The development of complex equipment faces prominent challenges, including unequal status among participating agents, multi-objective full confrontation, strong super-conflict among multi-indicators, dynamic risk evolution, delayed on-site perception, and the absence of collaborative negotiation. Traditional risk management and control methods, based on the ideal assumptions of equal subjects and independent indicators, struggle to characterize and resolve super-conflict games dominated by super decision-makers. Furthermore, they lack dynamic early warning and closed-loop execution mechanisms linked to real-time perception, commonly suffering from drawbacks such as low early warning accuracy, high decision-making conflict, delayed response, and inefficient collaboration. To address these issues, this paper integrates multi-agent conflict negotiation with intelligent perception and learning technologies to propose an intelligent risk early warning and closed-loop control method for complex equipment development. The three-dimensional risk evolution dynamics model and Virtual Risk Center (VRC) are employed to decouple super-conflict indicators, while the industrial inspection unmanned aerial vehicle (UAV) perception relative motion model enables the unified mapping of physical risks and decision-making games. A super-conflict gray target negotiation (SCGTN) model is constructed to achieve stable consensus decisions among multiple parties under conflicting indicators. Based on Markov Decision Processes and the PPO algorithm, the optimal virtual risk trajectory is generated, which is then combined with the Archimedean spiral convergence trajectory to synthesize executable control trajectories. This forms an integrated system of UAV real-time perception → super-conflict resolution → intelligent decision-making → closed-loop regulation. Validated through a case study of large-scale complex aviation equipment development, the proposed method achieves field-validated risk early warning accuracy of 94.7% evaluated against real-world on-site ground truth labels. The numerical simulation results, whose parameters are fully calibrated against real-world engineering datasets, indicate that under simulated test conditions, our method yields simulation-predicted performance: it reduces decision-making conflict intensity by 49.3%, controls risk deviation error within 1.38%, and shortens closed-loop response time to 158 ms. Note that conflict reduction level, risk deviation error, and closed-loop response time are pure simulation outputs and have not been directly measured from physical on-site closed-loop experiments. Under the same simulation setup, the end-to-end response speed is 7.6 times faster than the peer dynamic closed-loop Digital Twin-Proximal Policy Optimization (DT-PPO) benchmark algorithm with identical online sensing and reinforcement learning architecture and roughly 4700 times faster than simulated counterparts of traditional static offline evaluation modes that rely on periodic manual statistics and offline meetings. It is adaptable to complex equipment development scenarios characterized by strong super-conflict, high dynamics, and unequal subjects, providing a theoretical framework and technical support for intelligent risk prevention and control throughout the full lifecycle of complex equipment.

Ignoring real decision contexts in linguistic multi-criteria classification (MCC) causes unreasonable weights, rank reversal and unreliable classification. This paper develops a context-constrained MCC framework integrating case-based reasoning, grey target theory and cloud model. Unlike conventional data-only grey target methods, three practical contextual features structurally reshape all algorithm modules: linguistic dual fuzziness-randomness, labeled benchmark cases and expert weight preferences. First, cloud Hamming distance and cloud-based dual target centers are constructed to eliminate rank reversal, a property invalid without linguistic uncertainty. Second, benchmark case constraints build a target-score classification mechanism with case-calibrated grading thresholds. Third, expert weight bounds form a bi-objective optimization model fusing multi-source information to avoid data-biased weights. A riparian wetland ecological grading case validates the method. A rigorous mathematical proof, counterexample contrast and large-scale random simulation are further supplemented to verify that the proposed fixed dual cloud target center rule significantly mitigates rank-reversal phenomena triggered by adding new alternatives under the designed experimental settings. Comparisons show our framework yields domain-consistent weights, grading outputs with perfect agreement with independent blind expert judgments, and stable monotonic scoring performance (where larger target scores correspond to worse ecological conditions), which achieves more consistent grading results compared with cloud entropy weighting and virtual linguistic approaches in the tested wetland case.