Economic-environmental-resilient Equilibrium Oriented Optimal Design Towards Distributed Hybrid Energy System: Case Study From China

. The planning and transformation of existing energy systems through renewable energy sources and the cleanest fossil fuels is considered to be one of the most promising and e ﬀ ective strategies for achieving the transition to a low-carbon world. At the same time,due to the increasing penetration of renewable energy and the frequent occurrence of extreme disasters comes with decreasing system inertia and much faster frequency drop when contingency of large power loss occurs, which seriously threatens the security of power system operation, so more and more attention is being paid to the safety and security of power systems in key areas. In order to achieve energy transition and emergency security through optimal distributed energy, this paper proposes a hybrid energy system of photovoltaic - natural gas - energy storage. A multi-objective optimization model that simultaneously considers economic costs, environmental beneﬁts and system resilience is then developed, utilizing fuzzy satisfying approach to obtain the optimal system conﬁguration under the decision maker’s attitude parameters. A case study from an industrial park is conducted to demonstrate the practicality and e ﬃ ciency of the optimization method. The calculations show that multi-objective optimisation of distributed energy systems can reduce carbon emissions by 75,364 to 414,997 tonnes per year, and can also e ﬀ ectively respond to extreme disasters by ensuring the normal operation of critical loads with high priority and the safety of the system through appropriate load shedding for loads with low priority.


Introduction
great deal of research work has been carried out on the synergistic and optimal operation of distributed energy systems for electricity-gas integration. For example,Geidl et al. [2] discussed the regulation of multiple energy flows and proposed a solution to the coupled power flow optimization problem, Fu et al. [3] considered the random fluctuation characteristics of gas, electricity and heat as well as their multi energy interaction and dependence based on the Wasserstein distance metric. These studies have made important contributions to the integration of hybrid electric-gas energy systems, but few has simultaneously considered an economic, low carbon and resilient electricity system planning under the influence of extreme disasters and policy impacts.
Therefore, this paper designs a distributed energy system with economic, low carbon and high resilience by optimising the scale of the PV-Gas-Storage distributed energy system after taking into account policy implications, natural conditions, geographical location and disaster risks. A multi-objective model was developed to optimise the benefits of the economy-environment-resilience triad.
The remainder of this paper is organized as follows. The key problems are discussed in Section 2; The multi-objective optimization model that simultaneously considers economic, environmental, and system resilience is developed in the model solving process related to Section 3. In Section 4, a real world case is conducted to demonstrate the practicality and efficiency of the optimization method is conducted. Section 5 presents the conclusions and future study directions. With the growing popularity of renewable energy, the distributed energy system is developing rapidly, and the traditional distribution network is evolving from a passive distribution network to an active distribution network. Although renewable distributed generation (DGs) has brought significant economic and environmental benefits, it also poses a major challenge to the operation of today's distribution network [4]. The main problems with power systems based on renewable energy sources are the following: (1) The system inertia decreases and the active power reserve might be insufficient [5]. Thus, the speed of frequency decline would increase when a fault occurs [6], and the risk of low frequency becomes much higher. (2) The dynamic reactive power reserve and short-circuit capacity decreases, the voltage regulation performance deteriorates, and the risk of voltage collapse increases [7].

Optimising distributed energy systems
In order to alleviate the problems associated with large-scale grid integration of renewable energy sources,the addition of natural gas energy stations to the distributed energy system is considered. Natural gas plays an increasingly important role in energy balance. It has the advantages of environmental protection, high efficiency and abundant reserves, which makes natural gas power generation more and more important. Secondly, the gas unit has high operation efficiency, rapid adjustment, and can effectively cope with wind and photoelectric fluctuations. At the same time, the use of underground natural gas pipelines as part of the energy transmission system can mitigate the damage to infrastructure caused by natural disasters.
Therefore, this paper optimises the distributed energy system by combining renewable energy and natural gas to ensure that the distributed energy system can effectively cope with the impact of extreme disasters to meet the electricity demand on the customer side. The distributed energy system is shown in figure 1. Electricity comes from photovoltaic panels, energy storage batteries, gas turbines and the main grid. Natural gas is sourced from external delivery, with natural gas storage tanks installed as an emergency reserve supply. By combining renewable energy with natural gas, energy efficiency can be increased, renewable energy consumption can be promoted. Also the use of gas turbines can effectively prevent frequency/voltage collapse due to large-scale grid connection of renewable energy sources and maintain a certain level of emergency reserves to deal with the effects of extreme disasters [8]. Power outages caused by extreme weather events have disastrous consequences for economies and societies [9]. During natural disasters and post-disaster reconstruction, the amount of resources available is often limited and difficult to replenish due to damage to the grid or transportation network [10]. For the purpose of this paper, the resilience of distributed hybrid energy system is defined as the expected maximum supply situation that system can serve critical local loads without interruption given the amount of generation resources available.

