摘要
为科学评估矿山救援基地应急能力,提升其灾害事故响应与救援效能,提出了一种矿山应急救援能力评估模型,对山东某矿山救援基地进行评估并提出相应提升策略。首先,通过分析影响应急救援能力的关键因素,构建包含应急建设准备能力、协同能力和保障能力3项一级指标、8项二级指标及40项三级指标的救援能力评估体系;其次,基于序关系法(G1法)与标准间相关性法(CRITIC法)确定各项指标的主客观权重,结合博弈论将各权重融合得到最优综合权重;最后,采用核主成分分析(Kernel Principal Component Analysis,KPCA)法和冠豪猪优化(Crested Porcupine Optimizer,CPO)算法改进并构建支持向量机(Support Vector Machine,SVM)模型,以专家评分结合综合权重得到的输入样本进行回归预测。结果表明,KPCA-CPO-SVM模型对山东某矿山救援基地能力预测结果对比评估结果的准确率达到95%,评估指标平均绝对误差(Mean Absolute Error,MAE)为0.63%,均方根误差(Root Mean Square Error,RMSE)为0.79%,平均绝对百分比误差(Mean Absolute Percentage Error,MAPE)为0.84%,决定系数达到0.9698,验证了该模型的鲁棒性与可靠性。该救援基地需要在应急救援队伍建设、人员能力培训、应急物资调度和装备智能化等方面进行提升,研究结果为提升应急救援能力提供了参考。
This paper presents a mine emergency rescue capability evaluation model that integrates empowerment techniques with Kernel Principal Component Analysis(KPCA)and Collaborative Parameter Optimization-Support Vector Machine(CPO SVM).The model is specifically designed to evaluate the emergency response capabilities of a mine rescue base in Shandong and offers strategies for improvement.Initially,we analyze the key factors influencing emergency rescue capabilities to construct a comprehensive assessment system.This system comprises three first-level indicators,eight second-level indicators,and 40 third-level indicators,all of which are associated with emergency construction readiness,coordination ability,and support capability.Next,the subjective and objective weights of the indicators are determined using the G1 method and the CRITIC method,respectively.Subsequently,a game-theoretic approach is employed to integrate these weights and derive the optimal comprehensive weights.Following this,KPCA is utilized to reduce data dimensionality,thereby decreasing computational complexity while maintaining assessment accuracy.Concurrently,the Crested Porcupine Optimizer(CPO)is applied to optimize the kernel function parameter(g)and the penalty factor(C)of the Support Vector Machine(SVM),further enhancing the model s prediction performance and evaluation effectiveness.Finally,the evaluation model incorporating combined weighting and the KPCA CPO SVM framework is developed.Input samples are generated by integrating expert scores with the comprehensive weights,followed by regression analysis for prediction.The results indicate that the KPCA CPO SVM model achieves a prediction accuracy of 95%when assessing the rescue capacity of a mine rescue base in Shandong.The evaluation metrics include a Mean Absolute Error(MAE)of 0.63%,a Root Mean Squared Error(RMSE)of 0.79%,and a Mean Absolute Percentage Error(MAPE)of 0.84%.Additionally,the coefficient of determination(R^(2))reaches 0.9698,demonstrating the robustness and reliability of the assessment model.Furthermore,the model predicts that the emergency rescue capability of the rescue base is rated as“general,”indicating a need for enhancements in areas such as emergency rescue team development,personnel training,emergency material logistics,and equipment intelligence.The findings of this study offer valuable insights for improving emergency rescue capabilities。
作者
马砺
闫小庆
王旭
范晶
宋岩
MA Li;YAN Xiaoqing;WANG Xu;FAN Jing;SONG Yan(School of Safety Science and Engineering,Xi'an University of Science and Technology,Xi'an 710054,China;Yankuang Energy Group Company Limited,Zoucheng 273500,Shandong,China)
出处
《安全与环境学报》
北大核心
2025年第6期2259-2269,共11页
Journal of Safety and Environment
基金
陕西省秦创原“科学家+工程师”队伍建设项目(2022KXJ-087)
消防科学与技术创新团队项目(2023-CX-TD-42)。
关键词
公共安全
矿山应急救援基地
应急救援能力
组合赋权
评估模型
public safety
mine emergency rescue base
emergency rescue capability
combination empowerment
evaluation model
