Smart grid systems are seen as the next big thing to energy distribution and management with more efficiency, stability, and sustainability in power distribution and energy management systems. It is challenging to predict the amount of energy consumption in smart grid system because of its randomness resulting from various factors such as consumer's behavior, weather conditions. This is mainly because traditional forecasting techniques are usually inadequate in capturing these complexities and as a result resources are poorly allocated and organizational processes become inefficient. The work provides the energy consumption forecasting in smart grids through the application of Quantum Boltzmann Machine(QBM) in the smart grid utilizing historical data. Some of the variables incorporated in the data are reaction time, power balance, and price elasticity coefficients customized through smart meters. Once raw data is preprocessed, QBMs are developed to establish an empirical probability model that will predict actual future energy utilization characteristics. The QBM is tested against the actual consumption data and focuses on the visualization of energy profile and identification of points like local extremes. © 2025 Elsevier B.V., All rights reserved.