Power system resilience under extreme disasters
As the available supply resources cannot meet the recovery of all the loads, the choice was made to prioritise them and supply them in segments according to the order of priority. In [11], load priority is defined in isolated microgrid, and a strategy is designed to improve the viability of critical load in case of power interruption for a long time. In [12], a resiliency reinforcement method is proposed to define different weights for different load types. Therefore, in this paper, the electricity demand to be dispatched on the customer side is divided into four parts: first, the intelligent energy control centre; Second, the central data storage centre; Third is the daily electricity demand, such as lighting; The fourth is the energy-intensive electricity demand, such as industrial electricity.
Based on the wide-accepted resilience framework [13][14][15], many researchers have obtained typical resilience recovery diagrams for power systems [16][17][18]. In this paper, the typical performance recovery diagram is combined with load prioritisation to obtain a power system performance recovery diagram based on priority recovery, as shown in figure 2. At t 31 , the load with priority 1 is the first to recover; At t 32 , the load with priority 2 also begins to recover; At t 33 , the load with priority 3 begins to recover; At t 34 , the load with priority 4 begins to recover; At t 4 , the main grid and other power generation facilities in the system return to normal and are able to supply electricity to the industrial area normally.

Assumptions
The assumptions in this study are as follows: (1)In the event of a disaster, the energy storage equipment and gas turbines used as emergency energy supply in the distributed energy system will not be damaged.
(2)The duration of the recovery period following a disaster is determined by the amount of emergency reserve energy, with the aim of removing the impact of maintenance plans and focusing on the contribution of existing emergency reserve energy to the recovery strategy.

Notation
The following symbols are used in this paper.
N pv Number of solar arrays.

N b
Number of battery storage systems. N gt Number of gas turbines.

E e
The electricity purchased annually imported from the main grid. F gas The amount of natural gas consumed per year. c pv The annual equivalent investment cost per unit PV. c ba The annual equivalent investment cost per unit of energy storage battery. c gt The annual equivalent investment cost per unit gas turbine. c e The average price of unit electricity imported from the main grid. c gas The average price of unit natural gas imported from the natural gas network. η 1 The average carbon emissions parameter for the coal burning generation process(g/kWh). δ 1 The proportion of coal-fired power generation in the main grid. η 2 The rated capacity of the equipment j. P ba rate The rated power of unit energy storage battery. P gt rate The rated power of unit gas turbine.

P pv
The annual electricity generation per unit PV. P gt The annual electricity generation per unit gas turbine.
The power consumption per unit time of the divided priority customers A n .
The time that the divided priority customers A n can be supplied during the recovery phase.
The weight of critical load A n representing its level of priority.

System objectives
(1)Minimize the total annual costs of the district energy system.
where c pv is the annual equivalent investment cost per unit PV, c ba is the annual equivalent investment cost per unit of energy storage battery; c gt is the annual equivalent investment cost per unit gas turbine; N pv is the installed quantity of photovoltaic; N ba is the installed quantity of energy storage battery; N gt is the installed quantity of gas turbine;c e is the average price of electricity imported from the main grid , c gas is the average annual cost of natural gas imported from the natural gas network, E e is the electricity purchased annually imported from the main grid, F gas is the amount of natural gas consumed per year, (2)Minimize carbon emissions.
where η 1 is the average carbon emissions parameter for the coal burning generation process(g/kWh), δ 1 is the proportion of coal-fired power generation in the main grid, η 2 is the average carbon emissions parameter for the combustion of natural gas(g/kWh), E e the total annual power imported from the main power grid in the industrial park, F gas is the total amount of natural gas consumed annually.
(3)Maximize resilience guarantee Based on the research problem in this paper, it is practical and consistent with related studies [19] to choose Q(t) as the total power provided to the restoring load at the corresponding moment, weighted by its priority, as a function of the system's restoring power,maximize resilience is represented: Calculate the availability of electricity for each priority level based on the energy available, as follows: where h A1 ,h A2 ,h A3 ,h A4 are the time that the divided priority customers can be supplied during the recovery phase; P A1 ,P A2 ,P A3 ,P A4 are the power consumption per unit time of the divided priority customers respectively, P ba rate is the rated power of unit energy storage battery, P gt rate is the rated power of unit gas turbine, A denotes the set of critical load restored by hybrid energy system; A n represents the priority of n th level, A 1 ≥ A 2 ≥ · · · A n , A n ∈ A; w A n is the weight of critical load a representing its level of priority; P A n is the active power of critical load A n .

