Research on the Economic Optimization of an Electric–Gas Integrated Energy System Considering Energy Storage Life Attenuation

Abstract

Battery storage is one of the important units in the optimal scheduling of integrated energy systems. To give full play to the advantages of battery storage in stabilizing power quality and smoothing the output of intermittent new energy generation, the battery life decay problem needs to be considered in optimal scheduling. In this paper, we studied the energy storage life decay model and established an electric–gas integrated energy system model considering energy storage life decay to minimize the economic cost of system operations and to compare the optimal dispatch results with and without energy storage life decay through simulation analysis to verify the rationality and economy of the energy storage life decay model used in this paper.

1. Introduction

Along with the rapid development of the globalization trend in the world economy, the supply of energy cannot meet the growing production and living needs of people, and the situation of energy shortages and waste has become more and more serious in recent years. As the lifeblood of the national economy, energy is fundamental to national development, so a series of problems such as the excessive use of energy will be a common challenge for human beings [1,2,3]. An integrated energy system covers several energy subsystems, which can realize the complementary advantages of different energy sources in operation [4] and help improve energy utilization efficiency. They are widely valued, especially the increasing maturity of electricity to gas technology has further strengthened the coupling between energy sources, while gas turbines, heat exchangers, electric chillers, energy storage, and other technologies have been more widely used as important components in integrated energy systems [5,6]. At the same time, renewable energy is clean, environmentally friendly, and inexpensive, which makes the whole system more efficient and cleaner. However, solar, wind, and other renewable energy sources are susceptible to natural environmental factors, resulting in randomness, volatility, and uncertainty in the output of photovoltaic wind turbines [7,8]. As one of the most important parts of an integrated energy system, energy storage plays a big role in increasing grid flexibility, improving power quality, and promoting new energy consumption [9,10]. However, during the operation of the system, the life of energy storage is affected by the number of times of charging and discharging, the depth of discharge, temperature, etc. The capacity of energy storage, the number of cycles, etc., will change, as will the operating efficiency and life decay characteristics of energy storage. The impact of the reliability and the economy of power system operations is increasing [11]. Therefore, the impact due to the decay of the energy storage life must be considered in the process of the study of the economic optimal dispatch of the integrated energy system.

A lot of research has been carried out at home and abroad on integrated energy optimization scheduling that has included energy storage lifetime decay. In the literature [12,13,14] studies have been carried out on the economic dispatch of integrated energy systems, but they have not considered the impact of energy storage lifetime decay. The literature [15] proposes to use a large amount of experimental data to estimate the remaining capacity of the battery and thus characterize its health status. The literature [5] uses a rainflow-like counting method to simulate the energy operation state and calculate the lifetime of the energy storage. The literature [16] uses the rainflow-like counting method to analyze battery performance and calculate the battery aging cost. The literature [17] considered the full cell capacity life decay with instability on the surface of the silicon component. Among the existing studies conducted to study battery capacity decay, one model is based on physical and data-driven models, and the literature [18,19] uses raw input signals (current, voltage, and temperature) to train machine learning models. The literature [20,21,22,23,24] uses preprocessed features from voltage, current, temperature, impedance, or power curves as input to the machine learning model. Still, another model is a continuum approach using porous electrode theory, where the model can be extended to capture potential degradation mechanisms leading to capacity decay, including differential voltage analysis models [25] and first-principles degradation models based on porous electrode theory [26].

The literature [27] selected a commutation factor to measure energy storage life decay in terms of exchange power, but the fixed commutation ratio makes it difficult to measure the dynamic change process of the battery during the actual operation. A method for estimating the residual life of lithium-ion batteries based on data hybrid drive is proposed in the literature [28]. The literature [29] defines the main parameters affecting life decay, selects the suitable stress conditions for retired power cells, and identifies the life decay characteristics of retired power cell modules using an accelerated life test method based on an inverse power law model. The literature [30,31] is based on the accurate measurement of energy storage lifetime losses based on experimental data, and the method is again too complex and computationally overwhelming. The literature [32,33] limits the number of cycles of energy storage and the charging and discharging time, and the model’s accuracy is not accurate enough. Therefore, constructing a suitable energy storage lifetime decay model and using it to evaluate the lifetime decay of energy storage during operation is crucial in the process of the optimal scheduling of integrated energy systems.

