CN-121479165-B - Intelligent analysis method for steel support shaft force sensing data of assembled station

CN121479165BCN 121479165 BCN121479165 BCN 121479165BCN-121479165-B

Abstract

The invention relates to the technical field of engineering intelligent monitoring, in particular to an intelligent analysis method for steel support shaft force sensing data of an assembled station. The method comprises the steps of collecting a shaft force time sequence data set, an installation position information set and a construction time record set of each steel support in the assembled station, calculating a base shaft force reference value of each steel support based on the shaft force time sequence data set, calculating a space attenuation coefficient of each steel support based on the installation position information set and the shaft force time sequence data set and forming space attenuation coefficient data, so that the problem that base line setting relies on experience, space difference is difficult to quantify and time effect processing is discontinuous in the traditional method is solved, and scientificity, reliability and operability of shaft force monitoring of the steel support are improved.

Inventors

BAI YUNBO
GENG LIANG
WANG YUYU
GONG LONG
WU WEIXIANG
WANG JUN
WANG YONGMING
TANG QI
SONG FEI
LI SHUAI
LI ZHONGCHEN

Assignees

中铁上海工程局集团有限公司
深圳地铁建设集团有限公司
中铁南方投资集团有限公司

Dates

Publication Date: 20260508
Application Date: 20251014

Claims (8)

1. The intelligent analysis method for the steel support shaft force sensing data of the assembled station is characterized by comprising the following steps of: step S1, acquiring a shaft force time sequence data set, an installation position information set and a construction time record set of each steel support in an assembled station; s2, calculating a basic axis force reference value of each steel support based on the axis force time sequence data set; Step S3, calculating the spatial attenuation coefficient of each steel support based on the installation position information set and the axial force time sequence data set and forming spatial attenuation coefficient data, wherein the step S3 comprises the following steps: S31, calculating Euclidean distance of adjacent supports of each steel support by using an installation position information set, and selecting an adjacent support set in a neighborhood radius; s32, calculating a spatial variation coefficient for a baseline axial force reference value and a current axial force difference value of adjacent support sets; S33, calculating and normalizing an axial force attenuation factor based on the difference value of the distance and the axial force, wherein the obtained axial force attenuation factor is used as a spatial attenuation coefficient of the support and is recorded into spatial attenuation coefficient data; And step S4, calculating the time response coefficient of each steel support based on the construction time record set and the axial force time sequence data set and forming a time response coefficient set, wherein the step S4 comprises the following steps: Acquiring a construction time record set and an axial force time sequence data set of each steel support, and calculating the number of days between the current observation time and the installation completion time; setting the time response coefficient to be an inhibition coefficient in the range of 0.1 to 0.2 when the number of days of interval is less than 3 days; calculating a time response coefficient according to a delay Gaussian morphology function when the number of days of interval is between 4 days and 14 days, wherein the delay center is 10 days, and the width is 5 days; setting the time response coefficient to be 0.8-0.95 interval when the interval days are 15-30 days; And S5, calculating a basic axis force reference value, spatial attenuation coefficient data, a time response coefficient set and a current observed axis force value according to a preset combination rule to obtain an axis force estimated initial value of each steel support and generate an axis force estimated value.
2. The intelligent analysis method for steel support shaft force sensing data for an assembled station according to claim 1, wherein step S1 comprises: Controlling a distributed strain type axial force sensor network to collect axial force time sequence data of each steel support under static load and working condition load according to a uniform time base line, wherein an axial force time sequence data set comprises an isochronous sampling sequence of not less than ten continuous minutes; And synchronously recording the installation coordinates of each steel support to form an installation position information set, and recording the first stress time and the installation completion time of each steel support as a construction time record set.
3. The intelligent analysis method of steel support shaft force sensing data for an assembled station according to claim 1, wherein step S2 comprises: s21, extracting a sample window in a pre-installation or initial idle stage from the axial force time sequence data set and calculating a median value; S22, taking the median value as an initial baseline estimation, and calculating a stability coefficient in a window as a baseline stability criterion; And S23, when the baseline stability criterion is smaller than a preset threshold, determining the median value as a baseline axial force reference value of the support, and when the baseline stability criterion is larger than the preset threshold, adopting an expanded sample window to recalculate the median value until the stability criterion is met, wherein the obtained median value is used as the baseline axial force reference value.
4. The intelligent analysis method of steel support shaft force sensing data for an assembled station according to claim 1, wherein the determination of spatial coefficient of variation in step S32 comprises: And calculating the ratio of the standard deviation to the average value of the baseline axial force reference values within the range of three meters of adjacent support sets with the neighborhood radius as a spatial variation coefficient, and when the spatial variation coefficient is between 0.20 and 0.70 and the baseline axial force is in radial distribution with high center and low edge, confirming that the neighborhood affects the center and bringing the support in the neighborhood into the spatial attenuation coefficient calculation range.
5. The intelligent analysis method for steel support shaft force sensing data for an assembled station according to claim 1, wherein in step S4, the basic shaft force reference value, the spatial attenuation coefficient data, the time response coefficient set and the current observed shaft force value are calculated according to a predetermined combination rule specifically as follows: dividing the baseline deviation for each support by 1000N to obtain a normalized first input term; directly taking the spatial attenuation coefficient data as a second input item; Taking the product of the first input item and the second input item as a third input item; multiplying the first input item by 0.50 according to the pre-calibration weight to obtain a structural contribution item, multiplying the second input item by 0.30 to obtain a position contribution item, and multiplying the third input item by 0.20 to obtain a cooperative contribution item; The method comprises the steps of obtaining a comprehensive response index by adding a structure contribution term, a position contribution term and a cooperative contribution term, multiplying the comprehensive response index by space attenuation coefficient data and multiplying the comprehensive response index by a time response coefficient to obtain an axial force increment after space-time modulation, and obtaining an axial force estimated initial value by adding the axial force increment and a basic axial force reference value.
6. The intelligent analysis method for steel support shaft force sensing data for an assembled station according to claim 5, wherein adding the shaft force increment to the baseline shaft force reference value to obtain the initial shaft force estimation value further comprises: Extracting the initial value of the axial force estimation, the difference value between the initial value and the reference value of the axial force of the basic line, the spatial attenuation coefficient value and the time response coefficient value of each steel support item by item, and sequentially arranging the values in the same data record to obtain four corresponding joint characteristic data; Selecting a representative supporting point in the assembled station to acquire the actual measurement axial force to form a calibration data set, and pairing the calibration data set with four corresponding combined characteristic data to form a training sample pair; And establishing a vertical shaft force inversion relation by using training sample pairs, taking the actually measured shaft force as a target value, taking four combined characteristic data as input variables of the shaft force inversion relation, and outputting inversion shaft force values of all supports.
7. The intelligent analysis method for steel support axial force sensing data for an assembled station according to claim 6, wherein using four pieces of joint characteristic data as input variables of an axial force inversion relationship comprises: taking a training sample pair as input, solving a weight coefficient and a constant term in an axial force inversion relation by adopting a minimization method, expressing the axial force inversion relation in a linear combination form, and obtaining a final inversion weight coefficient; and applying the axial force inversion relation to four joint characteristic data of all steel supports in the assembled station, and outputting inversion axial force values of all the supports.
8. The intelligent analysis method for steel support shaft force sensing data for an assembled station according to claim 7, wherein outputting the inverted shaft force value of each support further comprises: calculating point-by-point difference values of inversion axial force values and basic axial force reference values of all supports in the assembled station to obtain an axial force abnormality index; support data with the axial force abnormality index greater than a preset abnormality index threshold value to generate an abnormality alarm list; and according to the abnormal alarm list, sequencing and outputting a priority treatment list according to the spatial attenuation coefficient data, and generating a steel support axial force sensing data analysis visual chart.

