Multi-objective optimal allocation strategy for the energy internet in Huangpu District, Guangzhou, China

Abstract

To improve the overall efficiency of the energy system, the basic structure for the energy internet of coordination and optimization of “generation-grid-load-storage” of Huangpu District, Guangzhou, China is designed, while the arrangement for the output of centralized and distributed energy module and energy storage are proposed. Taking economic benefit maximization, environmental benefit maximization and energy efficiency maximization as sub-objectives, the mathematical model of multi-objective optimal allocation and operation strategy of the energy internet is established considering supply-demand balance constraints, equipment characteristic constraints, operation mode constraints, and energy conditions constraints. The calculation results show that without considering the outsourced electricity, the balanced strategy, the economic development strategy, the environmental protection strategy, and the energy efficiency strategy are obtained by calculation, which are all superior to the traditional energy supply strategy. Moreover, considering the outsourced electricity, the proportion of outsourced electricity to total electricity is 19.8%, which is the system optimization of the energy internet under certain power demand. Compared with other strategies without outsourced electricity, the outsourced electricity strategy can have a certain emission reduction effect, but at the same time reduce the economic benefit. Furthermore, the huge difference in demand for thermal and cooling load between industrial and commercial areas results in the installed capacity of gas distributed energy stations in industrial areas being nearly twice as large as that in commercial areas. The distributed photovoltaic power generation is allocated according to the proportion of the installed roof areas of photovoltaic power generation system in residential, industrial, and commercial areas.

Keywords

generation-grid-load-storage / energy internet / economy-environment-energy efficiency / multi-objective optimal allocation / mathematical model

Introduction

The energy internet is a kind of new development mode of energy industry, which is highly coupled with the production, transmission, storage, and consumption of various energies, with renewability, distribution, interconnection, openness, intelligence and other characteristics [¹]. It will promote energy utilization in the direction of optimizing structure, improving efficiency, saving energy, reducing emissions, and advancing energy production and consumption revolution. Besides, it will activate potential of each link and every factor in energy industry, and establish a new energy production and consumption system, and a new coordinated management system.

As an efficient and comprehensive energy utilization system for users, domestic and foreign scholars have had preliminary studies on the application scenarios of the energy internet, among which the studies on the global energy internet and the park energy internet are relatively comprehensive. The discussion about establishing a global energy internet is mainly at the macro level, referring to energy configuration, energy planning, standard system, legal rules, environmental benefit evaluation and so on [²–⁵]. Comparatively speaking, the exploration of the park energy internet is more microcosmic. In combination with the specific circumstance of energy equipment or project construction in the park, proposing plans, building models, and making assessments [⁶–⁸] are to optimize the allocation of energy units in the park. The research content of the urban energy internet is relatively not sufficient. The existing literatures mainly focus on the description of the concept, operating framework, development direction, and other modules of the energy internet [⁹–¹¹]. There are still deficiencies in the discussion of collaborative optimization, operation mechanism, and implementation construction of energy internet.

The ultimate realization of the energy internet is inseparable from each sub-system and energy device in the energy system. Therefore, in order to solve the above problems, the research can be implemented into the concrete energy infrastructures in cities and the connection with the energy internet should be considered so that the results have a higher popularization value and maneuverability. In the section of energy supply, Xia et al. [⁶] demonstrated the role of natural gas power generation in the energy internet from the three dimensionalities of reliability, economy, and environmental protection. Ding et al. [¹²] conducted independent fine modeling of CCHP system equipment, based on which, discussed the optimal economy operation model of the energy internet. Liu et al. [¹³] designed a cogeneration system which is able to realize the comprehensive utilization of photovoltaic/thermal solar energy while meeting the system need of cooling, heating, and electricity load. In the section of energy transmission, Ren et al. [¹⁴] considered the collaborative optimization of energy transmission network. In the section of energy utilization, Zhang et al. [¹⁵] analyzed the load characteristic of the integrated energy system with cooling, heating, and electricity. In the section of energy storage, Guo et al. [¹⁶] proposed an optimal selection model for multi-class heterogeneous electrical energy storage systems based on the multi-attribute decision method. In general, the above studies discussed the optimization of the system from the entity each part of the “generation-grid-load-storage” of the energy internet. However, there is still a lack of research on the integration of all links and total factors of energy internet as well as the overall optimization of the energy internet. Moreover, most of the collaborative optimization studies have been conducted just from the perspective of economy, while considerable factor appears somehow single.

Combing each sub-system and composition of the existing energy system and comprehensively considering “generation-grid-load-storage,” based on the structure and operation mode of the energy internet in Huangpu District of Guangzhou, China, using the multi-objective nonlinear and linear programming methods, and taking the economic benefit maximization, the environmental benefit maximization and the energy utilization efficiency maximization as goals and assigning a certain weight coefficient to each goal, a mathematical model of multi-objective optimal configuration and operation of the energy internet in Huangpu District is established. From excluding the outsourced electricity and including the outsourced electricity these two aspects to calculate and analyze, the optimal energy technology portfolio strategies in a specific scenario are respectively acquired, which provides decision support for the programming and construction of the energy internet.

Overall design of the energy internet scheme in Huangpu District, Guangzhou

Practical basis and conditions for building the energy internet in Huangpu District, Guangzhou

Huangpu District (hereafter called Huangpu) is an important economic growth pole of Guangzhou with its vast economies of scale, whose GDP ranks second in Guangzhou. Since 2010, it has kept the total energy consumption above ten million tons of standard coal every year, far more than other administrative districts of Guangzhou. Vast economies and huge energy consumption make the benefit from the implementation of the energy internet remarkable. Besides, the urgent pursuit of energy efficiency and cleanliness makes the construction of the energy internet valuable and indispensable.