In the process of solving search and optimal scheduling problems, the most common approach is to use optimal scheduling algorithms for solving them, where particle swarm algorithms show good performance in multi-objective, multi-constrained, and nonlinear optimization problems and can be applied to nonlinear systems, computer science, robotics, engineering, and many other fields in real-world applications [34,35]. They have been widely used with the advantages of few constraints on the fitness function, simple structure, easy coding, and fast convergence. The literature [36] uses a particle swarm algorithm to optimize the scaling factor of a PID controller, thus improving the performance of a permanent magnet DC motor for a photovoltaic wire feeding system. The literature [37] uses a particle swarm algorithm combined with an MPPT controller to control the PV output power near the maximum point. The literature [38] uses a particle swarm algorithm to optimize the coefficients of a dynamic regulator for the purpose of increasing the power of a solar PV feeder system. The literature [39] evaluates different PSO approaches in the scheduling and control of different residential energy sources such as smart appliances, electric vehicles (EVs), heating/cooling equipment, and energy storage, so that particle swarm optimization algorithms can be widely used for distributed energy scheduling and control in real demand response applications.

In summary, this paper establishes a life decay model of energy storage, measures the life decay of energy storage by studying the changes in the capacity and charge state of energy storage during system operations, and introduces it to the changes in objective function and constraints in the process of system optimal scheduling, solves the integrated energy system optimization and dispatch model by the particle swarm algorithm, and finally verifies the reasonableness and the validity of the model by arithmetic analysis.

2. Integrated Electricity–Gas Energy System Model with Energy Storage

The structure of the electric–gas integrated energy system model including energy storage studied in this paper is shown in Figure 1. The source side of the model mainly consists of a wind turbine, PV, a grid, and a gas network, and the load side consists of an electric load, a gas load, a cooling load, and a thermal load, and the main equipment in the model includes an electric boiler, an electric chiller, P2G, and a gas turbine.

2.1. Micro Gas Turbine Mathematical Model

The micro gas turbine is one of the core equipment of an electric–gas integrated energy system, which has received great attention for its small size and low pollution emission [40]. The micro gas turbine can convert clean energy such as methane and natural gas into various forms of energy such as electricity, heat, and cold, etc. It can provide electricity to the electric load and the waste heat generated can also provide energy to the heat and cold load. The gas consumption of a micro gas turbine can be expressed by Equation (1) [41].

$$\begin{array}{l}{P}_{mt}(t)=\frac{q_{{}_{mt}}(t)\times {\eta }_{mt}(t)\times R_{LHVT}(t)}{\Delta t}\\ Q_{mt}(t)=\frac{P_{mt}(t)\times (1-{\eta }_{mt}(t)-{\eta }_L)}{{\eta }_{mt}}\end{array}$$

(1)

where $P_{mt}(t)$ is the output power; $Q_{mt}(t)$ is the output waste heat; $R_{LHVT}(t)$ is the low-level heat value, taken as 9.7 $KW\cdot h/m^3$; ${\eta }_{mt}(t)$ is the generation efficiency; ${\eta }_L$ is the heat loss rate; and $\Delta t$ is the unit dispatch time, taken as 1 $h$.

2.2. Mathematical Model of the Electric Boiler

The working principle of an electric boiler is the thermal effect of the electric current, which is a heating device that converts electrical energy into thermal energy. The mathematical model can be represented by Equation (2) [42].

$${\eta }_{eb}=\frac{Q_{eb}(t)}{P_{eb}(t)}$$

(2)

where ${\eta }_{eb}(t)$, $Q_{eb}(t)$, and $P_{eb}(t)$ are the thermal efficiency, heating power, and power consumption of the electric boiler at the time $t$, respectively.

2.3. Mathematical Model of Electric Chillers

The use of electrical energy to drive the compressor to convert electrical energy into cold energy is the main mode of operation of the electric refrigeration machine. The mathematical model can be represented by the following Equation (3).

$${\eta }_{ec}=\frac{Q_{ec}(t)}{P_{ec}(t)}$$

(3)

where ${\eta }_{ec}$, $Q_{ec}(t)$, and $P_{ec}(t)$ are the refrigeration efficiency, refrigeration power, and power consumption of the electric refrigeration machine at the moment $t$, respectively.

