(中南大學(xué) 資源與安全工程學(xué)院，長(zhǎng)沙 410083)
摘 要: 為提高采場(chǎng)聲發(fā)射事件率預(yù)報(bào)精度，將采場(chǎng)聲發(fā)射事件率不同的單個(gè)預(yù)測(cè)模型的預(yù)測(cè)值作為函數(shù)鏈神經(jīng)網(wǎng)絡(luò)的原始輸入值，并將原始輸入值按正交的三角函數(shù)擴(kuò)展得到的數(shù)值作為函數(shù)鏈神經(jīng)網(wǎng)絡(luò)擴(kuò)展輸入值，在分析函數(shù)鏈神經(jīng)網(wǎng)絡(luò)擬合充要條件的基礎(chǔ)上，結(jié)合模糊自適應(yīng)變權(quán)重算法計(jì)算函數(shù)鏈神經(jīng)網(wǎng)絡(luò)權(quán)重，對(duì)采場(chǎng)聲發(fā)射事件率進(jìn)行基于模糊自適應(yīng)變權(quán)重算法的函數(shù)鏈神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)，對(duì)其預(yù)測(cè)結(jié)果再進(jìn)行函數(shù)鏈神經(jīng)網(wǎng)絡(luò)算法擬合，然后結(jié)合采場(chǎng)冒頂尖點(diǎn)突變模型的判別式對(duì)采場(chǎng)冒頂進(jìn)行預(yù)報(bào)。某鉛鋅礦采場(chǎng)冒頂預(yù)報(bào)結(jié)果表明，基于模糊自適應(yīng)變權(quán)重算法的函數(shù)鏈神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)方法的預(yù)測(cè)誤差小于0.3%，可實(shí)現(xiàn)采場(chǎng)冒頂精確預(yù)報(bào)。
關(guān)鍵字: 函數(shù)鏈神經(jīng)網(wǎng)絡(luò)；模糊自適應(yīng)變權(quán)重算法；預(yù)測(cè)；采場(chǎng)冒頂；聲發(fā)射
fuzzy adaptive variable weight method
（School of Resource and Safety Engineering, Central South University, Changsha 410083, China）
Abstract:In order to enhance the predict precision about happening rate of acoustic emission in mine, the happening rate of acoustic emission in mine was forecasted based on functional link neural network due to fuzzy adaptive variable weight algorithm by using of making some forecasting values from different single forecasting model of happening rate of acoustic emission in mine as original input values of functional link neural network, making the original input values as patulous input values of functional link neural network after the original input values being extended according to the orthogonal trigonometric function, analyzing the necessary and sufficient conditions of functional link neural network fitting and calculating the weight of functional link neural network based on fuzzy adaptive variable weight algorithm. And the roof caving can be predicted when the forecasting results is fitted by functional link neural network algorithm and the discriminant of roof caving abrupt change model. The forecasting results of happening rate of acoustic emission in some lead and zinc mine reveal that the functional link neural network forecasting method based on fuzzy adaptive variable weight algorithm is higher than that of other forecasting model and its forecasting error is smaller than 0.3%. And the precision predicting roof caving is able to be realized due to the functional link neural network forecasting.
Key words: functional link neural network; fuzzy adaptive variable weight method; prediction; roof caving; acoustic emission