Furthermore, Huangpu has ample elements for the construction of the energy internet, including lots of energy facilities such as large power plants, thermal power plants, photovoltaic power generation, gas distributed energy stations, distribution network of various voltages, hydrogen refueling stations, charging stations, heating network and so on. The basic hardware condition for “generation-grid-load-storage” of multi-energy flow synergy is complete. Therefore, the energy internet framework has theoretically been built.

Basic structure and operation mode of the energy internet in Huangpu

Basic structure

Under the framework of the energy internet, the energy supply system in Huangpu will present a combined form of distributed and centralized energy module (Fig. 1). Relying on coal power, NG power, garbage power, and NG cogeneration to form a centralized energy module, some of the heat and electricity are provided to the users through the main heating network and the main power grid, some of the heat and electricity are transmitted to distributed energy module through the micro heating network and the micro power grid, and the rest of the energy flows into the energy storage module. Photovoltaic power generations, gas distributed energy stations, geothermal heating, and hydrogen fuel cell, which are directly oriented to users, with the characteristics of small scale and in situ production and energy supply, are configured as a distributed energy module. The distributed energy module is mainly to meet the personalized energy requirements of residential areas, industrial areas, commercial areas and other kinds of users. It is not only closely combined with the energy storage module, but also a powerful supplement of centralized energy module.

The energy internet can horizontally realize the multi-generation complementarity. Coal, natural gas, biomass, and other energy resources are complementary and coordinated in the centralized energy module, which highlight the substitutability among various energies [¹⁷]. Besides the centralized supply of energy, users can also choose different distributed energy combination plans according to their different energy requirements. The energy internet can longitudinally realize the “generation-grid-load-storage” interconnection, which can ensure the connectivity among the energy supply side, the energy transmission side, the energy demand side and the energy storage, and the effective delivery of the energy flow and information flow every part. The “generation-grid-load-storage” interconnection can also enhance the cooperative degree among each part of “generation-grid-load-storage” and effectively improve the efficiency and benefit of the whole energy system.

Operation mode

Generally, there are three typical components of the load duration curve of users. Most of the time, the system faces a relatively low load, called the base load which always exists. Sometimes, a high demand of load is faced in a short period of time, which is called the peak load, and usually exists for less than a fifth of a year [¹⁸]. Between the base load and the peak load, load demand usually rises gradually until it reaches the peak. The load during this period is referred to as the shoulder load (Fig. 2).

Under the framework of the energy internet, centralized energy module and distributed energy module are coordinated and supplemental. Due to the more extensive supply of the centralized energy module, the units need to operate all year round and usually do not have the ability to adjust the supply adopting to changing requirements, so the centralized energy module is suitable as the base load power plant. The distributed energy module cooperates with the orderly charge and discharge of the energy storage device, which not only suppresses the fluctuation and randomness of new energy power generation, but also effectively achieves peak load shifting. Flexibly responding to the change of shoulder load and peak load, the output can be adjusted frequently according to actual demand during operation of the distributed energy module and energy storage which play the role of the shoulder load power plant and the peak power plant (Fig. 3).

Research methods

Modeling approach

To achieve maximum economic benefit, the environmental benefit, energy utilization efficiency, and utilization of renewable energy, the development strategies and user configuration scheme can be scientifically and reasonably worked out by fully considering the climatic and geographic conditions, energy policies, emission and other information of Huangpu, and analyzing the various types of energy demand of Huangpu, based on equipment characteristics and energy market information of different regions, and then through building the multi-objective optimal allocation model for the energy internet, according to different development goals and regional requirements (Fig. 4). The mathematical model is extensive and complex. Therefore, the following assumptions about the model are made:

1) It is assumed that the power supply and heat source in Huangpu is only provided for the locality;

2) It is assumed that the same energy technology uses energy equipment with the same technical parameters;

3) It is assumed that all cogenerations and distributed energy stations use the plan of “fix heat/cooling based on power”;

4) It is assumed that in the centralized energy module, the benefits of NG cogenerations are significantly better than those of gas power generations. Therefore, only NG cogenerations are selected in centralized energy production with natural gas serving as its fuel;

5) It is assumed that since the application of hydrogen fuel cells and ground source heat pumps has not been popularized, only photovoltaic power generations and distributed energy stations are considered in the distributed energy module.

Objective function

Maximization of economic benefit

Based on the analysis of the cost-benefit of the total the energy internet in Huangpu, and by maximizing the economic benefits of the system, the objective function is established, as expressed in Eq. (1). The energy supply cost includes the equipment depreciation cost, operation and maintenance cost, and fuel cost. The depreciation cost includes the investment and construction costs of centralized energy equipment such as coal power, NG cogeneration, garbage power, and distributed energy equipment and energy storage equipment such as photovoltaic power generations and distributed energy stations. The operation and maintenance cost refer mainly to the regular maintenance costs of energy equipment. The fuel cost is determined by the purchase costs of coal, natural gas, and outsourced electricity.

