Intelligent evaluation system for tensioning quality of large-tonnage pre-stressed anchor cable

Abstract

The invention relates to the technical field of engineering monitoring, in particular to a large-tonnage pre-stressed anchor cable tensioning quality intelligent evaluation system which comprises a time sequence construction module, a track recognition module, a stability recognition module, a coordination analysis module and a quality evaluation module. According to the method, the trend direction and rate conjoint analysis is carried out on multiple types of response parameters in the anchor cable tensioning process, the response synchronism characteristics in the tensioning period and in the cross-period can be accurately recognized, the anchor cable response track with a time sequence structure is constructed on the basis, accurate recognition of the lag phenomenon and the trend closing relation is achieved, and the accuracy of the anchor cable tensioning process is improved. The method further extracts the dynamic correlation characteristics between the structural disturbance and the tensioning segments, enhances the substantive identification capability of the structural stability risk by means of the time domain coincidence analysis of the disturbance duration of the unstable segments and the falling time consumption, and improves the accuracy of the structural stability risk by normalizing the response rates of the segments with the time overlapping relationship. An index system reflecting the overall coordination level can be established in an anchor cable group.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering monitoring, in particular to a large-tonnage prestressed anchor cable tensioning quality intelligent evaluation system. BACKGROUND

[0002] The technical field of engineering monitoring relates to collecting, analyzing and evaluating various physical quantities of engineering structures during construction and operation to realize monitoring and early warning of engineering safety and stability.

[0003] Among them, the large-tonnage prestressed anchor cable tensioning quality evaluation refers to the quality judgment and record of the tension force value, tension length, anchor reverse force and rebound condition of the large-tonnage prestressed anchor cable used to stabilize the surrounding rock and lining structure in the construction tensioning process in the tunnel supporting structure.

[0004] The prior art only relies on recording and judging single technical indicators such as tension force value, tension length, anchor reverse force and rebound condition, and fails to establish a dynamic evolution relationship between response parameters, lacks a continuity analysis mechanism for response trends between multiple cycles, and easily ignores the hysteresis and coordination characteristics of tension response in the process of simultaneous construction of multiple anchor cables or long-period tensioning, especially under complex geological conditions. The structural disturbance caused by tensioning cannot be effectively corresponded to the specific section, resulting in the inability to clearly identify abnormal sections at the stability risk boundary, for example, in sections with frequent surrounding rock disturbance or densely arranged anchor cables. It is difficult to judge the coordination level of the anchor cable group only by the single-point tension force change value, thereby reducing the accurate judgment ability of the overall structure stability and tensioning quality, and easily causing uneven stress distribution of the structure and missing potential instability hazards. SUMMARY

[0005] The purpose of the present application is to solve the shortcomings in the prior art, and a large-tonnage prestressed anchor cable tensioning quality intelligent evaluation system is proposed.

[0006] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: the large-tonnage prestressed anchor cable tensioning quality intelligent evaluation system comprises: A time sequence construction module acquires response parameters in the anchor cable grading tensioning process, and constructs time sequence characteristic data of anchor cable change state according to the trend consistency and cross-cycle change rate of the response parameters in the same tensioning cycle. A trajectory recognition module refers to the time sequence characteristic data of the anchor cable change state, recognizes the hysteresis and multi-cycle response coordination closed chain of the anchor cable response parameters in the change trajectory, and obtains an anchor cable response evolution trajectory. A stability recognition module extracts tensioning section information in the anchor cable response evolution trajectory, recognizes the structure stability boundary under the influence of anchor cable tensioning, and marks unstable anchor cable tensioning sections with structure stability risks. The coordination analysis module extracts anchor cable segment corresponding information of the unstable anchor cable tensioning section in the anchor cable response evolution track, analyzes response coordination characteristics of the anchor cable group in the tensioning process, and constructs an anchor cable coordination evaluation parameter set; The quality evaluation module judges anchor cable overall tensioning quality according to the anchor cable coordination evaluation parameter set, summarizes all section judgment information, and outputs an anchor cable tensioning quality evaluation result.

[0007] As a further scheme of the present application, the timing feature data includes a periodic trend direction label, a periodic rate change value, and a trend rate coupling strength index; the anchor cable response evolution track includes a lag time interval, a trend closure structure type, and a multi-period coordinated track path; the unstable anchor cable tensioning section includes a stable boundary identification result, an abnormal section time window position, and a structure disturbance index overlap situation; the anchor cable coordination evaluation parameter set specifically includes a stress rate normalized value, a displacement rate normalized value, and a parameter synchronization deviation amount; and the anchor cable tensioning quality evaluation result includes abnormal section quantity statistics, abnormal section change type classification, and coordinated instability feature identification information.

[0008] As a further scheme of the present application, the timing construction module includes: The response parameter extraction submodule obtains three response parameters, including a displacement increment change value monitored by a fiber displacement meter, an average amplitude of a reverse oscillation of a peak tension stress monitored by a self-induction stress meter, and a cumulative response lag time from the initial to the final stage, and constructs an anchor cable staged tensioning response parameter set; The trend consistency identification submodule calls the anchor cable staged tensioning response parameter set, judges whether the trend direction of the response parameter in the same tensioning period remains consistent, analyzes the change rate of the response parameter between continuous periods, screens a section with consistent fluctuation direction and stable rate change, and obtains a trend and rate corresponding section between response parameters; The timing feature generation submodule extracts response parameter trend direction data and change rate data in each period section as input variables according to the trend and rate corresponding section between response parameters, calculates a non-linear correlation degree between the trend and the rate through a maximum normalized mutual information algorithm, and arranges to form timing feature data of an anchor cable change state in a tensioning time sequence.

