Coordinated Optimization of Hydrogen-Integrated Energy Hubs with Demand Response-Enabled Energy Sharing

Abstract

The energy hub provides a comprehensive solution uniting energy producers, consumers, and storage systems, thereby optimizing energy utilization efficiency. The single integrated energy system’s limitations restrict renewable absorption and resource allocation, while uncoordinated demand responses create load peaks, and global warming challenges sustainable multi-energy system operations. Therefore, our work aims to enhance multi-energy flexibility by coordinating various energy hubs within a hydrogen-based integrated system. This study focuses on a cost-effective, ecologically sound, and flexible tertiary hub (producer, prosumer, and consumer) with integrated demand response programs, demonstrating a 17.30% reduction in operation costs and a 13.14% decrease in emissions. Power-to-gas technology enhances coupling efficiency among gas turbines, boilers, heat pumps, and chillers. A mixed-integer nonlinear programming model using a GAMS BARON solver will achieve the optimal results of this study. The proposed model’s simulation results show reduced energy market costs, total emissions, and daily operation expenses.

1. Introduction

In response to global economic and political initiatives, hydrogen is gaining traction, driving worldwide efforts to transition to green energy and achieve carbon neutrality by 2030 [1]. The energy crisis has intensified the push for renewable energy, straining the energy balance and necessitating immediate solutions for better integration and efficiency. Multi-energy systems (MES) present promising models for future energy infrastructure by enhancing interactions among various energy carriers and offering effective, flexible methods to incorporate more renewable sources. This approach has garnered significant interest from both scholars and industries [2,3].

In fact, the supply, conversion, and storage components of integrated energy system (IES) core equipment enable the cooperative provision of electricity, heating, cooling, and gas [4]. Improving system flexibility reduces discrepancies between power supply and load demand, contributing to overall stability. This involves managing shifting load demands and the volatile behavior of green energy sources within the power system [5,6]. Therefore, the aim is to enhance overall flexible integration, improving multi-energy storage, sharing, and usage by incorporating these systems into the proposed model. This plan also promotes the widespread use of hydrogen as a clean energy source.

1.1. Literature Review

The author of [6] suggests coordinating multiple microgrid systems with demand response programs (DRP) to reduce costs and enhance flexibility. To promote economic and ecological sustainability, they avoid integrating hydrogen, limit power conversion units, and emphasize an advanced power-to-gas system with carbon capture and storage (P2G-CCS) over carbon capture and trading benefits. For supplying energy to electric and hydrogen fuel cell vehicles, electrolyzers and hydrogen storage tanks are incorporated into the charging stations, as demonstrated in [7,8]. However, this approach lacks precise demand prediction and overlooks the economic advantages of demand response. To mitigate operating costs and load fluctuations, the authors of [9,10] propose a multi-objective optimal operation model that includes DR. Nevertheless, their analysis does not consider carbon capture, internal energy replication, or cost-effective sharing. These aspects are addressed in [11,12,13], but the DR method for customer load flexibility is excluded. In [14], the author examines the impact of demand response on cooperative energy management in multi-carrier hubs, emphasizing electrical and thermal demand response programs for cost reduction. However, the author does not incorporate an advanced P2G-CCS system, hydrogen integration, or an efficient demand response unit to enhance the model’s sustainability. The MINLP with the BARON solver addresses deterministic real-world problems by combining nonlinear functions with advanced branch-and-bound optimization, constraint propagation, interval analysis, and duality to handle non-convex mathematical programming challenges [15]. In [16], the author focused on minimizing cost and emissions, considering energy trading returns and demand response advancements, and using low-carbon optimization methods. Their model, however, lacked hydrogen sourcing and had fewer conversion units compared to this study’s more environmentally friendly and cost-effective operation. In [17], the author optimizes integrated power and hydrogen networks using power-to-gas and fuel-cell electric vehicles to minimize daily costs. To enhance sustainability, they avoid P2G-CCS integration, limit energy sharing due to the fewer power conversion units, and overlook the economic benefits of demand response programs.

1.2. Research Gaps and Contribution

In conclusion, this review primarily focuses on the internal optimization of a single integrated energy system (IES), which limits the optimal allocation of resources since demand-side assets are not fully considered within the refined P2G-CCS method. The study highlights the value of demand response in linking demand-side resources with the rapid growth of green energy [16].

Table 1. Comparison of the main novelties with several related references.

Ref.	Objective Function			CCU System	Tri-Hub’s IES	Demand Response			Multi-Energy Storage				Prosumer-H2 Synergy Coalition	Schedule Model
	${\mathit{TC}}^{\mathit{min}}$	${\mathit{E}}^{\mathit{min}}$	$T^{\mathit{min}}$			${\mathit{L}}_{\mathit{E}}$	${\mathit{L}}_{\mathit{H}}$	${\mathit{L}}_{\mathit{C}}$	H2	${\mathit{P}}_{\mathit{E}}$	${\mathit{P}}_{\mathit{H}}$	${\mathit{P}}_{\mathit{C}}$
[6]	$✓$	$✓$	×	×	×	$✓$	$✓$	×	×	$✓$	$✓$	×	×	MILP
[9]	$✓$	×	×	×	×	×	$✓$	×	$✓$	$✓$	×	×	×	MILP
[10]	$✓$	×	×	×	×	$✓$	$✓$	×	×	$✓$	$✓$	×	×	MINLP
[11]	$✓$	$✓$	×	$✓$	$✓$	×	×	×	×	$✓$	$✓$	$✓$	×	MILP
[13]	$✓$	$✓$	×	$✓$	×	×	×	×	$✓$	$✓$	×	×	×	MINLP
[17]	$✓$	$✓$	×	×	×	×	×	×	$✓$	$✓$	×	×	×	MILP
[18]	$✓$	$✓$	×	$✓$	×	×	×	×	$✓$	$✓$	×	×	×	MINLP
[19]	$✓$	×	×	×	×	$✓$	$✓$	×	$✓$	$✓$	$✓$	×	×	MILP
[20]	$✓$	×	×	×	×	×	×	×	×	$✓$	$✓$	$✓$	×	MILP
[21]	$✓$	$✓$	×	$✓$	×	×	×	×	×	$✓$	×	×	×	MILP
This Work	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	$✓$	MINLP

$CCU$: Carbon capture and utilization; $T{C}^{min},E^{min},T^{min}$: Optimal operation of total cost, emission, and transportation cost function, respectively; $L_{E,H,C}$: load of electricity, heating, and cooling; $P_{E,H,C}$: electricity, heating, and cooling power.

contrasts the methods utilized in the existing literature and the validation section with the proposed strategy, illustrating its unique aspects.

This paper aims to broaden the scope of optimal resource allocation for individual IES through energy storage and sharing units, including P2G, combined heat pump (CHP), gas turbine (GT), gas boiler (GB), absorption chiller (AC), and electric chiller (EC) tools, as well as carbon capture and transfer mechanisms, by employing a planned tertiary hub-hydrogen-based integrated energy system (TH-IES). To enhance system flexibility and efficient green energy utilization, integrated demand response programming techniques are applied, considering the electrical, heating, and cooling sections of Hub-3 on the consumer side and taking into account the pricing of electricity and natural gas.

