三間房礦3.0Mta新井設(shè)計(jì)【含CAD圖紙+文檔】,含CAD圖紙+文檔,三間,mta,設(shè)計(jì),cad,圖紙,文檔 附錄A 應(yīng)用具有遺傳算法和模糊選擇模型的神經(jīng)網(wǎng)絡(luò)為全機(jī)械化采煤選擇設(shè)備 王新宇，吳瑞明，馮春華 摘要：根據(jù)典型的工程樣品，用遺傳算法來優(yōu)化權(quán)值的神經(jīng)網(wǎng)絡(luò)工作模式提出了預(yù)測(cè)回采工作面的生產(chǎn)能力和效率。通過這個(gè)模型，我們可以獲得一定地質(zhì)情況下綜采工作面設(shè)備組合可能的結(jié)果。模糊選擇理論被用于評(píng)估每組設(shè)備組合的各項(xiàng)效率。通過詳細(xì)的實(shí)證分析，該模型結(jié)合預(yù)測(cè)回采工作面的數(shù)據(jù)和選擇最佳的設(shè)備組合的功能，并且有利于完全機(jī)械化采煤設(shè)備組合的決策。 關(guān)鍵詞：遺傳算法；人工神經(jīng)網(wǎng)絡(luò)；模糊選擇；選擇設(shè)備組合 1前言 在煤礦生產(chǎn)過程中，為采煤工作面選擇適當(dāng)?shù)脑O(shè)備組合是最重要的決策任務(wù)之一。一般的解決方案，任務(wù)是通過專業(yè)知識(shí)和工程經(jīng)驗(yàn)完成的。設(shè)備與采煤工作面的地質(zhì)條件的匹配的重要性是為了使設(shè)備運(yùn)行良好并且實(shí)現(xiàn)高輸出性能。在本文中，我們應(yīng)用人工神經(jīng)網(wǎng)絡(luò)模型，根據(jù)地質(zhì)條件和回采設(shè)備，利用在典型的工程案例中學(xué)到的經(jīng)驗(yàn)性的知識(shí)來預(yù)測(cè)回采工作面的輸出和效率。模糊優(yōu)選理論被用于評(píng)估每個(gè)組合設(shè)備在某些綜采工作面的地質(zhì)條件下的性能。使用模糊決策模型可以得到最令人滿意的設(shè)備組合。在人工神經(jīng)網(wǎng)絡(luò)模型中，通過遺傳算法來完成網(wǎng)絡(luò)節(jié)點(diǎn)之間最優(yōu)權(quán)的搜索，這種算法可以通過反向傳播算法避免網(wǎng)絡(luò)目標(biāo)函數(shù)局部?jī)?yōu)化。通過詳細(xì)的實(shí)證分析，該模型集成預(yù)測(cè)回采工作面的成績(jī)和選擇最佳的設(shè)備組合的功能，并有利于綜采設(shè)備組合的的決策。 2選擇合適的設(shè)備組合系統(tǒng)模型 2.1用人工神經(jīng)網(wǎng)絡(luò)和遺傳算法預(yù)測(cè)輸出和采煤工作面的效率 遺傳算法模擬生物進(jìn)化的自然選擇的原則，它已被用于解決各種工程和科學(xué)的優(yōu)化問題來找到真正理想的最佳點(diǎn)。通過遺傳算法隨機(jī)運(yùn)營，遺傳算法控制對(duì)最佳點(diǎn)的搜索過程。大家都知道，反向傳播算法是梯度遞減算法，它不能得到節(jié)點(diǎn)之間的最佳權(quán)重。因此，我們應(yīng)用遺傳算法訓(xùn)練網(wǎng)絡(luò)，而不是被稱為GA-BP算法的BP算法。它可以提高網(wǎng)絡(luò)收斂速度的非線性映射能力。 在GA-BP算法中，用算法數(shù)字的字符串代替二進(jìn)制數(shù)字的字符串來直接表達(dá)節(jié)點(diǎn)之間的權(quán)重。由此，在二進(jìn)制編碼的字符串中實(shí)施相應(yīng)的編碼和解碼操作是可以避免的，因此收斂速度更高。另一個(gè)好處是，權(quán)重的計(jì)算精度要好得多。 詳細(xì)步驟如下： 1）隨機(jī)產(chǎn)生N組網(wǎng)絡(luò)的權(quán)重； 2）采取BP算法訓(xùn)練網(wǎng)絡(luò)與上述初始N組權(quán)重。如果至少有一組權(quán)重達(dá)到目標(biāo)函數(shù)規(guī)定的訓(xùn)練精度，算法結(jié)束;否則采取下一步驟； 3）尋找權(quán)重屬于上述訓(xùn)練有素的N組權(quán)重信息的可能的數(shù)字領(lǐng)域。我們?cè)谶@個(gè)數(shù)值區(qū)域設(shè)新的R×N組隨機(jī)權(quán)重。所有組權(quán)重由整個(gè)基因集落組成； 4）應(yīng)用遺傳操作，即選擇，交叉和變異（R+1）×N組權(quán)重； 5）如果有至少一組權(quán)重在第4步中滿足精度要求，算法結(jié)束；否則我們從（R﹢1）×N組權(quán)重中選擇最好的權(quán)重，然后返回到步驟2。 根據(jù)典型工程案例，人工神經(jīng)網(wǎng)絡(luò)的主要任務(wù)是學(xué)習(xí)地質(zhì)條件和設(shè)備與產(chǎn)出和效率的映射關(guān)系的經(jīng)驗(yàn)知識(shí)。然后，它被用來預(yù)測(cè)在一定地質(zhì)條件下可用的設(shè)備組合的未來業(yè)績(jī)。 2.2根據(jù)模糊選擇決策模型評(píng)估設(shè)備的性能 可用的設(shè)備組合及其未來業(yè)績(jī)，如通過人工神經(jīng)網(wǎng)絡(luò)構(gòu)建一個(gè)多目標(biāo)的決策系統(tǒng)來獲得輸出和效率。我們運(yùn)用模糊優(yōu)選理論來評(píng)估每個(gè)可用的設(shè)備組合。評(píng)價(jià)模型會(huì)幫我們找到最好的設(shè)備組合。假設(shè)有m個(gè)目標(biāo)和n可用的項(xiàng)目，我們使用矩陣X=（Xi）M×N表達(dá)基礎(chǔ)數(shù)據(jù)集，其中Xij表示第i個(gè)目標(biāo)第j個(gè)項(xiàng)目的值。對(duì)于輸出和效率，我們采取下列公式得到相對(duì)隸屬度： （i=1，2，…，m）， （1） 其中rij為第i個(gè)目標(biāo)第j個(gè)項(xiàng)目的相對(duì)隸屬度。 