Increasing electricity demands with limited existing utility grid infrastructure has led to electricity consumers dealing with operational constraints such as contractual energy violation penalty and and time of day pricing of electricity. Therefore, integration of renewable energy such as solar power generation is popular choice to fulfil additional demands. The Integration of storage elements such as battery gives scope to optimize for contractual energy violation and energy purchase cost by ’day-ahead’ planning of battery charging schedules. However, the uncertainty of day ahead variables such as load/demand and solar generation pose problems in reasonable optimization. In this paper, the EV charging loads(for C&I consumers) are considered as uncertain day ahead variable. At first, a day-ahead forecasting of EV charging loads by training iTransformer based model was performed which offers Mean Absolute Error (MAE) of 2.93%. Further these load forecasts were fed to three optimization algorithms- Linear Programming (LP), Stochastic Optimization (SO), and Robust Optimization (RO) to establish trade off between contractual energy violations and grid energy purchase cost. The SO shows good trade-off performance in terms of 3-5% reduction in grid energy costs compared to LP and 20-70% reduction in contractual energy violation (i.e. peak shaving) compared to LP. Also, the RO shows best performance of 90% reduction in contractual energy violations and poorest performance for grid energy purchase cost minimization.