The ability to predict the volume of methane outbursts in coal mines is critical for the prevention of methane outburst accidents and the assurance of coal-mine safety. This paper’s central argument is that existing prediction models are limited in several ways. These limitations include the complexity of the models and their poor ability to generalize. The paper proposes a methane outburst volume-prediction and early-warning method. This method is based on a secondary decomposition and improved TSMixer model. First, data smoothing is achieved through an STL decomposition–adaptive Savitzky–Golay filtering–reconstruction framework to reduce temporal complexity. Second, a CEEMDAN-Kmeans-VMD secondary decomposition strategy is adopted to integrate intrinsic mode functions (IMFs) using K-means clustering. Variational mode decomposition (VMD) parameters are optimized via a novel exponential triangular optimization (ETO) algorithm to extract multi-scale features. Additionally, a refined TSMixer model is proposed, integrating reversible instance normalization (RevIn) to bolster the model’s generalizability and employing ETO to fine-tune model hyperparameters. This approach enables multi-component joint modeling, thereby averting error accumulation. The experimental results demonstrate that the enhanced model attains RMSE, MAE, and R² values of 0.0151, 0.0117, and 0.9878 on the test set, respectively, thereby exhibiting a substantial improvement in performance when compared to the reference models. Furthermore, we propose an anomaly detection framework based on STL decomposition and dual lonely forests. This framework improves sensitivity to sudden feature changes and detection robustness through a weighted fusion strategy of global trends and residual anomalies. This method provides efficient and reliable dynamic early-warning technology support for coal-mine gas disaster prevention and control, demonstrating significant engineering application value.

Ventilation network feature graphs (Q-H graphs) are a key visualisation tool for mine ventilation systems, and their automated generation reduces to a combinatorial optimisation problem over independent-path permutations. Existing methods, however, exhibit three limitations: a single-dimensional evaluation criterion, inadequate nodal pressure-energy assignment, and unstable convergence in factorial-scale search spaces. This paper proposes an adaptive NSGA-II (A-NSGA-II) framework with coordinated enhancements at the evaluation, modelling, and algorithmic levels. A three-objective system that minimises split-block count, topological-spatial discrepancy, and layout fragmentation is established, together with an aggregate evaluation score (AES) for engineering decision-making; nodal pressure energies are reconstructed via the longest path on a directed acyclic graph; and topology-aware initialisation, Lagrange three-point interpolated adaptive operators, and periodic memetic local search are integrated within NSGA-II. Experiments on two mine ventilation networks (75 and 112 branches) over 30 independent trials show that A-NSGA-II consistently outperforms four benchmarks (NSGA-II, MOEA/D, SPEA2, and MOSA) in terms of split-block count, AES, and hypervolume; statistical tests confirm significant, large-effect HV advantages on the 112-branch network, while the 75-branch network shows a 56.6–71.5% reduction in HV standard deviation.

This study develops an interpretable point-and-interval prediction method for mine ventilation-parameter identification using limited field samples. Guided by ventilation-resistance theory, eight engineering attributes—roadway length, width, height, cross-sectional area, wetted perimeter, equivalent diameter, support type, and cross-sectional shape—were selected, and a prediction dataset was constructed from 260 field samples. The extended Random Feature Model (xRFM) establishes a nonlinear mapping, and the proposed Evolution-Path Guided Ivy Algorithm (EPG-IVY) optimizes hyperparameters within the training data. Prediction intervals are then constructed from out-of-fold residuals using Adaptive Bandwidth Kernel Density Estimation (ABKDE), while SHAP is used to analyze the model’s dependence on engineering features. With a fixed 7:3 split, EPG-IVY-xRFM achieved R² = 0.9754, RMSE = 0.001600, and MAE = 0.001162. Repeated nested five-fold cross-validation yielded R² = 0.9719 ± 0.0084, RMSE = 0.001675 ± 0.000267, and MAE = 0.001225 ± 0.000182, with lower RMSE and MAE than the optimized XGBoost and CatBoost baselines. External validation on 28 independent samples from Balasu Coal Mine yielded R² = 0.890, RMSE = 0.007072, and MAE = 0.003873. The empirical coverage of the 90% ABKDE prediction interval was 0.8846, and its Wilson 95% confidence interval included the nominal coverage level. SHAP results identified support type, roadway width, and equivalent diameter as the main contributing factors in the current model. These results provide a data-driven framework for friction resistance coefficient identification with point prediction, uncertainty estimation, and feature attribution under limited field samples.