Mobile ad-hoc network is an infrastructure-less spontaneously formed wireless environment that is deployed without any centralized unit. In this type of network, all nodes are assumed to be trustworthy but in the real scenario, a few can be malicious and, therefore, secure and reliable route for data transmission is always a matter of question. A new protocol called Secure Routing Algorithm Based on Differential Evolution (SRABDE) is proposed for Mobile Adhoc networks with effective fitness function. Differential Evolution (DE) is an evolutionary computational method inspired by the biological processes of evolution and mutation used to minimize routing overload in terms of battery power, node reliability and stability. The objective of this algorithm is to find the most suitable path for data transmission taking into account three influencing factors: reliability, battery power and node stability, then to resolve the malicious performing nodes such as black hole and gray hole attacks. The component indices for the effective fitness function involved are trust factor, residual battery power and node stability for the purpose of selecting the most promising routes through the proposed SRABDE algorithm. The trust is developed to predict the nature of the node, whether malicious or not; while the stability of the nodes ensures longer stable routes.The simulation is carried out in various scenarios to evaluate the performance of the differential evolutionary secure routing algorithm with packet delivery ratio, average end-to-end delay, routing overhead, energy consumption, detection accuracy, and throughput. Results are compared with three existing algorithms Intruder Free Route Discovery (IFRD), Trust Path Ant Colony Optimization (TPACO), and DE algorithm-based ad hoc on-demand multi-path distance vector (DE_AOMDV).