to develop the wind farms and photovoltaic plants coordinately in the GEI, which can help the stable operation of power systems;

3) The penalty coefficients of renewable energy generation curtailment reflect the selection preference between economy and sustainability, which should be chosen properly according to the reality. Specifically, when economy is preferred, the penalty coefficients should be small; on the contrary, when sustainability weighs more,the penalty coefficients should be large.

Acknowledgements

This work was supported in part by National Key Research and Development Program of China(2016YFB0901900), in part by the Science and Technology Project of SGCC-Research on Grid Dispatching and Transaction Mode for Global Energy Interconnection, in part by China Postdoctoral Science Foundation(2017T100748), in part by Natural Science Basis Research Plan in Shaanxi Province of China (2016JQ5015)and in part by the project of State Key Laboratory of Electrical Insulation and Power Equipment in Xi’an Jiaotong University (EIPE17205, EIPE16301).

References

[1] Dong Z, Zhao J, Wen F et al (2014) From Smart grid to Energy Internet. Automation of Electric Power Systems, 38(15): 1-11

[2] Sun H, Guo Q, Pan Z (2015) Energy Internet: Concept,Architecture and Frontier Outlook. Automation of Electric Power Systems, 39(19): 1-8

[3] Bie Z, Wang X, Hu Y (2017) Review and Prospect of Planning of Energy Internet. Proceedings of the CSEE. doi: 10.13334/j.0258-8013.pcsee.171188

[4] Yamaguchi S, Iitsuka T, Osada M et al (2018) Asian international grid connection and potentiality of DC superconducting power transmission. Global Energy Interconnection, 1(1): 11-19

[5] Tian S, Luan W (2015) Technical Forms and Key Technologies on Energy Internet. Proceedings of the CSEE, 35(14): 3482-3494

[6] Liu Z (2015) Global Energy Internet. China Electric Power Press, Beijing, pp 209-210

[7] Rooijers F, Amerongen R (1994) Static Economic Dispatch for Co-Generation Systems. IEEE Transactions on Power Systems,9(3): 1392-1398

[8] Wood WG (1982) Spinning Reserve Constrained Static and Dynamic Economic Dispatch. IEEE Transactions on Power Apparatus and Systems, PAS-101(2): 381-388

[9] Borghetti A, Bosetti M (2010) Short-Term Scheduling and Control of Active Distribution Systems with High Penetration of Renewable Resources. IEEE Systems Journal, 4(3): 313-322

[10] Gaing Z (2003) Particle Swarm Optimization to Solving the Economic Dispatch Considering the Generator Constraints.IEEE Transactions on Power Systems, 18(3): 1187-1195

[11] Wang C, Shahidehpour SM (1993) Effects of Ramp-Rate Limits on Unit Commitment and Economic Dispatch. IEEE Transactions on Power Systems, 8(3): 1341-1350

[12] He P, Ai X, Zhao J (2013) Survey on Economic Dispatch Optimization Model for Microgrid. Shaanxi Electric Power,41(7): 68-73

[13] Liu W, Wen J (2015) Multi-objective Optimal Method Considering Wind Power Accommodation Based on Sourceload Coordination. Proceedings of the CSEE, 35(5): 1079-1088

[14] Chen J, Yang X (2013) Microgrid Multi-objective Economic Dispatch Optimization. Proceedings of the CSEE, 33(19): 57-66