Generally, the following are the paper’s significant contributions:

• This work develops a tri-hub hydrogen-based integrated energy system that optimizes cost, reduces emissions, and facilitates energy-sharing coordination among producers, prosumers, and consumer hubs. This study aims to maximize the benefits of green hydrogen through electrolysis. Additionally, P2G can use a methanation reactor to increase energy efficiency and operating economics by preparing natural gas with less wind and photovoltaic electricity.
• This paper introduces an integrated demand response mechanism into the proposed energy-sharing model between energy hubs.
• This study evaluates the proposed model, including CCS-P2G and energy-sharing technology, by comparing it with the traditional multistage operating model. The proposed model enhances the coupling efficiency of electricity, natural gas, and carbon trading markets. TH-HIES can reduce overall costs and emissions by sharing them across nearby hubs.

1.3. Paper Layout

The remaining sections of this paper are organized as follows: Section 2 presents the suggested energy hub architecture. Section 3 discusses problem framing and optimal modeling. Section 4 introduces the simulation results discussion and validation analysis. Finally, Section 5 summarizes the paper’s conclusions.

2. Model Structure

There are three IES hubs, each of which can be expressed as hub 1, hub 2, and hub 3. Each subsystem exhibits distinct functional properties and operational qualities and is installed in a hierarchical manner, enabling better design, optimization, and integration of the overall tertiary hub.

Integrated Energy Hubs Description

Hub-1 is an integrated energy system serving as an energy supplier, offering hydrogen, electricity, cooling, heating, and carbon storage capabilities, as shown in Figure 1. Electrolyzer (EL), fuel cell (FC), and MR units operate to meet demands for hydrogen, electricity, and natural gas, respectively. Additionally, a furnace produces heat for reference heat loads, using natural gas as fuel through the MR tool. If natural gas is unavailable, it must be purchased from the natural gas market.

Energy prosumer Hub-2 and consumer Hub-3 were better designed and integrated into the overall tertiary hub to meet consumer needs. Each hub exhibits distinct functional properties and operational qualities, installed hierarchically as illustrated in Figure 2. Energy prosumer hub generates power and shares it with nearby hubs, including capabilities for heat, carbon, and electricity storage but excluding cooling and hydrogen energy. Gas turbines (GT), gas boilers (GB), and waste heat boilers (WHB) are operated to meet the demands for heat, electricity, and CO₂, respectively. The CCS tool is utilized for trading carbon with nearby Hub-1 to meet the demand of the MR unit and reduce emissions. Additionally, the GT and GB share their electricity and heat energy with adjacent hubs.

As illustrated in Figure 2, Hub-3 can store heating, cooling, and electricity but cannot store carbon or hydrogen energy. A combined heat pump (CHP) system and a gas boiler (GB) are operated to meet the demands for electricity, heat loads, and storage. Cool storage and load consumption are managed by energy converter components such as the electric chiller (EC) and the absorption chiller (AC), which are powered by the electricity from the CHP and the heat from the GB, respectively. A demand response unit is used to manage various elastic loads on the consumer side. This connectivity enables the TH-HIES hub to engage in communication and resource exchange with the external energy market.

Different types of energy storage, such as hydrogen, electricity, heating, cooling, and gas, play unique roles in facilitating efficient energy management and utilization. These storage methods enhance the overall effectiveness and reliability of the proposed model. These measures are implemented to strengthen the integration and coordination of different energy sources and subsystems, thereby facilitating a more efficient and flexible energy network. The proposed model is formulated as an MINLP problem, as presented in the following section.

3. TH-HIES Operating Framework

3.1. Objective Function

To analyze the cost-effective and environmentally friendly gains of the proposed model operation, including the two main objectives of cost and emission functions that are written below in Equations (1) and (2), respectively, the total cost (TC) function encompasses facility cost, demand response cost, and energy share or sale gain. The first nine components like electrical energy buying cost ($C^{P^E}$), hydrogen energy purchased cost ($C^{P^{H_2}}$), gas energy purchased cost ($C^{P^G}$), battery charging–discharging costs of hydrogen, electricity, heating, and cooling ($C^{P^{BCD(E,H2,Ht,C)}}$), wind curtailment cost ($C^{P^{WCC}}$), optimal transportation cost ($C^{OPT}$), carbon trading cost ($C^{CT}$), carbon storage cost ($C^{CS}$), carbon dioxide purchased cost ($C^{C{O}_2^P}$) of Equation (1) explain facility costs, which are clearly listed in Equation (3). Equation (1)’s ten-component representation of the demand response cost illustrates how to programme low- and high-demand responses for adjusting heating, electricity, and cooling loads. Finally, the energy share or sale gain described in Equation (63) is explained by the eleven parts of Equation (1). The effects of planned (TH-HIES) operations on the economy and environment can be fully assessed by taking these elements into account. Equation (1) illustrates the cost-objective function and is consistent with the points that are stated.

$$T{C}^{min}(t,c)=\sum _{t=1}^{24}\sum _{c=1}^N{\mathsf{\Pi }}_c\left\{\begin{array}{c}\begin{array}{c}\hfill [C^{P^E}+C^{P^{H_2}}+C^{P^G}+C^{P^{BCD(E,H2,Ht,C)}}\\ \hfill +C^{P^{WCC}}+C^{OPT}+C^{CT}+C^{CS}+C^{C{O}_2^P}]\\ \hfill +\sum _{j\in (e,h,c)}{\gamma }_{j^{DR}}\left(t\right)[{{DR}^{up}}_j\left(t\right)+\\ \hfill {{DR}^{dn}}_j\left(t\right)]-\sum _{s\in (e,h,c)}{\gamma }_s\left(t\right)\left[P_s^{sale}\left(t\right)\right]\end{array}\hfill \end{array}\right.$$

(1)

$$E^{min}\left(t\right)=\sum _{t=1}^{24}[{\mathsf{\Phi }}_{CHP}\times P_{CHP}^g\left(t\right)+{\mathsf{\Phi }}_B\times P_B^g\left(t\right)+{\mathsf{\Phi }}_E\times P_E\left(t\right)+{\mathsf{\Phi }}_{GT}\times P_{GT}\left(t\right)]$$

(2)

where ${\mathsf{\Phi }}_{CHP}$, ${\mathsf{\Phi }}_B$, ${\mathsf{\Phi }}_E$ and ${\mathsf{\Phi }}_{GT}$ are CO₂ emissions coefficients for CHP, GB, electricity network and GT, respectively.