2.4. P2G Mathematical Model

The electricity-to-gas equipment electrolyzes water into hydrogen and oxygen, and hydrogen can be synthesized with carbon dioxide into methane and other gases, thus converting electrical energy into a more storable gas and realizing the graded use of energy, which is of vital importance for the consumption of new energy. The mathematical model can be represented by the following Equation (4) [43].

$$\begin{array}{l}{q}_{P2G}(t)=\frac{Q_{P2G}(t)}{R_{LHVT}(t)}\\ Q_{P2G}(t)={\eta }_{P2G}\cdot P_{P2G}(t)\end{array}$$

(4)

where ${\eta }_{P2G}$ is the conversion factor of the gas-to-electric equipment and $q_{P2G}(t)$, $Q_{P2G}(t)$, and $P_{P2G}(t)$ are the gas flow rate, heat generation, and power consumption of the gas-to-electric equipment at the time $t$, respectively.

2.5. Mathematical Model of Heat Exchangers and Absorption Chillers

The heat exchanger and absorption chiller can process the waste heat generated by the micro gas turbine to provide energy for the heat load and the cooling load, respectively. The mathematical model can be represented by Equation (5) [44].

$$\begin{array}{l}{Q}_{he}(t)={\eta }_{he}\cdot Q_{he,in}(t)\cdot CO{P}_{hot}\\ Q_{ac}(t)={\eta }_{ac}\cdot Q_{ac,in}(t)\cdot CO{P}_{cool}\end{array}$$

(5)

where $Q_{he,in}(t)$, $Q_{he}(t)$, $Q_{ac,in}(t)$, and $Q_{ac}(t)$ are the heat production and cooling capacity of the heat exchanger and absorption refrigerant at the input side and output side, respectively and ${\eta }_{he}$, ${\eta }_{ac}$, $CO{P}_{hot}$, and $CO{P}_{cool}$ are the heat production efficiency and cooling efficiency as well as the heat production coefficient and cooling coefficient of the heat exchanger and absorption refrigerator, respectively.

3. Energy Storage Life Decay Model

Energy storage is one of the important units of an integrated energy system, and the deployment of energy storage in the system has multiple roles. On the one hand, it can avoid the fluctuation of the power grid caused by the uncertainty of the new energy output; on the other hand, it can maintain the transient power balance of the new energy [45], and in addition, it can be used as an emergency power source to ensure the reliability of the whole system.

Battery energy storage is a widely used energy storage method in the current power system. As the cycle proceeds, the capacity gradually decreases due to an increase in the positive and negative polarization of the battery, which is selected as the object of study in this paper, and its life decay model is investigated. The important parameters to measure the decay of energy storage life are capacity and charge state.

3.1. Capacity Decay Model

Capacity is a key parameter to measure the battery life decay model. The main factors affecting capacity decay are calendar capacity decay and cycle capacity decay, calendar capacity decay is small, and this paper focuses on the study of battery cycle capacity decay. Battery cycle capacity decay is influenced by the charge/discharge rate, ambient temperature, and cycle times. During the use of the battery, the temperature can be maintained at a constant temperature by the corresponding equipment; as the time-sharing tariff mechanism is used in the dispatching process, the charge/discharge multiplier can also be constrained to about 1C, so this paper mainly studies the influence of the number of cycles on the battery capacity decay [46].

During the operation of the battery, the capacity of the battery will keep changing as the time and the frequency of charging and discharging increase; it was found through the research that the capacity loss rate of the battery shows a positive relationship with the testing time [47]. As shown in Equation (6).

$$C(t)=C_R(1-b(t-1))$$

(6)

where $C(t)$ indicates the capacity of the battery at the time $t$, $C_R$ is the initial capacity of the battery, and $b(t)$ is the capacity loss rate at the moment of 1C discharge of the battery.