System constraints
(1)Installation area restriction Due to the limitation of the available installation area , there is a maximum limit on the number of PV arrays,energy storage batteries and natural gas units that can be installed, as shown in that.
(2)Demand and supply balance The main objects of energy balance in the system are electricity. The following formula ensures that the electric energy generated by the integrated energy system and the electric energy input from the power grid can meet the power demand of the industrial park.
where D e is the annual power demand of the industrial park, P pv is the annual electricity generation per unit PV, P gt is the annual electricity generation per unit gat turbine, E e the total annual power imported from the main power grid in the industrial park.

Global model
The combination of electricity and gas integrated energy generation system reduces the dependence of the industrial park on the main power grid and reduces the carbon emission caused by the energy consumption of industrial development. Based on the overall planning of electric power system in park, a mixed integer multi-objective model is established. The economic objective evaluation is to reflect the total cost of deployment and operation of the hybrid energy system with the balanced power cost. The environmental goal is to reduce carbon emissions from conventional energy sources. When considering the impact of disaster risk, emergency reserve energy is used to express the resilience of the system, that is, the ability of the system to maintain normal operation of the load in the event of a disaster. The decision variables in the system planning phase are the number of photovoltaic panels, energy storage batteries and gas turbines. The global model is obtained by combining the constraints and objectives: min f 1 = c pv N pv + c ba N ba + c gt N gt + c gas F gas + c e E e min f 2 = E e η 1 δ 1 + F gas η 2 max

Case Study
In this section, a practical example from an industrial zone is given to demonstrate the practicality and efficiency of the proposed hybrid renewable system optimization method.

Economic feasibility analysis
Based on certain expected costs, an appropriate trade-off between economic costs and environmental benefits and system resilience capacity is determined using a designed multiobjective optimisation model. However, the cost of putting in place a hybrid energy system to meet electricity demand varies under different scenario changes in natural resources and natural gas prices. In the natural gas market, changes in the price of natural gas may lead to higher costs; in the variation of solar radiation, the more abundant the solar radiation resource, the lower the cost is likely to be. Therefore, economic costs are analysed by discussing solar radiation variations and natural gas price fluctuations. In this subsection,influence of changing solar irradiation and natural gas price on the hybrid energy system are conducted,with the result shown in figure 3. The vertical coordinate in figure 3 shows the annual consumption cost(ACC) of the electricity system, while the horizontal coordinate shows the self-sufficiency of photovoltaic and gas-fired power generation for the electricity supply of the park. Figure 3(a) shows that the better the solar radiation conditions, the lower the total annual consumption cost of electrical energy. For example, when the solar radiation speed increases to 1.2 times of the basic case, the ACC for the hybrid generation system when SSR is 80.7% (point C) is 1.84 ×10 8 CNY, indicating that the economically feasible SSR of this system is within 80.7%. The total cost of electricity consumption tends to fall and then rise as the proportion of self-generation rises. This is due to the fact that the cost of electricity generation is lower than that of electricity purchased from the main grid, mainly through photovoltaic power generation, but as the proportion of self-generation rises, the limited photovoltaic resources need to be supplemented by gas-fired power generation, which increases due to the high one-off investment price of gas turbines and the high price of natural gas.Therefore, it is not economical to install gas turbines in industrial areas with low electricity demand, when putting in a gas turbine to supply the electricity system, it needs to be done in areas with intensive and high consumption of electricity.
As can be seen in figure 3(b), the lower the natural gas price, the lower the annual cost of electricity consumption. The curves overlap when the proportion of self-generation is below 30%, indicating that changes in natural gas prices have no effect on costs when the proportion of self-generation is below 30%, mainly because only the lower-cost PV is chosen. When the natural gas price is 0.9 times the base price, as shown at point A, the ACC for the hybrid generation system when SSR is 90.6% is 1.84 ×10 8 CNY, indicating that the economically feasible SSR of this system is within 90.6%.