Furthermore, ${\gamma }_{j^{DR}}\left(t\right)$ and ${\gamma }_s\left(t\right)$ are the DRP tool usage cost and energy sale price for electricity, heating, and cooling loads at time t. Here, ${\mathsf{\Pi }}_c$ represents the demand response programming cases.

$$SubE{q}^{\prime }s\Rightarrow \left\{\begin{array}{c}\begin{array}{c}\hfill C^{P^E}=\sum _{t=1}^{24}\{{\gamma }_e\left(t\right)\times \left(P_e\left(t\right)\right)\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{P^G}=\sum _{t=1}^{24}\{{\gamma }_g\left(t\right)\times \left(P_g\left(t\right)\right)\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{P^{H_2}}=\sum _{t=1}^{24}\{{\gamma }_{H_2}\left(t\right)\times \left(P_{H_2}\left(t\right)\right)\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{WCC}=\sum _{t=1}^{24}\{{\gamma }_{curt}\left(t\right)\times \left(P_{WT}^{curt}\left(t\right)\right)\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{P^{BCD(E,H2,Ht,C)}}=\sum _{t=1}^{24}\{{\gamma }_{E,H2,Ht,C,C^{st}}\left(t\right)\times \\ \hfill [P_{E,H2,Ht,C,C^{dis}}\left(t\right)+P_{E,H2,Ht,C,C^{ch}}\left(t\right)]\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{OPT}=\sum _{sh}\{C_{sh}\left(t\right)\times \left(X_{sh}^2\left(t\right)\right)\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{CT}=\sum _{t=1}^{24}\{P_c\times [(E_{net}^{C{O}_2}\left(t\right)-\left(E_{ceq}^{C{O}_2}\left(t\right)\right]\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{CS}=\sum _{t=1}^{24}\{K_{cs}\times \left[\left(E_{stored}^{C{O}_2}\left(t\right)\right]\right\}\end{array}\hfill \\ \begin{array}{c}\hfill C^{CS}=\sum _{t=1}^{24}\{P_{C{O}_2}\left(t\right)\times \left[\left(E_{MR}^{C{O}_2}\left(t\right)\right]\right\}\end{array}\hfill \end{array}\right.$$

(3)

Utilizing neighboring hubs and efficient energy sharing minimizes dependence on energy market purchases and reduces overall operational costs. The objective function, constructed based on

Table 2. Overall cost of purchasing hydrogen, natural gas and electricity from the grid.

Hub’s	Devices	Process of Mentioned Units for This Proposed Model	Cost Paid
$Hub-1$:	$MR$	$Producenaturalgas$	$Nill$
$Hub-2$:	$GT,GB$ $WHB$	$Convertnaturalgasto$$electricityandheat$	$40\%$
$Hub-3$:	$CHP$  $AC$ $CHP$  $EC$	$Convertnatural$$gastoelectricity$$Convertheat$$tocooling$ $Convertnaturalgastoheat$$Convertelectricitytocooling$	$40\%$$60\%$ $60\%$$40\%$

’s presumptions, details the cost implications of energy procurement for various devices from the corresponding grid, enhancing the cost-effectiveness of the proposed model.

The second objective function explains the quantity of emissions emitted by the proposed model and energy networks minimized, which is shown above in Equation (2).

The proposed energy hub system covers various types of equipment such as CHP, GT, GB and the electric grid. During their operation, these components play a crucial role in managing the distribution of greenhouse gases. Specifically, within the system, the component responsible for handling internal emissions is commonly referred to as the energy hub.

This study analyses the dispersion of greenhouse gases in energy systems, specifically in energy hubs, with a focus on emissions from energy carriers and their effects on the environment.

3.2. Facility Cost Level Operation Constraints

3.2.1. Advanced Power to Hydrogen Gas and Methanation Reactor Constraints

This study advances P2G technology by developing processes for hydrogen-based energy production and methanation reactors. MR, which converts hydrogen into natural gas, is facilitated by P2G [13]. The aim is to optimize green hydrogen utilization by introducing a hydrogen storage subsystem and hydrogen fuel cell (HFC) technology for electricity generation. An electrolyzer converts electricity into hydrogen, particularly during periods of high hydrogen prices, using power from the grid or renewable sources like wind or solar turbines.

The fuel cell operates by converting hydrogen into electricity, particularly during periods when the price of electricity is high. Equation (4) can be used to express the relationship between the volume of hydrogen produced and the electricity used by EL. According to Equation (5), EL’s output power must adhere to ramp-up and ramp-down limits. Equations (6) and (9) set limits on the input and output power of EL and FC, respectively. Hydrogen consumption constraints across FC are represented in Equations (7) and (8), respectively. Equations (9) and (10) are the operation restrictions for HFC. Furthermore, Equation (10) indicates that both the fuel cell and the electrolyzer cannot be operational simultaneously [22].

$$H_{EL}\left(t\right)=[\frac{{\lambda }_{EL}\times P_{EL}\left(t\right)}{H{2}_{LHV}}]$$

(4)

$$P_{EL}^{down}\left(t\right)\le P_{EL}\left(t\right)\le P_{EL}^{up}\left(t\right)$$

(5)

$$0\le P_{EL}\left(t\right)\le P_{EL}^{max}\left(t\right)\times I_{EL}\left(t\right)$$

(6)

$$H_{FC}\left(t\right)=[\frac{{\lambda }_{FC}\times P_{FC}\left(t\right)}{H{2}_{LHV}}]$$

(7)

$$0\le H_{FC}^2\left(t\right)\le H_{FC}^2\left(t\right)$$

(8)

$$0\le P_{FC}\left(t\right)\le P_{FC}^{max}\left(t\right)\times I_{FC}\left(t\right)$$

(9)

$$I_{EL}\left(t\right)+I_{FC}\left(t\right)\in [0,1]$$

(10)

To store electricity during off-peak times or when wind/solar turbines produce excess power, hydrogen storage systems are used.The hydrogen tank pressure’s limitation and the beginning pressure value are shown in Equations (11), (12) and (14) [22]. The hydrogen storage tank’s pressure levels are shown in Equation (13). Equation (15) is used to stop a hydrogen storage tank from being charged and discharged at the same time.

$$0\le H_{EL}\left(t\right)\le H_{HSS}^{ch}\left(t\right)\times I_{ch{H}_2}\left(t\right)$$

(11)

$$0\le H_{EL}\left(t\right)\le H_{HSS}^{dis}\left(t\right)\times I_{dis{H}_2}\left(t\right)$$

(12)

$$SO{C}_{H_2}\left(t\right)=[SO{C}_{H_2}(t-1)\times H_{HSS}^{ch}\left(t\right)-\frac{H_{HSS}^{dis}\left(t\right)}{{\lambda }_{dis}}]\times \frac{Cons{t}_{H_2}}{Vo{l}_{H_2}}$$

(13)

$$SO{C}_{H_2}^{initial}=SO{C}_{H_2}^{t=24}$$

(14)

$$I_{c{h}_{H_2}}\left(t\right)+I_{di{s}_{H_2}}\left(t\right)\in [0,1]$$

(15)

3.2.2. Methanation Reactor Constraints

The correlation between synthetic natural gas (SNG) and hydrogen consumption in the MR process can be expressed through Equation (16). Equations (17) and (18) incorporate the operational, ramp-up and ramp-down constraints of the MR process [13].

$$S_{MR}\left(t\right)=[\frac{{\lambda }_{MR}\times H_{MR}\left(t\right)}{H{2}_{LHV}}]$$

(16)

$$0\le H_2^{MR}\left(t\right)\le {H_2^{MR}}^{max}\left(t\right)$$

(17)

$${H_2^{MR}}^{down}\left(t\right)\le H_2^{MR}\left(t\right)-H_2^{MR}(t-1)\le {H_2^{MR}}^{up}\left(t\right)$$

(18)

3.2.3. Energy Supply Equipment’s Model

The correlation between wind speed and turbine output power in a day is represented by Equation (19) [23,24]. This equation signifies that the wind turbine generates its maximum output power within the wind speed range of $v_r$ to $v_{co}$. Furthermore, Equation (20) illustrates that the electrical power generated by a photovoltaic system depends on the solar irradiation and weather temperature at different hours of the day [25]. Operation constraints of wind, solar power curtailment and energy buying from the grid are represented by Equation (21) to Equation (23), respectively [11]. All these parameter values $v_r$, $v_{co}$, $v_{cn}$, $S_{max}$, $T_{NOC}$, $S_{iro}$, $P_{max}$ for the wind and solar models are shown online https://doi.org/10.6084/m9.figshare.23804517, accessed on 6 June 2024.