[0009] As a further scheme of the present application, the track identification module includes: The lag track extraction submodule refers to the timing feature data of the anchor cable change state, extracts a time position and a change trend of the response parameter in each tensioning period, identifies a period paragraph with response delay, and obtains a lag track interval of the response parameter; The trend comparison submodule calls the lagging track interval of the response parameter, compares the response parameter change direction in each lagging period, screens the period sequence with consistent trend direction, and obtains the response parameter consistent period; The closed chain construction submodule determines the continuous response track closed feature according to the response parameter consistent period, and obtains the anchor cable response evolution track.

[0010] As a further scheme of the present application, the stability recognition module comprises: The tension segment extraction submodule extracts the tension segment time position in the anchor cable response evolution track, sorts the time window corresponding information of each tension segment, and obtains the tension segment time window. The stable boundary recognition submodule detects the disturbance occurrence time interval, disturbance peak duration and disturbance stable falling time recorded by the surrounding rock monitoring point in each tension segment time window, judges the coincidence length of the three disturbance data, and obtains the structure stable boundary response judgment result. The unstable segment marking submodule identifies the tension segment whose disturbance delay coincidence length exceeds the falling time according to the structure stable boundary response judgment result, marks that the target tension segment is in the structure stability risk state, and obtains the unstable anchor cable tension segment.

[0011] As a further scheme of the present application, the coordination analysis module comprises: The segment information extraction submodule extracts the anchor cable tension segment corresponding time window position of the unstable anchor cable tension segment in the anchor cable response evolution track, and screens the anchor cable tension segment corresponding time information. The response correlation analysis submodule calls the anchor cable tension segment corresponding time information, compares the time window of the unstable anchor cable tension segment with the response peak time in the anchor cable response evolution track, judges whether there is overlap under the same tension order, and extracts the anchor cable tension segment with overlapping relationship. The parameter set construction submodule obtains the stress change rate and displacement change rate in the response parameter in the tension order corresponding to the anchor cable tension segment with overlapping relationship, and sums up the coordination index reflecting the overall change trend to obtain the anchor cable coordination evaluation parameter set.

[0012] As a further scheme of the present application, the quality evaluation module comprises: The coordination parameter analysis submodule identifies the trend direction of the stress change rate and displacement change rate in the anchor cable coordination evaluation parameter set, judges whether there is obvious reverse change, and obtains the anchor cable coordination change trend information. Segment state judgment submodule, filtering the anchor cable coordination change trend information exists the number of times of reverse cumulative over limit anchor cable tension segment, judging that the target tension segment exists tension response instability characteristics, obtaining the anchor cable tension segment with abnormal change; The evaluation result generation submodule aggregates the anchor cable tension segments with abnormal changes, combines the numerical change of the corresponding tension segment corresponding to the anchor cable coordination evaluation parameter set, and constructs the anchor cable tension quality evaluation result.

[0013] Compared with the prior art, the advantages and positive effects of the present application are that: In the present application, by jointly analyzing the trend direction and rate of multiple response parameters in the anchor cable tension process, the response synchronization characteristics within the tension period and across the period can be accurately identified, and a time sequence structure of the anchor cable response trajectory is constructed, the lag phenomenon and the trend closed relationship are accurately identified, the dynamic correlation characteristics between the structure disturbance and the tension segment are further extracted, the time domain coincidence analysis of the disturbance duration and the falling time of the unstable segment is used to enhance the substantial identification ability of the structure stability risk, and the response rate of the segments with time overlap relationship is normalized to establish an index system reflecting the overall coordination level in the anchor cable group, thereby improving the collaborative identification efficiency of local instability and response abnormalities in the anchor cable tension process. Ultimately, on the basis of comprehensive analysis of the abnormal segment evolution characteristics and the coordination fluctuation trend, quantitative evaluation and abnormal classification statistics of the overall tension quality are completed, and the tension quality judgment deviation caused by local disturbance not being timely identified or multi-period collaborative instability not being found is effectively avoided. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The system flowchart of the present application; Figure 2 The flowchart of the time sequence construction module of the present application; Figure 3 The flowchart of the trajectory identification module of the present application; Figure 4 The flowchart of the stability identification module of the present application; Figure 5 The flowchart of the coordination analysis module of the present application; Figure 6 The flowchart of the quality evaluation module of the present application. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0016] Please refer to Figure 1 The intelligent evaluation system for large-tonnage prestressed anchor cable tensioning quality comprises: A time sequence construction module acquires response parameters in the anchor cable staged tensioning process, and constructs time sequence characteristic data of the anchor cable change state according to the trend consistency and cross-cycle change rate between the response parameters in the same tensioning cycle. A trajectory recognition module refers to the time sequence characteristic data of the anchor cable change state, recognizes the lagging situation and multi-cycle response coordinated closed chain of the anchor cable response parameters in the change trajectory, and obtains an anchor cable response evolution trajectory. A stability recognition module extracts tensioning section information in the anchor cable response evolution trajectory, recognizes the structural stability boundary under the influence of anchor cable tensioning, and marks unstable anchor cable tensioning sections with structural stability risks. A coordination analysis module extracts anchor section corresponding information of the unstable anchor cable tensioning section in the anchor cable response evolution trajectory, analyzes the response coordination characteristic of the anchor cable group in the tensioning process, and constructs an anchor cable coordination evaluation parameter set. A quality evaluation module judges the overall tensioning quality of the anchor cable according to the anchor cable coordination evaluation parameter set, summarizes all section judgment information, and outputs an anchor cable tensioning quality evaluation result. The time sequence characteristic data comprises a cycle trend direction label, a cycle rate change value, and a trend rate coupling strength index. The anchor cable response evolution trajectory comprises a lagging time interval, a trend closed structure type, and a multi-cycle coordinated trajectory path. The unstable anchor cable tensioning section comprises a stable boundary recognition result, an abnormal section time window position, and a structure disturbance index overlap situation. The anchor cable coordination evaluation parameter set specifically comprises a stress rate normalized value, a displacement rate normalized value, and a parameter synchronization deviation amount. The anchor cable tensioning quality evaluation result comprises abnormal section quantity statistics, abnormal section change type classification, and coordinated instability characteristic recognition information.

