Construction of source load uncertainty economic dispatch model based on distributed robust opportunity constraints

Abstract

With the increasing demand for electricity, the power system is facing enormous challenges. To ensure the equilibrium between supply and demand in the electricity market and the safety and stability of the power grid, a source load uncertainty economic dispatch model based on distributed robust opportunity constraints is proposed to cope with the uncertainty of sustainable energy resources such as wind power and photovoltaics. By introducing an improved Elman network and grey wolf optimization algorithm, high-precision prediction of short-term loads is achieved, providing data support for scheduling models. The experiment outcomes indicate that the prediction model grounded on the improved Elman network and grey wolf optimization algorithm performs the best in scheduling performance on both the training and testing sets, with the lowest cost, the highest utilization rates of wind and solar power, and the lowest probability of constraint default. In addition, the economic dispatch model proposed by the research has significant advantages in reducing total dispatch costs, improving wind and photovoltaic utilization rates, and constraining default probability control. In typical load scenarios, the total scheduling cost of the model is $1,308,469, with wind and photovoltaic utilization rates reaching 90.5% and 86.1% respectively, and a default probability of only 0.9%. The research results indicate that the model exhibits superiority in real-time response time, especially suitable for scenarios with high load fluctuations. The research provides important theoretical basis and application value for the economic dispatch of power systems.