技术领域Technical field
本发明属于土木工程领域，具体涉及一种钻探数据分析评估方法。The invention belongs to the field of civil engineering, and specifically relates to a drilling data analysis and evaluation method.
背景技术Background technique
桩基是结构的主要承重部分，其质量直接关系到结构的安全性和长久性。灌注桩由于其使用范围广、单桩承载力大、经济等特点，被广泛应用于各种大型建筑、桥梁施工。但因地质条件复杂，灌注方法等原因，混凝土桩身质量较难控制，而桩身质量的好坏直接影响着桩的承载力。在这种情况下，一个科学，实用的检测方法就显得尤为重要。The pile foundation is the main load-bearing part of the structure, and its quality is directly related to the safety and durability of the structure. Cast-in-place piles are widely used in the construction of various large buildings and bridges due to their wide range of use, large single pile bearing capacity, and economy. However, due to complex geological conditions, pouring methods and other reasons, the quality of the concrete pile body is difficult to control, and the quality of the pile body directly affects the bearing capacity of the pile. In this case, a scientific and practical detection method is particularly important.
目前传统的检测方法有超声波探伤、钻芯取样等方法。其中钻芯取样法因其具有不用预埋声管、能较为直观的检测混凝土的强度和较好的完整性等特点，被广泛采用。At present, traditional detection methods include ultrasonic flaw detection, core drilling and other methods. Among them, the drill core sampling method is widely used because it does not require embedded sound pipes and can more intuitively detect the strength and better integrity of concrete.
作为桩基检测中应用最为广泛的方法之一，钻芯法能够客观直接的反映出灌注桩的完整性，桩底沉渣厚度以及桩端持力层的情况。但钻芯法对桩身完整性的判断依赖于现场施工人员对所采集的芯样进行完整性比对，对桩底沉渣厚度的评估依赖于人工肉眼观测钻孔机动力头的下降的速度，对桩身混凝土强度的判断更是只能依据对采集的芯样进行混凝土强度破坏性实验。As one of the most widely used methods in pile foundation testing, the core drilling method can objectively and directly reflect the integrity of the cast-in-place pile, the thickness of the sediment at the bottom of the pile, and the condition of the bearing layer at the pile end. However, the core drilling method's judgment on the integrity of the pile body relies on on-site construction personnel to compare the integrity of the core samples collected. The assessment of the sediment thickness at the bottom of the pile relies on manual visual observation of the descending speed of the drilling machine's power head. The judgment of the concrete strength of the pile body can only be based on the destructive concrete strength test of the collected core samples.
因此，目前的钻芯法虽然能够对桩基进行检测，但是其方法中大量的人为因素对检测结果的影响较大，并且缺少对检测数据科学的记录方法，这对检测结果的严谨性有一定局限。此外，对灌注桩混凝土强度的检测，也一定程度上增加了工程的成本。Therefore, although the current core drilling method can detect pile foundations, a large number of human factors in the method have a greater impact on the detection results, and there is a lack of scientific recording methods for detection data, which has a certain impact on the rigor of the detection results. limitations. In addition, testing the concrete strength of cast-in-place piles also increases the cost of the project to a certain extent.
发明内容Contents of the invention
本发明的目的在于提供一种可靠性高、成本较低且实用性较好的钻探数据分析评估方法。The purpose of the present invention is to provide a drilling data analysis and evaluation method with high reliability, low cost and good practicability.