$$P_{WT}\left(t\right)=\left\{\begin{array}{cc}0\hfill & \hfill if:V_s\left(t\right)\ge v_{co}or{V}_s\left(t\right)\le v_{cn}\\ \frac{V_s\left(t\right)-v_{cn}}{v_r-v_{cn}}\times P{r}_{WT}\hfill & \hfill if:v_{cn}\le v_s\left(t\right)\le v_r\\ P{r}_{WT}\hfill & \hfill Otherwise\end{array}\right.$$

(19)

$$P_{PV}\left(t\right)=\{P^{max}+S_{max}(T\left(t\right)+S_{ir}\left(t\right)\times \frac{T_{NOC}-20}{0.8}-T_r)\}\times \frac{S_{ir}\left(t\right)}{S_{iro}}$$

(20)

$$0\le P_{WT}^{cur}\left(t\right)\le P_{WT}\left(t\right)$$

(21)

$$0\le P_{PV}^{cur}\left(t\right)\le P_{PV}\left(t\right)$$

(22)

$$0\le P_{Grid}^E\left(t\right)\le P_{bu{y}_{Grid}}^{max}\left(t\right)$$

(23)

A specific ratio exists between the amount of electricity and heating power produced by GT considering natural gas as a fuel [26]. Equations (24) and (25) explain the generated power and heat through GT, respectively, considering CH₄ as a fuel. The electric power ramp-up and ramp-down constraint is shown in Equation (26). The maximum generated power limitation constraint is shown in Equation (27) and $u_{gt}$(t) is a binary variable. Here, the efficiency parameter value ${\lambda }_{GT}$ and heating loss coefficient of GT are 35% and 10%, respectively.

$$P_{GT}\left(t\right)={\lambda }_{GT}\times S_{GT}\left(t\right)$$

(24)

$$Hea{t}_{GT}\left(t\right)=\frac{P_{GT}\left(t\right)[1-{\lambda }_{GT}-{\lambda }_{loss}]}{{\lambda }_{GT}}\times \left(CO{P}_{Ht}\right)$$

(25)

$$P_{down}^{GT}\left(t\right)\le P^{GT}\left(t\right)-P^{GT}(t-1)\le P_{up}^{GT}\left(t\right)$$

(26)

$$0\le P^{GT}\left(t\right)\le P_{max}^{GT}\left(t\right)\times u_{gt}\left(t\right)$$

(27)

Heating power generated using GB, WHB, and CHP in hubs 1 to 3, respectively, by burning CH₄ represents Equation (28) to Equation (30) [26]. The same set of constraints is also used for WHB and CHP units. Here, efficiencies parameter ${\lambda }_{GB}$, ${\lambda }_{WHB}$ and ${\lambda }_{CHP}$ values are 75%, 75% and 35%, respectively. Maximum heat limit’s values are given in data online https://doi.org/10.6084/m9.figshare.23804517, accessed on 6 June 2024.

$$H{t}_{GB}\left(t\right)={\lambda }_{GB}\times S_{GB}\left(t\right)$$

(28)

$$0\le H{t}^{GB}\left(t\right)\le H{t}_{max}^{GB}\left(t\right)\times u_{gb}\left(t\right)$$

(29)

$$0\le u_{gb}\left(t\right)\le 1$$

(30)

With increasing usage of RERs, energy storage systems are essential for reducing the ecological impact and meeting the rising demand for electricity. By injecting or extracting energy, it controls the power flow in RESs, reducing their periodic nature [27]. Equations (31)–(37) model heat charge–discharge, battery unit commitment, storage limits, heat loss, and initial charging status. They are integral to the proposed model, encompassing constraints for cooling, electricity storage banks, and variable DRP’s tools [28].

$$\begin{array}{c}\hfill {\mathsf{\Psi }}_H^{min}\times \frac{P_{H{t}_f}^{max-ch}}{{\eta }_H^{ch}}\times I_{H_f}^{ch}(t,c)\le P_{H_f}^{ch}(t,c)\le {\mathsf{\Psi }}_H^{max}\times \frac{P_{H{t}_f}^{max-ch}}{{\eta }_H^{ch}}\times I_{H_f}^{ch}(t,c)\end{array}$$

(31)

$$\begin{array}{c}\hfill I_{H_{f_{(t,c)}}}^{dis}(P_{H_f}^{max-dis}\times {\eta }_{di{s}_H}){\mathsf{\Psi }}_H^{min}\le P_{H_{f_{(t,c)}}}^{dis}\le P_{H_f}^{max-dis}({\eta }_{di{s}_H}\times {\mathsf{\Psi }}_H^{max})I_{H_{f_{(t,c)}}}^{dis}\end{array}$$

(32)

$$I_{H_f}^{ch}(t,c)+I_{H_f}^{dis}(t,c)\in [0,1]$$

(33)

$$P_H^{max}\times {\phi }_H^{min}\le P_H(t,c)\le P_H^{max}\times {\phi }_H^{max}$$

(34)

$$P_H^{loss}(t,c)={\gamma }^{loss}\times P_H(t,c)$$

(35)

$$P_{Ht}(t,c)=P_{Ht}(t-1,c)+P_{H{t}_f}^{ch}(t,c)-P_{H{t}_f}^{dis}(t,c)-P_H^{loss}(t,c)$$

(36)

$$P_{ht}^{initial}=P_{ht}^{t=24}$$

(37)

Additionally, the energy conversion equipment’s (EC and AC) mathematics model is represented by these equations from (38) to (39) [11].

$$P_C^{EC}\left(t\right)={\lambda }^{EC}\times P_E^{EC}\left(t\right)$$

(38)

$$P_C^{AC}\left(t\right)={\lambda }^{AC}\times P_{Ht}^{EC}\left(t\right)$$

(39)

where ${\lambda }_{EC}$ and ${\lambda }_{AC}$ are efficiencies for EC and AC in hub 3. AC has a chilling efficiency of 1.2. EC has a cooling efficiency of 3.5 [11].

3.2.4. Carbon Trading, Emission, and Government Quota Allocation Constraints

Carbon capture storage in the CCS-P2G system captures CO₂ emissions from GT, GB, and WHB. “Storage” refers to the long-term isolation of some CO₂ from the atmosphere. In the P2G process, water electrolysis produces hydrogen, which reacts with CO₂ via a catalyst, yielding natural gas used as fuel for GT, CHP, WHB, and GB, measured in kilograms (kg) and cubic meters (m³). Equations (40) and (41) quantify overall carbon emissions and captured CO₂ by CCS, respectively, and are used to calculate net CO₂ emissions as expressed in Equation (42) [16,29].

$$E_{total}^{co2}\left(t\right)={\sigma }_{gas}^{co2}\times S_{GB}\left(t\right)+{\sigma }_{gas}^{co2}\times S_{GT}\left(t\right)+{\sigma }_{grid}^{co2}\times P_{buy}\left(t\right)$$

(40)

$$E_{cap-css}^{co2}\left(t\right)={\lambda }_{ccs}\times E^{C{O}_2}\left(t\right)$$

(41)

$$E_{net}^{co2}\left(t\right)=E_{total}^{co2}\left(t\right)-E_{cap-css}^{co2}\left(t\right)$$

(42)

Equation (43) illustrates carbon trading or how much CO₂ the MR uses to create natural gas.

$$E_{MR}\left(t\right)=\phi \times H_2^{MR}\left(t\right)$$

(43)

where $\phi$ is the CSS unit’s carbon capture efficiency and has a value of 1.35 [13].

