Technical Barriers for Harnessing the Green Hydrogen: A Power System Perspective (2010.06384v1)

Abstract: Extracting green hydrogen from renewable energy sources is a new concept in the energy industry. As an energy carrier, hydrogen is well capable of facilitating a strong coupling between various energy sectors, as well as integration of renewable energy sources. This paper investigates the system-wide technical factors that might limit the amount of producible hydrogen in a given power system. A non-linear programming formulation is proposed to quantify the impact of voltage security constraints, the location and size of power to hydrogen facilities, and finally the wind penetration levels on the harvest-able green hydrogen. The applicability of the proposed framework is demonstrated on the IEEE 39 bus system.

• The paper introduces a nonlinear programming model to maximize hydrogen extraction by balancing P2H allocation with voltage security limits.
• It demonstrates that optimal P2H facility size decreases as higher loading margin requirements restrict grid capacity.
• The study shows that increased wind penetration enhances green hydrogen production until grid security constraints become binding.

This paper, "Technical Barriers for Harnessing the Green Hydrogen: A Power System Perspective" (2010.06384), investigates the technical factors within the electrical power system that limit the amount of green hydrogen that can be produced using renewable energy sources (RES). The core concept is that Power-to-Hydrogen (P2H) facilities, which use electricity to produce hydrogen via electrolysis, act as a new, potentially flexible load on the grid. While beneficial for integrating high levels of variable RES and decarbonizing other sectors, these facilities can push the power system closer to its operational limits, particularly concerning voltage security.

The paper proposes a non-linear programming (NLP) formulation to quantify the impact of several factors on the harvestable green hydrogen:

1. Location and Size of P2H Facilities: Where should large-scale electrolyzers be placed in the transmission network, and what should their capacity be?
2. Voltage Security Constraints: How do requirements to maintain system stability under heavy loading conditions (avoiding voltage collapse) limit hydrogen production? This is modeled using the concept of Loading Margin (LM), which represents the distance between the current operating point (COP) and the security limit point (SLP).
3. Wind Penetration Levels: How does the amount of renewable energy available influence the potential for green hydrogen production?

The objective of the optimization problem is to maximize the total hydrogen extracted from P2H units over a specified time horizon (e.g., 24 hours). The model includes detailed AC power flow equations for both the COP and the SLP, simultaneously considering nodal active and reactive power balances.

Key constraints in the formulation cover:

• P2H Efficiency: The amount of hydrogen produced ($H_{b,t}$) is proportional to the active power consumed by the electrolyzer ($PH_{b,t,c_1}$) multiplied by an efficiency factor ($\eta_{b,t}$). For practical purposes near nominal capacity, a constant efficiency is assumed.
• Generator Capability Curves: A detailed model of generator reactive power limits, including armature current, field current, and under-excitation limits, is incorporated to accurately capture voltage security dependencies.
• Network Security (LM): The relationship between the COP and SLP is defined by scaling factors ($K^P, K^Q, K^G$) applied to demands and non-slack generation, reaching the SLP at a loading parameter $\lambda$. Constraints ensure that the system can withstand this increased loading up to the SLP while respecting operational limits (voltages, line flows, generator reactive power limits). The voltage deviation between COP and SLP due to generators hitting reactive power limits is modeled using auxiliary variables ($v^{up}, v^{dn}$).
• Operational Limits: Standard constraints are included for active and reactive power generation limits, voltage magnitude limits, wind farm active/reactive power output limits, P2H active/reactive power consumption limits, and line apparent power flow limits.
• System Reserve: A constraint ensures sufficient generation reserve is available.
• Wind Penetration Limit: The total wind power injection is limited to a percentage ($\alpha$) of the total system demand.
• Generator Ramp Rates: Limits on the hourly change in active power output for conventional generators are considered in the multi-period model.

To handle the non-linearity and non-differentiability introduced by the $\min$ operator in generator reactive power limits and the scaling of generator active power output with $\lambda$, the paper reformulates these constraints using auxiliary binary-like variables ($y, z$) and Big-M constants, resulting in an NLP problem solvable by specialized solvers.

The model is demonstrated on the IEEE 39-bus test system over a 24-hour period, using hourly load and wind power availability data. Potential P2H locations are considered at all load buses. The problem is solved using GAMS with the KNITRO solver.

Practical insights from the numerical studies include:

• Optimal P2H Allocation: For the studied system, optimal P2H locations tend to coincide with or be near high wind penetration areas. However, the paper also notes that P2H facilities can serve as flexible loads anywhere in the network, potentially located closer to hydrogen demand centers rather than just RES sources, leveraging the existing electricity transmission network for power delivery.
• P2H Size vs. LM: The optimal size of P2H units that can be installed is inversely related to the required Loading Margin. Higher LM requirements (i.e., requiring the system to be more secure against voltage collapse) necessitate smaller P2H capacities, as large P2H loads stress the system.
• Hydrogen Harvest vs. LM and Wind Penetration:
  • For a fixed wind penetration level, increasing the required LM beyond a certain threshold leads to a decrease in attainable green hydrogen. This is because tighter voltage security constraints limit the total load (including P2H) the system can support.
  • For a fixed LM, increasing wind penetration generally increases the amount of exploitable green hydrogen, up to the point where security constraints (like LM) or other network limits become binding.

In essence, the paper provides a framework for power system planners and operators to analyze the trade-offs between integrating large P2H loads for decarbonization/RES integration and maintaining the voltage security of the electrical network. It highlights that while P2H can absorb RES, its placement and size must be carefully considered within the context of grid loadability and reactive power capabilities, especially in future grids with high RES penetration.

Future practical considerations suggested by the authors include integrating the impact of hydrogen production on the coupled gas network and developing more detailed security-constrained optimal power flow models that explicitly account for the P2H dynamics and the inherent uncertainty of RES.