本发明提供的这种钻探数据分析评估方法，包括如下步骤：The drilling data analysis and evaluation method provided by the present invention includes the following steps:
S1.在钻探过程中，获取钻孔机在单位时间内移动距离的时间序列数据集；S1. During the drilling process, obtain a time series data set of the drilling machine's movement distance per unit time;
S2.对步骤S1获取的单位时间内移动距离的时间序列数据集进行数据处理，得到处理后的时间序列数据集；S2. Perform data processing on the time series data set of moving distance per unit time obtained in step S1 to obtain a processed time series data set;
S3.根据步骤S2获取的处理后的时间序列数据集，确定钻进速度，以及处理后的时间序列数据集中每个数据所对应的钻探深度；S3. Based on the processed time series data set obtained in step S2, determine the drilling speed and the drilling depth corresponding to each data in the processed time series data set;
S4.根据步骤S3获取的钻进速度和处理后的时间序列数据集中每个数据所对应的钻探深度，构建钻探序列数据集；S4. Construct a drilling sequence data set based on the drilling speed obtained in step S3 and the drilling depth corresponding to each data in the processed time series data set;
S5.构建钻探数据评估网络模型，并利用步骤S4得到的钻探序列数据集对构建的钻探数据评估网络模型进行优化；S5. Construct a drilling data evaluation network model, and use the drilling sequence data set obtained in step S4 to optimize the constructed drilling data evaluation network model;
S6.采用步骤S5得到的优化后的钻探数据评估网络模型进行基桩桩身混凝土完整性及桩底沉渣厚度分析，并取特定某段芯样进行相关实验，输入钻探数据评估网络模型后得到完整的基桩桩身混凝土强度报告；S6. Use the optimized drilling data evaluation network model obtained in step S5 to analyze the concrete integrity of the pile body and the sediment thickness at the bottom of the pile, and take a specific core sample to conduct relevant experiments. After inputting the drilling data evaluation network model, the complete Concrete strength report of foundation pile body;
S7.根据步骤S6的分析结果，进行最终的钻探数据分析评估。S7. Carry out final drilling data analysis and evaluation based on the analysis results of step S6.
步骤S1所述钻孔机在单位时间内移动距离的时间序列数据集由布置在钻孔机动力头上的红外激光测距传感器所采集，所述红外激光测距传感器根据设定的采样频率采集单位时间内钻孔机处于工作状态下动力头在垂直方向上的位置变化，并将此位置变化数据进行解析、存储后形成钻孔机在单位时间内移动距离的时间序列数据集。In step S1, the time series data set of the moving distance of the drilling machine in unit time is collected by an infrared laser ranging sensor arranged on the power head of the drilling machine. The infrared laser ranging sensor collects data according to the set sampling frequency. The position change of the power head in the vertical direction when the drilling machine is in working condition per unit time is analyzed and stored to form a time series data set of the moving distance of the drilling machine within the unit time.
步骤S2所述的对步骤S1获取的单位时间内移动距离的时间序列数据集进行数据处理，得到处理后的时间序列数据集，具体为采用如下步骤进行处理：As described in step S2, perform data processing on the time series data set of moving distance per unit time obtained in step S1 to obtain a processed time series data set. Specifically, the following steps are used for processing:
A.将单位时间内移动距离的时间序列X＝{x1,x2,...,xn}表示为X(t)＝f(w)+e(t)，其中w为时间序列的模式，f(w)为时间序列的模式表示，e(t)为时间序列表示与对应的模式表示之间的误差；从而得到处理后的时间序列为L(x)＝{L(xi1,xi2),L(xi2,xi3),...,L(xi(k-1),xik)}，其中L(xi(k-1),xik)为连接点xi(k-1)和点xik之间的线段；A. Express the time series of moving distance in unit time X = {x1 , x2 ,..., xn } as X(t) = f(w) + e(t) , where w is the mode, f(w) is the mode representation of the time series, and e(t) is the error between the time series representation and the corresponding mode representation; thus the processed time series is obtained as L (x) = {L (xi1 , xi2 ),L(xi2 ,xi3 ),...,L(xi(k-1) ,xik )}, where L(xi(k-1) ,xik ) is the connection point x The line segment betweeni(k-1) and point xik ;
B.对步骤A得到的序列L(x)进行线性拟合，从而得到拟合后的时间序列B. Linearly fit the sequence L(x) obtained in step A to obtain the fitted time series.
C.根据步骤B得到的拟合后的时间序列判定得到有效钻探数据集。C. The fitted time series obtained according to step B It is determined that a valid drilling data set is obtained.
