The workflow scheduling in cloud computing systems has gained significance in both commercial and artificial operations. However, generating effective and affordable scheduling methods under deadline constraints remains extremely challenging, particularly for large-scale workflow operations. To address this issue, this study analyzes the workflow scheduling problem in the cloud with the goal of reducing the overall execution cost while keeping the processing time within a set deadline. An Intelligent Adaptive Optimization (IAO) algorithm is developed based on the problem-specific knowledge of workflow scheduling in the cloud. In the proposed IAO, an operator for discrete propagation is created using knowledge of idle time in an hourly cost model to dynamically explore the extensive search space. The adaptive refraction operator is designed to reduce stagnation and increase available resources. The dynamic grouping-based breaking operator is created to leverage the excellent block structure knowledge of the task allocation scheme and speed up convergence. The IAO approach has been shown to outperform several state-of-the-art algorithms in extensive simulation experiments conducted on a well-known scientific workflow.