NLP-based concrete pile body integrity grade intelligent determination method and system

Abstract

The invention discloses an NLP-based intelligent judgment method and system for the integrity grade of a concrete pile body, and the method comprises the steps: detecting a to-be-detected concrete pile body through a sound wave transmission method, obtaining the acoustic parameters of the to-be-detected concrete pile body, and receiving waveform signal data; and inputting the acoustic parameters of the to-be-detected concrete pile body and the received waveform signal data into the trained integrity level judgment model, and outputting the integrity level of the to-be-detected concrete pile body. According to the method, the natural language processing technology and the ultrasonic image recognition technology are utilized, and the technical problems that the manual judgment efficiency of the integrity level of the concrete pile body based on the acoustic transmission method is low, the integrity of all to-be-detected piles under the same project cannot be judged at the same time, and the judgment results are inconsistent are solved.

Description

Technical Field

[0001] The present invention relates to the application of natural language processing and ultrasonic image recognition in the field of pile foundation detection integrity level determination, and in particular to an NLP-based intelligent determination method and system for the integrity level of a concrete pile body. Background Art

[0002] The acoustic wave transmission method is a method for testing the integrity of concrete piles. Its main principle is to use the characteristics of diffraction, reflection and refraction of ultrasonic waves when encountering defects in concrete to measure the acoustic parameters of the tested concrete piles, and then use the characteristics of acoustic parameters and ultrasonic signal characteristics to judge the integrity of the concrete piles and assess the location, scope and degree of pile defects. At present, the acoustic wave transmission method for testing the integrity of concrete piles is widely used in the fields of highway engineering pile foundation testing, railway engineering pile foundation testing, and building pile foundation testing. The national, provincial and municipal governments have issued relevant testing technical specifications for the acoustic wave transmission method for testing the integrity of concrete piles.

[0003] The requirements for pile integrity level determination in the technical specifications for highway engineering pile foundation detection, railway engineering pile foundation detection, building pile foundation detection, etc. vary due to different application scenarios and regional soil structures. However, each technical specification contains a pile integrity level determination table, which is a comparison table that describes the pile integrity level with acoustic parameter characteristics and ultrasonic signal characteristics in natural language.

[0004] The traditional method of judging the pile body integrity level based on the acoustic wave transmission method for concrete pile body integrity detection is manual judgment, that is, the detection personnel determine the integrity level of the tested concrete pile based on their own understanding of the pile body integrity level judgment table in various national, local, provincial and municipal detection technical specifications, combined with the acoustic parameter characteristics and ultrasonic signal characteristics of the tested concrete pile. However, as the country pays more and more attention to the quality of pile foundations in construction projects, the manual workload has increased dramatically, and the traditional method has individual subjective differences, resulting in inconsistent judgment results. Summary of the invention

[0005] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a method and system for intelligently determining the integrity level of concrete piles based on NLP, which utilizes natural language processing technology and ultrasonic image recognition technology to solve the technical problems of low efficiency in manual determination of the integrity level of concrete piles based on the acoustic wave transmission method, inability to simultaneously determine the integrity of all piles to be tested in the same project, and inconsistent determination results.

[0006] According to one aspect of the present invention, there is provided a method for intelligently determining the integrity level of a concrete pile body based on NLP, comprising:

[0007] Using the acoustic wave transmission method to detect the concrete pile body to be tested, and obtaining the acoustic parameters of the concrete pile body to be tested and receiving waveform signal data;

[0008] Input the acoustic parameters of the concrete pile body to be tested and the received waveform signal data into the trained integrity level determination model, and output the integrity level of the concrete pile body to be tested; wherein the training of the integrity level determination model includes:

[0009] Identify the corpus of several testing technical regulations corresponding to the integrity test of concrete piles, and obtain the acoustic parameters and waveform feature description information corresponding to the integrity level;

[0010] Based on the acoustic parameters and waveform feature description information corresponding to the integrity level, the judgment object and its feature keyword set are extracted, and the fuzzy quantity feature keywords are counted;

[0011] Based on the pre-built quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained classification model of receiving waveform state characteristics, the extracted judgment objects and their characteristic keyword sets are quantified to form an integrity level judgment model;

[0012] A training sample set for integrity level determination is constructed, an integrity level determination model is trained with the training sample set, and fuzzy quantitative feature keywords are used as adjustment parameters to obtain a trained integrity level determination model.

[0013] As a further technical solution, the extracted judgment objects include: acoustic parameters, received waveform, sound speed range and first wave.