The captured CO₂ emissions at time t, as indicated in Equation (44), are computable. A fraction of the captured CO₂ serves as raw material for the MR process, while the rest is allocated for storage, as specified in Equation (45) [29]. Additionally, the electricity consumption linked to the CCS process at time t is expressed in Equation (46) [11,13].

$$E_{cap-css}^{co2}\left(t\right)={\lambda }_{ccs}\times E^{C{O}_2}\left(t\right)$$

(44)

$$E_{stored}^{co2}\left(t\right)=E_{cap-ccs}^{co2}\left(t\right)-E_{MR}^{co2}\left(t\right)$$

(45)

$$P_{ccs}\left(t\right)={\lambda }_{ccs}\times E_{cap-ccs}^{co2}\left(t\right)$$

(46)

Under the carbon market framework, each emission source is allocated a specific carbon emission quota by the government, as depicted by Equation (47) for the planned model [30].

$$E_{ceq}^{CO2}\left(t\right)={\sigma }_{ceq-h/gt}^{co2}[P_{GT}\left(t\right)+H_{GT}\left(t\right)]+{\sigma }_{ceq-WT/buy}^{co2}[P_{WT}\left(t\right)+P_{buy}\left(t\right)]$$

(47)

where ${\sigma }_{ceq-pgt}^{co2}$, ${\sigma }_{ceq-hgt}^{co2}$ and ${\sigma }_{ceq-WT/Grid}^{co2}$ are the ordinary values for per-unit output of power-heat from GT, WT and grid power plant (CO₂-t/MWh), and these data are provided online https://doi.org/10.6084/m9.figshare.23804517, accessed on 6 June 2024.

3.2.5. Energy Balance Constraints

In the context of electricity, the supply is comprised of various sources, including power obtained from the main grid, fuel cell, wind turbine, photovoltaic, gas turbines and CHP units, and discharged power from ESS. Conversely, the demand encompasses the charged power of the ESS, the power consumed by CCS systems, electrical loads and the overall power load, as shown in Equation (48). Similarly, heating, cooling, H₂ and CH₄ energy balance constraints are considered from Equation (49) to Equation (52), respectively.

$$L_E\left(t\right)=\left\{\begin{array}{c}[P_{WT}\left(t\right)+P_{PV}\left(t\right)+P_{GT}\left(t\right)+P_{FC}\left(t\right)+P_{Grid}\left(t\right)]\hfill \\ +[P_e\left(t\right)+P_{ESS}^{dis}\left(t\right)+P_E^{DR-low}\left(t\right)-P_E^{DR-high}\left(t\right)]\hfill \\ -[P_{ESS}^{ch}\left(t\right)+P_{CCS}\left(t\right)+P_{WT}^{curt}\left(t\right)+P_{EL}\left(t\right)+P_{chill}^e\left(t\right)]\hfill \end{array}\right.$$

(48)

$$L_{ht}\left(t\right)=\left\{\begin{array}{c}[P_{ht}^{CHP}\left(t\right)+P_{ht}^B\left(t\right)+P_{ht}^{dis}\left(t\right)+P_{ht}^{DR-low}\left(t\right)]\hfill \\ -[P_{ht}^{DR-high}\left(t\right)+P_{f_{ht}}^{ch}\left(t\right)+P_{ht}^{AC}\left(t\right)]\hfill \end{array}\right.$$

(49)

$$L_C\left(t\right)=\left\{\begin{array}{c}[P_C^{EC}\left(t\right)+P_C^{AC}\left(t\right)+P_{C_f}^{dis}\left(t\right)+P_C^{DR-low}\left(t\right)]\hfill \\ -[P_C^{DR-high}\left(t\right)+P_{f_C}^{ch}\left(t\right)]\hfill \end{array}\right.$$

(50)

$$L_{H_2}\left(t\right)=\left\{\begin{array}{c}[H_2^{EL}\left(t\right)+H_2^{buy}\left(t\right)+H_2^{dis}\left(t\right)]\hfill \\ -[H_2^{FC}\left(t\right)+H_2^{ch}\left(t\right)+H_2^{MR}\left(t\right)]\hfill \end{array}\right.$$

(51)

$$S_{MR}\left(t\right)=[S_{GB}\left(t\right)+S_{GT}\left(t\right)-S_{buy}\left(t\right)]$$

(52)

In Figure 3, the proposed model illustrates unified energy flow and sharing among equipment using matrix approach methods, enhancing grasp.

3.2.6. Optimal Transportation Cost Constraints

The optimal transportation problem and its model represent from Equation (53) to Equation (58). Here, it is assumed that the transportation cost is directly proportional to the square of the transported quantity from a supply node (S) to the corresponding demand hub (H) and influenced by route length. Every route’s cost coefficient, maximum flow, demand, and producer capacity are shown online https://doi.org/10.6084/m9.figshare.23896524, accessed on 6 June 2024. [31].

$$T{r}_{min}^{SH}=\sum _{SH}[C_{SH}\times X_{SH}^2)]$$

(53)

$$P_S^{min}\le P_S\left(t\right)\le P_S^{max}$$

(54)

$$P_S\left(t\right)=0$$

(55)

$$\sum _S{X}_{SH}\ge D_H$$

(56)

$$\sum _H{X}_{SH}=P_S$$

(57)

$$0\le X_{SH}\le X_{SH}^{max}$$

(58)

3.3. Integrated Demand Response Programming’s (IDRPs) Model

The mathematical model of IDRP’s tool is useful for facilitating suitable load shifting and ensuring optimally planned model management, which is mentioned as the tenth component in the cost objective function [32]. Within the energy hub structure, it helps to reduce the use of electricity, natural gas, and total model operation costs [25,33]. To evaluate the impact of IDRPs in the proposed model, the load-sharing rates in this paper are considered, exactly at levels of 0% and 20%. Consequently, the following constraints from Equation (59) to Equation (62) relate to IDRPs techniques for electrical, heating, and cooling elastic loads respectively.