3.2. Charge State Decay Model

The state of charge is another key parameter to measure battery life decay. The state of charge at the time $t$ $SO{C}_{ES}(t)$ and its value are related to two parameters, which are the state of charge at the time $t-1$ $SO{C}_{ES}(t-1)$ and the charge/discharge rate at the time $t$. Without taking into account battery life decay, the state of charge can be expressed as Equation (7).

$$SO{C}_{ES}(t)=SO{C}_{ES}(t-1)+({\eta }_C\frac{P_{ES,C}}{C_R}-{\eta }_D\frac{P_{ES,D}}{C_R})\Delta t$$

(7)

Taking into account battery life decay, the charge state is expressed as Equation (8).

$$SO{C^{\prime }}_{ES}(t)=SO{C}_{ES}(t-1)+({\eta }_C\frac{P_{ES,C}}{C(t)}-{\eta }_D\frac{P_{ES,D}}{C(t-1)})\Delta t$$

(8)

where ${\eta }_C$ and ${\eta }_D$ are the charging and discharging efficiency of the battery, respectively, and $P_{ES,D}$ are the charging and discharging power of the battery, respectively, and $\Delta t$ is the unit dispatch time interval.

4. Integrated Energy System Optimization Scheduling

In this paper, an energy storage life decay model was established and based on this, an optimal dispatching model of an electric–gas integrated energy system with energy storage life decay was built, and the lowest daily operating cost was the objective function to meet its operating constraints, and the output of each device was analyzed according to the known load data.

4.1. Optimal Scheduling Model without Accounting for Energy Storage Lifetime Decay

4.1.1. Objective Function

Not taking into account the decay of energy storage life, energy storage does not generate direct freight costs when the system is operating, the system’s operating costs include the initial investment cost of energy storage, power purchase cost, gas purchase cost, system operating costs, and abandoned scenery generation costs, and the objective function that the sum of all types of operating costs is the lowest.

$$\min F=F_e(t)+F_g(t)+F_{OC}(t)+F_q(t)+F_R$$

(9)

where $F_e(t)$ is the cost of electricity purchase; $F_g(t)$ is the cost of gas purchase; $F_{OC}(t)$ is the cost of system operation, including operation and maintenance costs and depreciation costs; $F_q(t)$ is the cost of abandoned scenery penalty; and $F_R$ is the initial investment cost of energy storage.

(1) Cost of electricity purchase $F_e(t)$.

$$F_e(t)={\displaystyle \sum _{t=1}^{24}{C}_{grid}(t)\cdot P_{grid}(t)}$$

(10)

where $C_{grid}(t)$ is the purchase price of electricity and $P_{grid}(t)$ is the electric power purchased from the power grid.

(2) Gas purchase cost $F_g(t)$.

$$F_g(t)={\displaystyle \sum _{t=1}^{24}{C}_g(t)\cdot q_{grid}(t)}$$

(11)

where $C_g(t)$ is the price of purchased natural gas and $q_{grid}(t)$ is the flow rate of gas purchased from natural gas.

(3) Abandoned scenery penalty cost $F_q(t)$.

$$F_q(t)={\lambda }_{PV}{\displaystyle \sum _{t=1}^{24}(P_{PV}^{\mathsf{\Gamma }}(t)-P_{PV}(t))}+{\lambda }_{WT}{\displaystyle \sum _{t=1}^{24}(P_{WT}^{\mathsf{\Gamma }}(t)-P_{WT}(t))}$$

(12)

where ${\lambda }_{PV}$ is the penalty factor for light abandonment; ${\lambda }_{WT}$ is the penalty factor for wind abandonment; $P_{PV}^{\mathsf{\Gamma }}(t)$ is the photovoltaic power; $P_{WT}^{\mathsf{\Gamma }}(t)$ is the wind power; $P_{PV}(t)$ is the photovoltaic power used to supply the electricity–gas integrated energy system; and $P_{WT}(t)$ is the wind power used to supply the electricity–gas integrated energy system.