[0017] Please refer to Figure 2 The time sequence construction module comprises: A response parameter extraction submodule acquires three response parameters, namely, a displacement increment change value monitored by a fiber displacement meter in the anchor cable staged tensioning process, an average amplitude of reverse oscillation after peak tensioning stress monitored by a self-induction stress meter, and a cumulative response lag time from the initial to the final stage, and constructs an anchor cable staged tensioning response parameter set. The displacement increment change value monitored by the fiber displacement meter in the anchor cable grading tensioning process is obtained. In practical application, the fiber displacement meter is first arranged in the displacement sensitive area of the anchor cable, and the displacement readings of the corresponding time nodes before and after each tensioning stage are recorded. The displacement increment is calculated, which is the difference between the last tensioning reading and the last tensioning end reading. A periodic displacement change sequence is formed by successive difference calculation. The average amplitude of the reverse oscillation after the peak tensioning stress monitored by the self-induced stress meter is obtained. A tensioning peak value identification threshold is set in the stress meter. For example, the local maximum value of the continuous three readings is taken as the peak stress reference point. Then the stress fluctuation readings within a fixed time period after the peak point are intercepted, and the average value is taken as the average amplitude of the reverse oscillation. The evaluation reference is formed by analyzing the change trend of the average value. The cumulative response lag time from the initial to the final period is calculated depending on the time difference between the start time of each tensioning load and the monitoring response time. The total response lag time is obtained by accumulating the time difference in all tensioning periods. In the calculation, the load application instruction time in the tensioning operation record is compared with the first response time of the sensor. For example, if the tensioning instruction sending time is 10:00:00 and the first effective displacement change detection of the sensor is 10:00:02, the lag time of this time is 2 seconds. The whole period is accumulated in this way, and finally the anchor cable response parameter set including displacement increment, oscillation amplitude and response lag is formed.

[0018] The trend consistency identification submodule calls the anchor cable grading tensioning response parameter set, judges whether the trend direction of the response parameters in the same tensioning period is consistent, and analyzes the change rate of the response parameters between consecutive periods. The section with consistent fluctuation direction and stable rate change is selected, and the trend and rate corresponding sections between the response parameters are obtained. The constructed response parameter set is called to judge whether the trend direction of the response parameters in the same tensioning period is consistent. In the actual process, the displacement increment sequence, stress reverse oscillation amplitude sequence and lag time sequence in the same period are synchronized and aligned. The first order difference value of each parameter is calculated, and then it is judged whether the difference signs are consistent. If the first order differences of the three groups of parameters in a period are all positive, it is considered that the trend direction is consistent, otherwise it is not consistent. The change rate of the response parameters between consecutive periods is analyzed by constructing the period difference sequence and calculating the average change amplitude per unit time. For example, the displacement increment increases by 0.5 mm per second, the oscillation amplitude changes by 0.2 kN per second, and the lag time decreases by 0.1 second per second. The change standard deviation of each continuous three periods is taken for the three kinds of rates respectively, and it is judged whether it is less than the set rate stability threshold. If it is less than the set rate stability threshold, it is considered that the rate change is stable. The threshold can be set according to the empirical data of previous anchor cable tensioning experiment results, such as taking the change standard deviation less than 0.05 as the stability judgment basis. Finally, the period section with consistent trend and rate change standard deviation meeting the set threshold is selected as the trend and rate corresponding section between the response parameters.

[0019] The time sequence characteristic generation sub-module extracts the trend direction data and the change rate data of the response parameters in each period segment as input variables according to the corresponding segment of the trend and the rate between the response parameters, calculates the nonlinear correlation degree between the trend and the rate by using the maximum normalized mutual information algorithm, and arranges the time sequence characteristic data of the anchor cable change state in the order of the tensioning time. Based on the identified consistent segment of the trend and the rate, the response parameter characteristics of each tensioning period are extracted in sequence, including the trend direction data and the change rate data, and the nonlinear coupling strength between the two is calculated by using the maximum normalized mutual information algorithm to form the time sequence characteristics of the anchor cable change.

[0020] In the operation flow, first, each selected period is numbered as to , each period corresponds to three response parameters: the displacement increment (denoted as ), the average amplitude of oscillation (denoted as ), and the response lag time (denoted as ). For each parameter, in the period , the trend direction data (denoted as ) and the rate data (denoted as ) are extracted to form a three-dimensional trend vector and a three-dimensional rate vector . The trend direction is represented by a standardized value ±1: if the parameter value in the period rises, the value is +1, and if it falls, the value is -1; the rate value is obtained by linear fitting and normalized to the interval [0, 1], for example, if the original rate , and if the historical maximum and minimum rates of this type of parameter are 0.1 and 0.5 mm / s, respectively, the normalized value is , and .