(1)\mathrm{M}\mathrm{a}\mathrm{x}S=({{\displaystyle I}}_{\mathrm{e}}+{{\displaystyle I}}_{\mathrm{h}}+{{\displaystyle I}}_{\mathrm{c}})-({{\displaystyle C}}_{\mathrm{d}}+{{\displaystyle C}}_{\mathrm{o}}+{{\displaystyle C}}_{\mathrm{f}}),

(2)\{\begin{array}{c}\hfill {{\displaystyle I}}_{\mathrm{e}}=\mathrm{\Sigma }{C}^{j,i}{T}^{j,i}{{\displaystyle p}}_{\mathrm{e}}^{j,i},\hfill \\ \hfill {{\displaystyle I}}_{\mathrm{h}}=\mathrm{\Sigma }{C}^{k,\mathrm{G}}{T}^{k,\mathrm{G}}{\mathrm{H}\mathrm{E}\mathrm{R}}^{k,\mathrm{G}}{{\displaystyle \epsilon p}}_{\mathrm{h}}^{k,\mathrm{G}},\hfill \\ \hfill {{\displaystyle I}}_{\mathrm{c}}=C^{\mathrm{C}\mathrm{C}\mathrm{H}\mathrm{P},\mathrm{G}}{T}^{\mathrm{C}\mathrm{C}\mathrm{H}\mathrm{P},\mathrm{G}}{\mathrm{H}\mathrm{E}\mathrm{R}}^{\mathrm{C}\mathrm{C}\mathrm{H}\mathrm{P},\mathrm{G}}{{\displaystyle p}}_{\mathrm{c}}^{\mathrm{C}\mathrm{C}\mathrm{H}\mathrm{P},\mathrm{G}},\hfill \end{array}

(3)\{\begin{array}{c}\hfill {{\displaystyle C}}_{\mathrm{d}}=\mathrm{\Sigma }{C}^{j,i}{{\displaystyle \nu }}^{j,i}{\displaystyle \frac{{r\left(\mathrm{1}+r\right)}^n}{{\left(\mathrm{1}+r\right)}^n-\mathrm{1}}},\hfill \\ \hfill {{\displaystyle C}}_{\mathrm{o}}=\mathrm{\Sigma }{C}^{j,i}{\beta }^{j,i},\hfill \\ \hfill {{\displaystyle C}}_{\mathrm{f}}=\Sigma C^{j,i}{T}^{j,i}{\alpha }^{j,i}{{\displaystyle p}}_{\mathrm{f}}^{j,i}+\Sigma C^{k,\mathrm{G}}{T}^{k,\mathrm{G}}{\mathrm{H}\mathrm{E}\mathrm{R}}^{k,\mathrm{G}}{\alpha }^{j,i}{{\displaystyle p}}_{\mathrm{f}}^{k,\mathrm{G}}+{{\displaystyle p}}_{\mathrm{e}}^{\mathrm{b}}E,\hfill \end{array}

where C represents the rated capacity of the main energy equipment; T represents the annual utilization hours of the energy equipment; I_e, I_h, and I_c represent the income from providing electricity, heat, and cooling load for the system; C_d, C_o, and C_f, represent the annual equipment investment cost, operation and maintenance cost, and fuel cost; p_e,p_h, p_c, p_f, and {{\displaystyle p}}_{\mathrm{e}}^{\mathrm{b}} represent the unit price of electricity, heat, cooling, fuel, and outsourced electricity; HER HER represents the heating to power ratio (or the cooling to power ratio); e represents the unit conversion coefficient; u represents the unit capacity cost; b represents the annual unit capacity operation and maintenance cost; a represents standard coal or natural gas consumption used in the power generation (heat or cooling supply); r represents the discount rate; n represents the service life of the corresponding equipment; E represents the amount of outsourced electricity; k represents the energy equipment of NG cogenerations or distributed energy stations; j represents the type of main energy equipment; and i represents the type of energy.

Maximization of environmental benefit

The impact of CO₂ on the environment is considered to be primary. Therefore, the objective function is established with the goal of minimizing CO₂ emissions. The direct CO₂ emissions plus the indirect CO₂ emissions equals the total amount of CO₂ emissions.

(4)\mathrm{MinAE=}{{\displaystyle \mathrm{AE}}}_{\mathrm{com}}+{{\displaystyle \mathrm{AE}}}_{\mathrm{e}},

(5){{\displaystyle \mathrm{AE}}}_{\mathrm{com}}={{\displaystyle \mathrm{AD}}}_{j,i}{{\displaystyle O}}_{j,i}{{\displaystyle \mathrm{EF}}}_{j,i}\times {\mathrm{10}}^{-\mathrm{6}},

(6){\mathrm{AD}}_{j,i}={\displaystyle \mathrm{\Sigma }{C}^{j,i}{T}^{j,i}{a}^{j,i}}+{\displaystyle \mathrm{\Sigma }{C}^{k,\mathrm{G}}{T}^{k,\mathrm{G}}{\mathrm{HER}}^{k,\mathrm{G}}{a}^{j,i}},

(7){\mathrm{AE}}_{\mathrm{e}}=\mathrm{A}{\mathrm{D}}_{\mathrm{e}}{\mathrm{EF}}_{\mathrm{e}},

where AE represents the total amount of CO₂ emissions (ton); AE_com represents the direct CO₂ emissions from fuel combustion; AE_e represents the indirect CO₂ emissions caused by outsourced electricity or heat; AD (t or 10⁴m³) represents fuel consumption; Q (MJ/t or MJ/(MJ·10⁴m³)) represents low heat value; EF (g–CO₂/MJ) represents the emission factor; and 10^-6 represents the conversion of grams to tons.