(4) System operation cost $F_{OC}(t)$.

$$F_{OC}(t)={\displaystyle \sum _{i=1}^n{f}_i(t)}$$

(13)

where $f_i(t)$ is the operating cost of the $i$-th device at the time $t$.

$$\{\begin{array}{c}\begin{array}{c}\begin{array}{l}{f}_{P2G}(t)=(K_{om}^{P2G}+C_{dp}^{P2G}){\displaystyle \sum _{t=1}^{24}{P}_{P2G}(t)}+{\gamma }_{P2G}^{H_{2}O}{\displaystyle \sum _{t=1}^{24}{Q}_{P2G}^{H_{2}0}(t)-}{\gamma }_{P2G}^{O_2}{\displaystyle \sum _{t=1}^{24}{Q}_{P2G}^{O_2}(t)}\\ f_{ES}(t)={\displaystyle \sum _{t=1}^{24}[K_{om1}^{ES}\cdot C_R+(K_{om2}^{ES}+C_{dp}^{ES})\left|P_{ES}(t)\right|]}\end{array}\\ f_{mt}(t)=(K_{om}^{mt}+C_{dp}^{mt}){\displaystyle \sum _{t=1}^{24}{P}_{mt}(t)}\\ f_{eb}(t)=(K_{om}^{eb}+C_{dp}^{eb}){\displaystyle \sum _{t=1}^{24}{P}_{eb}(t)}\end{array}\\ f_{ec}(t)=(K_{om}^{ec}+C_{dp}^{ec}){\displaystyle \sum _{t=1}^{24}{P}_{ec}(t)}\\ \begin{array}{l}{f}_{he}(t)=(K_{om}^{he}+C_{dp}^{he}){\displaystyle \sum _{t=1}^{24}{Q}_{he,in}(t)}\\ f_{ac}(t)=(K_{om}^{ac}+C_{dp}^{ac}){\displaystyle \sum _{t=1}^{24}{Q}_{ac,in}(t)}\\ f_{PV}(t)=(K_{om}^{PV}+C_{dp}^{PV}){\displaystyle \sum _{t=1}^{24}{P}_{PV}(t)}\\ f_{WT}(t)=(K_{om}^{WT}+C_{dp}^{WT}){\displaystyle \sum _{t=1}^{24}{P}_{WT}(t)}\end{array}\end{array}$$

(14)

In the formula, $K_{om}^i$ is the operation and maintenance coefficient of the $i$ equipment, $C_{dp}^i$ is the depreciation coefficient of the $i$ equipment, ${\gamma }_{P2G}^{H_{2}O}$ is the unit price of water consumed by the P2G equipment during operation, ${\gamma }_{P2G}^{O_2}$ is the oxygen unit price sold during the operation of the P2G equipment, $K_{om1}^{ES}$ is the energy storage capacity maintenance cost coefficient [48], and $K_{om2}^{ES}$ is the energy storage power maintenance cost coefficient.

(5) Initial investment cost of energy storage $F_R$.

The initial investment cost of energy storage $F_R$ is a fixed value.

4.1.2. Constraints

(1) Coupling device constraint:

$$\begin{array}{l}{P}_i^{\min }\le P_i\le P_i^{\max }\\ Q_i^{\min }\le Q_i\le Q_i^{\max }\end{array}$$

(15)

where $P_i$ and $Q_i$ are the electric and thermal power of the coupled equipment of the electric–gas integrated energy system, respectively.

(2) Energy storage constraint:

$$\begin{array}{l}0\le P_{ES,C}\le P_{ES,C}^{\max }\\ 0\le P_{ES,D}\le P_{ES,D}^{\max }\\ SO{C}_{ES}^{\min }\le SO{C}_{ES}\le SO{C}_{ES}^{\max }\end{array}$$

(16)

where the superscripts $\max$ and $\min$ correspond to the upper and lower limits of the physical quantity, respectively.