由方程（1）我們得到的相對(duì)隸屬度矩陣R=（rij）m×n。最好的項(xiàng)目的相對(duì)隸屬度是：g=（g1，g2，…，gm）T=（1，1，…，1）T，最差的項(xiàng)目的相對(duì)隸屬度為b=（b1，b2，…，bm）T=（0，0，…，0）T，所有決策目標(biāo)的權(quán)重向量w=（w1，w2，…，wm）T，。第j個(gè)項(xiàng)目可表示為：Rj=（r1j’，r2j'，…，rmj’）T。對(duì)于第j個(gè)項(xiàng)目的最好的項(xiàng)目的一般距離為。根據(jù)模糊集合論，如果設(shè)第j個(gè)項(xiàng)目中最好的項(xiàng)目的相對(duì)隸屬度為μj，第j個(gè)項(xiàng)目中最差的項(xiàng)目的相對(duì)隸屬度必須是。 第j個(gè)項(xiàng)目中最好的項(xiàng)目的加權(quán)距離定義為；第j個(gè)項(xiàng)目的最差的項(xiàng)目的加權(quán)距離是。我們采取決策規(guī)則得到μj：最小值{}，則有是第j個(gè)項(xiàng)目的最終評(píng)估值。μj越大，第j個(gè)項(xiàng)目越好。 3實(shí)證分析 我們?cè)谏綎|省收集了一些采煤工作面的典型工程案例?？紤]到綜采的不同方法，我們將所有樣品分成四組訓(xùn)練樣本。 3.1用人工神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)輸出和效率 3.1.1輸入和輸出變量 地質(zhì)因素對(duì)選擇合適的設(shè)備很重要，也對(duì)設(shè)備性能有影響。相應(yīng)的主要地質(zhì)因素是神經(jīng)網(wǎng)絡(luò)的輸入對(duì)象包括：煤層的高度X1（米），煤層傾角X2（度），煤炭硬度系數(shù)X3，綜采工作面的長(zhǎng)度X4（米），老巖石峰的水平X5，直巖石峰的水平X6，氣體的水平X7（立方米每噸每天），采煤工作面機(jī)械的類型X8，支架的類型X9。我們只考慮采煤工作面機(jī)械和支架兩種煤機(jī)。神經(jīng)網(wǎng)絡(luò)的輸出對(duì)象是輸出Y1（t）和效率Y2（t?I-1）。 3.1.2神經(jīng)網(wǎng)絡(luò)的輸入和輸出對(duì)象的轉(zhuǎn)型 神經(jīng)網(wǎng)絡(luò)中的節(jié)點(diǎn)的核心功能往往是乙狀結(jié)腸功能。當(dāng)節(jié)點(diǎn)的輸入接近0或1時(shí)，節(jié)點(diǎn)的輸出改變速度是非常緩慢的。為了避免這種情況，我們通過線性變換把所有的輸入和輸出數(shù)據(jù)轉(zhuǎn)換到[0.5，0.95]內(nèi)。 1）連續(xù)輸入和輸出對(duì)象 輸入變量X1，X2，X3，X4，X7，XJQ和輸出變量Y1，Y2是連續(xù)的。以煤層的高X1為例，線性變換公式是： . （2） 為了收回實(shí)際輸出，我們采取另一種線性變換： . （3） 2）離散輸入和輸出對(duì)象 輸入變量X5，X6，X8，X9是離散的。例如，變量X5包括四個(gè)層次：第I，II，III和IV。我們用0.05，0.35，0.65和0.95標(biāo)志這些層次作為網(wǎng)絡(luò)的輸入。變量X8，X9代表設(shè)備。訓(xùn)練樣本中有7種采煤工作面機(jī)械： MLS3-170，MXA-300，MG-300，AM-500，MG-150，MD-150，MXA-600用0.05，0.20，0.35，0.50，0.65，0.80，0.95編碼。有10種支架：BY360-25/20，BC-480，QY320-13/32，QY200-14/31，ZY35，QY240-26/10，BY240-16/35，ZY28，ZY560K/ L.12，ZYQ1700，同樣可以編碼。 3.1.3神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)結(jié)果 建立一個(gè)有9個(gè)輸入節(jié)點(diǎn)，30個(gè)隱藏節(jié)點(diǎn)和2個(gè)輸出節(jié)點(diǎn)的神經(jīng)網(wǎng)絡(luò)。我們選擇18個(gè)樣本使用GA-BP算法訓(xùn)練網(wǎng)絡(luò)，并使用其他的7個(gè)樣品測(cè)試訓(xùn)練有素的網(wǎng)絡(luò)的可靠性。預(yù)測(cè)結(jié)果如表1。所有樣品的所有相對(duì)百分比誤差在士10％之內(nèi)。這表明，該模型是有效的。 3.1.4從網(wǎng)絡(luò)學(xué)到的知識(shí)的存儲(chǔ) 我們用C語言開發(fā)GA-BP算法的計(jì)算機(jī)軟件。當(dāng)訓(xùn)練過程結(jié)束時(shí)，我們使用文本文件來存儲(chǔ)網(wǎng)絡(luò)的信息包括：網(wǎng)絡(luò)結(jié)構(gòu)，網(wǎng)絡(luò)中任意兩個(gè)節(jié)點(diǎn)之間的所有權(quán)重。當(dāng)推理任務(wù)需要時(shí)，我們可以使用存儲(chǔ)在文本文件中的網(wǎng)絡(luò)信息重建的計(jì)算機(jī)網(wǎng)絡(luò)，然后新的輸入信息通過神經(jīng)網(wǎng)絡(luò)進(jìn)行處理，以獲得預(yù)測(cè)的輸出結(jié)果。神經(jīng)網(wǎng)絡(luò)中的權(quán)重如表2。 3.2設(shè)備組合的評(píng)價(jià) 3.2.1原始地質(zhì)條件 有一個(gè)采煤工作面，它的地質(zhì)條件是：煤層高2.5米，煤層傾角5度，煤炭硬度系數(shù)1.8，挖掘長(zhǎng)度面臨180米，老巖石峰的水平1，直巖石峰的水平2，氣體水平6立方米/（噸·天）。我們?cè)岢龌谀：畔⒎峙淅碚摰哪Ｐ蛠磉x擇采煤方法。