A Quantitative Risk Evaluation Method for Emitter Clogging Integrating Fuzzy Comprehensive Evaluation and Dynamic Bayesian Network (FCE‐DBN)

ABSTRACT Assessing emitter clogging risk is essential for the safe operation of drip irrigation systems. This study developed and validated a quantitative clogging risk assessment model integrating fuzzy comprehensive evaluation with a dynamic bayesian network (FCE‐DBN). Applied to field data from the Ulanbuhe Irrigation District, the model delivered empirically grounded, numerically precise risk predictions. For four non‐pressure‐compensating emitters (FE1‐FE4), the baseline clogging probabilities, derived from structural parameters, water quality, and management practices, were quantified as 8.64 × 10 −2 , 7.67 × 10 −2 , 1.01 × 10 −1 and 6.94 × 10 −2 , respectively. The DBN component captured temporal dynamics, revealing that the clogging risk increased linearly with cumulative irrigation time ( t ), achieving an exceptional goodness‐of‐fit ( R 2 = 0.987) across all emitters. The slope was governed by emitter‐specific influence coefficients ( λ ), directly linking physical design to accelerated risk accumulation. Combined with the measured clogging degree ( C ) via the relative discharge ratio (Dra), the risk index Rc = P × C ( P denotes the probability of emitter clogging) yielded dynamic trajectories that quantitatively matched the field‐observed clogging progression. The resulting five‐level classification provided actionable thresholds; for instance, moderate clogging (Level 3) triggered at Rc ≥ 0.045 enables timely intervention (e.g., acid flushing). This work shifts emitter clogging assessment from qualitative expert judgement to a rigorously validated, time‐resolved and numerically explicit framework for precision irrigation management.