[0021] After obtaining the trend sequence and the rate sequence , the following maximum normalized mutual information formula is used: ; wherein : the maximum normalized mutual information value between the trend vector sequence and the rate vector sequence , which is used to measure the degree of correlation between the two groups of data in the nonlinear space; : the mutual information value between the trend and the rate, which is used to quantify the information gain of the joint distribution of and relative to their respective marginal distributions; : the trend data Shannon entropy of trend data, representing the uncertainty of each combination appearing in trend data; : rate data Shannon entropy of rate data, reflecting the distribution information of normalized rate combination; : take the larger value in , used for normalizing mutual information, so that the result is limited in the range of [0, 1]; : the trend direction vector extracted in the th tension cycle, consisting of three elements, respectively , , , where the superscripts , represent the trend direction corresponding to the displacement increment, the average amplitude of oscillation and the response lag time respectively; : the normalized rate vector in the th tension cycle, containing three components: , , ; : represents the cycle index, ranging from , corresponding to each valid cycle in the aforementioned tension process; , , : respectively corresponding to three response parameters, representing the specific dimension in the trend or rate vector; The calculation formula of mutual information value is as follows: ; Where: : represents a specific value vector of trend direction combination, for example ; : represents a normalized value vector of rate combination, for example ; : represents the summation of all trend combination values in the combination set ; : represents the summation of all rate combination values in the set ; : represents the joint probability of the simultaneous occurrence of trend combination and rate combination ; : represents the probability of the occurrence of trend combination in the sample; : represents the probability of the occurrence of rate combination in the sample; : logarithm function with base 2, used to convert probability information into information quantity unit bit; trend entropy and rate entropy The calculation formulae are respectively: , ; wherein: the negative sign indicates that the probability is multiplied by the inverse of its logarithm, so that the information entropy is positive; : indicates that the sum is taken over all trend combinations; : indicates that the sum is taken over all rate combinations.

[0022] The final result is a normalized value reflecting the degree of non-linear matching between the trend direction and the change rate of the response parameter in the current tensioning period. The closer the value is to 1, the stronger the correlation between the two, and finally a characteristic sequence arranged in chronological order according to the period time is formed .

[0023] Suppose that in the staged tensioning process of a section of anchor cable, after 4 consecutive effective tensioning periods (i.e. ), the response parameters are collected and normalized for each period, the following normalized data are obtained: Trend direction data (standardized to ±1): , , , ; Rate vector data (normalized to [0, 1]): , , , ; First, the trend combinations and rate combinations are discretely processed, the rate vectors are divided into normalized rate segments (for example: low 0-0.4, medium 0.4-0.7, high 0.7-1.0), and then each vector is mapped to a rate level. For example: , , , Next, the trend-rate combinations are counted: Combination A: , : appears once; Combination B: , : appears once; Combination C: , : appears once; Combination D: , : appears once; Thus the joint probability is: , , , ; Marginal probability (trend): , , ; Marginal probability (rate): , , .

[0024] Calculate mutual information , using the formula: ; Calculate each term in turn (only non-zero terms are listed): A term: ; B term: ; C term: ; D term: ; So: .

[0025] Calculate entropy , : ; Rate entropy is calculated as follows: ; Calculate normalized mutual information ; For a data segment consisting of 4 cycles, the time series characteristic value is: , such as processing each segment in time sequence , the full cycle sequence can be generated.

[0026] In the process of anchor cable hierarchical tensioning, the overall correlation between the trend direction and the rate of change of the response parameters can be quantified by calculating the mutual information value, and the larger the value is, the tighter the coupling between the two is; then the entropy of the trend direction and the entropy of the rate of change are calculated to measure the complexity of the respective numerical distribution and the information uncertainty, and the larger the entropy is, the higher the diversity and volatility of the data sequence is; the mutual information value is normalized with the larger of the two entropies to obtain the maximum normalized mutual information value, so that the value is limited between zero and one, which is convenient for horizontal comparison and trend analysis between multiple periods. The whole calculation process includes: first, the trend direction of the three response parameters in each tensioning period is converted into a standardized symbol, and the rate is calculated and normalized at the same time; then the trend and the rate are combined for frequency statistics to construct joint probability distribution and marginal probability distribution, and the mutual information value and the entropy are calculated accordingly; finally, the time series feature index is formed by normalization, so as to realize the dynamic quantitative expression of the anchor cable stress process.

[0027] Please refer to Figure 3 , the trajectory recognition module comprises: The lag trajectory extraction submodule refers to the time series feature data of the anchor cable change state, extracts the time position and change trend of the response parameter in each tensioning period, identifies the period paragraph with response delay, and obtains the lag trajectory interval of the response parameter; According to the time series feature data of the anchor cable change state, the response parameter in each tensioning period is positioned, and the time difference between the actual time point when the parameter changes and the starting time of the tensioning action is extracted as the lag time. The acquisition of this difference value depends on the high-frequency sampled time series, which is realized by comparing the starting mark of the force signal with the position where the response parameter changes significantly. The judgment standard of significant change is that the change rate breaks through the set change rate threshold in continuous sampling points, and the threshold should be calculated based on the average rate of the most obvious point in the response curve of the previous tensioning test, and the empirical mean plus twice the standard deviation of the change rate is usually selected as the identification benchmark. If the judged lag time is greater than the set value, it is considered that the period has response lag. After collecting all the periods that meet the lag standard, the complete trend of the response parameter is extracted according to the time position where the lag occurs and the change trajectory, the waveform of the trend is extracted by continuous curve fitting, such as rising, falling, platform or fluctuation type, and the corresponding time period and trend structure are combined to form the lag trajectory interval set.