(3) Energy balance constraint:

a. Electrical balance constraint:

$$P_{grid}(t)+P_{PV}(t)+P_{WT}(t)+P_{mt}(t)=P_{ec}(t)+P_{eb}(t)+P_{ES}(t)+P_{P2G}(t)+L_e(t)$$

(17)

where $L_e(t)$ is the electrical load.

b. Natural gas balance constraint:

$$q_{grid}(t)+q_{P2G}(t)=q_{mt}(t)+L_q(t)$$

(18)

where $L_q(t)$ is the gas load.

c. Thermal equilibrium constraint:

$$Q_{eb}(t)+Q_{he}(t)+Q_{P2G}(t)=L_h(t)$$

(19)

where $L_h(t)$ is the heat load.

d. Cold equilibrium constraint:

$$Q_{ac}(t)+Q_{ec}(t)=L_c(t)$$

(20)

where $L_c(t)$ is the cold load.

e. Balance of waste heat output from gas turbines and heat input from heat exchangers and absorption chillers.

$$Q_{ac,in}(t)+Q_{he,in}(t)=Q_{mt}(t)$$

(21)

4.2. Optimal Scheduling Model for Energy Storage Life Decay and Calculation

4.2.1. Objective Function

When taking into account the decay of energy storage life, the system in operation, energy storage will be continuously charged and discharged resulting in the decay of capacity, resulting in life loss costs, and the capacity of energy storage operations and maintenance costs will also change accordingly.

$$\begin{array}{l}{F}_{ES}(t)=\frac{\Delta C(t)}{C_R}{F}_R\\ \Delta C(t)=C(t)-C(t-1)\\ f_{ES}{}^{\prime }(t)={\displaystyle \sum _{t=1}^{24}[K_{om1}^{ES}\cdot C(t)+(K_{om2}^{ES}+C_{dp}^{ES})\left|P_{ES}(t)\right|]}\end{array}$$

(22)

where $F_{ES}(t)$ is used for the consideration of energy storage life decay after the cost of life loss, $\Delta C(t)$ represents the decay of energy storage capacity from the moment $t-1$ to the moment $t$, and $f_{ES}{}^{\prime }(t)$ is the energy storage operation cost after the operation and maintenance costs of energy storage has changed. When the cost of energy storage life loss and the cost of operation and maintenance are taken into account, the operation goal of the system is as follows Equation (23):

$$\min F^{\prime }=F_e(t)+F_g(t)+{F^{\prime }}_{OC}(t)+F_q(t)+F_R+F_{ES}(t)$$

(23)

where ${F^{\prime }}_{OC}(t)$ represents the system operating cost after the change in energy storage operations and maintenance costs and $F^{\prime }$ represents the total system cost after taking into account the decay of energy storage life.

4.2.2. Constraints

The charge state of the energy storage will change when the energy storage life decay is taken into account; therefore, the operating constraint of energy storage becomes Equation (24):

$$\begin{array}{l}0\le P_{ES,C}\le P_{ES,C}^{\max }\\ 0\le P_{ES,D}\le P_{ES,D}^{\max }\\ SO{C}_{ES}^{\min }\le SO{C^{\prime }}_{ES}\le SO{C}_{ES}^{\max }\end{array}$$

(24)

The rest of the constraints are the same as in Section 4.1.2.

5. Example Analysis

In this paper, the typical daily load data of a park are selected, and the structural model diagram of an electricity–gas integrated energy system shown in Figure 1 is used for simulation analysis, and the parameters of each piece of equipment are shown in

Table 1. System equipment parameters.

Parameters of Each Equipment	Value/Unit
Rated power of electric chillers	$200 \mathrm{kW}$
Rated power of the electric boiler	$150 \mathrm{kW}$
The power rating of P2G	$200 \mathrm{kW}$
Rated power of the micro gas turbine	$200 \mathrm{kW}$
The power rating of heat exchangers	$180 \mathrm{kW}$
The power rating of absorption chillers	$200 \mathrm{kW}$
The initial capacity of the battery	$200 \mathrm{kWh}$
Maximum battery state of charge	0.9
Minimum battery state of charge	0.2
Battery charging and discharging efficiency	0.97
The efficiency of electric chillers	4
The efficiency of electric boilers	0.94
The efficiency of P2G	0.6
Micro gas turbine efficiency	0.32
Heat exchanger efficiency	0.9
Unit maintenance cost of electric chillers	$0.01 \mathrm{yuan}/\mathrm{kWh}$
Unit maintenance cost of electric boilers	$0.02 \mathrm{yuan}/\mathrm{kWh}$
Unit maintenance cost of P2G	$0.12 \mathrm{yuan}/\mathrm{kWh}$
Unit maintenance cost of micro gas turbines	$0.23 \mathrm{yuan}/\mathrm{kWh}$
Heat exchanger unit maintenance cost	$0.025 \mathrm{yuan}/\mathrm{kWh}$
Unit maintenance cost of absorption chillers	$0.025 \mathrm{yuan}/\mathrm{kWh}$
Energy storage per day investment cost	1000 yuan

[49,50,51,52]. The system’s new energy generation and load forecast data are shown in Figure 2, using peak and valley tariffs as shown in

Table 2. Peak and valley period electricity prices.