通過這種模式，適當(dāng)?shù)牡刭|(zhì)條件下的煤炭開采方法是普通綜采。 3.2.2 可用的設(shè)備組合 有兩種采煤工作面機(jī)械：MXA-600，MLS3-170;三種支架：BY240-16/35，ZY35，QY200-14/31。所以，有六個(gè)可用的設(shè)備組合。 3.2.3不同的設(shè)備組合的評(píng)價(jià)結(jié)果 不同的設(shè)備組合評(píng)價(jià)結(jié)果如表3。從表3可以看到，第2個(gè)項(xiàng)目得分最高。所以，它是我們的最佳選擇。這表明MXA-600和QY240-14/31是最好的設(shè)備組合。 表1 神經(jīng)網(wǎng)絡(luò)的預(yù)測(cè)值 編號(hào) X1 X2 X3 X4 X5 X6 X7 X8 X9 Y1 預(yù)測(cè)值 Y1/t 誤差/% Y2 預(yù)測(cè)值 Y2/t 誤差/ % 1 2.6 10 1.6 180 Ⅰ 3 8 7 1 88.940 93.876 5.55 46.8 45.0 -3.75 2 2.2 8 2.5 172 Ⅱ 2 7 7 2 70.555 64.840 -8.10 24.6 26.6 8.10 3 2.4 8 1.8 130 Ⅱ 1 8 3 3 78.569 77.830 -0.94 29.3 30.1 4.70 4 2.0 12 2.4 130 Ⅱ 1 13 2 4 48.100 50.601 5.20 15.5 15.4 -0.55 5 3.1 6 2.2 156 Ⅱ 2 7 4 5 63.223 66.447 5.10 28.4 26.6 -6.30 6 2.6 10 1.5 144 Ⅱ 1 5 1 6 75.617 76.146 0.70 30.2 29.6 -2.05 7 2.7 8 1.4 138 Ⅱ 2 8 1 4 76.175 78.003 2.40 32.8 30.8 -6.15 8 3.0 5 1.8 165 Ⅱ 1 10 3 7 73.019 74.991 2.70 29.2 28.7 -1.70 9 2.8 11 1.5 142 Ⅱ 1 6 1 6 75.584 75.463 -0.16 30.2 29.2 -3.20 10 3.2 7 2.0 152 Ⅰ 2 4 2 6 86.216 83.543 -3.10 37.3 39.3 5.30 11 2.2 5 2.2 180 Ⅱ 3 11 1 8 53..079 52.798 -0.53 16.8 17.0 1.40 12 2.6 13 1.6 165 Ⅱ 1 5 2 7 86.407 84.592 -2.10 37.4 38.5 2.90 13 2.8 3 2.0 149 Ⅰ 2 8 7 5 86.170 89.651 4.04 47.1 45.5 -3.50 14 2.0 10 2.3 150 Ⅱ 2 10 4 9 46.349 47.508 2.50 16.7 65.9 -4.50 15 2.8 9 2.0 165 Ⅱ 2 12 5 5 58.642 57.938 -1.20 19.6 18.9 -3.70 16 2.0 10 2.0 180 Ⅱ 2 12 2 10 54.306 53.980 -0.60 17.4 17.9 2.70 17 2.6 14 1.9 145 Ⅱ 3 6 6 8 46.411 50.634 9.10 16.7 17.4 4.70 18 2.4 9 1.6 180 Ⅰ 2 8 7 4 94.302 95.292 1.05 46.7 50.0 7.10 19 2.5 12 1.8 170 Ⅱ 1 6 2 1 78.020 76.148 -2.40 25.3 26.5 4.90 20 2.2 8 1.8 148 Ⅱ 1 8 1 4 68.749 69.265 0.75 23.2 24.8 7.10 21 2.4 11 2.0 190 Ⅰ 1 6 2 6 90.548 91.906 1.50 51.2 48.0 -6.20 22 2.2 8 1.8 160 Ⅱ 2 8 5 3 69.302 67.362 -2.80 23.1 23.7 2.80 23 2.5 5 2.5 168 Ⅱ 2 11 6 10 54.898 54.678 -0.40 18.2 17.9 -1.90 24 1.8 13 1.8 154 Ⅱ 1 9 4 1 69.070 65.271 -5.50 22.8 23.0 0.90 25 3.2 7 2.3 158 Ⅱ 2 7 4 5 62.342 61.158 -1.90 27.1 28.4 4.80 表2 神經(jīng)網(wǎng)絡(luò)的權(quán)值 表3不同的設(shè)備組合的評(píng)價(jià)結(jié)果 編號(hào). X8 X9 Y1 Y2 評(píng)價(jià)結(jié)果 1 MXA-600 ZY-35 67.913 28.4 0.1948 2 MXA-600 QY240-14/31 82.445 39.2 0.9977 3 MXA-600 BY240-16/35 83.971 36.4 0.8834 4 MLS3-170 ZY-35 60.718 26.1 0.0266 5 MLS3-170 QY240-14-31 75.482 30.3 0.5951 6 MLS3-170 BY240-16/35 65.329 22.6 0.0234 4結(jié)論 本文提出了用遺傳算法和模糊優(yōu)選模型來選擇基于神經(jīng)網(wǎng)絡(luò)的綜采工作面設(shè)備的一個(gè)框架，這個(gè)框架可以整合工程知識(shí)，為煤炭開采提供智能的決策支持。這個(gè)框架可以整合工程知識(shí)，為煤炭開采提供的智能決策支持。在今后的工作中，為了使整個(gè)決策支持系統(tǒng)更可靠和有效的，我們可以接受更多典型的工程案例，并擴(kuò)大輸入和輸出對(duì)象的數(shù)量，如經(jīng)濟(jì)因素。 