模糊系统方法在混凝土无损检测抗压强度预测中的应用

APPLICATION OF FUZZY SYSTEM METHOD TO THE PREDICTION OF COMPRESSIVE STRENGTH IN CONCRETE UNDAMAGED INSPECTION

将模糊系统方法应用于混凝土无损检测技术中,以回弹值、超声值、碳化深度值和含水率作为模型输入变量,抗压强度值作为模型输出变量值,通过对样本数据训练,提取模糊规则,形成完备的模糊规则库。采用单值模糊器、乘积推理机、中心平均解模糊器,建立混凝土无损检测模糊系统模型。实验结果表明:模糊系统模型预测结果的平均相对误差为7.98%,相对标准差为10.53%。该模型的预测精度明显高于目前常用的回归模型预测精度。进而提出一种评定各输入变量重要性的方法,实验结果表明,影响混凝土抗压强度的各因素的重要性评定依次为:回弹值、超声值、碳化深度和含水率,其中回弹值最重要。该方法可以帮助我们选择较重要的变量作为模型的输入变量,实现模型的效率优化。

Fuzzy system method is applied to concrete undamaged inspection. In the model, the inputs are rebound value, ultrasonic value, carbonation depth value and moisture rate. The output is the compressive strength. By the sample data training, fuzzy rules can be extracted. All rules constructed a complete base of fuzzy IF-THEN rules. Using the singleton fuzzifier, the product inference engine, and the center-average defuzzifier, the fuzzy system model of concrete undamaged inspection is obtained. The experiment shows the average relative error of the predicted results is 7.89%, and the relative standard error is 10.53%. The prediction accuracy of the fuzzy model is obviously higher than that of the regression model. A method is further proposed to rank the importance of the input variables. The result we obtained shows that the rank is： rebound value, ultrasonic value, carbonation depth value, and moisture rate, with rebound value being the most important. It can help us to select the more important ones as the inputs to the model so as to achieve the efficiency optimization of the model.