Project Category (Tariff)	Time Period $\left(\mathbf{h}\right)$	Unit Price $(\mathbf{yuan}/\mathbf{kWh})$
Peak	8:00–11:00 18:00–20:00	0.886
Flat value	11:00–18:00 20:00–21:00	0.559
Valley value	23:00–8:00 the next day	0.223

. In this paper, a particle swarm algorithm is used to simulate and analyze the optimal dispatching model of an integrated electricity–gas energy system to minimize the daily operating cost of the system.

From Figure 3, it can be seen that during the valley tariff period, the new energy output is not enough to provide the electric load required, so it is necessary to purchase power from the grid, and the battery is charged at this time so that it can be discharged during the peak tariff period to reduce the power purchased from the grid; during the peak tariff period, the new energy and gas turbine coordinate to supply power, and when the new energy output is insufficient, the battery starts to discharge to provide the rest of the required power together with the gas turbine.

From the thermal power optimization results in Figure 4, it can be seen that the heat load demand is high, and the electric boiler and heat exchanger are needed to provide heat energy at the same time. In the period of the flat tariff, the heat load demand is relatively small, and the heat energy is mainly provided by the gas turbine and the heat exchanger, and in the period of the valley tariff, the electric boiler is subject to the upper limit of output, and the heat exchanger is required to provide heat energy together.

From the cold power optimization results in Figure 5, it can be seen that the energy required for the cold load is mainly provided by the electric chiller, which is because the electric chiller has lower operating costs than the gas turbine. When the waste heat generated by the gas turbine is not fully utilized by the heat exchanger, the absorption chiller uses the remaining waste heat to provide energy to the cold load at this time, avoiding the excess cost of the gas turbine due to heat dissipation.

From the gas power optimization results in Figure 6, it can be seen that the electric-to-gas equipment mainly participates in system optimization and dispatch during the valley tariff, which is because, during the valley tariff, the operating costs of electric-to-gas equipment are lower than the costs of gas purchase, while during the peak tariff, the operating costs of electric-to-gas equipment are higher, so the energy required by the gas load mainly comes from the gas grid.

From the optimization results in Figure 7, Figure 8 and Figure 9, it can be seen that the battery charging power is equal to the discharging power in a scheduling cycle, and there is no simultaneous charging and discharging; when storage life decay is not taken into account, the stored energy is frequently charging and discharging, and the change is more drastic; the SOC curve after taking into account storage life decay is more gentle, and the relative charging and discharging depth is lower, which achieves the effect of shallow charging and shallow discharging and prolongs the service life of the stored energy.

Figure 10 shows the daily operating cost of the system under the two cases. The daily operating cost of the system is CNY 9139.3656 when storage life decay is not taken into account and CNY 9081.6026 when storage life decay is taken into account, with a comprehensive cost reduction of CNY 57.763. During the operation of the system, the battery not only effectively plays the role of peak-shaving and valley-filling, shortening the peak-to-valley difference of the system, but also reduces the operating costs of the system and improves the economy of the system.

6. Conclusions

In this paper, we studied the energy storage life decay model and considered the influence of this factor in integrated energy system optimization scheduling, and we used the typical daily load data to analyze the calculation cases and optimize the power output of each piece of equipment in an electric–gas integrated energy system. We can obtain the following conclusions: comparing the results of the electric–gas integrated energy system optimization scheduling with and without energy storage life decay, the formerly established energy storage life decay model can not only measure the life decay of the energy storage battery during system operations but also optimize the charging and discharging process of the energy storage battery to extend its service life. Considering the influence of energy storage life decay in the process of system optimization and scheduling can also reduce system scheduling costs, which is of great significance for the economic operation of the system.