[Paper Review] A Stochastic-Robust Approach to Hierarchical Generation-Transmission Expansion Planning.

Maria de Lujan Latorre, G.C. Oliveira+3 more|arXiv (Cornell University)|Oct 3, 2019

Electric Power System Optimization10 references4 citations

TL;DR

This paper proposes a stochastic-robust hierarchical approach to generation and transmission expansion planning, decomposing the problem into three steps: integrated generation and interconnection planning, production costing simulation to generate operation scenarios, and robust TEP using enhanced Benders decomposition with a greedy warm-up. The method yields a least-cost, scenario-resilient transmission expansion plan for Central America’s multi-country electricity market.

ABSTRACT

In this paper, the differences between an integrated and hierarchical generation and transmission expansion planning approaches are first described. Then, the hierarchical approach is described in detail. In general terms, in this scheme the investment decisions are made in steps, instead of an overall optimization scheme. This paper proposes a stochastic hierarchical generation-transmission expansion planning methodology based on a three-step procedure, as follows: 1. In this step, an integrated expansion planning problem of generation and interregional interconnections is solved; 2. Taking the optimal expansion plan of (generation and interconnections) into account, a production costing simulation with the detailed network representation is performed without monitoring circuit flow limits (except the ones in the interconnections, which are monitored); this simulation produces a set of optimal dispatch scenarios (vectors of bus loads and generation); 3. A Transmission Expansion Planning (TEP) model is then applied to determine the least-cost transmission expansion plan that is robust with respect to all operation scenarios of step 2, using an enhanced Benders decomposition scheme that: (a) incorporates a subset of the operation scenarios in the investment module; and (b) presents a warm-up step with a greedy algorithm that produces a (good) feasible solution and an initial set of feasibility cuts. The application of the hierarchical planning scheme is illustrated with a realistic multi-country generation and transmission planning case study of the Central America's electricity market.

Motivation & Objective

• To address the computational complexity and scalability issues of integrated generation and transmission expansion planning (GTEP) by decomposing the problem into a hierarchical framework.
• To develop a robust transmission expansion planning (TEP) model that accounts for operational uncertainties from diverse generation and load scenarios.
• To improve solution quality and convergence in large-scale TEP by incorporating a warm-start greedy algorithm and selective scenario subset usage in Benders decomposition.
• To validate the methodology on a realistic, multi-country case study of Central America’s electricity market, reflecting real-world interregional interconnections and operational constraints.

Proposed method

• Step 1 solves an integrated generation and interregional interconnection expansion problem to determine optimal generation and interconnection investments.
• Step 2 performs a detailed production costing simulation using the optimal generation and interconnection plan, generating a set of dispatch scenarios without enforcing all network flow limits (except in interconnections).
• Step 3 formulates a TEP model that identifies the least-cost transmission expansion plan robust to all scenarios from Step 2.
• An enhanced Benders decomposition scheme is employed, where only a subset of operation scenarios is used in the master problem to improve computational efficiency.
• A warm-up step applies a greedy algorithm to generate an initial feasible solution and an initial set of feasibility cuts, accelerating convergence.
• The method ensures robustness by requiring the final transmission plan to be feasible across all generated dispatch scenarios.

Experimental results

Research questions

• RQ1 How can the computational burden of integrated GTEP be reduced while preserving solution quality?
• RQ2 What is the impact of using a subset of operation scenarios in the Benders decomposition framework on solution robustness and convergence?
• RQ3 Can a greedy warm-start strategy significantly improve the efficiency of Benders-based TEP in large-scale systems?
• RQ4 How does the hierarchical approach compare to integrated planning in terms of cost and robustness for multi-country electricity systems?
• RQ5 What level of operational scenario diversity is necessary to ensure robustness in transmission expansion planning?

Key findings

• The hierarchical approach significantly reduces computational complexity compared to solving the full integrated GTEP problem in one step.
• The use of a scenario subset in Benders decomposition maintains robustness while improving computational efficiency.
• The greedy warm-up step produces a high-quality initial feasible solution, reducing the number of Benders iterations required for convergence.
• The final transmission expansion plan is robust across all dispatch scenarios generated from the production costing simulation.
• The method successfully identifies a least-cost transmission expansion plan that satisfies all operational constraints under diverse system conditions.
• The case study demonstrates the method’s applicability and effectiveness in a real-world, multi-country electricity market context.