Applying Neural Network with Genetic Algorithm and Fuzzy Selection Models to Select Equipments for Fully-Mechanized Coal Mining WANG Xin-yu， WU Rui-ming， FENG Chun-hua Abstract: According to the typical engineering samples. A neural net work model with genetic algorithm to optimize weight values is put forward to forecast the productivities and efficiencies of mining faces. By this model we can obtain the possible achievements of available equipment combinations under certain geological situations of fully mechanized coal mining faces. Then theory of fuzzy selection is applied to evaluate the performance of each equipment combination. By detailed empirical analysis，this model integrates the functions of forecasting mining faces' achievements and selecting optimal equipment combination and is helpful to the decision of equipment combination for fully-mechanized coal mining. Key words: genetic algorithm; artificial neural network; fuzzy selection; selection of equipment combination. 1 Preface Selecting propr equeipment combinations for coal mining face is one of the most important decision-making tasks during the process of coal mine production. Ordinary solution to that task is completed by expert knowledge, engineering experience. The matching of equipments and the geological conditions in coal mining face is of importance in order to make the equipments work well and achieve high performance of output. In this paper，we apply artificial neural network model to predict the outputs and efficiency of mining faces according the geological conditions and mining equipments with the learned empirical knowledge from typical engineering cases. The theory of fuzzy selection is applied to evaluate the performance of each equipment combination under certain geological conditions of mining face. Using the fuzzy decision model the most satisfied equipment combinations can be obtained. In the artificial neural network model the search for optimal weights between nodes of networks is finished by genetic algorithm which can avoid the objective function of networks into the local optimization by back- propagation algorithm. By detailed empirical analysis, this model integrates the functions of forecasting mining faces' achievements and selecting optimal equipment combination and is helpful to the decision of equipment combination for fully- mechanized coal mining. 