步骤A所述的将单位时间内移动距离的时间序列X＝{x1,x2,...,xn}表示为X(t)＝f(w)+e(t)，具体为采用如下分段模式进行表示：As described in step A, the time series X = {x1 , x2 ,..., xn } of the moving distance per unit time is expressed as X(t) = f(w) + e(t) , specifically using It is represented by the following segmentation mode:
式中wi为时间区间[ti-1,ti]的两个端点坐标，fi(t,wi)为连接wi的两个端点的线性函数；ek(t)为该段时间内时间序列与对应的模式表示之间的误差。In the formula, wi is the coordinates of the two endpoints of the time interval [ti-1 ,ti ], fi(t,wi) is the linear function connecting the two endpoints of wi ; ek(t) is the period of time The error between the internal time series and the corresponding model representation.
步骤C所述的判定得到有效钻探数据集，具体为采用如下步骤进行判定：The determination described in step C obtains a valid drilling data set. Specifically, the following steps are used to determine:
a.针对步骤B得到的拟合后的时间序列中的任意一点/>计算点/>与前一个点/>确定线段的斜率tgk，以及点/>与后一个点/>确定线段的斜率tgk+1；a. For the fitted time series obtained in step B Any point in/> Calculation point/> Same as previous point/> Determine the slope tgk of the line segment, and the point/> with the next point/> Determine the slope of the line segment tgk+1 ;
b.计算斜率tgk与斜率设定值tg的第k差值，以及斜率tgk+1与斜率设定值tg的第k+1差值，并与设定阈值进行比较：b. Calculate the kth difference between the slope tgk and the slope setting value tg, and thek+1 difference between the slope tg k+1 and the slope setting value tg, and compare them with the set threshold:
若第k差值小于设定阈值且第k+1差值小于设定阈值，则线段和线段判定为疑似非钻进线段，并将疑似非钻进线段的数据从拟合后的时间序列中剔除；If the kth difference is less than the set threshold and the k+1th difference is less than the set threshold, then the line segment and line segments It is determined to be a suspected non-drilling line segment, and the data of the suspected non-drilling line segment is extracted from the fitted time series. eliminated;
c.重复步骤a和步骤b，直至所有的拟合后的时间序列中的点均判定完毕，从而得到最终的有效钻探数据集。c. Repeat steps a and b until all the fitted time series are All points in have been determined, thus obtaining the final effective drilling data set.
步骤S4中所述的钻探序列数据集，具体为钻探序列数据集表示为X＝{x1＝(v1,d1),x2＝(v2,d2),...,xk＝(vk,dk)}，其中元素xi＝(vi,di)表示在钻探深度di时的钻进速度为vi。The drilling sequence data set described in step S4 is specifically expressed as X={x1 =(v1 ,d1 ),x2 =(v2 ,d2 ),...,xk =(vk ,dk )}, where the element xi =(vi ,di ) represents the drilling speed at drilling depthdi asvi .
步骤S5所述的构建钻探数据评估网络模型，并利用步骤S4得到的钻探序列数据集对构建的钻探数据评估网络模型进行优化，具体为采用如下步骤进行优化：Construct the drilling data evaluation network model as described in step S5, and use the drilling sequence data set obtained in step S4 to optimize the constructed drilling data evaluation network model. Specifically, the following steps are used for optimization:
(1)采用循环神经网络模型作为钻探数据评估模型；(1) Use the recurrent neural network model as the drilling data evaluation model;
(2)利用步骤(1)得到的循环神经网络模型的隐藏节点提取钻探序列数据集的钻进信息，从而完成钻探数据评估网络模型的优化。(2) Use the hidden nodes of the recurrent neural network model obtained in step (1) to extract the drilling information of the drilling sequence data set, thereby completing the optimization of the drilling data evaluation network model.