Maximization of energy efficiency

The total energy efficiency of the energy internet is the sum of the utilization efficiency of energy equipment multiplied by the weight [¹⁹], and the weight value is the ratio of the energy consumed by the equipment to the total energy consumption. Under feasible economic and environmental conditions, the energy efficiency of each energy equipment should be considered to ensure that high energy efficient equipment and limited energy are reasonably made full use of.

(8)\mathrm{Max}U={\displaystyle \frac{{\displaystyle \mathrm{\Sigma }{E}_{\mathrm{in}}^{j,i}{\eta }^{j,i}}}{E_{\mathrm{in}}^{\mathrm{total}}}},

where U represents the total energy efficiency of the energy internet; {{\displaystyle E}}_{\mathrm{in}}^{\mathrm{total}} represents the total energy input of the energy equipment; {{\displaystyle E}}_{\mathrm{in}}^{j,i} represents the energy input of different kinds of energy equipment; and h represents the utilization efficiency of the energy equipment.

Constraint conditions

Constraints of supply-demand balance

From the perspective of energy supply, the construction of the energy internet in Huangpu can provide electricity, heat, cooling and other final energy for specific areas such as residential areas, industrial areas and commercial areas in the locality, and meet the energy demand of different users, which is the balance between energy supply and demand.

(9)\{\begin{array}{l}{\displaystyle \mathrm{\Sigma }{{\displaystyle P}}_{\mathrm{e}}^{j,i}}+{\displaystyle \mathrm{\Sigma }{{\displaystyle P}}_{\mathrm{e,buy}}}\ge {{\displaystyle D}}_{\mathrm{e}},\\ {\displaystyle \mathrm{\Sigma }{{\displaystyle P}}_{\mathrm{h}}^{j,i}}\ge {{\displaystyle D}}_{\mathrm{h}},\\ {\displaystyle \mathrm{\Sigma }{{\displaystyle P}}_{\mathrm{c}}^{j,i}}\ge {{\displaystyle D}}_{\mathrm{c}},\end{array}

where {{\displaystyle P}}_{\mathrm{e}}^{j,i} represents power generation; {{\displaystyle P}}_{\mathrm{h}}^{j,i} represents heat output; {{\displaystyle P}}_{\mathrm{c}}^{j,i} represents cooling output; and D_e, D_h, and D_c represent electricity demand, heat demand and cooling demand, respectively.

Constraints of equipment characteristics

The installed capacity of main energy equipment must be greater than or equal to the rated capacity [²⁰].

(10)C^{j,i}\ge \max P^{j,i}\mathrm{.}

The output power of each main energy equipment is constrained by its own upper and lower limits. {{\displaystyle P}}_{\min }^{j,i}\hspace{0.17em}\mathrm{a}\mathrm{n}\mathrm{d}\hspace{0.17em}{{\displaystyle P}}_{\max }^{j,i} represent the minimum power and maximum power of the corresponding energy equipment, respectively:

(11){{\displaystyle P}}_{\min }^{j,i}\le P^{j,i}\le {{\displaystyle P}}_{\max }^{j,i}.

Constraints of operation mode

Centralized energy modules are used for meeting the users’ base load, whose constraints can be set as

(12){{\displaystyle \mathrm{Load}}}_{\mathrm{base}}\le {\displaystyle \mathrm{\Sigma }{C}^{j,i}},

where Load_base represents the base load of the load duration curve.

Distributed energy modules and energy storage modules are used for the energy supply during the shoulder load period and peak load period. Thus, the constraints can be set as

(13)\int \mathrm{Load}\hspace{0.17em}\mathrm{d}t-{{\displaystyle \mathrm{Load}}}_{\mathrm{m}}\cdot t=\mathrm{0},

(14){{\displaystyle \mathrm{Load}}}_{\mathrm{m}}-{{\displaystyle \mathrm{Load}}}_{\mathrm{base}}\le {\displaystyle \mathrm{\Sigma }{C}^{j,i}},

(15)\max \left|{{\displaystyle \mathrm{Load}}}_{\mathrm{t}}-{{\displaystyle \mathrm{Load}}}_{\mathrm{m}}\right|\le C^{\mathrm{s}},

where Load represents the function of load duration curve; Load_m represents a certain medium load, which makes that the charging energy of the energy storage equipment equals to its own releasing energy. The performance of energy storage achieves the most revenue at this time. C^s represents the energy storage capacity; and Load_t is any point on the load duration curve.

Constraints of energy conditions

(1) Constraint of total energy consumption

The total energy consumption constraint can be expressed as

(16){\displaystyle \mathrm{\Sigma }{{\displaystyle E}}_{\mathrm{in}}^{j,i}}\le {{\displaystyle E}}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}},

where E_total represents the control objective of total energy consumption in Huangpu, Guangzhou.

(2) Constraint of coal power units

There is a constraint on coal power units, since Guangdong province will no longer approve new coal power projects in the future [²¹]. The constraint of coal power units can be expressed as

(17)C^{j,\mathrm{c}}\le \mathrm{1080000.}

(3) Constraint of garbage power units

There is a constraint on garbage power units, since Guangzhou will no longer approve new garbage power projects in the future [²²]. The constraint of garbage power units can be expressed as

(18)C^{j,w}\le \mathrm{100000.}

(4) Constraint of photovoltaic power generations

It is clearly pointed out that the installed capacity of photovoltaic power generations in Huangpu will increase to 480 MW by 2020 in Guangzhou by the “13th Five-Year Plan” for energy development. Meanwhile, the upper limit of the PV installation scale in Huangpu has been evaluated [²³]. The maximum installed capacity will not exceed 649 MW. Therefore, there is a constraint on photovoltaic power generations, which can be expressed as

(19)\mathrm{480000}\le C^{j,\mathrm{p}}\le \mathrm{649000.}

Solution

To transform the multi-objective model into a single objective model, assigning a certain weight coefficient to each objective, the weighted partial coefficient method is used for solving the mathematical model of multi-objective optimal configuration and operation.

