Power Systems Reliability Assessment

A comprehensive MATLAB-based framework for assessing the reliability of bulk power systems using Monte Carlo Simulation (MCS). This package supports both Hierarchical Level I (HL1) and Hierarchical Level II (HL2) reliability studies, utilizing the IEEE RTS-24 bus test system.

Overview

Reliability assessment is critical for planning and operating power systems. This repository implements two primary Monte Carlo simulation techniques:

1. Non-Sequential MCS (State Sampling): Evaluates independent system states to determine probability-based indices.
2. Sequential MCS (Chronological): Simulates the system evolution over time to capture frequency and duration indices.

Hierarchical Levels

HL1: Generation Adequacy

Focuses solely on the ability of generation resources to meet the total system load. Transmission constraints are ignored (Copper Sheet model).

• Key Question: Is there enough generation capacity?
• Indices: LOLE (Loss of Load Expectation), EENS (Expected Energy Not Supplied).

HL2: Composite System Adequacy

Evaluates the combined generation and transmission system. It accounts for network topology, line limits, and voltage constraints using DC Optimal Power Flow (DC-OPF).

• Key Question: Can the energy be delivered to the load points?
• Indices: LOLE, EENS, plus Nodal Indices (Bus-specific reliability).

Methodologies Implemented

1. Non-Sequential Monte Carlo (State Sampling)

Located in Montecarlo_nsq_single/.

• Approach: Randomly samples component states (Up/Down) based on unavailability probabilities ($U$). Each sampled state is independent.
• Process:
  1. Sample state vector.
  2. Run DC-OPF (minimizing load shedding).
  3. Accumulate failure statistics.
• Best for: Fast calculation of expected values (EENS, LOLE) when chronological correlation is not critical.

2. Sequential Monte Carlo (Chronological)

Located in Montecarlo_seq/.

• Approach: Simulates the system hour-by-hour over many years. Component states transition based on Time-To-Failure (TTF) and Time-To-Repair (TTR) distributions.
• Features:
  • Chronological Load: Uses the IEEE RTS-79 hourly load profile (8736 hours/year).
  • Frequency & Duration: Calculates how often failures occur and how long they last.
• Best for: Detailed analysis requiring frequency (occ/yr) and duration (hr/occ) indices.

Prerequisites

• MATLAB: (Tested on R2020b+)
• MATPOWER: Required for power flow and OPF calculations. Download here.
• Optimization Solver:
  • MIPS: Included with MATPOWER (Default).
  • CPLEX / Gurobi: Highly recommended for faster performance, especially for the Sequential simulation which solves thousands of OPFs.

Usage

Running Non-Sequential Simulation

1. Navigate to the folder:

   cd Montecarlo_nsq_single

2. Run the main script:

   nsqMain

3. Outputs: Convergence plots, reliability_results.mat, and nodal_results.csv.

Running Sequential Simulation

1. Navigate to the folder:

   cd Montecarlo_seq

2. Run the main script:

   seqMain

3. Outputs: Time-series plots, frequency/duration indices, and seq_nodal_results.csv.

Key Files

• nsqMain.m / seqMain.m: Master scripts for simulation control.
• mc_sampling.m / seq_mcsampling.m: State generation logic.
• mc_simulation.m / seq_mcsimulation.m: System evaluation (OPF) logic.
• case24_ieee_rts.m: MATPOWER case file for the test system.
• anloducurve.m: Generates the chronological load profile (Sequential only).

License

This project is intended for educational and research purposes.