所述的利用步骤(1)得到的循环神经网络模型的隐藏节点提取钻探序列数据集的钻进信息，具体为采用如下算式进行提取：The hidden nodes of the recurrent neural network model obtained in step (1) are used to extract the drilling information of the drilling sequence data set. Specifically, the following calculation formula is used to extract:
hi＝f(Uxi+Whi-1+b),i∈{1,2,...,n}hi =f(Uxi +Whi-1 +b),i∈{1,2,...,n}
式中hi为隐藏节点输出的钻进信息；n为钻探序列数据集中的元素个数；i为第i个钻探数据；hi-1为第i-1个钻探数据包含的钻进信息；xi为输入的第i个钻探数据；f()为非线性激活函数；U为连接输入层与隐藏层的权重；W为连接第i个隐藏节点与第i-1个隐藏节点之间的权重，b为偏置。In the formula, hi is the drilling information output by the hidden node; n is the number of elements in the drilling sequence data set; i is the i-th drilling data; hi-1 is the drilling information contained in the i-1 drilling data; xi is the i-th drilling data input; f() is the nonlinear activation function; U is the weight connecting the input layer and the hidden layer; W is the weight connecting the i-th hidden node and the i-1 hidden node Weight, b is the bias.
步骤S6所述的采用步骤S5得到的优化后的钻探数据评估网络模型进行基桩桩身混凝土完整性及桩底沉渣厚度分析，并取特定某段芯样进行相关实验，输入钻探数据评估网络模型后得到完整的基桩桩身混凝土强度报告，具体为采用如下步骤进行分析：As described in step S6, the optimized drilling data evaluation network model obtained in step S5 is used to analyze the concrete integrity of the pile body and the sediment thickness at the bottom of the pile, and a specific core sample is taken to conduct relevant experiments and the drilling data evaluation network model is input. Finally, a complete concrete strength report of the foundation pile body is obtained. Specifically, the following steps are used for analysis:
1)采用如下公式，利用优化后的钻探数据评估网络模型的输出层将钻进信息转化为基桩桩身混凝土强度参考值：1) Using the following formula, the output layer of the optimized drilling data evaluation network model is used to convert the drilling information into the concrete strength reference value of the foundation pile body:
yi＝Softmax(Vhi+c),i∈{1,2,...,n}yi =Softmax(Vhi +c),i∈{1,2,...,n}
式中yi为输出的第i个基桩桩身混凝土强度参考值，hi为第i个隐藏层输出的钻进信息，Softmax()为非线性激活函数，V为权重值，c为偏置；In the formula, yi is the concrete strength reference value of the i-th foundation pile pile body output, hi is the drilling information output by the i-th hidden layer, Softmax() is the nonlinear activation function, V is the weight value, and c is the partial value. set;
2)对所采集的芯样选取特定段进行相关实验后，将此特定段的混凝土强度参数输入1)中钻探数据评估网络模型中，得到模型中其余数据段的混凝土强度值；2) After selecting a specific section of the collected core sample to conduct relevant experiments, input the concrete strength parameters of this specific section into the drilling data evaluation network model in 1) to obtain the concrete strength values of the remaining data sections in the model;
3)利用自然断点法将基桩混凝土强度值划分为8个等级：C60、C50、C45、C40、C35、C30、C20、沉渣；3) Use the natural break point method to divide the foundation pile concrete strength values into 8 levels: C60, C50, C45, C40, C35, C30, C20, and sediment;
4)根据步骤1)得到的基桩桩身混凝土强度值，和步骤2)划分的等级，进行基桩桩身混凝土强度与桩底沉渣厚度分析。4) Based on the concrete strength value of the foundation pile body obtained in step 1) and the grade classified in step 2), conduct an analysis of the concrete strength of the foundation pile body and the sediment thickness at the bottom of the pile.