Owing to the difference between the dimensions of the sub-objective functions, it is necessary to eliminate the influence of different physical dimensions on the optimal results in advance. By using the min-max standardization method, the values are mapped to [0, 1]. The main function can be expressed as

(20)F_i(x)={\displaystyle \frac{f_i(x)-\mathrm{0}}{\left|f_i{(x)}_{\max }\right|-\mathrm{0}}}（i=\mathrm{S,\; AE,\; U）}.

Since there is a nonlinear function in the objective function, the mathematic model for solving the nonlinear multi-objective programming problem can be expressed as

(21)\mathrm{Min}F(x)={\displaystyle \mathrm{\Sigma }{{\displaystyle \mathrm{weight}}}_i{{\displaystyle F}}_i(x),i=S,\mathrm{AE},\mathrm{U}}

(22)\mathrm{s},\mathrm{t},\{\begin{array}{l}Ax\le B,\\ {{\displaystyle A}}_{\mathrm{e}\mathrm{q}}x={{\displaystyle b}}_{\mathrm{e}\mathrm{q}},\\ C(x)\le \mathrm{0},\\ C_{\mathrm{e}\mathrm{q}}(x)=\mathrm{0},\\ {{\displaystyle b}}_{\mathrm{1}}\le x\le {{\displaystyle b}}_{\mathrm{u}},\end{array}

where A represents coefficient matrix of inequality objective function; B represents constant term of inequality objective function; A_eq represents coefficient matrix of equality objective function; b_eq represents constant term of equality objective function; C(x) represents vector function of inequality constraints; C_eq(x) represents vector function of equality constraints; b_l represents the lower limit of x; b_u represents the upper limit of x.

In Eqs. (20)-(22), the decision variables “x”of multi-objective programming include the installed capacity of coal power, garbage power, NG cogeneration, photovoltaic power generations, gas distributed energy stations, and energy storage. The inequality constraints are supply-demand balance constraints, equipment characteristics constraints, operation mode constraints, and energy conditions constraints. There is no nonlinear function in the constraints of the model.

For the above model, the program is solved by using MATLAB optimization toolbox and calling FMINCON function.

Calculation and results

Characteristic of load demand

Huangpu of Guangzhou is mainly composed of residential, industrial, and commercial areas. According to the electric power data and related documents provided by Guangzhou Power Supply Bureau [²⁴–²⁶], the energy demand of various types of users in Huangpu in 2020 can be sorted out, as listed in Table 1.

Huangpu is located in the tropics, where the coldest temperature is between 0°C and 15°C. Therefore, there is no demand for heat load in the residential area in winter. Besides, the cooling load in the residential area in summer is mainly supplied by electricity load conversion. Therefore, there is only electricity load in residential areas. The cooling load in the industrial area can be supplied by the heat load [²⁷]. Therefore, the demand of heat load is huge.

Taking 2020 as the study year, the reduced graph of prediction of typical daily electrical load in Huangpu can be depicted in Fig. 5.

Parameter setting

The energy internet is equipped with various energy technologies such as coal power, garbage power, NG cogeneration, photovoltaic power generations, and gas distributed energy stations, whose relevant parameters are tabulated in Table 2.

CO₂ emissions from various energies are presented in Table 3. The energy price and other parameters are given in Table 4.

Result and analysis without considering the outsourced electricity

Based on the mathematic model, under the given operation mode, different weight coefficient distribution plan with the maximization of economic benefits, environmental benefits, and energy efficiency is selected as sub-objectives. Thus the balanced strategy, the economic development strategy, the environmental protection strategy, and the energy efficiency strategy are formulated. The different weight coefficient distribution plan is shown in Table 5.

Equipment configuration layout of different strategies is demonstrated in Fig. 6.

From Fig. 6, it can be seen that the installed scale of coal power and garbage power with serious environmental pollution and low energy efficiency in the centralized energy modules of the four strategies has reached the upper limit permitted by the relevant policies, mainly due to the considerable economic benefits. The limited installed scale of coal power and garbage power makes NG cogeneration popular, which is mainly attributed to its relatively higher comprehensive efficiency and lower initial investment.

The balanced strategy comprehensively takes into account economy, environment, and energy efficiency. Consequently, the three sub-objective functions are given the same weight coefficient. Therefore, the installed scale of photovoltaic power generations and gas distributed energy stations of the balanced strategy is in the middle of that of other strategies. The economic development strategy focuses on considering the economic factor, and aims at maximizing economic benefits. Therefore, the total installed scale of photovoltaic power generations and gas distributed energy stations is larger than that of other strategies. The environmental protection strategy emphasizes on CO₂ emissions, and as a result, environment-friendly photovoltaic power generations are allocated the maximum capacity. Gas distributed energy stations not only achieves multipurpose utilization of energy to satisfy diverse energy requirements from different users, but also gives consideration to economy and environment. In general, gas distributed energy stations prove to be outstanding. Particularly, their synthetical efficiency for energy utilization is as high as 70%. Therefore, the installed scale of gas distributed energy stations of the energy efficiency strategy is the largest.

