The shift from conventional energy sources to renewable sources leads to high penetration of Distributed Generators (DGs) in Active Distribution Networks (ADNs). This trend along with other factors causes major problems in ADNs such as energy management, voltage imbalances, poor efficiency, and reliability. To overcome these challenges the article suggests a novel approach by optimally scheduling energy using advanced methods for Dynamic Economic Dispatch (DED) algorithms with machine learning based load forecasts. Initially, the proposed system deploys Short-Term Load Forecasting (STLF) on the load buses based on the previously observed patterns using Multilayer Artificial Neural Networks (MANN). Then, the system implements a Genetic Algorithm based DED with major operational and network constraints. The proposed methods are integrated and tested on the IEEE 30-bus network using Kazakhstan's power plant and electricity demand parameters. The proposed system has shown improvement in the reliability and efficiency of the distributed system. Since the proposed methodology is generic and was verified on the standard radial network it can be further implemented in other systems.