Embedding Grid Security Constraints in Generation Expansion Planning Models: A Real Network Case Study.

Generation expansion planning (GEP) models rely on linear programming to determine an optimal generation plan and schedule that satisfies the demand-supply balance criterion. They are computationally efficient but do not consider the constraints related to grid stability and operational limits.

In addition, renewable energy sources are characterized by near-zero marginal generation costs.

Therefore, when available, they are chosen preferentially for the dispatch over traditional means of generation by the GEP algorithm. However, in high-renewable scenarios the resulting dispatch solution may be economically optimal from a cost point of view but not acceptable from a grid operation standpoint. It can violate stability and operational requirements due to the absence of the previously mentioned grid constraints in the models.

This paper proposes a methodology to incorporate the grid stability and operational constraints in the GEP mathematical modelling. The objective is to bridge the gap between economic optimisation and operational security requirements, associated with the generation mix. To this end, the study considers, in the calculation of the optimal generation mix, the constraints related to voltage margins in steady states and frequency stability. The method is tested through simulations conducted on the future New Caledonia power transmission network, which features a high penetration of renewable energy sources. First, a generation plan is computed using PLEXOS, without the integration of grid constraints in the model. Then, using DIgSILENT PowerFactory, a static and dynamic study of this mix allows to evaluate the type and severity of the stability and operational violations it induces.

Based on this diagnosis, constraints that need to be integrated into the economic model to ensure system security are formulated and linearized.

First a constraint was derived from static network simulations and was used to guide dispatch decisions and to reduce violations of operational security limits. This approach led to a twentyfold decrease in voltage-related violations, demonstrating a clear improvement in system robustness.

The second analysis addresses frequency stability, where a nadir threshold was implemented to prevent the system frequency from falling below critical levels, following a major disturbance.

This measure successfully reduced the risk of instability by 50%, reinforcing the system’s resilience. Overall, the results show that integrating economic optimisation with voltage and frequency security constraints leads to dispatch solutions that are both cost-efficient and technically reliable.

By embedding grid security constraints into the GEP model, this work contributes to the identification of an optimal portfolio mix of transmission, generation, and storage, while ensuring system adequacy, resilience, and flexibility. The integration of security constraints ensures that each solution in the portfolio meets economic objectives and respects the operational and stability limits of the grid. This security-aware approach supports the development of a resilient and flexible energy system, capable of handling variability and uncertainty induced by renewable energy sources.