Recently, the emergence of distributed energy resources in the consumer premises has given rise to microgrids. It
has become crucial to consider the impact of microgrids on the operation of existing power systems. Additionally,
the reforms in the electricity market also play a major role in the optimal generation and scheduling of electrical
energy in existing systems to avoid power scarcity and power surplus. Hence, in the proposed research, the
electricity market with spot pricing is analyzed through different scenarios, with the microgrid integrated into
the existing modified IEEE 5 bus system. The first step of the proposed work considers the price estimation using
neural networks-based algorithms such as Levenberg-Marquardt, Bayesian Regularization, Scaled Conjugate
Gradient, machine learning technique namely Random Forest Regressor (RFR), and deep learning technique
namely Long Short Term Memory (LSTM), and Ensemble (AdaBoost + Levenberg-Marquardt) technique to
obtain the estimated price with minimal error. The second step of the proposed work involves the economic
dispatch of five generating units to maximize revenue using both the estimated and actual tariffs. The third step
of the proposed work addresses the economic scheduling of power flow between the generators and load-serving
entities for different scenarios, such as with/ without power transfer limits and with network and congestion fees.
The simulation results of the scenarios show that following the proposed optimal bidding strategy with
congestion fees effectively addresses price volatility issues in spot pricing, achieving less than a 1 % deviation
between the estimated and actual spot prices with improved profit maximization in the power market.
