IoT-activated WSNs require energy gratification as one of their primary scheme conditions. A lack of efficient data collection by Sensor Nodes (SNs) threatens network durability due to energy conservation issues. In order to enhance network durability, the energy of SNs must be utilized in an efficient manner. The load can be balanced among sensors through clustering techniques to optimize the energy efficiency of SNs. Automatically, the lifetime and the scalability of the network are elevated. In this article, we examine a few lavish AI approaches, alongside supervised, unsupervised, and reinforcement learning (RL). We propose in this paper a new RL with Q-learning and a Low Energy Adaptive Clustering Hierarchy (LEACH) built using clustering and GA-assist routing for WSNs (RiLgQL-LEACH-GA). The solution to this is to elect a Machine Learning (ML) technique for energy management in the network. This RL-based LEACH can be used in conjunction with other modern ML techniques to cluster data. It is particularly important for this paper to decrease energy consumption in WSNs and to improve network sustainability. Genetic Algorithms (GAs) are used to choose optimal paths among Cluster Heads (CHs) and Base Stations (BSs). Simulation results indicate that the recommended methodology leads to outstretched energy efficiency and network existence.