张强, 李耀庄, 刘保华, 徐志胜. 秸秆灰混凝土力学性能试验及强度预测[J]. 农业工程学报, 2017, 33(2): 259-265. DOI: 10.11975/j.issn.1002-6819.2017.02.036
    引用本文: 张强, 李耀庄, 刘保华, 徐志胜. 秸秆灰混凝土力学性能试验及强度预测[J]. 农业工程学报, 2017, 33(2): 259-265. DOI: 10.11975/j.issn.1002-6819.2017.02.036
    Zhang Qiang, Li Yaozhuang, Liu Baohua, Xu Zhisheng. Mechanical properties and strength prediction of straw ash concrete[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(2): 259-265. DOI: 10.11975/j.issn.1002-6819.2017.02.036
    Citation: Zhang Qiang, Li Yaozhuang, Liu Baohua, Xu Zhisheng. Mechanical properties and strength prediction of straw ash concrete[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(2): 259-265. DOI: 10.11975/j.issn.1002-6819.2017.02.036

    秸秆灰混凝土力学性能试验及强度预测

    Mechanical properties and strength prediction of straw ash concrete

    • 摘要: 为了优化混凝性能,减少水泥产业耗能,尝试采用以部分秸秆灰替代水泥制备混凝土。该文通过试验对油菜秸秆灰混凝土拉压性能进行了研究,得到秸秆灰质量分数和水胶比对秸秆灰混凝土拉压性能的影响规律,如当秸秆灰质量分数增大时,混凝土拉压性能呈下降趋势;当水胶比过大时,混凝土力学性能急剧下降。同时提出秸秆灰混凝土抗拉性能与抗压性能间的线性函数关系以及混凝土轴心抗压强度计算公式,并与其他混凝土抗压强度公式进行比对验证。采用小波神经网络的预测方法,引入随机函数,对试验数据抽样进行训练,而后预测数据并与试验数据进行比对,计算误差,并将预测数据用于该文提出的拉压公式进行验证,结果表明验证较好。最后试验结果表明:当秸秆灰替代掺量为10%时,秸秆灰混凝土劈裂抗拉强度下降了25%,抗压强度仅下降了8%;当替代掺量为20%时,抗压强度下降了31%。

       

      Abstract: Abstract: In order to enhance the coagulation performance and reduce the energy consumption of the cement industry, part of the cement is replaced by straw ash to produce concrete. In this paper, biomass stalk ashes were acquired through microthermal incineration of rape stalks (500 ℃ for 5 h), and the stalk ash samples were obtained through grinding and screening of the preliminary ashes. Stalk ash was used as the concrete admixture to replace the same quantity of cement to produce experimental specimens, different amounts of rape stalk ash admixtures and concrete water-binder ratios were selected as the affecting variables of concrete property, and the effects of stalk ash on the concrete were discussed. The results indicated that when the amount of rape stalk ash admixture was 5%, the splitting tensile strength property of stalk ash concrete was 12% lower and the compression resistance was only 4% lower than normal concrete (28 d); when the amount of rape stalk ash admixture was 10%, the splitting tensile strength property of rape stalk ash concrete was 25% lower and the compression resistance was 8% lower than normal concrete; when the amount of admixture was 15%, the compression resistance was 13% lower than normal concrete, which met the use requirement of structural concrete (Code for Design of Reinforced Concrete Structures); and when the amount of rape stalk ash admixture was 20%, the splitting tensile strength property of rape stalk ash concrete was 45% lower and the compression resistance was 23% lower than normal concrete. And the rate of descent accelerated when the amount of rape stalk ash admixture exceeded over 20%. The experiments proved that stalk ash was somewhat water-absorbing, and therefore the best water-binder ratio of stalk ash concrete fell in the range of 0.45-0.55, and the best water-binder ratio was 0.5 for rape stalk ash concrete. The relationship between straw ash concrete's tensile and compressive properties was given, which was further verified by comparing with other concrete's tensile and compressive strength formulas including American Concrete Association's recommended formula, Yuan Biao's empirical formulas and Wang Dehui's empirical formulas. The splitting tensile strength ratio could be concluded from the fitting of experimental data, and the fitting result was good. Due to the lack of experimental data of straw ash concrete, however, the next step of the research focused on the verification of the reasonability of this relationship. The method of the random function was introduced to conduct the random sampling on the experimental data. And the prediction method of wavelet neural network was used to improve the training samples, and automatically modify the network structure parameters and predict stalk ash concrete's experimental data. Then, the predicted data were used for verifying the tensile and compressive formulas proposed in this study and the predicted and the test data were compared for error calculation. The wavelet neural network forecast data indicated that the maximum forecast error was 8% and the minimum was only 0.8%, so it was appropriate to forecast the mechanical property of stalk ash concrete.

       

    /

    返回文章
    返回