(1. 中南大學(xué) 資源與安全工程學(xué)院，長沙 410083；
2. 中南大學(xué) 湖南省深部資源開發(fā)與災(zāi)害控制重點實驗室，長沙 410083)
摘 要: 應(yīng)用改進(jìn)的物元可拓理論，并選取影響巖爆烈度的主要因素，如硐室最大切向應(yīng)力σθ、巖石單軸抗壓強(qiáng)度σc、巖石單軸抗拉強(qiáng)度σt以及沖擊傾向指數(shù)Wet，以σθ/σc、σc/σt及Wet作為巖爆烈度評價指標(biāo)，建立巖爆烈度預(yù)測的改進(jìn)物元可拓模型。該模型首先確定巖爆烈度分級標(biāo)準(zhǔn)，通過隸屬函數(shù)對分級標(biāo)準(zhǔn)進(jìn)行隸屬化，并確定節(jié)域隸屬化范圍，彌補(bǔ)特征值可能超出節(jié)域而使關(guān)聯(lián)函數(shù)失效的缺陷；然后運用博弈論，將評價指標(biāo)客觀動態(tài)權(quán)重與主觀靜態(tài)權(quán)重優(yōu)化、融合，克服傳統(tǒng)物元可拓評價中單純依靠特征值賦權(quán)而忽略特征本身對評價結(jié)果的重要性的弊端；最后通過最大關(guān)聯(lián)度準(zhǔn)則對巖爆烈度等級進(jìn)行預(yù)測，從而建立巖爆烈度預(yù)測的改進(jìn)物元可拓模型。為進(jìn)一步考察該模型有效性及實用性，運用該模型對國內(nèi)外20個巖爆工程進(jìn)行分析。應(yīng)用結(jié)果表明：該模型預(yù)測結(jié)果與實際情況吻合較好，且與傳統(tǒng)物元可拓評價方法和模糊數(shù)學(xué)綜合評價方法得到的結(jié)果相比，具有更高的準(zhǔn)確度。
關(guān)鍵字: 巖爆烈度預(yù)測；博弈論；改進(jìn)物元可拓模型；隸屬函數(shù)；關(guān)聯(lián)度
（1. School of Resources and Safety Engineering, Central South University, Changsha 410083, China;
2. Key Laboratory of Deep Mining and Disaster Control in Hunan Province, Central South University,）
Abstract:Based on the improved matter-element extension theory, an improved matter-element extension model used to predict the rockburst intensity was established. Firstly, the main factors of rockburst intensity, such as the maximum tangential stress of the cavern wall σθ, uniaxial compressive strength σc, uniaxial tensile strength σt, and the elastic energy index of rock Wet, were taken into account in the analysis. Three factors, σθ/σc, σc/σt and Wet were defined as the criterion indices for rockburst intensity prediction in the proposed model. Secondly, the classification standards of rockburst intensity were confirmed. In order to remedy the defect of the correlation function which may exceed the controlled field, the rockburst intensity classification standards were normalized. And then, based on the game theory, the synthetic weight values of eigenvalue were determined by integrating the objective-dynamic weight and subjective-static weight, solving problems occurred in traditional matter-element extension assessment method like indicator weight only depends on eigenvalue, ignoring the significance of feature. Lastly, the rockburst intensity level (RIL) was predicted by the maximum incidence degree criterion, then are improved matter-element extension model used to predict the rockburst intensity was proposed. 20 cases of the engineering project examples of the domestic and foreign were analyzed by using the proposed model to study the effectiveness and practicality of the model. The results show that the prediction results agrees well with the practical situations, and have higher accuracy compared with the results of the traditional matter-element extension method and the fuzzy mathematics comprehensive evaluation method.
Key words: prediction of rockburst intensity; game theory; improved matter-element extension model; membership function; incidence degree
