MANET is a dynamic self-configuring, decentralized network, manet here stand for a mobile ad-hoc network which commonly described as node who serves as routers aimed to route the packets throughout the network. The energy of all the nodes in MANETs depletes with utilization due to which network lifetime is considered a major challenge, that no one can take part in any operation of network except those operations whose main goal is not related with data transmission. This paper suggests a set of energy-efficient techniques to control the routing, data transmission and node operations that aim at longer network lifetime. These are realized through adaptive energy-aware routing protocols which tend to choose paths with homogeneous energy utilization and refrain from unnecessarily loading specific nodes. The design leverages a novel energy prediction model for modeling future energy consumption based on workload and traffic patterns, allowing reconfigurations ahead of time. In addition, with our optimized transmission scheduling strategy, we avoid redundant transmissions by adjusting transmission power according to the distance between a pair of nodes and link quality over time. These techniques can be further leveraged with collaborative sleep-scheduling mechanisms, in which dormant nodes transition to low-power states while preserving network connectivity. Simulation — Detailed simulations were performed using ns3 to evaluate the algorithms under different mobility scenarios and traffic loads. The findings show that these methods can lead to substantial reductions — up to 35% lower overall energy consumption compared with baseline protocols — in the amount of energy used. Average improvement of 40% in network life time is gained by the algorithms and their packet delivery ratio is ensured with reliability while being energy efficient. By providing the fundamental trade-offs between energy efficiency and network performance, our results contribute to the design of sustainable MANETs for a wider range of applications. The proposed algorithms serve as a strong framework to prolong the network lifetime and are well-suited for energy constrained environments such as disaster recovery, military operations, and remote sensor networks. Future work may investigate the application of machine learning models into energy-efficient decision-making in dynamic MANET scenarios.