2 Systematic Model of Selecting Proper Equipment Combination 2.1 Predicting output and efficiency of mining face with artificial neural network and genetic algorithm. Genetic algorithm mimics the natural selection principle in evolution of biology. It has been used to solve all kinds of optimal problems of engineering and science to find the really and ideally optimal point. Genetic algorithm controls the search process towards the optimal points by genetic arithmetic operators randoml. As we all known，the back-propagation algorithm can not get the best weights between nodes for it is a gradient-decreasing algorithm. So we apply the genetic algorithm to train the networks instead of BP algorithm， which is called GA-BP algorithm. It can enhance the nonlinear mapping capability of networks and convergence speed. In GA-BP algorithm，the algorism string of number takes place of binary string of number to express the weights between nodes directly. By that the corresponding coding and decoding operations in binary coding string is avoided， so the convergence speed is higher. Another benefit is that the calculation precision of weights can be much better. The detailed steps are as following: 1) Randomly producing N groups of weights of networks; 2) Take BP algorithm to train the networks with above initial N groups of weights. If there is at least one group of weights that makes the prescribed training precision of objective function， the algorithm is over; or take the next steps; 3) Finding the possible numerical areas which the weights are belong to with the above trained N groups of weights information. In this numerical area we crate other new r×N groups of weights randomly. All the groups of weights make up of the whole gene colony. 4) Apply genetic operations namely selection， crossover and mutation to (r+1)×N groups of weights; 5) If there is at least one groups of weights satisfies the precision demand in step 4， the algorithm is over; or we select the best N groups of weights from (r+1)×N groups of weights and return to step 2. The main task of artificial neural network is to learn the empirical knowledge on the mapping relations from geological conditions and equipments to outputs and efficiency，according to the typical engineering cases. Then it is used to predict the available equipment combinations' future performance in certain geological condition. 2.2 Evaluating equipments' performance based on fuzzy selection decision-making model Available equipment combinations and their future performance such as output and efficiency obtained by artificial neural networks construct a multi-objective decision-making system. We apply fuzzy selection theory to evaluate each available equipment combination. The evaluation model can help us to find which equipment combination is best. Suppose that there are m objectives and n available projects， we use matrix X=（Xi）m*n to express the basis data set，in which Xij means the value of i-th objective of j-th project. For output and efficiency，we take the following