Optimal operation of electric-heat-gas-hydrogen integrated energy system considering source-load uncertainty

Faced with the threat of global climate change crisis, increasing the use of hydrogen-enriched compressed natural gas (HCNG) and the utilization rate of renewable energy can both reduce the carbon emissions of the integrated energy system (IES). This paper establishes an electric-heat-gas-hydrogen integrated energy system (EHGH-IES) framework considering source-load uncertainty, and the natural gas pipe network in the traditional IES is transformed into a HCNG pipe network with a certain hydrogen blending ratio. An optimal operation model of the EHGH-IES considering source-load uncertainty is proposed in this paper. The chance constrained programming method is adopted to deal with source-load uncertainty. We pay attention to comparing different confidence coefficient and hydrogen blending ratio in terms of the changes in the cost of the IES. The results show that the total cost of the EHGH-IES will increase by considering source-load uncertainty and using HCNG pipeline, but will bring some stability and environmental benefits to the IES by them.


Introduction
Faced with the threat of global climate change crisis, increasing the use of HCNG can both reduce the carbon emissions of IES and the utilization rate of renewable energy.countries are seeking to reduce carbon emissions to mitigate the further rise in global average temperature.Therefore, on the basis of the traditional electric-heat-gas-IES, the construction of EHGH-IES can further reduce the carbon emissions of the IES and improve the environmental benefits of the IES.There have been many achievements in the research on the source-load uncertainty [1]- [3] .At present, the research on the optimal operation of the electric-heat-gas-IES has been relatively perfect development [4]- [6] .In recent years, many related studies have been carried out on the optimization operation of the IES considering HCNG [7]- [9]   .Most studies focus on the benefits of renewable energy under the influence of source-load uncertainty, or the environmental benefits of hydrogen energy itself in the IES containing hydrogen.However, there are few relevant studies considering the benefits brought by both of them and the interaction between them, so it is necessary to study the change of IES cost and benefits under the influence of source-load uncertainty and hydrogen energy.
This paper studies the optimal operation problem of EHGH-IES considering source-load uncertainty.The optimal operation model of the EHGH-IES considering source-load uncertainty is established, and the chance constraint programming method to deal with source-load uncertainty is adopted in this paper.
Finally, this paper sets up scenarios with different confidence coefficient and hydrogen blending ratios to analyse the optimal operation model of the EHGH-IES considering source-load uncertainty, and studies the changes in the cost of the IES under the influence of the source-load uncertainty and hydrogen energy.The results show that the total cost of the EHGH-IES will increase by considering source-load uncertainty and using HCNG pipeline, but will bring some stability and environmental benefits to the IES by them.

Optimal operation model of EHGH-IES considering source-load uncertainty
The framework of EHGH-IES is shown in Figure 1.The bus configuration of EHGH-IES is made up of wind turbine (WT), photovoltaic panel (PV), battery (BT), electrolysis device(P2H), hydrogen storage tank (HS) and combined heat and power generation (CHP).

Chance constrained model under source-load uncertainty
The output power of wind turbine, photovoltaic panel and the load of electric network, thermal network, HCNG network are uncertain.The actual value of them can be expressed with the predicted value plus the error value.(1) According to the central limit theorem, the error value of them obeys normal distribution.Its mean is zero, and its standard deviation can be expressed as follows [10,11] .
In formula (2), /, WT PV N P represent rated power of wind turbine output or photovoltaic panel output.The electric power, thermal power and HCNG balance constraints are presented in ( 26)-(28), respectively.

Objective function
The total cost includes purchasing cost of electricity, thermal power and HCNG, penalty of photovoltaic power curtailment and wind power curtailment, operation cost of CHP and P2H.

Constraints
The constraints of the optimal operation model in this paper includes operation constraints of electric network, thermal network, HCNG network, CHP unit, and BT [12] .

Case description
The proposed optimal operation model is tested on an EHGH-IES with an IEEE 33-bus system, a modified Belgian 14-node gas system and a 12-node heating system.The topology of the test system is shown in figure 2. The upper power grid(PG), natural gas source(NGS), and thermal source(TS) interact with the E1 bus, G1 node, and T1 node, respectively.The PV, WT and BT are located at E3, E6, E2 buses, respectively.The P2H is located at E2 bus, and connect to G1 node.The CHP is located at G2 node, and connect to E2 buses and H1 node.The HS is located at the output pipeline of P2H connected to G1 node.
The operation parameters of each system device are shown in Table 1.

