Construction and application of optimized model for mine water inflow prediction based on neural network and ARIMA model

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 15, 2025

Mine water influx is a significant geological hazard during mine development, influenced by various factors such as conditions, hydrology, climate, and mining techniques. This phenomenon characterized non-linearity high complexity, leading to frequent accidents in coal mines. These not only impact production quality but also jeopardize the safety of staff. In order better predict amount surging mines provide an important basis for damage prevention work, based on time series data from January 2020 February 2023 Northern Guizhou Province Longfeng Coal Mine, BP-ARIMA prediction model was established combining BP neural network ARIMA autoregressive sliding average model, It predicted total 6 months July 2022 2023, compared results with four models, namely, traditional method Large well method, GM(1,1) grey used absolute relative error calculation accuracy. The show that BP-ARIMA(3,1,1) much closer actual value, 1.02% maximum 3.036%, goodness fit R² 0.93, which than other single substantially improves accuracy influx. Furthermore, utilizing future predictions were made, offering scientific foundation effective control measures.

Language: Английский

Comparing the Efficiency of Particle Swarm and Harmony Search Algorithms in Optimizing the Muskingum–Cunge Model

Water, Journal Year: 2025, Volume and Issue: 17(1), P. 104 - 104

Published: Jan. 2, 2025

Climate change-induced alterations in monsoon patterns have exacerbated flooding challenges Balochistan, Iran. This study addresses the urgent need for improved flood prediction methodologies data-scarce arid regions by integrating Muskingum–Cunge model with advanced optimization techniques. Particle swarm (PSO) and harmony search (HS) algorithms were applied compared across eight major rivers each distinct hydrological characteristics. A comprehensive multi-metric evaluation framework was developed to assess performance of these algorithms. The results demonstrate PSO’s superior performance, particularly complex terrain conditions. For instance, at Kajou station, PSO Coefficient Residual Mass (CRM) 0.01, efficiency (EF) 0.92, Agreement Index (d) 0.98, Normalized Root Mean Square Error (NRMSE) 0.10 HS. Correlation coefficients ranging from 0.6558 0.9645 validate methodology’s effectiveness environments. research provides valuable insights into algorithm under limited data conditions offers region-specific parameter guidelines similar geographical contexts. By advancing routing science providing a validated selection, this contributes management vulnerable climate change.

Language: Английский