The future is expected to bring increased loads to the grid, which will result in system peak loads. Traditionally, the solution to this issue has been to strengthen the grid. However, an alternative option is to offer flexibility services, which can help avoid or delay the need for grid development. As the total load continues to rise, the distribution transformer can become overloaded, causing the insulation life of the transformer to deteriorate. To tackle this problem, the distribution system control employs the transactive Energy paradigm, which focuses on enhancing the deterioration of the transformer. In this study, Transactive Energy Control is specifically utilized to reduce the overall cost for the electricity consumption to the end user. This approach involves the use of a Machine Learning Algorithm to predict future loads, thereby reducing transformer overloading and ultimately reducing the electricity cost to the consumer. The experimental validation of this approach is conducted using MATLAB software.