Operation cost analysis in different scenario
In order to analyse the effectiveness and economic efficiency of the proposed optimization operation model of the EHGH-IES considering source-load uncertainty, the following four scenarios are set up in this paper.The division of sub-scenario is shown in Table 2. Scenario 1: Not considering source-load uncertainty without using HCNG.Scenario 2: Not considering source-load uncertainty but using HCNG.Scenario 3: Considering source-load uncertainty without using HCNG.Scenario 4: Considering source-load uncertainty and using HCNG.Scenario 4 is divided into the following five sub-scenarios.

Cost comparison of renewable energy under different confidence coefficient.
From the figure 3(a) and figure 4(a) we can see when hydrogen blending ratio is set to 0.2 and sourceload uncertainty is taken into account, abandoned photovoltaic power and abandoned wind power decreases.With the increase of confidence coefficient, Abandoned photovoltaic power and abandoned wind power decreases further.This shows that improving the confidence coefficient can help improve the utilization rate of renewable energy.Considering the uncertainty of the photovoltaic power and the wind power, the energy supply of them reduces, and thus reduce the abandoned photovoltaic power and the abandoned wind power.

Daily total cost comparison under different confidence coefficient.
As shown in figure 5(a), when hydrogen blending ratio is set to 0.2 and source-load uncertainty is taken into account, the daily total cost of the system increases.With the increase of confidence coefficient, the daily total cost of the system increases further.This shows that improving confidence coefficient is not conducive to improving system economic efficiency.Considering source-load uncertainty, the energy supply of the photovoltaic power and the wind power reduce and the load energy consumption increase, thereby increase the daily total cost of the system.However, considering source-load uncertainty can deal with some extreme cases to a certain extent, thus improving the stability of the system.

Cost comparison of renewable energy under different hydrogen blending ratio.
From the figure 3(b) and figure 4(b) we can see when confidence coefficient is set to 0.95 and using HCNG, Abandoned photovoltaic power and abandoned wind power have almost no change.Abandoned photovoltaic power and abandoned wind power still have no marked change with the increase of hydrogen blending ratio.This shows that blending hydrogen has little impact on the utilization rate of renewable energy.The change of energy supply power brought about by the increase of hydrogen blending ratio, which is equivalent to the small renewable energy supply power, so the effect on renewable energy supply power is not obvious.

Daily total cost comparison under different hydrogen blending ratio.
The figure 5(b) show that when confidence coefficient is set to 0.95 and using HCNG, the daily total cost of the system increases.The daily total cost of the system further increases with the increase of hydrogen blending ratio.This shows that increasing the hydrogen blending ratio is not conducive to improving the economic efficiency of the system.With the increase of hydrogen blending ratio the cost of P2H increases, so that the daily total cost of the system increases.However, increasing the hydrogen blending ratio can reduce the carbon emission of the system, thus improving the environmental benefit of the system.The optimal operation results show that considering source-load uncertainty and using HCNG will increase the total cost of the system, but it will bring some stability and environmental benefits to the system.

Conclusion
In this paper, the framework of the EHGH-IES considering source-load uncertainty is firstly established.Secondly, the optimal operation model of the EHGH-IES considering source-load uncertainty is proposed.The chance constraint programming method is adopted to deal with source-load uncertainty.Finally, the optimization operation model of the EHGH-IES considering source-load uncertainty is analysed with a case.The results show that the total cost of the system will increase by considering source-load uncertainty and using HCNG pipeline, but will bring some stability and environmental benefits to the system by them.In the follow-up study, carbon trading mechanism can be introduced into the optimal operation model of the EHGH-IES considering source-load uncertainty, so as to further quantify the environmental benefits of HCNG and renewable energy.
/thermal/HCNG power at node i, abandoned wind/photovoltaic power, /, , HS BT c it P represent the charge power of HS/BT, /, , HS BT d it P represent the discharge power of HS/BT, ,/ , CHP E T it P represent the electric/thermal power generation of CHP, , CHP it V represent the input flow of CHP, HCNG q represent the calorific value of HCNG.

Figure 2 .
Figure 2. Structure of example system.

Table 2 .
Division of sub-scenario.