formula to get the relatively subjection degree: （i=1，2，…，m）， （1） where rij is relative subjection degree for i-th objective of j-th project. By equation (1) we get the relative subjection degree matrix R=(rij)m*n. The relative subjection degree of the best project is g=(g1,g2,…，gm)T=(1，1，…，1)T,and the relative subjection degree of the worst project is b=(b1,b2,…，bm)T=(0，0，…，0)T. The weight vector of all decision objectives is w=(w1，w2，…，wm)T，. The j-th project can be expressed as rj=(r1j',r2j',…，rmj)T. The general distance to the best project for j-th project is .If we set the relative subjection degree to best project of j-th project as μj according the fuzzy set theory，the relative subjection degree to the worst project of j-th project must be . The weighted general distance to the best project of j-th project is defined as; The weighted general distance to the worst project of j-th project is. We take the following decision rule to get μj: min{},then we have is the final evaluation values of j-th project. The bigger μj，the better j-th project. 3 Empirical analysis We collected some typical engineering cases of coal mining faces in Shandong province. In regard to the different methods of fully mechanized coal mining， we divided all samples into four groups of training samples. 3.1 Predicting output and efficiency with artificial neural networks Geological factors are of importance to select proper equipments， and also have an effect on the performance of equipments. The main corresponding geological factors are input targets of neural network including: height of coal bed X1 (m), obliquity of coal bed X2(°), rigidity coefficient of coal X3，length of mining face X4 (m)，levels of old rock peak X5 levels of straight rock peak X6，levels of gas X7(m3?t-1?d-1),types of coal face machinery X8，types of bracket X9. We only consider two kinds of coal machines, which are coal face machinery and bracket. The output targets of neural networks are output Y1(t) and efficiency Y2(t×I-1). 3.1.2 Transformation of input and output targets in neural networks The core function of nodes in neural networks is often sigmoid function. When the input of nodes is near to 0 or 1, the changing speed of output of nodes is very slow. In order to avoid this, we convert all input and output data into areas [0.5，0.95] by linear transformations. 