In addition, according to the characteristics of load duration curve of typical daily power load forecasting simplified graph in Huangpu, it can be estimated that the installed capacity of energy storage is at least 753.2 MW. Because photovoltaic power generations have typical power fluctuation and randomness, it is necessary to introduce energy storage to suppress the fluctuation of photovoltaic power. Ding et al. (2014) designed an improved method of energy storage allocation that collects the meteorological data from NREL photovoltaic observation station and uses the data to calculate the installed capacity of 1 MW photovoltaic devices, which need 0.2545 MW energy storage devices to achieve power stabilization [⁴¹]. Therefore, based on the 753.2 MW, the capacity of energy storage devices should be increased by 144.6 MW, 165.2 MW, 165.2 MW, and 122.2 MW respectively under the balanced strategy, economic development strategy, environmental protection strategy, and energy efficiency strategy.

Different equipment configuration and operation strategies will lead to different operation results. According to the case of Huangpu, the economic benefits, CO₂ emissions, and energy efficiency of the four strategies are exhibited in Table 6.

An analysis of Table 6 indicates that all aspects of the balanced strategy are in the middle of those of other strategies. Although the economic development strategy gains the highest economic benefits, CO₂ emissions and energy efficiency are lower in the four strategies. The environmental protection strategy guarantees the minimum CO₂ emissions, but the energy efficiency and economic benefits are the lowest of all the strategies. The main reason for this is that the zero-pollution photovoltaic power generations are limited by the annual utilization hours and technology, resulting in the need to sacrifice some benefits and energy efficiency for a clean environment. The economic benefits of the energy efficiency strategy are very close to those of the economic development strategy and the CO₂ emissions of the energy efficiency strategy are the highest of all the strategies, which indicates that the energy technology with a high energy efficiency brings about a strong economy but a weak environment.

Besides, it can be clearly found that the combined effect of economy-environment-energy efficiency of the traditional energy supply strategy is lower than that of the four strategies of the energy internet in various degrees. A horizontal comparison of the comprehensive performance of the energy internet strategies and the traditional energy supply strategies is shown in Table 7.

Considering the space-time differences, regional differences and, energy demand differences, to achieve a cross-regional and cross-energy balance, the residential, industrial, and commercial areas in Huangpu need to rationally allocate photovoltaic power generations and gas distributed energy stations according to their available roof areas and energy needs (Table 8).

As far as distributed energy is concerned, 129.7 MW gas distributed energy stations are needed to meet the cooling load of commercial areas, while 247.0 MW, 259.4 MW, 237.1 MW, and 268.0 MW gas distributed energy stations are allocated under the four strategies to meet the huge thermal and power demand of industrial areas, and gas distributed energy stations are not required in residential areas.

The distributed photovoltaic power generation is allocated according to the proportion of the installed roof areas of the photovoltaic power generation system in residential, industrial and commercial areas.

Result and analysis considering the outsourced electricity

Considering the outsourced electricity, in addition to local power supply, Huangpu receives power from the outer region through the 500 kV AC channel. Because the outsourced electricity belongs to exotic energy, which does not involve the energy efficiency of local energy technologies, only economic benefits and environmental protection should be considered in the situation. This paper uses MATLAB optimization toolbox for multi-objective linear programming. The results are presented in Table 9.

An analysis of Tables 9 and 10 suggests that the proportion of outsourced electricity to total electricity is 19.8%, which is the optimization of the energy internet system. About 78% of the outsourced electricity is used to meet the basic load and 22% of the outsourced electricity is used to cope with the peak load. Compared with the four strategies of excluding outsourced electricity, the strategy considering the outsourced electricity can reduce the installed scale of coal power, NG cogeneration, and gas distributed energy stations, thus greatly reduce the consumption of fossil energy. Moreover, the strategy considering the outsourced electricity can achieve more excellent environmental benefit. The reductions in CO₂ emissions account for about 5% of that of environmental protection strategy. Meanwhile, the small profit of outsourced electricity from the price differential affects the economic benefits of the energy internet system to a certain degree, which leads to a lower profit than other strategies without considering the outsourced electricity.

Conclusions and discussion

Conclusions

An analysis of the multi-objective optimal allocation strategy for the energy internet in Huangpu District, Guangzhou, China leads to the following conclusions:

First, the relation among economic development, environmental protection, and energy efficiency should be correctly handled. Therefore, without taking into consideration the outsourced electricity, the balanced strategy of all kinds of power supply at equilibrium, the economic development strategy of the largest installed capacity, the environmental protection strategy of the largest installed capacity of photovoltaic power generations and the energy efficiency strategy of the largest installed capacity of gas distributed energy stations are calculated for the energy internet from the aspect of economy, environment, and energy efficiency. Different strategies can be chosen for implementation according to the current development objectives. In addition, each of these strategies is superior to the traditional energy supply strategy. Taking into consideration the outsourced electricity, the proportion of outsourced electricity to total electricity is 19.8%, which is the system optimization of the energy internet under the certain power demand. Compared with other strategies without the outsourced electricity, the outsourced electricity strategy can have a certain emission reduction effect, while the economic benefits are reduced.

