Machine learning plays a substantial role in the resilience of electrical secondary distribution system. For advanced power distribution system monitoring and control applications, accurate Φ connection information is crucial. Φ identification methods based on data currently available lack a precise physical interpretation and theoretical performance guarantee. As network complexity, the quantity of Φ connections, and the degree of load balancing rise in current techniques, performance typically degrades. Furthermore, correct Φ connection is critical for the integration of Distributed Energy Resources (DER) at the Low Tension (LT) side of distribution network. This decreases the excessive line losses, minimizes the voltage rise issue, and maintains voltage value in the prescribed limit. In this regard, Machine learning optimization techniques are utilized for the accurate Φ identification to interconnect DER such as Small Wind Turbine Generator (SWTG) in power distribution system. In this paper, development of algorithm for Φ identification to integrate DERs such as SWTG in BEC distribution system using Artificial Intelligence and Machine Learning (AIML) is carried out. The K-Nearest Neighbors (KNN) classifier had a better impact in the identification of data in supervised machine learning algorithm. This method predicts a target variable using one or multiple independent variables. Also, Φ voltages data of Distribution Transformer Center (DTC) are collected from SCADA system and KNN algorithm is useful to find the better Φ in the LT side of distribution system. Result shows that, Y-Φ is more accurate to interconnect the DERs compared to other Φs. This work is helpful to install and obtain optimum placement of the DERs at LT side of distribution system.