[0028] The trend comparison submodule calls the lag trajectory interval of the response parameter, compares the change direction of the response parameter in each lag period, selects the period sequence with consistent trend direction, and obtains the response parameter consistent period; Based on the hysteresis trajectory interval, the direction of change of the response parameters contained therein is extracted periodically and compared in chronological order. The trend direction of each period needs to be extracted from the continuous change results of the parameters. Typically, the direction of change of the response values ​​at both ends of the period is used for classification: an upward trend is recorded as positive, a downward trend as negative, and a unchanged direction is considered a stable state. The identification of trend consistency relies on the consistency requirement of the three parameters' directions. If the displacement, stress, and hysteresis parameters all show the same direction of change or stability in adjacent periods, the two periods are judged to have a consistent trend; otherwise, the trend is considered interrupted. To eliminate occasional misjudgments, only when three or more consecutive periods meet the trend direction consistency requirement can it be retained as a valid trend segment. Finally, all period segments that meet the continuous consistency requirement are extracted and organized into a set of trend-consistent period segments.

[0029] The closed-chain construction submodule determines the continuity relationship and periodic arrangement structure of response parameters between multiple periods based on the consistent periodic segments of response parameter trends, identifies the closed characteristics of continuous response trajectories, and obtains the anchor cable response evolution trajectory. Continuing with consistent trend segments, the continuity between these segments is determined. Continuity refers to the consistency of parameter changes in trend direction and temporal continuity between segments. First, the trend direction at the end of the current segment is compared with the starting trend direction of the next segment. If they are consistent and the interval between the two segments does not exceed two periods, the continuity criterion is met. Once the continuity is confirmed, the preceding and following segments can be joined into a longer trend segment. After constructing all possible continuous segments, these trend chains are identified for closure characteristics. Closure identification conditions include consistent trend directions at the beginning and end, and the trend changes falling into the same interval in terms of rate or amplitude. This interval is divided based on the historical distribution of the response parameters, typically using 20% ​​of the parameter fluctuation range as the dividing line. If the beginning and end values ​​fall within this interval, it is considered to have a closed structure. All trend chains that meet the closure requirements are recorded as a time series, forming a complete evolutionary trajectory set of the anchor cable response.

[0030] Please see Figure 4 The stability identification module includes: The tensioning segment extraction submodule extracts the time position of the tensioning segment in the anchor cable response evolution trajectory, organizes the time window information corresponding to each tensioning segment, and obtains the time window of the tensioning segment. Based on the evolution trajectory of the anchor cable response, the time positioning processing is performed on the response segment with continuous trend characteristics to identify the start and end time of the segment with typical tension characteristics. During the operation, first, the trend change curve of each closed chain in the evolution trajectory is segmented to find the time point where the trend changes significantly as the segment boundary. The identification criterion for the turning point is that the slope change value exceeds thirty percent of the original change rate during the continuous change of the response parameter, and the symmetry of the fluctuation on both sides is modified to ensure that the segment division has logical boundary significance. The start and end time of each segmented segment is used as the boundary of the time window, and then the response parameter values, trend markers, and disturbance event information in the time window are synchronized and stored as tension segment time window information.

[0031] The stable boundary recognition submodule detects the disturbance occurrence time interval, disturbance peak duration, and disturbance stable decline time recorded by the surrounding rock monitoring points in each tension segment time window, judges the length of the overlapping section of the three disturbance data, and obtains the structure stable boundary response determination result; Taking each tension segment time window as the analysis object, the disturbance data recorded by the surrounding rock monitoring points within the time window range are detected, the time interval between the start of the disturbance and the next disturbance, the duration of the disturbance peak, and the time used for the disturbance amplitude to fall back to the stable state are extracted, and the three data are extracted through waveform analysis: the time interval is obtained from the difference between the start points of two consecutive wave peaks, the duration is defined by the time used for a single disturbance to reach the maximum value from the start, and the stable decline time is the time difference from the disturbance peak to the stable zero point of the disturbance response. By aligning the three time data on the time axis, the overlapping time period of the three data is judged, and the degree of overlap between the disturbance concentration and the stable interval is identified. If the three disturbance sections have obvious overlapping parts and the overlapping time exceeds sixty percent of the total disturbance time, it is determined that the segment is in the stable boundary; otherwise, if the overlapping time between the three is too short or does not constitute an intersection, it is considered that the stable condition has not been reached, and the determination result of the structure stable boundary response is output.

