In large-scale networks such as the Internet of Things (IoT), devices seek multihop communication for longdistance communications, which considerably impacts their power exhaustion. Hence, this study proposes an energy harvesting-enabled, relay-based communication in multihop clustered IoT networks in a bid to conserve the battery power in multihop IoT networks. Initially, this study proposes an efficient, hierarchical clustering mechanism in which entire IoT devices are clustered into two types: the closest cluster (CC) and remote clusters (RCs). Additionally, Euclidean distance is employed for the CC and fuzzy c-means for the RCs. Next, for cluster head (CH) selection, this study models a fitness function based on two metrics, namely residual energy and distance (device-to-device distance and device-to-sink distance). After CH selection, the entire clustered network is partitioned into several layers, after which a relay selection mechanism is applied. For every CH of the upper layer, we assign a few lower-layer CHs to function as relays. The relay selection mechanism is applied only for the devices in the RCs, while for devices in the CC, the CH functions as a relay. Finally, several simulation experiments are conducted to validate the proposed method's performance. The results show the method's superiority in terms of energy efficiency and optimal number of relays in comparison with the state-of-the-art methods.