Operational optimisation of integrated campus energy systems considering integrated demand response

Against the backdrop of the rapid development of the energy internet in the park, the multi energy coupling and complementary characteristics of integrated energy systems provide more space for optimizing the participation of demand parties in their coordinated planning. Establishing a effective demand side model with multiple energy flows and response types has become an effective means to improve system performance. For this reason, this paper takes the comprehensive energy system of the park with multiple electricity, gas, heat and cold complements as the research object, establishes a complete model of multi load and multi type demand response based on the improvement, incentive and substitution of real-time pricing mechanism, and uses multi-attribute decision-making method to obtain the optimal configuration of the system by establishing an optimization framework for iteration of the main problem and sub problem. The simulation results show that compared with traditional energy supply system configurations, the complete model established in this paper for various controllable resources on the demand side achieves the coordinated and complementary operation of the system’s multi energy and low-carbon economy, fully utilizing the regulatory potential of the demand side, effectively reducing load fluctuations and energy supply costs.


Introduction
With the increasingly prominent global energy crisis and environmental problems, how to promote energy transformation and fundamental changes in production and consumption patterns has become an urgent issue in the world today. [1,2]China is in the context of energy policies such as "carbon peaking", "carbon neutral" and new energy-based power systems. Therefore, in order to achieve a positive interaction between the supply and demand sides, it is imperative to study the integrated demand response (IDR) for the synergy of multiple heterogeneous energy sources. [5]The literature [6][7] proposes a two-layer collaborative optimal dispatch model for the uncertainty of multiple loads and renewable energy sources in IES operation, and the simulation results show that the proposed model can effectively improve the economy and stable operation of the system.The literature [8][9][10] introduces the demand response characteristics of multi-energy coupling and constructs a demand response model for horizontal complementary substitution and vertical time-shifting of electricity, gas, heat and cooling loads to facilitate flexible switching between energy sources.
On the basis of the above discussion, taking an industrial park in the north as the research object, taking into account the demand response strategy based on incentives, substitution and improved prices, a multi energy and multi type IDR refined model was constructed, and a multi-objective optimal allocation and multi attribute decision-making model of PIES was established.The impact of demand side response on the park system at the current stage was compared and analyzed through simulation results, The simulation results are used to compare and analyze the impact of demand side response on the current park system, and the effect of the improved price based demand response strategy based on RTP mechanism, which provides a reference for the construction and operation of PIES.

Basic architecture of the park's integrated energy system
The diagram shows a typical PIES structure, dividing the various types of energy equipment into four parts: energy supplier, energy conversion equipment, energy storage equipment and end energy user, as shown in Figure 1.

Improving price-based demand response on the basis of RTP
TOU is a widely used price-based demand response mechanism, but its peak-to-valley time division and energy price level are fixed.RTP is highly likely to be effectively promoted and implemented.In this paper, RTP is used to guide TSL active shifting and vertical adjustment to smooth out load fluctuations through differentiated prices.As different types of loads have different sensitivities to the same tariff signal and hot and cold loads are characterised by perceptual ambiguity and time delays, only demand response based on electricity and gas prices are considered in this paper.This paper improves the RTP mathematical model as Eq (1).,0 ,0 ,0 The DR characteristic is described by the price elasticity of demand matrix, where the element , t j e in row j of E(t,j) represents the coefficient of elasticity of load at moment t with respect to the price of electricity at moment j.

Alternative demand response
The mathematical model of substitutable load (RL) is r,e , where:,

Improving PSO algorithm
In order to cope with the problem of low convergence accuracy and loss of diversity of particles due to "premature maturity" of the standard PSO algorithm, this paper adopts the dynamic adaptive update method for the inertia weight factor and learning factor in the basic PSO algorithm, as shown in Eq (3).
Where: is the initial weight, taken  =0.9; is the end weight, taken  = 0.4; is the total number of iterations;、 denotes the current particle fitness function value and the group fitness function average, respectively.

Simulation results and analysis
In order to compare the impact of introducing multiple load demand responses on the system operation and to verify the validity of the proposed model, five scenarios in Table 1 were set up and the results were analysed in this paper.
The individual target values for the optimal configuration of the system under the corresponding scenarios were obtained separately using the method proposed in the paper as shown in Table 2.The PIES system mainly purchases gas and heat energy demand from the upper tier throughout the day,as shown in Figure (2), while the electrical energy demand throughout the day gives priority to electricity consumption from clean energy WT and PV.During low tariff periods (01:00-07:00, 23:00-24:00), there is no electricity output from PV and the system mainly purchases electricity from the upper tier to meet the electrical energy demand, when the EC converts part of the electricity to cold energy to meet the The EC converts part of the power into cold energy to meet the cold load demand.During this period (08:00-22:00) the price of electricity gradually rises and the demand for electricity increases.In order to reduce purchased electricity and ensure low carbon economic operation, the system CHP and ES equipment output increases significantly.At the same time, with the reduction of electricity, the system gas purchase increases, the gas to electricity and gas to heat conversion of CHP equipment also increases, and AC becomes the main equipment for cooling.
Time-shifted demand response is mainly influenced by energy prices, so this section provides a comparative analysis of the time-shifted demand response of the improved RTP-based TSL and TOU, as shown in Figure (3), with positive values indicating load reduction after participation in demand response and negative values indicating increase.Overall, the improved RTP-based electric and natural gas loads have a higher demand response compared to TOU, and the effect of peak and valley reduction is significant.The operating costs of the RTP-based system optimisation have been reduced.Therefore, the improved RTP with real-time differentiated pricing based on energy supply and demand in the system can better stimulate the potential of the PIES demand side to optimise the system, curb load fluctuations and ensure low carbon and economic operation of the park's energy internet.Fig. 3. Based on RTP and TOU time-shifted demand response volumes for electric and gas loads

Conclusion
This paper proposes a multi-objective optimal allocation and decision-making method for the park system based on a multi-energy coupled park system of electricity, gas, heat and cooling.Compared to TOU, the RTP mechanism uses hourly price update steps to implement differentiated pricing based on real-time energy supply and demand in the system, and dynamically directs user loads to participate in demand response, which has better economic and incentive effects.Compared with a single demand response strategy, the improved price-based demand response model, which integrates substitution, incentive and RTP, achieves horizontal complementary substitution and vertical time-shifting of multienergy loads, and suppresses fluctuations in the load curve more effectively, resulting in an optimal operation scheme for the integrated energy system of the park.

Figure 1
Figure 1 PIES structure diagram limit of the replaceable load substitution; and, limit of the replaceable load substitution.

Fig. 2 .
Fig.2.(a) Electric power balance figure (b) Gas power balance figure (c) Heat power balance figure (d) Cold power balance figure

Table 2
System operating cost of each scenario ) shows the supply and demand balance for each energy subsystem under scenario S3.