Equation (59) represents the stability between adjustable high and low customer demands, considering a high and low price time, respectively. Equations (60) and (61) further describe the permissible limits for the shifted high and low specific load demands during low and high price times, respectively, based on the extent of their presence in the IDRPs. Equation (62) imposes limitations on the use of IDRPs for load-shifting purposes within specific time periods [28,34].

$$\sum _{t=1}^{24}\{P_{DR{P}_{low}}^{E,H,C}\left(t\right)\}=\sum _{t=1}^{24}\{P_{DR{P}_{high}}^{E,H,C}\left(t\right)\}$$

(59)

$$0\le P_{DR{P}_{high}}^{E,H,C}\left(t\right)\le [P_{DR{P}_{high}}^{E,H,C}\left(t\right)\times L^{E,h,C}\left(t\right)]\times U_{high}^{E,H,C}\left(t\right)$$

(60)

$$0\le P_{DR{P}_{low}}^{E,H,C}\left(t\right)\le [P_{DR{P}_{low}}^{E,H,C}\left(t\right)\times L^{E,h,C}\left(t\right)]\times U_{low}^{E,H,C}\left(t\right)$$

(61)

$$U_{low}^{E,H,C}\left(t\right)+U_{high}^{E,H,C}\left(t\right)\in [0,1]$$

(62)

3.4. Energy Sharing Model

In this paper, energy sharing is defined by the power sales equation in Equation (63) of the TH-HIES model, detailed as eleven components in Equation (1). This model integrates the electricity and heat power equations for CHP, GT, GB, and WHB units and includes natural gas sharing via MR tools and cooling power from EC or AC units. By utilizing these elements in conjunction with RERs for electricity, gas, carbon trading, and heat transfer, including the use of internal heat or electricity for cooling power production and IDRPs, the model effectively reduces costs associated with procuring electricity and natural gas from upstream grid markets.

$$P_{sale}\left(t\right)=\sum _{t=1}^{24}\sum _{c=1}^N{\Pi }_c\left\{\begin{array}{c}40\%c_e\left(t\right)[P^{CHP}(t,c)+P_{GT}(t,c)+H{t}_{GT}(t,c)\hfill \\ +H{t}_{GB}(t,c)+H{t}_{WHB}(t,c)+{P_C}^{EC}(t,c)]\hfill \\ +[60\%.c_g\left(t\right)[{P_{ht}}^{CHP}(t,c)+{P_C}^{AC}(t,c)]\hfill \end{array}\right.$$

(63)

In conclusion, the modeling plan outlined in this study is depicted as a flowchart in Figure 4.

4. Simulation Results and Discussion

The effectiveness of the proposed technique is validated using the TH-HIES structure, as illustrated in Figure 2. Additionally, the simulations were executed using GAMS version 24.4.6 on a system equipped with a 3.9 GHz Intel Core i3-7100 dual-core processor, 4 GB of 2400 MHz DDR4 RAM, and a 128GB SSD.

4.1. Data Description

This study uses numerical simulations to evaluate the proposed model’s viability and superiority within a one-day dispatch cycle. Figure 5a,b displays demand profiles for electricity, heating, and cooling, featuring both fixed (0% DRPs) and (20% DRPs) adjustable loads, respectively, represented by distinct bar segments in blue, orange, and aqua on the y-axis.

The technical parameters, comprehensive details, and associated costs for all hub equipment are provided online in https://doi.org/10.6084/m9.figshare.23804517, accessed on 6 June 2024, as well as in references [13,22,28]. The simulated values for the maximum and minimum allowable electricity, heating, and cooling power factor parameters ${\mathsf{\Psi }}_E^{max/min}$, ${\phi }_E^{max/min}$, ${\mathsf{\Psi }}_H^{max/min}$, ${\phi }_H^{max/min}$, ${\mathsf{\Psi }}_C^{max/min}$,${\phi }_C^{max/min}$ are 0.14, 0.05, 0.92, 0.05, 0.15, 0.05, 0.92, 0.09, 0.16, 0.05, 0.92, 0.05, respectively [28]. The cost coefficients, maximum flow, demand, and producer capacity for each route to achieve optimal transportation data are shown online in https://doi.org/10.6084/m9.figshare.23896524, accessed on 6 June 2024. [31]. Predicted wind speed (m/s), solar radiation (KW/m²), ambient temperature (K), and imported power based on market prices for the planned model are shown online in https://doi.org/10.6084/m9.figshare.24147339, accessed on 6 June 2024 [35].

4.2. Case Study

The advantageous outcomes of energy sharing are explored in the following two cases:

• Evaluate the proposed integration of multiple networks encompassing power, gas, heating, cooling, and hydrogen while incorporating carbon capture and storage and power-to-gas technologies. This assessment includes the analysis of power transportation costs, the presence of an energy storage system (ESS), energy sharing, and carbon trading, all under a fixed load condition with a 0% DRP.
• Investigate system flexibility and energy utilization efficiency by implementing a 20% DRP across electricity, heating, and cooling loads on the consumer side, with all other conditions remaining consistent with those in case 1.

4.3. Scheduling Results Analysis

For simplicity, the proposed (TH-HIES) are classified as producer, prosumer, and consumer hubs. The model is developed by integrating multiple subsystems in the simulated example.

4.3.1. Electricity, Heating, Cooling, and Hydrogen Supply vs. Demand Balance Analysis

In each result displayed in Figure 6, Figure 7, Figure 8 and Figure 9, the up and down bars of varying colors represent the combined power output of different generating and consuming units. The supply and demand equations are satisfied when supply and demand are visually represented in the same manner.

Figure 6 illustrates the electrical energy supply and customer demand for Cases 1 and 2.

In the top section, the orange bars indicate the total power generation from electricity sources, while in the bottom section, purple bars represent the total electricity usage across various equipment and electric loads. With integrated internal generating resources, including fuel cells, batteries, and shared CHP units, the 0% DRP-based load in Case 1 of Figure 6a is primarily served by WT, PV, GT, and the upstream grid. This energy-sharing technique reduces reliance on the grid, ensuring cost-effectiveness and alleviating grid stress. Customer demand fluctuates throughout the day, depending on whether it is peak hours (11 a.m. to 15 p.m. and 19 p.m. to 21 p.m.), mid-time (7 a.m. to 10 a.m. and 16 p.m. to 18 p.m.), or off-peak time (1 a.m. to 6 a.m. and 22 p.m. to 24 p.m.).

Integrating DRP methods at 20% in Case 2 enables load shifting to off-peak and mid-peak hours, enhancing flexibility, as illustrated by the down-brown bars in Figure 6b. This method maximizes flexible load utilization, promoting grid stability, cost savings, emissions reduction with green energy, and efficient use of energy-sharing technology and multi-energy storage. These benefits align with environmental and economic goals, as shown in

Table 3. Carbon emissions compared with an optimization-based IES with and without considering H₂.

Cases	This Work	[35] P : 10	[36] P : 10	[37] P : 11–14	[13] P : 12	[38] P : 10
	em/t	em/t	em/t	em/t	em/t	em/t
1	$2.343$	$11.352$	$7.26$	$26.8$	1580	$1512.76$
2	$2.035$	$12.192$	$6.68$	$50.1$	1600	$1365.80$

P: Represent page number; em/t: Emission per tonne.

and

Table 4. Optimal TH-HIES scheduling cost and execution time for Cases (1–2).

Cases	Model Operation	Imported	MES Sharing	Power	Total	Exe-
	MES, and	Carbon and	and CT	Transport	Model	Cution
	Imported Cost	Storage Cost	Return	Cost	Cost	Time
	( ¥ × ${\mathbf{10}}^{\mathbf{3}}$)	( ¥ × ${\mathbf{10}}^{\mathbf{3}}$)	( ¥ × ${\mathbf{10}}^{\mathbf{3}}$)	( ¥ × ${\mathbf{10}}^{\mathbf{3}}$)	( ¥ × ${\mathbf{10}}^{\mathbf{3}}$)	(ms)
1	$238.0705$	$8.6580$	$-27.1835$	$2.3741$	$221.919$	32
2	$224.0630$	$6.9180$	$-27.3496$	$2.3741$	$206.006$	31

$MES$: Multi-energy (H₂, electricity, heating, and cooling) storage; $CT$: carbon trading.

, respectively.