Next, the “generation-grid-load-storage” of the energy internet should be correctly coordinated. As far as “generation-generation” complementary and coordination is concerned, NG cogeneration should be vigorously developed to deal with the shortage of power load caused by the limited installation of coal power and garbage power. Building roofs should be made full use of for solar photovoltaic layout, and large-scale gas distributed energy stations should be built in industrial parks with large demand for heat and cooling. As far as “generation- grid-load” coordination is concerned, and according to the demand of cooling and heat load in the target area, it is necessary to construct the supply-side resources, equivalent to the demand side, so as to achieve the balance of supply and demand through the transmission of energy network. As far as “generation-load-storage” coordination is concerned, the arrangement of the output that the centralized energy module supply the base load, the distributed energy module and the energy storage module cooperate to track the peak load and the shoulder load, can reduce the peak-valley difference coefficient of the system.

Finally, the differences between function divisions in Huangpu should be correctly handled. As a way of nearby energy utilization, distributed energy should be allocated scientifically and reasonably according to the characteristics and conditions of energy consumption in function divisions. The result shows that, as far as the distributed energy equipment is concerned, the huge difference in demand for thermal and cooling load between industrial and commercial areas results in the installed capacity of gas distributed energy stations in industrial areas being nearly twice as large as that in commercial areas, while residential areas do not need to be equipped with gas distributed energy stations. The distributed photovoltaic power generation should be allocated according to the proportion of the installed roof areas of photovoltaic power generation system in residential, industrial, and commercial areas.

Discussion

This paper just solves the problem of the optimal allocation of the energy internet in Huangpu, Guangzhou, China. There are other problems worthy of further study, which can be summarized as follows:

(1) The mathematical model of multi-objective optimal allocation and operation only solves the structural optimization problems of centralized energy module, distributed energy module, and energy storage module. When the energy internet project is completed, it involves the space allocation of the energy module. Therefore, this paper cannot provide data information for the location of distributed and centralized energy modules. Besides, it is necessary to analyze the climate, geography, energy demand, and energy conditions of specific locations in Huangpu in combination with the geographic information system, so as to provide decision support for the layout of the energy technology.

(2) The mathematical model of multi-objective optimal allocation only calculates five different forms of energy technology using economic logic. However, the cost change caused by joint operation of different energy technologies is neglected. Therefore, fine modeling of the energy internet is the key and difficulty to overcome in the next stage.

(3) The mathematical model just simply links the “generation-grid-load-storage” together. The mechanism of the interconnection between the energy flow and the information flow of “generation-grid-load-storage” has not been explored yet.

References

[1] Zha Y, Zhang T, Tan S, Huang Z, Wang W G. Understanding and thinking of the energy internet. National Defense Technology, 2012, 33(05): 1–6 (in Chinese)

[2] Li J, Song F, Yu X. Research on global energy interconnection backbone grid planning. Journal of Global Energy Interconnection, 2018, 1(5): 527–536 (in Chinese)

[3] Zhang J, Liu X, Zhang S, Wang W, Shen P, Zhang J, Li B, Qin B, Dai P. Studies on priority areas of standard system for global energy interconnection (GEI). Distribution & Utilization, 2018, 35(08): 61–66 (in Chinese)

[4] Chang Y. On the legal constraints and the opportunities of global energy interconnection. Hebei Law Science, 2018(08): 14–24 (in Chinese)

[5] Zhao Q, Feng J, Jin Y, Wang X, Tan X. Assessment of environmental benefits under the background of global energy interconnection. Journal of Global Energy Interconnection, 2018, 1(S1): 257–262 (in Chinese)

[6] Xia Y, Dong C, Li P, Jiang Y, Zou Y, Zhou B, Jia J. Construction scheme for park energy internet based on natural gas power generation. New Energy, 2017, 45(11): 19–23 (in Chinese)

[7] Xu X, Jia H, Jin X, Yu X, Mu Y. Study on hybrid heat-gas-power flow algorithm for integrated community energy system. Proceedings of the CSEE, 2015, 35(14): 3634–3642 (in Chinese)

[8] Yang P, Chen W, Zhang J, Kang L, Luo Z. Research on energy efficiency evaluation index system of the energy internet in industry park. Science and Technology Innovation Herald, 2017, 14(25): 157–160 (in Chinese)

[9] Liu D, Cao M, Li W, Chen F, Deng Y, Weng J. Development trends in urban energy internet technology. Distribution & Utilization, 2018, 35(11): 34–37 (in Chinese)

[10] Pu T, Liu K, Chen N, Ge X,Yu J, Wang D, Wang W. Design of ADN based urban energy internet architecture and its technological issues. Proceedings of the CSEE, 2015, 35(14): 3511–3521 (in Chinese)

[11] Federal Ministry of Economics and Energy of German. E-Energy. 2016–01–20, available at website of e-energy.de/en/index.php

[12] Ding Z, Chen B, Zhou Y, Xia T, Zou J, Hu X. The research on operation optimization of regional energy internet based on detailed model. Renewable Energy Resources, 2018, 36(09): 1362–1368

[13] Liu X, Wu H. A Control strategy and operation optimization of combined cooling heating and power system considering solar comprehensive utilization. Automation of Electric Power Systems, 2015, 39(12): 1–6 (in Chinese)

[14] Ren H, Deng D, Wu Q, Liu J. Collaborative optimization of distributed energy network based on electricity and heat interchanges. Proceedings of the CSEE, 2018, 38(14): 4023–4034 (in Chinese)

[15] Zhang Y, Shen H, Ma J. Load characteristic analysis and application research on integrated energy system. Electric Power Construction, 2018, 39(09): 18–29 (in Chinese)

[16] Cuo S, Zhao H, Zhao M, Zhang Y, Zhao H. Optimal selection model for multi-class heterogeneous electrical energy storage systems. New Energy, 2018, 46(10): 11–17 (in Chinese)