[0032] The unstable segment marking submodule identifies the tension segment with disturbance delay overlapping time longer than the decline time according to the structure stable boundary response determination result, marks the target tension segment in the structure stability risk state, and obtains the unstable anchor cable tension segment; According to the determination result of the structural stability boundary, the tensioning section in the boundary judgment failure state is further analyzed. The specific way is to compare the delay coincidence time length in the disturbance data with the disturbance recovery time. If the delay coincidence time lasts more than the recovery time, it means that the structure cannot quickly recover after the disturbance response, and there is a risk of structural stability. The comparison benchmark of the coincidence time length and the recovery time is the proportion of the difference between the two to the recovery time. If the proportion exceeds one, that is, the delay coincidence time is longer than the recovery time, the tensioning section is determined to be an unstable section. All sections meeting this determination condition are marked as stability risk paragraphs, and the output is unstable anchor cable tensioning section.

[0033] Please refer to Figure 5 , the coordination analysis module comprises: The segment information extraction submodule extracts the anchor cable tensioning section corresponding time window position of the unstable anchor cable tensioning section in the anchor cable response evolution track, and screens the anchor cable tensioning section corresponding time information; Taking the unstable anchor cable tensioning section as the input basis, the time window position of each unstable section in the overall response process is determined according to the anchor cable response evolution track formed in the early stage, and the time start and end information covered by the section is extracted. In the specific process, first, the number of the unstable section is compared with the tensioning section in the evolution track in time, and whether its start and end time completely falls into any evolution track section is screened. If there is a complete inclusion relationship, the corresponding time window information of the evolution track section is extracted as the output result. The time window information needs to include the tensioning start time, tensioning action end time, response maximum value appearance time and other key time points, which are automatically identified and aligned through time sequence data. At the same time, the sections with insufficient time span, that is, the total duration time does not reach the minimum section reference value, are not retained. The reference value needs to be determined in combination with the shortest effective loading time in the tensioning control process, for example, the minimum reaction time required for tensioning loading to response generation as the judgment threshold. The final output result is the time information set corresponding to all unstable sections meeting the condition.

[0034] The response correlation analysis submodule calls the anchor cable tensioning section corresponding time information, compares the time window of the unstable anchor cable tensioning section with the response peak time in the anchor cable response evolution track, judges whether there is overlap under the same tensioning level, and extracts the anchor cable tensioning section with overlap relationship; Based on the time window of unstable anchor cable tensioning section, combined with the response peak time of each stage in the response evolution trajectory, specific time comparison analysis is performed, and the response time overlap strength between unstable section and peak time is calculated to identify whether there is a physical response coupling relationship. During the execution, first, the time window interval of each unstable section is extracted, that is, the tensioning start and end time, and the corresponding time points of the response peak appearing in each tensioning order in the evolution trajectory are extracted synchronously; then the time position difference model of the two is constructed to calculate the overlap measurement index.

[0035] To realize quantitative comparison, the following calculation formula is defined to represent the time overlap strength: ; Wherein, : represents the response correlation index, which is used to measure the coincidence degree between the unstable section and the peak time; : represents the coincidence time, that is, the time intersection range formed by the unstable section time window and the response peak occurrence time; : represents the total time length of the unstable tensioning section, that is, the overall duration from the section start time to the end time.

[0036] In the calculation, the coincidence time is obtained by judging whether the peak time point falls within the section time window, and on this basis, a certain offset tolerance is expanded to accommodate sampling errors or delay effects, which can be set as a fixed time range before and after the peak value, for example, the experience is set to not more than one average loading cycle duration. If the peak drop point and the section time window exist intersection, and the intersection interval exceeds twenty percent of the total time window, it is considered that the effective coupling exists.

[0037] Suppose that in a certain anchor cable tensioning process, a unstable tensioning section is identified by the system, and the tensioning action starts from a certain time and forms a complete time window until the tensioning is completed: Unstable section time window starting point: tensioning start time is 08:00:00; Unstable section time window end point: tensioning end time is 08:00:10; Response peak time point: the response peak recorded by the sensor occurs at 08:00:07; At the same time, the tolerance time window of this project is expanded to 2 seconds before and after, which is used to deal with the system delay or local disturbance lag phenomenon caused by tensioning when detecting the response peak.

[0038] The total length of the unstable section is calculated as : End time-start time=10 seconds; Determine the extension peak time range: the peak response time is 08:00:07, considering a tolerance of 2 seconds before and after, the peak time interval is: ; Calculate the coincidence time , the peak tolerance interval and the unstable segment time window are overlapped on the time axis: the segment time window is 08:00:00 to 08:00:10, the peak response interval is 08:00:05 to 08:00:09, and the intersection of the two is 08:00:05 to 08:00:09, corresponding to the coincidence time: ; Substitute the formula to calculate the response correlation index : ; The final response correlation index is: .

[0039] According to the preset threshold value judgment standard, for example, set to It is considered that there is effective coupling, then the unstable segment and the response peak have obvious time overlap relationship, which belongs to the tension segment with physical correlation in the structural response evolution track.

[0040] The parameter set construction submodule obtains the stress change rate and displacement change rate in the response parameters in the tension stage corresponding to the anchor cable tension segment with overlapping relationship, and summarizes the coordination index reflecting the overall trend, to obtain the anchor cable coordination evaluation parameter set; Based on the anchor cable tension segment identified as having overlapping relationship, the stress change rate and displacement change rate in the response parameters in the corresponding tension stage are extracted as core data items, and a coordination index reflecting the overall response trend consistency in the tension process is constructed. In the specific operation process, first, for each anchor cable tension segment, the stress and displacement monitoring value change in the corresponding tension stage time range is extracted, and the average change rate of the two parameters is calculated in the time range by using linear fitting method, that is, the result is the change rate by subtracting the initial value from the final value and dividing by the corresponding time difference. To ensure the comparability of the rate values of different segments, all rate values need to be normalized after calculation. The normalization is based on the maximum and minimum change rate range of the entire historical tension process, and the linear interval scaling method is used to convert to the standard [0, 1] interval. After normalization, the stress rate and displacement rate of the same segment are calculated, and a coordination index is constructed according to the closeness of the two, which measures the synchronization of the two response parameters by difference, ratio or angle, to judge the coupling consistency in the tension process. Finally, the coordination indexes of all segments with overlapping relationship are summarized to form the anchor cable coordination evaluation parameter set.