Similarly, the graphs balancing heating and cooling supply and demand for Cases 1 and 2 are displayed in Figure 7 and Figure 8, respectively. In Case 1, Figure 7a and Figure 8a demonstrate that the required fixed heating and cooling loads are primarily satisfied by the GB, GT, WHB, and CHP for heating and the EC and AC for cooling, without the employment of the IDRPs tool.

Conversely, Case 2 operates efficiently using the IDRP tools at specific flexible (20%) heating and cooling loads. During periods of excess heat output, the CHP and GB units are also used to operate the AC for cooling generation and the heat storage bank. During heat peak hours, the cool generation is satisfied by the EC unit, which is powered by electricity shared from a nearby CHP unit. The upward-shifted yellow bars in Figure 7b, referencing Figure 7a, illustrate how the proposed model facilitates a reduction in peak-period heat energy consumption. This is achieved through benefits such as financial incentives, energy-saving initiatives, and heightened consumer awareness, all of which contribute to improved energy usage efficiency. Peak demand hours for heating and cooling are between 11:00 a.m. and 15:00 p.m, and 7:00 p.m. and 2:00 a.m. Mid-peak hours are from 7:00 a.m. to 9:00 a.m. and 4:00 p.m. to 6:00 p.m., while off-peak hours are from 1:00 a.m. to 7:00 a.m. and 10:00 p.m. to 22:00 a.m.

The results of the optimal hydrogen and natural gas power dispatching for Cases 1 and 2 are illustrated in Figure 9. The HIES produces hydrogen through an electrolyzer and stores it in a hydrogen storage bank. The minimal discharge of hydrogen, and limited allocation to the MR and FC indicated by gray, yellow, and red bars in the top and bottom halves of Figure 9a, respectively, results in their sparse representation. This illustrates an optimized production process, where output closely aligns with demand, minimizing excess discharge and ensuring efficient utilization. However, it occasionally necessitates purchasing hydrogen from the market, as depicted in Figure 9a.

The operating costs of the model escalate due to the requirement to procure hydrogen from the market and the need to activate additional components within the hub for energy producers and consumer hubs. Consequently, we also acquired natural gas and CO₂ to operate the GT, GB, CHP, and MR units, respectively. All these imported fuels significantly impact the operating costs of the proposed model.

Utilizing CCS technologies for hydrogen and carbon trading, the MR unit is operated to produce natural gas. Without the MR plant, the reliance on imported natural gas would be significantly higher, as depicted by the red bars in Figure 9b. However, the implementation of MR reduces the necessity for purchasing CH₄, as illustrated by the green bars in the same figure, thereby enhancing the model’s efficiency. This demonstrates that advancements in P2G technology, energy exchange systems, IDRPs, and CCS positively impact power matching. Furthermore, carbon emission trading can augment the utilization rate of the hydrogen storage system facilitated by the MR.

Figure 10a illustrates the energy sharing among the hubs. Hub-1 uses a GB to meet its heat demand, as depicted by the purple bars.

During peak times, Hub-1’s GB output is insufficient due to natural gas sharing with nearby hubs. To compensate, Hub-1 purchases heat from Hub-2, utilizing power from the GT and WHB, as shown by the blue and black bars. Hub-1’s average daily heat load is 292.375 kW, but its GB produces only 0.0527 kW. The proposed energy-sharing method balances this deficiency. Hub-2 shares its excess heat, averaging 239.737 kW, at a lower price, represented by the green bars. Additionally, Figure 6, Figure 7, Figure 8, Figure 9 and Figure 10 graphically represent the behaviors of energy production, sharing, and consumption units among these proposed hubs.

Figure 10b shows the electricity, heating, cooling, and hydrogen SOC during one day for cases 1–2, with 0% and 20% DRPs. Hydrogen SOC is minimal initially due to limited renewable energy or higher early consumption. As midday approaches and renewable energy peaks, hydrogen production and storage increase, raising SOC hourly. In the final hours, SOC stabilizes, indicating balanced production and consumption, ensuring stable energy supply and optimal operation. This strategy maximizes renewable energy use and maintains a steady hydrogen supply. Similarly, electricity, heating, and cooling SOC fluctuate with varying loads, dynamically adjusting during DRPs to balance supply and demand, optimizing efficiency, reducing emissions, and ensuring reliable battery service.

4.3.2. Assessing Integrated CCS, IDRP, and RERs to Optimize Grid Power

On the upper side of Figure 11a, the total carbon emissions by TH-HIES are displayed as gray bars. The orange bars represent the state-allocated carbon emission rate for the model, while the dark green bars indicate the CO₂ stored after the MR process. In the negative section of Figure 11a, a cyan bar represents the electricity consumed by all CCS devices, and a purple bar denotes the carbon captured by these CCS devices. The blue bars show net emissions; the upper half depicts CO₂ emissions exceeding capture within state quotas, while the lower half shows CO₂ capture surpassing emissions, resulting in net negative emissions and reduced atmospheric CO₂ levels.

Excess CO₂ emissions are used for the state CO₂ quota, marketing, storage, and distribution by applying CCS technology in the prosumer hub. In addition, it is used in an energy system with a P2G plant to produce CH₄ through an MR facility using the chemical reaction CO₂ + H₂ → CH₄. An energy system must procure its CO₂ feedstock on the open market if the planned CCS technology is not implemented. As the volume of stored carbon increases, the cost of carbon storage also rises.

The operational costs of this model are directly influenced by the amount of electricity purchased from the grid during peak hours. Utilizing the IDRP tools and the concept of energy sharing through internal generation and prosumer hubs, Figure 11b demonstrates the efficiency of these planned energy scheduling techniques, which integrate WT, PV, DRP, and the main grid. For Case 1, the gray bars in Figure 11b represent the power obtained from the main grid, indicating that the DRP tools are not utilized. During this period, peak demand is managed using WT and PV energy, as shown by the yellow and red bars at the top. Peak demand can be met by energy-sharing units, depicted by the green bars at the bottom, which are less impacted by the main grid during weather changes. The application of DRP tools in Case 2, primarily through RERs and energy sharing via producer and prosumer hubs, shifts the load from peak to off-peak and mid-peak hours. This strategy effectively eliminates the aqua bars during peak hours, aiming to reduce dependence on the main grid and ensure optimal operation.

4.4. Environment–Economic–Energy Benefit Analysis

This section analyzes Cases 1 and 2, discussing how an integrated system impacts the economy, the environment, and the utilization of green energy. Additionally, it examines the use of IDRP tools for elastic loads and the updated P2G and CCS systems that incorporate hydrogen into carbon trading.

4.4.1. Environment Benefit Analysis

The carbon capture and emissions results, depicted in Figure 11a and represented as an orange bar on the top side, demonstrate levels below state emission rates. Alongside green power sources and renewable resources, these results underscore their vital role in ensuring the proposed TH-HIES operates with maximum efficiency and environmental sustainability.

By leveraging the full potential of green energy within this model, carbon emissions were reduced by 13.14% in Case 2 through the implementation of the proposed CCS and 20% IDRP tools. Conversely, Case 1, as shown in Table 3, illustrates a scenario with 0% IDRP.

4.4.2. Economic Benefit Analysis

As detailed in the validation section,

Table 5. Validation analysis with existing work.