[17] Zeng M, Yang Y, Liu D, Zeng B, Ouyang S, Lin H, Han X. “Generation-grid-load-storage” coordinative optimal operation mode of energy internet and key technologies. Power System Technology, 2016, 40(1): 114–124 (in Chinese)

[18] Bhattacharyya S C. Energy Economics—Concepts, Issues, Markets and Governance. Beijing: Economy & Management Publishing House, 2015

[19] Xu J. Regional energy internet operation control technology research. Dissertation for the Master’s Degree. Chengdu: Southwest Jiaotong University, 2017 (in Chinese)

[20] Shen Y. Method of energy supply system configuration optimization in industrial park. Dissertation for the Master’s Degree. Chongqing: Chongqing University, 2016 (in Chinese)

[21] Guangdong Provincial Department of Ecological Environment. Notice of the Guangdong Provincial Department of Ecological Environment on the Opinions on Soliciting the Action Plan for Winning the Blue Sky Defense in Guangdong Province (2018–2020). 2018-7-19, available at website of Guangdong Provincial Department of Ecological Environment (in Chinese)

[22] Guangzhou Power Supply Bureau Co. Ltd. Study on Guangzhou Power Supply Balance and Self-sufficiency Rate. Research Report. Guangzhou: Guangzhou Power Supply Bureau Co., Ltd, 2018 (in Chinese)

[23] Zhang Y, Liao C, Zhao D, Wang P. Solar energy resource potential assessment for industrial rooftop area: a case study of Huangpu. In: 2018 International Conference on Power, Energy and Environmental Engineering, Wuhan, China, 2018

[24] Guangzhou Research Institute of Environmental Protection Science. Environmental Impact on Guangzhou 13th Five-Year Plan for Energy. Research Report. Guangzhou: Guangzhou Research Institute of Environmental Protection Science, 2016 (in Chinese)

[25] Statistics Bureau of Guangzhou Municipality. Guangzhou Statistical Yearbook. Guangzhou: China Statistics Press, 2016–2018 (in Chinese)

[26] Liang Z, Chen Y, Xiao X, Fan L, Luo X. Analysis on characteristics of cooling, heating and power loads of three commercial buildings in Guangzhou. Building Science, 2012, 28(8): 13–20

[27] Zhang X. The method of industrial park heating load prediction. In: Proceedings of 2005 Academic Conference of Shanghai Society of Refrigeration. Shanghai Society of Refrigeration, Chinese association of Refrigeration, Shanghai, 2005: 4

[28] Polaris Power Network News Center. Super complete! Cost statistics of various types of power plants. 2018–7-25, available at website of news.bjx.com.cn (in Chinese)

[29] EIA. Annual Energy Outlook. 2019, available at website of eia.gov

[30] Guangdong Provincial Development and Reform Commission. Notice of the Guangdong Provincial Development and Reform Commission on Improving the On-grid Electricity Price of Coal-fired Power Generation Enterprises in Our Province (Guangdong Provincial Development and Reform Commission. Price [2017] No. 507). 2017-7-14, available at website of Guangdong Provincial Development and Reform Commission (in Chinese)

[31] Guangdong Provincial Development and Reform Commission. Notice on Matters Related to Reducing the On-grid Price of Natural Gas Power Generation. 2018-7-1, available at website of Guangdong Provincial Development and Reform Commission (in Chinese)

[32] National Development and Reform Commission. Notice of the National Development and Reform Commission on Improving the Price Policy for Waste Incineration Power Generation (Guangdong Price [2012] No. 170). 2012–4-1, available at website of National Development and Reform Commission (in Chinese)

[33] National Development and Reform Commission. Notice of the National Development and Reform Commission, Ministry of Finance, National Energy Administration on Matters Related to Photovoltaic Power Generation in 2018 (Development and Reform Energy [2018] No. 823). 2018–5-31, available at website of National Development and Reform Commission (in Chinese)

[34] Lai J, Ma J. The comparison between natural gas distributed energy and natural gas cogeneration project. Technology Innovation and Application, 2018(31): 50–52

[35] Guangzhou Development and Reform Commission. Guangzhou Distributed Photovoltaic Power Development Plan (2013–2020). 2013, available at website of Guangzhou Development and Reform Commission (in Chinese)

[36] Guangdong Provincial Development and Reform Commission. Guangdong Province Enterprise Carbon Dioxide Emissions Information Reporting Guide (2014 Edition). 2014, available at website o Guangdong Provincial Development and Reform Commission (in Chinese)

[37] South China Coal Trade Center. Coal guiding price of Guangzhou port. 2019–2-26, available at website of huanancoal

[38] Guangzhou Development and Reform Commission. Notice of the Guangzhou Municipal Development and Reform Commission on Reducing the Price of Non-residential Gases for Pipeline Natural Gas and the Gas Distribution Price of the Second Line of the West-East Gas Pipeline. 2018–6-1, available at website of Guangzhou Development and Reform Commission (in Chinese)

[39] Polaris Power Network News Center. Analysis of cost and economic benefit of gas distributed energy station. 2018-7-3, available at website of news.bjx.com.cn

[40] Ministry of Construction, National Development and Reform Commission. Economic Evaluation Methods and Parameters for Construction Projects. 3rd ed. Beijing: China Planning Press, 2006

[41] Ding Z, Du C, Zhang C. Configuration method of energy storage to suppress the PV power fluctuation. Journal of Power Supply, 2014, (06): 24–30

RIGHTS & PERMISSIONS

Higher Education Press