Description

Intelligent analysis method for steel support shaft force sensing data of assembled station Technical Field The invention relates to the technical field of engineering intelligent monitoring, in particular to an intelligent analysis method for steel support shaft force sensing data of an assembled station. Background The prior art still has more defects in the aspects of monitoring and analyzing the force of the steel support shaft. On the one hand, the existing sensing monitoring is dependent on single-point or small quantity of distributed sensors to collect data, and a unified time reference and space cooperative mechanism is lacked, so that the data among different sensing points are difficult to directly compare, and the time sequence integrity is insufficient. On the other hand, the axial force baseline is usually set by using manual experience or a single static load value as a reference, and the fluctuation in the initial stage cannot be judged by a statistical method, so that baseline deviation accumulation is easy to occur, and the subsequent judgment is influenced. In the space dimension, the mutual influence between supports is not considered enough by the traditional method, and the stress difference of adjacent components often fails to be quantized into an attenuation relation, so that the propagation of local anomalies is difficult to accurately describe. In the time dimension, the stress evolution rules of different stages after construction is completed are complex, but most of the existing methods are approximated by adopting fixed reduction coefficients or empirical formulas, and lack of continuous description of a time response process leads to indistinct short-term and medium-long-term effects. In addition, in terms of data fusion and anomaly judgment, the baseline value and the measured value are often directly compared in the prior art, and when deviation occurs, whether the baseline value and the measured value are caused by a position factor, a time factor or a synergistic factor cannot be clearly distinguished, false alarm or missing alarm is easily caused, and a comprehensive judgment mechanism of multi-dimensional indexes is lacked. Finally, in the visualization and alarm links, most of the existing systems only output a single numerical value or a simple curve, lack a traceable data link associated with spatial position and time evolution, and are not beneficial to priority ordering and quick decision of exception handling. Disclosure of Invention Based on this, there is a need to provide an intelligent analysis method for steel support shaft force sensing data for an assembled station, so as to solve at least one of the above technical problems. In order to achieve the above purpose, the intelligent analysis method for the steel support shaft force sensing data of the fabricated station comprises the following steps: step S1, acquiring a shaft force time sequence data set, an installation position information set and a construction time record set of each steel support in an assembled station; s2, calculating a basic axis force reference value of each steel support based on the axis force time sequence data set; s3, calculating a spatial attenuation coefficient of each steel support based on the installation position information set and the axial force time sequence data set and forming spatial attenuation coefficient data; S4, calculating a time response coefficient of each steel support based on the construction time record set and the axial force time sequence data set and forming a time response coefficient set; And S5, calculating a basic axis force reference value, spatial attenuation coefficient data, a time response coefficient set and a current observed axis force value according to a preset combination rule to obtain an axis force estimated initial value of each steel support and generate an axis force estimated value. The method has the beneficial effects that the distributed sensor network is utilized to acquire the axial force time sequence data of the steel supports, and the installation position information and the construction time record are synchronized, so that each support has a definite space-time label, and the comparability and the integrity of the data are ensured. The baseline linear force reference value is obtained by carrying out windowing processing and statistical judgment on the time sequence data, and the stability coefficient is used as a quality criterion, so that the limitation that the baseline depends on a single measuring point is avoided. And meanwhile, a time response coefficient set is constructed according to the construction completion time and the interval days of the observation time, and a time effect is characterized in a mode of combining a piecewise function and a continuous function. Further, the baseline deviation, the spatial attenuation coefficient and the time response coefficient are fused according to a preset rule, the c