本发明提供的这种钻探数据分析评估方法，基于循环神经网络RNN能够处理序列数据的特点构建针对钻孔机钻进的序列数据并评估基桩桩身混凝土强度等级及桩底沉渣厚度；因此，本发明方法能够克服现有方法中人为因素较多的主观性问题，本发明方法可靠性高、成本较低且实用性较好。The drilling data analysis and evaluation method provided by the present invention is based on the characteristics of the recurrent neural network RNN that can process sequence data to construct sequence data for drilling by the drilling machine and evaluate the concrete strength grade of the pile body and the sediment thickness at the bottom of the pile; therefore, The method of the present invention can overcome the subjectivity problem of many human factors in the existing methods. The method of the present invention has high reliability, low cost and good practicability.
附图说明Description of the drawings
图1为本发明方法的方法流程示意图。Figure 1 is a schematic flow chart of the method of the present invention.
具体实施方式Detailed ways
如图1所示为本发明方法的方法流程示意图：本发明提供的这种钻探数据分析评估方法，包括如下步骤：Figure 1 is a schematic flow chart of the method of the present invention: the drilling data analysis and evaluation method provided by the present invention includes the following steps:
S1.在钻探过程中，获取钻孔机在单位时间内移动距离的时间序列数据集；S1. During the drilling process, obtain a time series data set of the drilling machine's movement distance per unit time;
钻孔机在单位时间内移动距离的时间序列数据集由布置在钻孔机动力头上的红外激光测距传感器所采集，所述红外激光测距传感器根据设定的采样频率采集单位时间内钻孔机处于工作状态下动力头在垂直方向上的位置变化，并将此位置变化数据进行解析、存储后形成钻孔机在单位时间内移动距离的时间序列数据集；The time series data set of the moving distance of the drilling machine in unit time is collected by an infrared laser ranging sensor arranged on the power head of the drilling machine. The infrared laser ranging sensor collects drilling data in unit time according to the set sampling frequency. The position change of the power head in the vertical direction when the drilling machine is in working condition, and the position change data is analyzed and stored to form a time series data set of the moving distance of the drilling machine in unit time;
S2.对步骤S1获取的单位时间内移动距离的时间序列数据集进行数据处理，得到处理后的时间序列数据集；具体为采用如下步骤进行处理：S2. Perform data processing on the time series data set of moving distance per unit time obtained in step S1 to obtain a processed time series data set; specifically, the following steps are used for processing:
A.将单位时间内移动距离的时间序列X＝{x1,x2,...,xn}表示为X(t)＝f(w)+e(t)，其中w为时间序列的模式，f(w)为时间序列的模式表示，e(t)为时间序列表示与对应的模式表示之间的误差；从而得到处理后的时间序列为L(x)＝{L(xi1,xi2),L(xi2,xi3),...,L(xi(k-1),xik)}，其中L(xi(k-1),xik)为连接点xi(k-1)和点xik之间的线段；具体为采用如下分段模式进行表示：A. Express the time series of moving distance in unit time X = {x1 , x2 ,..., xn } as X(t) = f(w) + e(t) , where w is the mode, f(w) is the mode representation of the time series, and e(t) is the error between the time series representation and the corresponding mode representation; thus the processed time series is obtained as L (x) = {L (xi1 , xi2 ),L(xi2 ,xi3 ),...,L(xi(k-1) ,xik )}, where L(xi(k-1) ,xik ) is the connection point x The line segment betweeni(k-1) and point xik ; specifically, it is represented by the following segmentation mode:
式中wi为时间区间[ti-1,ti]的两个端点坐标，为连接wi的两个端点的线性函数；ek(t)为该段时间内时间序列与对应的模式表示之间的误差；In the formula, wi is the coordinates of the two endpoints of the time interval [ti-1 ,ti ], is the linear function connecting the two endpoints of wi ; ek(t) is the error between the time series and the corresponding pattern representation within this period of time;
B.对步骤A得到的序列L(x)进行线性拟合，从而得到拟合后的时间序列B. Linearly fit the sequence L(x) obtained in step A to obtain the fitted time series.