1) Continuous input and output targets The input variables X1 X2 X3 X4 ，X7 XJQ and the output variables Y1，Y2 are continuous. Take height of coal bed X1 as an example，the linear transformation formula is . （2） In order to recover the factual output， we take another linear transformation: . （3） 2) Discrete input and output targets The input variables X5 X6 X8 X9 are discrete. For example，variables X5 includes four levels: I, II， III， and IV. We code these levels with 0.05, 0.35, 0.65, and 0.95 as inputs of networks. Variables X8 X9 denote equipments. There are seven types of coal face machineries in training samples: MLS3-170, MXA-300, MG-300, AM-500, MG-150， MD-150， MXA-600, which are coded with 0.05, 0.20, 0.35, 0.50, 0.65, 0.80, 0.95. There are ten types of brackets: BY360-25/20BC-480, QY320-13/32, QY200-14/31, ZY35, QY240-26/10, BY240-16/35 ZY28, ZY560K/l.12, ZYQ1700, which can be coded similarly. 3.1.3 Predicting results of neural networks A neural network with 9 input nodes, 30 hidden nodes and 2 output nodes are built. We choose 18 samples to train the network by GA-BP algorithm, and use other 7 samples to test the reliability of trained networks. The predicting result is as Table 1. All relative percentage errors for all samples are within ±10%. It indicates the model is effective. 3.1.4 Storage of knowledge learned from net-works We develop the computer software of GA-BP algorithm with C language. When the training process is over, we use a text file to store the networks information including: networks structure, all weights between any two nodes in networks. When a reasoning task is needed, we can make use of the stored network information in that text file to reconstruct the network in computer, and then the new input information is processed by neural networks to obtain the predicting output results. The weights in neural network are as Table 2. Table 1 Forecasting values of neural network No. X1 X2 X3 X4 X5 X6 X7 X8 X9 Y1 Predicted Y1/t Error/ % Y2 Predicted Y2/t Error/ % 1 2.6 10 1.6 180 Ⅰ 3 8 7 1 88.940 93.876 5.55 46.8 45.0 -3.75 2 2.2 8 2.5 172 Ⅱ 2 7 7 2 70.555 64.840 -8.10 24.6 26.6 8.10 3 2.4 8 1.8 130 Ⅱ 1 8 3 3 78.569 77.830 -0.94 29.3 30.1 4.70 4 2.0 12 2.4 130 Ⅱ 1 13 2 4 48.100 50.601 5.20 15.5 15.4 -0.55 5 3.1 6 2.2 156 Ⅱ 2 7 4 5 63.223 66.447 5.10 28.4 26.6 -6.30 6 2.6 10 1.5 144 Ⅱ 1 5 1 6 75.617 76.146 0.70 30.2 29.6 -2.05 7 2.7 8 1.4 138 Ⅱ 2 8 1 4 76.175 78.003 2.40 32.8 30.8 -6.15 8 3.0 5 1.8 165 Ⅱ 1 10 3 7 73.019 74.991 2.70 29.2 28.7 -1.70 9 2.8 11 1.5 142 Ⅱ 1 6 1 6 75.584 75.463 -0.16 30.2 29.2 -3.20 10 3.2 7 2.0 152 Ⅰ 2 4 2 6 86.216 83.543 -3.10 37.3 39.3 5.30 11 2.2 5 2.2 180 Ⅱ 3 11 1 8 53..079 52.798 -0.53 16.8 17.0 1.40 12 2.6 13 1.6 165 Ⅱ 1 5 2 7 86.407 84.592 -2.10 37.4 38.5 2.90 13 2.8 3 2.0 149 Ⅰ 2 8 7 5 86.170 89.651 4.04 47.1 45.5 -3.50 14 2.0 10 2.3 150 Ⅱ 2 10 4 9 46.349 47.508 2.50 16.7 65.9 -4.50 15 2.8 9 2.0 165 Ⅱ 2 12 5 5 58.642 57.938 -1.20 19.6 18.9 -3.70 16 2.0 10 2.0 180 Ⅱ 2 12 2 10 