With the global shift in energy structures, remediating radioactive contamination has emerged as a critical challenge in environmental engineering and atomic science. This paper presents Purify-Enterprise-Ultimate, a comprehensive governance platform designed to overcome the limitations of traditional remediation methods, such as single detection dimensions and low execution efficiency in extreme environments. ​The core of this system lies in the integration of the "Shuiquan Scientific Philosophy System," specifically the "collision of sensibility and rationality." This translates to the engineering fusion of fuzzy logic/intuition-inspired heuristic searches with rigorous physical modeling and numerical computation. Technically, the system utilizes heterogeneous edge computing and 384-dimensional composite vector indexing to process multi-modal data. It integrates atomic-level forensics, including Atomic Force Microscopy (AFM) and mass spectrometry, to construct high-fidelity Digital Twin environments for pollutant modeling. ​Furthermore, the platform incorporates a decay model library and employs Monte Carlo simulations for precise radioactive risk assessment. The execution phase is driven by a Human-in-the-Loop (HIL) framework coordinating heterogeneous swarm robots (UAVs and UGVs) for automated monitoring, sampling, and disposal. This research provides a codified, verifiable, and industrial-grade execution framework for the total resolution of nuclear waste management.