C.根据步骤B得到的拟合后的时间序列判定得到有效钻探数据集；具体为采用如下步骤进行判定：C. The fitted time series obtained according to step B It is determined that a valid drilling data set is obtained; specifically, the following steps are used to determine:
a.针对步骤B得到的拟合后的时间序列中的任意一点/>计算点/>与前一个点/>确定线段的斜率tgk，以及点/>与后一个点/>确定线段的斜率tgk+1；a. For the fitted time series obtained in step B Any point in/> Calculation point/> Same as previous point/> Determine the slope tgk of the line segment, and the point/> with the next point/> Determine the slope of the line segment tgk+1 ;
b.计算斜率tgk与斜率设定值tg的第k差值，以及斜率tgk+1与斜率设定值tg的第k+1差值，并与设定阈值进行比较：b. Calculate the kth difference between the slope tgk and the slope setting value tg, and thek+1 difference between the slope tg k+1 and the slope setting value tg, and compare them with the set threshold:
若第k差值小于设定阈值且第k+1差值小于设定阈值，则线段和线段判定为疑似非钻进线段，并将疑似非钻进线段的数据从拟合后的时间序列中剔除；If the kth difference is less than the set threshold and the k+1th difference is less than the set threshold, then the line segment and line segments It is determined to be a suspected non-drilling line segment, and the data of the suspected non-drilling line segment is extracted from the fitted time series. eliminated;
c.重复步骤a和步骤b，直至所有的拟合后的时间序列中的点均判定完毕，从而得到最终的有效钻探数据集；c. Repeat steps a and b until all the fitted time series are All points in have been determined, thus obtaining the final effective drilling data set;
S3.根据步骤S2获取的处理后的时间序列数据集，确定钻进速度，以及处理后的时间序列数据集中每个数据所对应的钻探深度；S3. Based on the processed time series data set obtained in step S2, determine the drilling speed and the drilling depth corresponding to each data in the processed time series data set;
S4.根据步骤S3获取的钻进速度和处理后的时间序列数据集中每个数据所对应的钻探深度，构建钻探序列数据集；具体为钻探序列数据集表示为X＝{x1＝(v1,d1),x2＝(v2,d2),...,xk＝(vk,dk)}，其中元素xi＝(vi,di)表示在钻探深度di时的钻进速度为vi；S4. Construct a drilling sequence data set based on the drilling speed obtained in step S3 and the drilling depth corresponding to each data in the processed time series data set; specifically, the drilling sequence data set is expressed as X = {x1 = (v1 ,d1 ),x2 =(v2 ,d2 ),...,xk =(vk ,dk )}, where the element xi =(vi ,di ) represents the drilling depth di The drilling speed at is vi ;
S5.构建钻探数据评估网络模型，并利用步骤S4得到的钻探序列数据集对构建的钻探数据评估网络模型进行优化；具体为采用如下步骤进行优化：S5. Construct a drilling data evaluation network model, and use the drilling sequence data set obtained in step S4 to optimize the constructed drilling data evaluation network model; specifically, the following steps are used for optimization:
(1)采用循环神经网络模型作为钻探数据评估模型；(1) Use the recurrent neural network model as the drilling data evaluation model;
(2)利用步骤(1)得到的循环神经网络模型的隐藏节点提取钻探序列数据集的钻进信息，从而完成钻探数据评估网络模型的优化。(2) Use the hidden nodes of the recurrent neural network model obtained in step (1) to extract the drilling information of the drilling sequence data set, thereby completing the optimization of the drilling data evaluation network model.