54.306 53.980 -0.60 17.4 17.9 2.70 17 2.6 14 1.9 145 Ⅱ 3 6 6 8 46.411 50.634 9.10 16.7 17.4 4.70 18 2.4 9 1.6 180 Ⅰ 2 8 7 4 94.302 95.292 1.05 46.7 50.0 7.10 19 2.5 12 1.8 170 Ⅱ 1 6 2 1 78.020 76.148 -2.40 25.3 26.5 4.90 20 2.2 8 1.8 148 Ⅱ 1 8 1 4 68.749 69.265 0.75 23.2 24.8 7.10 21 2.4 11 2.0 190 Ⅰ 1 6 2 6 90.548 91.906 1.50 51.2 48.0 -6.20 22 2.2 8 1.8 160 Ⅱ 2 8 5 3 69.302 67.362 -2.80 23.1 23.7 2.80 23 2.5 5 2.5 168 Ⅱ 2 11 6 10 54.898 54.678 -0.40 18.2 17.9 -1.90 24 1.8 13 1.8 154 Ⅱ 1 9 4 1 69.070 65.271 -5.50 22.8 23.0 0.90 25 3.2 7 2.3 158 Ⅱ 2 7 4 5 62.342 61.158 -1.90 27.1 28.4 4.80 Table 2 Weight values of neural network 3.2 Evaluation on equipment combinations 3.2.1 Original geological conditions There is a coal face, which geological conditions are: height of coal bed 2.5m, obliquity of coal bed 5 radians, rigidity coefficient of coal 1.8; length of mining faces 180m; levels of old rock peak 1; levels of straight rock peak 2; levels of gas 6 m3/(t?d). We have ever put forth a model based on fuzzy information distribution theory to select coal mining methods. By that model the proper coal mining method for the geological conditions is general fully mechanized coal mining. 3.2.2 Available equipment combinations There are two kinds of coal face machinery: MXA-600 MLS3-170 and three kinds of brackets: BY240-16/35 ZY35 QY200-14/31. so we have six available equipment combinations. 3.2.3 The result of evaluation of different equipment combinations The evaluation results of different equipment combinations are as Table 3. From table 3, we can see that 2-th project has the highest scores. So we make it as the best choice. It indicates that MXA-600 and QY240-14/31 is the best equipment combination. Table 3 Evaluation results of different equipment combinations No. X8 X9 Y1 Y2 Evaluation value 1 MXA-600 ZY-35 67.913 28.4 0.1948 2 MXA-600 QY240-14/31 82.445 39.2 0.9977 3 MXA-600 BY240-16/35 83.971 36.4 0.8834 4 MLS3-170 ZY-35 60.718 26.1 0.0266 5 MLS3-170 QY240-14-31 75.482 30.3 0.5951 6 MLS3-170 BY240-16/35 65.329 22.6 0.0234 4Conclusions This paper puts forth a framework to select equipment for coal faces based on neural network with genetic algorithm and fuzzy selection models. This framework can integrate the engineering Knowledge, and supply intelligent decision support for coal mining. In future work, we can embrace more typical engineering cases and expanded the number of input and output targets such as economic factors. In order to make the whole decision support system more reliable and effective. 附錄B 小煤礦安全生產(chǎn)的開采方法和技術(shù)改進(jìn) 翟莘縣，邵強(qiáng)，李永明，李華明 摘要：河南省平頂山市新華區(qū)是中國國家重點(diǎn)煤炭生產(chǎn)縣之一，其生產(chǎn)能力是每年超過0.6M