Integrated Energy System Planning Method Considering Integrated Demand Response and Reliability Constraints

At present, most comprehensive energy planning methods aim at economy. A distributed integrated energy system planning method considering reliability and integrated demand response is proposed. This method considers that IDR can effectively realize the peak valley cutting of load characteristics, improve the system economy, and increase the reliability constraint penalty cost to make it more realistic. The example results show that the proposed method can consider the economy and reliability of configuration results under different conditions, and realize the selection of equipment.


Introduction
Distributed Integrated Energy System (DIES) is an integrated energy system located in and coupled with various distributed energy terminals. It breaks the original mode of separate design and operation of various energy supply systems and achieves the goal of multienergy complementary and energy cascade utilization through coordinated planning and operation of different energy supply systems [1].
IES can flexibly utilize different types of energy and various coupling devices, so it has a variety of operational strategies and configuration methods [2]. The literature [3][4][5] established an integrated energy system planning optimization model considering a variety of different equipment and energy types, and it is verified by simulation that the rational allocation of integrated energy system equipment types and capacities can not only achieve "multi-energy complementary".
However, most of the current planning models take economy as the goal. In this paper, considering both economy and reliability, a distributed integrated energy system planning method considering reliability and integrated demand response is established.

Distributed Integrated Energy System Model
The distributed integrated energy system proposed in this paper consists of CHP system, gas boiler, electric refrigerator and absorption chiller, including four loads of cold, heat, electricity and gas. In IDR, loads are divided into fixed loads and response loads based on their ability to participate in the demand response and their priority.
The fixed load is: representing three load types of electricity, heat and cold respectively, FL ,t k P represents the demand for fixed energy at the t-time of the k-th fixed energy, and FL0 ,t k P represents the demand under the benchmark price as the benchmark value.
The response load is: 1) Can reduce load [6]: Where, CL ,t k P represents the reduction of the the k-th load that can be cut at the t-time under the dynamic electricity price, e t  is the electricity purchase price of the user at the t-time, 0 e t  and CL0 ,t k P is the benchmark electricity price of the t-time and its corresponding reduction amount of load that can be cut respectively, CL ,t k  represents the price elasticity factor, reflecting the impact degree of price change on the user' s participation in the comprehensive demand response at the t-time.
2) the transferable load is [7]: Where, SL ,t k  represents the price elasticity coefficient of the k-th transferrable load in the t-time, price respectively. In this paper, it is assumed that the load will be transferred to adjacent time periods and linearly decrease in the maximum duration. The mathematical model of this assumption is as follows: Where, SLC ,t k P represents the transfer of the k-th transferable load from from the period t to the period ' t .

R
T represents the maximum duration of the transferable load transfer, k  represents the transfer attenuation coefficient indicating the effect of the load transfer amount decreasing with time.
3)The alternative load is: Where, TL ,t k  represents the price elasticity coefficient of the k-th alternative load of the user in the t-time,

Distributed Integrated Energy System Equipment Optimization Configuration Model
In this paper, a distributed integrated energy system equipment optimization configuration model is established. For a variety of electricity price schemes, the optimal electricity price and the optimal allocation results of distributed integrated energy system equipment can be obtained by comparing the allocation results and costs under different electricity price schemes.

setting of electricity price scheme
According to the time-of-use electricity pricing method proposed in the literature [8], under the condition of fixed peak-valley electricity price ratio and peacetime electricity price, we can get a variety of different participating alternative electricity price schemes by changing peak-valley pull ratio △. Considering the comprehensive interests of operators and customers, △ needs to: Where, △is the ratio between peak-flat electricity price difference and flat-valley electricity price difference, P L and V L represents the total load of peak and valley period respectively.

3.2.1Upper-level planning
(1) objective function The goal of upper-level planning is to achieve optimal game relationship between the economy and reliability of the operator's construction of integrated energy system, which can be expressed as: Where, inv C is the annual investment equivalent cost, op C is DIES annual operating cost, I C is the reliability constraint penalty cost, and numerically, k is the reliability penalty cost coefficient, reflecting the importance of system reliability in planning. ① Investment cost: mainly including investment cost, operation cost and residual value of equipment, which can be expressed as: Where, i represents the equipment type, which means 4 , Where, r represents the discount rate, which is 6.7% in this paper, ij l represents the life expectancy of the j-th alternative type of the i-th equipment.
②Operating cost: This paper selects three typical days of summer, winter, spring and autumn to optimize operation: ③Reliability constraint penalty cost: Where, I is the comprehensive energy reliability impact assessment index.
(2) Constraint condition Equipment capacity needs to meet the requirements of the maximum cold load and heat load, which can be expressed as:

R
represents the maximum value of the energy deficiency duration of system.

3.2.2Lower-level planning
Lower-level optimization optimizes the output of a variety of devices with the lowest operating cost of the day: Where, Cg represents the cost of gas purchase, Ce represents the cost of transaction with the superior power grid.
1)Natural gas purchase costs include the cost of natural gas consumed by the gas-fired boiler, the cost of natural gas consumed by the system, and the cost of direct natural gas supply after the replacement load is converted into the gas load: Where, g  represents the purchase price.
2)The interaction cost with the power grid is the difference between the electricity purchase cost and electricity sales revenue: 24 e e,in PG,in e,out PG,out 1 Where, , and , respectively represent the unit income of the distributed integrated energy system from the power grid at time t, , and , respectively represent the amount of electricity purchased from the grid at any time and the spare amount online.

Study Analysis
In this paper, an industrial park in north China is selected as an example to optimize the allocation of comprehensive energy system.

4.1Analysis of calculated results
The reliability penalty cost coefficient k is set as 1, and the optimal peak, valley and level electricity prices are respectively 1.26 yuan /kWh, 0.3 yuan /kWh and 0.77 yuan/kWh. Under the electricity price solutions, integrated energy system optimization configuration results as shown in table 1.

Analysis of the impact of reliability and IDR on optimized configuration results
This paper sets up the following four scenarios for comparative analysis: Scenario 1) no integrated energy system, no reliability, no IDR; Scenario 2) an integrated energy system is established without considering reliability or IDR; Scenario 3) an integrated energy system is established considering the reliability and setting k as 1 but without considering IDR.
Scenario 4) an integrated energy system is established considering reliability and IDR and setting k as 1. Through comparative analysis, the following conclusions can be drawn: 1) Compared with scenario 1, scenario 2 is provided by the traditional distribution system.
2) Compared with scenario 1, the reliability of scenario 2 is greatly improved. It shows that the integrated energy system is more economical and reliable than the distribution system.
3) Compared with scenario 2, the construction cost and the operating cost increased, indicating that the selection result of the distributed integrated energy system can be optimized both economically and reliably by including reliability in the selection model. 4) Compared with scenario 3, the initial investment cost and the operating cost decreased, indicating that considering IDR can effectively improve the economy of the system.

Conclusions
Considering that IDR can effectively realize the peak and valley cutting of load characteristics and improve the system economy, adding the reliability constraint penalty cost into the planning model can realize the selection of equipment considering the economy and reliability of configuration results under the condition of sufficient degree.