所述的利用步骤(1)得到的循环神经网络模型的隐藏节点提取钻探序列数据集的钻进信息，具体为采用如下算式进行提取：The hidden nodes of the recurrent neural network model obtained in step (1) are used to extract the drilling information of the drilling sequence data set. Specifically, the following calculation formula is used to extract:
hi＝f(Uxi+Whi-1+b),i∈{1,2,...,n}hi =f(Uxi +Whi-1 +b),i∈{1,2,...,n}
式中hi为隐藏节点输出的钻进信息；n为钻探序列数据集中的元素个数；i为第i个钻探数据；hi-1为第i-1个钻探数据包含的钻进信息；xi为输入的第i个钻探数据；f()为非线性激活函数；U为连接输入层与隐藏层的权重；W为连接第i个隐藏节点与第i-1个隐藏节点之间的权重，b为偏置；In the formula, hi is the drilling information output by the hidden node; n is the number of elements in the drilling sequence data set; i is the i-th drilling data; hi-1 is the drilling information contained in the i-1 drilling data; xi is the i-th drilling data input; f() is the nonlinear activation function; U is the weight connecting the input layer and the hidden layer; W is the weight connecting the i-th hidden node and the i-1 hidden node Weight, b is the bias;
S6.采用步骤S5得到的优化后的钻探数据评估网络模型进行进行基桩桩身混凝土完整性及桩底沉渣厚度分析，并取特定某段芯样进行相关实验，输入钻探数据评估网络模型后得到完整的基桩桩身混凝土强度报告，具体为采用如下步骤进行分析：S6. Use the optimized drilling data evaluation network model obtained in step S5 to analyze the concrete integrity of the foundation pile pile body and the sediment thickness at the bottom of the pile, and take a specific core sample to conduct relevant experiments. After inputting the drilling data evaluation network model, we get A complete concrete strength report of the foundation pile body is analyzed using the following steps:
1)采用如下公式，利用优化后的钻探数据评估网络模型的输出层将钻进信息转化为基桩桩身混凝土强度参考值：1) Using the following formula, the output layer of the optimized drilling data evaluation network model is used to convert the drilling information into the concrete strength reference value of the foundation pile body:
yi＝Softmax(Vhi+c),i∈{1,2,...,n}yi =Softmax(Vhi +c),i∈{1,2,...,n}
式中yi为输出的第i个基桩桩身混凝土强度参考值，hi为第i个隐藏层输出的钻进信息，Softmax()为非线性激活函数，V为权重值，c为偏置；In the formula, yi is the concrete strength reference value of the i-th foundation pile pile body output, hi is the drilling information output by the i-th hidden layer, Softmax() is the nonlinear activation function, V is the weight value, and c is the partial value. set;
2)对所采集的芯样选取特定段进行相关实验后，将此特定段的混凝土强度参数输入1)中钻探数据评估网络模型中，得到模型中其余数据段的混凝土强度值；2) After selecting a specific section of the collected core sample to conduct relevant experiments, input the concrete strength parameters of this specific section into the drilling data evaluation network model in 1) to obtain the concrete strength values of the remaining data sections in the model;
3)利用自然断点法将基桩混凝土强度值划分为8个等级：C60、C50、C45、C40、C35、C30、C20、沉渣；3) Use the natural break point method to divide the foundation pile concrete strength values into 8 levels: C60, C50, C45, C40, C35, C30, C20, and sediment;
3)根据步骤1)得到的基桩桩身混凝土强度值，和步骤2)划分的等级，进行基桩桩身混凝土强度与桩底沉渣厚度分析；3) Based on the concrete strength value of the foundation pile body obtained in step 1) and the grade classified in step 2), conduct an analysis of the concrete strength of the foundation pile body and the sediment thickness at the bottom of the pile;
S7.根据步骤S6的分析结果，进行最终的钻探数据分析评估。S7. Carry out final drilling data analysis and evaluation based on the analysis results of step S6.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011072945.6ACN112199889B (en) | 2020-10-09 | 2020-10-09 | Drilling data analysis and evaluation method |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011072945.6ACN112199889B (en) | 2020-10-09 | 2020-10-09 | Drilling data analysis and evaluation method |
| Publication Number | Publication Date |
|---|---|
| CN112199889A CN112199889A (en) | 2021-01-08 |
| CN112199889Btrue CN112199889B (en) | 2024-02-02 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011072945.6AExpired - Fee RelatedCN112199889B (en) | 2020-10-09 | 2020-10-09 | Drilling data analysis and evaluation method |
| Country | Link |
|---|---|
| CN (1) | CN112199889B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116305409A (en)* | 2023-02-07 | 2023-06-23 | 中国山东国际经济技术合作有限公司 | Time sequence prediction system and method for pile foundation construction state of rotary drilling rig in bridge construction |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102606151A (en)* | 2012-04-01 | 2012-07-25 | 中国石油大学(北京) | Method and device for predicting rock drillability of wildcat well before drilling |
