A Faster Optimal Solution for Economic Dispatch Problem in A Smart Grid Environment using Tensorflow

This paper has a description of an approach that is apt for generating the solution of classical Economic Dispatch (ED) problem in a smart grid environment. There is a need for designing newer set of machine learning algorithms which can be applied for smart grid applications in cloud computing and bigdata specific situations. There are numerous advantages that are offered by a cloud computing environment, which must be addressed in a suitable manner to apply these techniques in a smart grid application. Big-data application is identified by a huge volume, velocity, variety and veracity. This paper gives a description of a scheme that can generate a simplified solution of the classical ED problem in a smart grid environment. The implementation of the scheme is done using tensor-flow. Since the classical model of ED is simplified in the final state by this scheme, this method is bound to generate a faster solution when it is implemented online. A simple sample system is used to show the implementation of the proposed scheme based on tensor-flow.