Ref.	Total Cost Reduction Rate	Total Emission Reduction Rate	Multi-Energy Sharing and Carbon Trading Return Rate	Integrated DRP Progress	Hydrogen Fuel or Electrolysis Precess
[6]	15.75%	×	×	15.75%	×
[11]	6.93%	9.57%	1.85%	×	×
[13]	31.68%	×	17.59%	×	$✓$
[16]	14.45%	0.84%	3.62%%	6.44%	×
[19]	16.8%	×	×	4.33%	$✓$
[21]	0.51%	0.51%	1.64%	×	$✓$
This work	17.30%	13.14%	13.28%	7.17%	$✓$

✓: This implies; ×: This does not imply.

shows that the proposed model significantly enhances asset capacity and reduces operating costs compared to other models. The integration of import–export multi-energy systems, IDRP, and hydrogen extraction via MR reduces costs. Furthermore, multi-energy sharing during peak hours, carbon trading, and advanced P2G technology enable producer and prosumer hubs to efficiently convert more gas for energy sharing, resulting in superior market pricing, as shown in Table 4. It should be noted that CO₂ must be procured from an external source, incurring a cost of 8.658 × ${10}^3$ ¥.

Regarding the overall cost, Table 4 demonstrates that in Case 1, the total operational cost was reduced by 10.95% and 11.35% due to stored energy sharing and carbon trading in Cases 1 and 2, respectively. In Case 2, there was a total cost reduction of 7.17% compared to Case 1, attributable to the usage of a 20% IDRP unit. Consequently, by eliminating the need to acquire gas from the external market, the integration of these two techniques substantially enhances the economic efficiency of TH-HIES. Furthermore, the simulation results of Cases 1 and 2 indicate that considering 20% IDRP tools for the proposed model operations, which include a hydrogen source and energy sharing with efficient integration among these hubs in China, is economically advantageous.

The optimal operation of decentralized energy networks (residential, commercial, and industrial) is a critical emerging application in real-world sustainable energy management. Therefore, the proposed simulation model optimizes renewable energy usage for hydrogen production, consumption, and storage through multi-energy static conversion units, significantly reducing operational costs. It also incorporates demand response to mitigate peak energy prices, lowers energy costs through local energy exchanges, and enhances sustainability by maximizing the use of high-penetration renewable energy resources.

4.5. Validation Analysis of Proposed Model

4.5.1. Method Comparison

To evaluate this model’s effectiveness, ecological sustainability, and customer flexibility, this study assessed a variety of proposed methods. These methods include the demand response program, P2G-CCS, an integrated model with hydrogen-based energy sharing, and another integrated model without energy sharing, utilizing either hydrogen or another fuel. The main novelty of this study is illustrated in Table 1, which compares these methods with references [6,9,10,11,13,17] through [21]. Additional sources support the validation of the statistical model’s economic and ecological benefits, as highlighted in Table 5, showcasing the disparities addressed in this study. Extensive planning of the components and capacities of various energy hubs has significantly reduced the investment costs associated with energy devices within the energy flow domain [39,40].

4.5.2. Statistical Outcome Validation Considering Non-Hydrogen Fuels

In a study [6], a 15.75% reduction in total costs was achieved through demand response programs in grid-connected microgrid systems. However, the study did not account for carbon trading, total emission reduction, or access to abundant free hydrogen for high renewable energy penetration. Table 5 presents the statistical outcomes, indicating that their proposed strategy is costlier and less environmentally friendly compared to this proposed model. Similarly, reference [11] discusses the integrated energy system group, which incorporates energy sharing and carbon transfer but excludes hydrogen fuel and demand flexibility on the customer side. Table 5 shows a 6.92% reduction in total costs, a 9.57% reduction in emissions, and a 1.85% return rate due to energy trading, highlighting that their strategy is more expensive than the proposed model. According to [16], a 14.45% cost reduction and a 0.84% emission decrease were achieved, with return rates of 3.62% and 6.44% from energy trading and demand response advancements, respectively, using a low-carbon optimal learning and scheduling method for the power system. Notably, their model lacked hydrogen sourcing and had fewer coupled conversion units compared to the proposed model, as outlined in Table 5, indicating a higher cost compared to this work.

4.5.3. Statistical Outcome Validation Considering Hydrogen Fuel

The author of the study [13] presents a hydrogen-based integrated energy system (HIES) operational framework that incorporates the P2G-CCS process within a carbon trading framework. This approach achieves a 31.38% reduction in costs and a 17.59% return rate through advancements in P2G-CCS techniques, resulting in a 20.19% decrease in carbon emissions. However, the framework does not account for multi-energy transportation costs, total optimal emission objectives, multi-hub synergy coordination, cold energy utilization, or demand response units, which would enhance grid sustainability and customer demand flexibility, leading to improved cost-effectiveness and environmental benefits. Another study achieved a 16.80% reduction in total costs, including a 4.33% improvement in electrical and thermal demand response, by considering a hydrogen and ammonia fuel-based energy hub structure [19]. However, their model lacks a CCS unit and a minimal emission function, and it is further constrained by the limited integration of energy conversion units due to its single-hub structure. Consequently, their statistical outcomes, as delineated in Table 5, indicate that their proposed strategy is comparatively more expensive and environmentally unsound relative to the methodology presented in this study. The author of the paper [21] achieved a 0.51% reduction in total operational costs and emissions, along with a 1.64% return rate, by integrating hydrogen and oxygen energies within a low-carbon economic dispatch energy hub. Nevertheless, their strategy omitted IDRP and CCS units and suffered from limited coupling of conversion units due to the single-hub structure. The statistical outcomes in Table 5 reveal their approach to be more costly and less flexible compared to the method proposed in this study.

This proposed model surpasses the current approach by achieving a 17.30% reduction in operational expenses and a 13.14% decrease in emissions. It integrates multi-energy strategies, carbon storage and reuse, sharing, and advancements in electricity, heating, and cooling demand response, yielding return rates of 13.28% and 7.17%, respectively. This approach enhances both the economic and environmental aspects of multi-interconnected green energy systems.

5. Conclusions

To overcome the limitations of small-scale and inflexible individual integrated energy systems and to advance energy saving, emission reduction, and modern power systems, this paper introduces a coordinated optimization strategy for hydrogen-integrated energy hubs. This strategy utilizes demand response program-enabled energy sharing by synchronizing tertiary hubs (producers, prosumers, and consumers). Two case studies are conducted, each emphasizing different objective function weights. The simulation and theoretical analysis yield the following key findings:

• To enhance operational effectiveness and reliability, hydrogen energy dispatch, advanced P2G, and CCS technologies can meet hydrogen requirements. The integrated energy system group optimizes resource allocation and increases energy utilization efficiency through energy sharing and carbon transfer. These techniques, along with 20% customer demand response flexibility, reduce system operating costs by 17.30% without renewable power curtailments. Hydrogen and natural gas, combined with demand response methods for heating, electricity, and cooling, meet elastic consumer demand during peak, mid-peak, and off-peak periods, enhancing system performance.
• The P2G-CCS system optimizes the integration and coordination of electricity and carbon flows across energy supply, conversion, transmission, and trade by capturing, storing, and converting CO₂ in a methanation reactor. This approach achieves a 13.14% reduction in total emissions, surpassing state-mandated quotas and substantially augmenting environmental advantages.

The proposed model illustrates a future study addressing the impact of weather and load uncertainties on long- and short-term hydrogen storage and carbon trading. It demonstrates the reduction in real-time operational costs through energy management across multi-seasonal scenarios in integrated energy systems.