[0041] Please refer to Figure 6 , the quality evaluation module includes: The coordination parameter analysis submodule identifies the trend direction of the stress rate of change and the displacement rate of change in the anchor cable coordination evaluation parameter set, judges whether there is obvious reverse change, and obtains the coordination change trend information of the anchor cable; For the stress rate of change and the displacement rate of change of each tensioning section in the anchor cable coordination evaluation parameter set, a trend direction identification operation is performed to identify the synchronization and deviation characteristics of the changes of the two. In the specific process, first, the direction relationship is judged according to the sign of the change rate of the two parameters in each section. When both are positive or negative, it is considered that the directions are consistent, and if one is positive and the other is negative, it is considered that the directions are reversed. In the continuous section, the direction matching between the stress rate and the displacement rate is tracked. If the direction between the adjacent two sections changes from consistent to reverse, or from reverse to consistent, it is recorded as a reverse event. The identification of the reverse is not only based on the sign change, but also needs to judge whether the change degree exceeds the preset deviation threshold after normalization of the rate value. Usually, the threshold is set to be that the absolute value of the difference between the two normalized values exceeds 0.3, that is, it is considered that there is a significant difference. By traversing all the section sequences with coordination indicators, the trend change state is recorded one by one to form a coordination trend sequence containing reverse information and change direction labels, which is used as a basis for further determination of the stability of the anchor cable response.

[0042] The section state judgment submodule screens the anchor cable tensioning sections with reverse number exceeding the limit in the anchor cable coordination change trend information, determines that the target tensioning section has unstable tensioning response characteristics, and obtains the anchor cable tensioning sections with abnormal changes; The coordination change trend sequence is statistically analyzed, and the tensioning sections with the cumulative number of reverses exceeding the set limit are mainly identified. The specific method is to statistically analyze the trend change in a sliding window of a certain number of periods before and after each anchor cable tensioning section, record the number of reverse events in a fixed number of consecutive periods, and if the cumulative number of reverses exceeds the preset number threshold, it is considered that the coordination of the section has significant unstable characteristics. The setting of the reverse number threshold needs to be combined with the statistical characteristics of the stable tensioning process in typical engineering scenarios. For example, in actual cases, it can be set to allow at most two reverses in five consecutive periods, and more than this upper limit is marked as abnormal. After judgment, all sections that meet the reverse limit condition are extracted to form a set of anchor cable tensioning sections with abnormal changes.

[0043] The evaluation result generation submodule summarizes the anchor cable tensioning sections with abnormal changes, combines the numerical change of the corresponding tensioning section in the anchor cable coordination evaluation parameter set, and constructs the anchor cable tensioning quality evaluation result; The anchor cable tensioning segments determined to have abnormal changes are collected and combined with the stress and displacement rate numerical characteristics of each corresponding segment of the anchor cable coordination evaluation parameter to construct a comprehensive tensioning quality evaluation result. This process needs to summarize the average level, fluctuation range and change trend of the rate parameter in the abnormal segment, identify whether there is a phenomenon of significant deviation from the conventional range, especially whether it is concentrated in a certain tensioning level and whether it shows a gradual deterioration trend, etc. In the quality evaluation process, the scoring benchmark needs to be set, and the average value of the coordination index in the normal segment is usually taken as the reference value, and the grading interval is constructed combined with the standard deviation, such as the segment within two standard deviations below the average value is normal, and the segment exceeding two standard deviations is marked as deteriorated. The final tensioning quality evaluation result includes the following contents: the time window range of each abnormal segment, the identification label of unstable change trend, the corresponding stress and displacement rate comparison value, whether there is a response peak overlap relationship label, and the position description of the segment in the tensioning evolution track. These evaluation results are used as the basis for subsequent construction condition review, anchor structure safety judgment and response optimization strategy formulation.

[0044] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can use the disclosed technical content to make changes or modifications to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application still belongs to the protection scope of the technical solution of the present application.

Claims

1. A large-tonnage prestressed anchor cable tensioning quality intelligent evaluation system, characterized in that, The system comprises: a time sequence construction module, which acquires response parameters in the anchor cable grading tensioning process, and constructs time sequence characteristic data of the anchor cable change state according to the trend consistency and cross-period change rate among the response parameters in the same tensioning period; a trajectory recognition module, which refers to the time sequence characteristic data of the anchor cable change state, recognizes the lagging situation and multi-period response coordinated closed chain of the anchor cable response parameters in the change trajectory, and obtains an anchor cable response evolution trajectory; a stability recognition module, which extracts tensioning section information in the anchor cable response evolution trajectory, recognizes the structural stability boundary under the influence of anchor cable tensioning, and marks unstable anchor cable tensioning sections with structural stability risks; a coordination analysis module, which extracts anchor cable section corresponding information of the unstable anchor cable tensioning sections in the anchor cable response evolution trajectory, analyzes response coordination characteristics of the anchor cable group in the tensioning process, and constructs an anchor cable coordination evaluation parameter set; a quality evaluation module, which judges the overall tensioning quality of the anchor cable according to the anchor cable coordination evaluation parameter set, summarizes all section judgment information, and outputs an anchor cable tensioning quality evaluation result.