Resilience assessment
In this subsection, the resilience of the installed hybrid energy system and the accompanying energy storage equipment is analysed by discussing its impact on industrial areas in the face of extreme disasters. As an example of high electricity demand on a typical summer day, considering extreme disasters occurring at different times, the emergency reserve energy provides a two-hour emergency supply, which is restored according to the priority of the different loads in the industrial area. The hybrid energy system chosen is a photovoltaic panel capacity of 59MW, 12MW of energy storage batteries, and 26MWkw of gas turbines; the emergency power generated by the energy reserves is 37.2MW. Load restoration at different times is shown in Table  1.
Extreme disaster situations occur randomly at any point in the summer day. As shown in Table 1,for this hybrid energy system selected, it is possible to guarantee that the demand for electricity of priority 1 and priority 2 in this industrial area does not require load shedding in the event of a disaster at any time of the day. For priority 3 electricity demand, when facing a disaster between 9:00 and 15:00, the 60% load reduction is required, except for the demand that must be maintained for operation, while at all other times priority 3 load can be met; for priority 4 load, during the low peak period of electricity consumption between 23:00 and 5:00, a load reduction of less than 80% is required, at other times, due to the demand for electricity is high and the emergency reserve energy is limited, so that when faced with a disaster, more than 80% of the load has to be reduced.

Conclusions
This paper focuses on energy system planning in industrial parks to facilitate energy transition and reduce reliance on traditional coal combustion, contributing to China's dual carbon goals. A multi-objective model is built to allow decision makers to decide the optimal configuration of energy system.The mathematical model uses annual electricity consumption costs, carbon emissions, and power system resilience to evaluate economic, environmental, and system resilience objectives.The widely used ε-constraint method is applied to transform the proposed multi-objective model into its equivalent single-objective form. The prepared data is then substituted in the mathematical model by solving it in the lingo software. A case study from an industrial park, demonstrates the practicality and efficiency of the model. Analysis and comparison of the different scenarios of hybrid energy systems in the face of changes in solar radiation and natural gas prices. The results show that the hybrid energy system can effectively face changes in external conditions, reduce the total annual cost of electricity consumption and achieve economic viability. In addition, it is clear that the installation of hybrid photovoltaic-gas energy systems can help to reduce carbon emissions and facilitate the energy transition in energy-intensive industrial park. As a high-quality fuel and relatively clean fossil energy.Regardless of the greenhouse effects of extraction, transportation, liquefaction and gasification processes (because methane itself is a greenhouse gas), natural gas produces half as much carbon as coal-fired power, and one-tenth air pollutants of coal-fired power. Therefore, for regions with abundant natural gas resources, the natural gas industry should be fully developed. Further, by installing corresponding emergency energy storage facilities and prioritising the industrial park's electricity loads and supplying them in order of priority, the system can effectively withstand damage to the industrial area's power system from extreme disasters, maintain the normal operation of critical loads and carry out corresponding load shedding for loads of lower priority to ensure the safe operation of the system.
For future studies, the following directions are critical. First, the energy system of an industrial park has a cooling-heating system in addition to the electricity system, which can also have an impact on the economy, environment and resilience of the industrial area, however, these are not considered in this paper.Secondly, the dispatch strategy for power system resilience under extreme disasters generally includes system planning, response and recovery. In this paper, only system resilience under system planning is considered, for real-time response and recovery strategies in the event of disasters should also be considered. Thirdly, due to the rapid development of renewable energy sources, other renewable energy sources besides photovoltaics, such as wind and biomass, should also be considered in hybrid energy systems. Therefore,a more comprehensive approach needs to be developed to integrate more renewable energy sources for the different energy needs of cooling-heating-electricity. This research is funded by Sichuan university (Grant No. 2022CX23)