| CN110952456A (en)* | 2019-12-06 | 2020-04-03 | 中铁十八局集团建筑安装工程有限公司 | Pile type bridge construction process |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070185696A1 (en)* | 2006-02-06 | 2007-08-09 | Smith International, Inc. | Method of real-time drilling simulation |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102606151A (en)* | 2012-04-01 | 2012-07-25 | 中国石油大学(北京) | Method and device for predicting rock drillability of wildcat well before drilling |
| CN110952456A (en)* | 2019-12-06 | 2020-04-03 | 中铁十八局集团建筑安装工程有限公司 | Pile type bridge construction process |
| Title |
|---|
| 基于人工神经网络的不同混凝土实时强度预测研究;方涛;王俊锋;张鑫月;金光日;;混凝土(第12期);全文* |
| 钻芯法在桩基检测中的应用;王少众;;科技促进发展(第11期);全文* |
| Publication number | Publication date |
|---|---|
| CN112199889A (en) | 2021-01-08 |
| Publication | Publication Date | Title |
|---|---|---|
| CN111291934A (en) | A real-time classification prediction and self-inspection method for surrounding rock during tunnel construction | |
| CN104880544A (en) | Method for detecting and evaluating reinforcing effect on weak surrounding rock grouting during underground construction | |
| CN118392662B (en) | Rock mass compressive strength determination method and system based on rock core rebound and while-drilling parameters | |
| CN115267905B (en) | A method for predicting fractured lost circulation in drilling engineering in complex structural areas | |
| CN119026403B (en) | Pile foundation load influence assessment method, medium and system by pile sinking soil compaction effect | |
| CN117350145A (en) | Rock mechanical parameter while-drilling prediction method based on vibration signals | |
| CN114139381B (en) | Pile foundation damage census and evaluation method considering pile soil parameter uncertainty | |
| CN112199889B (en) | Drilling data analysis and evaluation method | |
| CN114091339A (en) | Method for predicting bow blowing instantaneous yield of drag suction dredger based on GRU | |
| CN119885395B (en) | A rock and soil stress monitoring data analysis method and system | |
| CN119475543B (en) | Rapid evaluation method and system for bearing capacity of grouting at the bottom of prefabricated road slabs | |
| CN118774969B (en) | Tunnel excavation support management method and system based on advanced geological prediction information | |
| Cao et al. | Investigation of interfacial debonding identification for concrete filled steel tube columns based on acoustic signals | |
| CN119985711A (en) | Method for selecting the optimal prediction model of shear wave velocity based on field demand evaluation | |
| CN111706322B (en) | Rock drilling response prediction method and prediction system | |
| CN110750863B (en) | Foundation pile wave impedance inversion analysis method based on cubic B spline interpolation function | |
| CN117451448A (en) | A method for prediction of rock mechanical parameters while drilling based on concrete specimens | |
| CN115573396A (en) | A Method for Evaluation of Pile Foundation Low Strain Waveform Data | |
| CN116205515A (en) | Construction evaluation method for driven precast pile based on static sounding and BP neural network | |
| Alavi | Developing an Artificial Intelligence Model for Estimating the Compressive Strength of Concrete Based on Combined Non-Destructive Test Results | |
| Yousefpour et al. | Determination of unknown foundation of bridges for scour evaluation using artificial neural networks | |
| Zhang et al. | Application of artificial neural network for diagnosing pile integrity based on low strain dynamic testing | |
| Omar | Condition Assessment of Concrete Bridge Decks Using Ground and Airborne Infrared Thermography | |
| CN116383697B (en) | Bituminous pavement compaction degree prediction method based on double-gate circulation network | |
| CN119335045B (en) | Concrete bridge pile foundation risk detection method and system |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20240202 |