Enhancing stochastic optimal power flow with modified cheetah optimizer for integrating renewable energy sources

Area of Science:

• Electrical Engineering
• Optimization Algorithms
• Renewable Energy Systems

Background:

• Optimal Power Flow (OPF) and economic dispatch are critical for efficient power grid operation.
• Integrating renewable energy sources (RES) like wind turbines (WTs) and photovoltaics (PVs) introduces cost uncertainties.
• Existing optimization methods may struggle with the complexity and scale of modern power grids with RES.

Purpose of the Study:

• To introduce a Modified Cheetah Optimizer (MCO) algorithm for solving OPF problems in RES-integrated power grids.
• To accurately model and account for cost uncertainties in WTs and PVs.
• To evaluate the MCO's performance across various objective functions and large-scale dynamic economic dispatch problems.

Main Methods:

• The Modified Cheetah Optimizer (MCO) algorithm was developed and applied to OPF and economic dispatch problems.
• Uncertainty in RES cost models was addressed by incorporating it into the direct cost calculation.
• The MCO was tested on multiple objective functions including operating cost, voltage deviation, emissions, and power loss.
• Backward-forward correction was used to enhance reserve constraint dynamic economic dispatch solutions.

Main Results:

• The MCO achieved optimal solutions for various OPF cases, including valve point effects ($781.9862), emission costs ($810.6655), and POZs ($781.7165).
• Minimum network loss was recorded at 2.0738 MW, and voltage deviations were mitigated effectively.
• For large-scale dynamic economic dispatch, the MCO significantly outperformed previous studies, achieving $1,016,361 for a 10-unit system and $3,048,405 for a 30-unit system.

Conclusions:

• The MCO algorithm is a robust and effective tool for solving complex optimization problems in power systems with RES.
• It accurately handles cost uncertainties and improves the efficiency of power grid operations.
• The MCO demonstrates superior performance compared to existing methods, especially for large-scale dynamic economic dispatch problems.