2. The large-tonnage prestressed cable tension quality intelligent evaluation system according to claim 1, characterized in that, The time sequence characteristic data comprises period trend direction labels, period rate change values, and trend rate coupling strength indicators. The anchor cable response evolution trajectory comprises lagging time intervals, trend closed structure types, and multi-period coordinated trajectory paths. The unstable anchor cable tensioning sections comprise stability boundary recognition results, abnormal section time window positions, and structural disturbance indicator overlap situations. The anchor cable coordination evaluation parameter set specifically comprises stress rate normalized values, displacement rate normalized values, and parameter synchronization deviation amounts. The anchor cable tensioning quality evaluation result comprises abnormal section quantity statistics, abnormal section change type classifications, and coordinated instability feature recognition information.

3. The large-tonnage prestressed cable tension quality intelligent evaluation system according to claim 1, characterized in that, The time sequence construction module comprises: a response parameter extraction submodule, which acquires three response parameters, i.e., a displacement increment change value monitored by a fiber displacement meter, an average amplitude of reverse oscillation of a peak tensioning stress monitored by a self-induction stress meter, and a cumulative response lagging time from the initial to the final stage, in the anchor cable grading tensioning process, and constructs an anchor cable grading tensioning response parameter set; a trend consistency recognition submodule, which calls the anchor cable grading tensioning response parameter set, judges whether the trend direction of the response parameters in the same tensioning period remains consistent, analyzes the change rate of the response parameters between consecutive periods, screens sections with consistent fluctuation directions and stable rate changes, and acquires corresponding sections of the trend and rate among the response parameters; a time sequence characteristic generation submodule, which extracts response parameter trend direction data and change rate data in each period section as input variables according to the corresponding sections of the trend and rate among the response parameters, calculates the non-linear correlation degree between the trend and the rate by a maximum normalized mutual information algorithm, and arranges them in the order of tensioning time to form the time sequence characteristic data of the anchor cable change state.

4. The large-tonnage prestressed cable tension quality intelligent evaluation system according to claim 3, characterized in that, The trajectory recognition module comprises: a lagging trajectory extraction submodule, which refers to the time sequence characteristic data of the anchor cable change state, extracts the time position and change trend of the response parameters in each tensioning period, recognizes period paragraphs with response delays, and acquires the lagging trajectory interval of the response parameters; The trend comparison submodule calls the lag trajectory interval of the response parameter, compares the response parameter change direction in each lag period, screens the period sequence with consistent trend direction, and obtains the response parameter consistent period with consistent trend; The closed chain construction submodule determines the continuous response trajectory according to the continuous relationship and period arrangement structure of the response parameter between multiple periods, determines the closed characteristics of the continuous response trajectory, and obtains the anchor cable response evolution trajectory.

5. The large-tonnage prestressed cable tension quality intelligent evaluation system according to claim 4, characterized in that, The stability recognition module comprises: The tension segment extraction submodule extracts the tension segment time position in the anchor cable response evolution trajectory, organizes the time window corresponding information of each tension segment, and obtains the tension segment time window; The stable boundary recognition submodule detects the disturbance occurrence time interval, disturbance peak duration and disturbance stable falling time recorded by the surrounding rock monitoring point in each tension segment time window, judges the coincidence length of the three disturbance data, and obtains the structure stable boundary response judgment result; The unstable segment marking submodule identifies the tension segment with disturbance delay coincidence time longer than the falling time according to the structure stable boundary response judgment result, marks that the target tension segment is in the structure stability risk state, and obtains the unstable anchor cable tension segment.

6. The large-tonnage prestressed cable tension quality intelligent evaluation system according to claim 5, characterized in that, The coordination analysis module comprises: The segment information extraction submodule extracts the anchor cable tension segment corresponding time window position of the unstable anchor cable tension segment in the anchor cable response evolution trajectory, and screens the anchor cable tension segment corresponding time information; The response correlation analysis submodule calls the anchor cable tension segment corresponding time information, compares the time window of the unstable anchor cable tension segment with the response peak time in the anchor cable response evolution trajectory, judges whether there is overlap under the same tension order, and extracts the anchor cable tension segment with overlapping relationship; The parameter set construction submodule obtains the stress change rate and displacement change rate in the response parameter in the tension order corresponding to the anchor cable tension segment with overlapping relationship, and summarizes the coordination index reflecting the overall change trend to obtain the anchor cable coordination evaluation parameter set.

7. The system according to claim 6, wherein, The quality evaluation module comprises: The coordination parameter analysis submodule identifies the trend direction of the stress change rate and displacement change rate in the anchor cable coordination evaluation parameter set, judges whether there is obvious reverse change, and obtains the anchor cable coordination change trend information; The segment state judgment submodule screens the anchor cable tension segment with reverse number accumulation exceeding the limit in the anchor cable coordination change trend information, judges that the target tension segment has tension response instability characteristics, and obtains the anchor cable tension segment with abnormal change; The evaluation result generation submodule summarizes the anchor cable tension segment with abnormal change, combines the numerical change of the corresponding tension segment in the anchor cable coordination evaluation parameter set, and constructs the anchor cable tension quality evaluation result.

Images