[0014] As a further technical solution, the pre-construction of the quantitative discrimination formula of the state characteristics corresponding to the acoustic parameters includes:

[0015] Obtain a historical data set of manually identified integrity levels of concrete pile bodies, and classify and count the quantified values ​​of acoustic parameter state characteristics according to the integrity levels identified in the historical data set;

[0016] According to the classification and statistical results, the characteristic quantitative limit values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD in different states are calculated respectively;

[0017] According to the calculated characteristic quantization limit values ​​of different states, a characteristic quantization discrimination formula corresponding to the state of acoustic parameters is constructed.

[0018] As a further technical solution, the pre-construction of the quantitative discrimination formula corresponding to the distribution characteristics of the acoustic parameters includes:

[0019] Classifying and counting the characteristic quantitative values ​​of the distribution state according to the integrity level determined in the historical data set;

[0020] According to the classification statistics, the characteristic quantitative limit values ​​of the longitudinal and transverse distribution directions in different distribution states are calculated respectively;

[0021] According to the calculated characteristic quantization limit values ​​of different distribution states, a discriminant formula for the distribution characteristic quantization value corresponding to the acoustic parameter is constructed.

[0022] As a further technical solution, the training of the received waveform state feature classification model includes:

[0023] Obtain historical data of concrete pile inspection;

[0024] Constructing a training sample set and a test sample set according to the historical data, and configuring a classification label for each waveform data in the sample set;

[0025] Construct a receiving waveform state feature classification model, including a CNN model for extracting high-level features of the receiving waveform data and an LSTM model for capturing the timing information of the receiving waveform data;

[0026] The receiving waveform state feature classification model is trained using the training sample set, and tested using the test sample set to obtain a trained receiving waveform state feature classification model.

[0027] As a further technical solution, the acoustic wave transmission method is used to detect the concrete pile body to be tested, and the acoustic parameters of the concrete pile body to be tested and the received waveform signal data are obtained, including:

[0028] The concrete pile body to be tested is tested by using the acoustic wave transmission method to obtain the test data set saved in a set format;

[0029] Calculate the characteristic quantization values ​​of the acoustic parameter state of the data set to be tested, including the characteristic quantization values ​​of each calculated based on the real-time values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD;

[0030] Calculating the acoustic parameter distribution characteristic quantization value of the data set to be tested, including the characteristic quantization values ​​of the longitudinal distribution and the transverse distribution respectively calculated based on the acoustic parameter state characteristic quantization value of the data set to be tested;

[0031] All the measured point data of the data set to be tested are converted into time series signals and the corresponding state characteristics are predicted using the trained receiving waveform state feature classification model to obtain the quantized value of the receiving waveform state feature.

[0032] As a further technical solution, the integrity level determination model includes multiple determination levels, each determination level corresponds to one or more determination conditions. When making level determinations, the determination is made in a manner of decreasing from high level to low level. If a certain condition is met, it is determined to be the integrity level corresponding to this condition; wherein, one or more of the determination conditions include: a quantitative determination formula for state characteristics corresponding to acoustic parameters, a quantitative determination formula for distribution characteristics corresponding to acoustic parameters, a trained classification model for receiving waveform state characteristics, and a logical relationship between acoustic parameters in the detection technical regulations corresponding to the concrete pile body to be tested.

[0033] According to one aspect of the present invention, there is provided a concrete pile body integrity level intelligent determination system based on NLP, comprising:

[0034] The data acquisition module is used to detect the concrete pile body to be tested by using the acoustic wave transmission method, and obtain the acoustic parameters of the concrete pile body to be tested and receive waveform signal data;

[0035] The grade determination module is used to input the acoustic parameters of the concrete pile body to be tested and the received waveform signal data into the trained integrity grade determination model, and output the integrity grade of the concrete pile body to be tested; wherein the training of the integrity grade determination model includes:

[0036] Identify the corpus of several testing technical regulations corresponding to the integrity test of concrete piles, and obtain the acoustic parameters and waveform feature description information corresponding to the integrity level;

[0037] Based on the acoustic parameters and waveform feature description information corresponding to the integrity level, the judgment object and its feature keyword set are extracted, and the fuzzy quantity feature keywords are counted;

[0038] Based on the pre-built quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained classification model of receiving waveform state characteristics, the extracted judgment objects and their characteristic keyword sets are quantified to form an integrity level judgment model;

[0039] A training sample set for integrity level determination is constructed, an integrity level determination model is trained with the training sample set, and fuzzy quantitative feature keywords are used as adjustment parameters to obtain a trained integrity level determination model.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention utilizes natural language processing technology and ultrasonic image recognition technology to propose an NLP-based intelligent determination method and system for the integrity level of concrete piles, which solves the technical problems of low efficiency and inconsistent determination results of manual determination of the integrity level of concrete piles based on the acoustic wave transmission method. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction is given below to the drawings used in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A schematic flow chart of a method for intelligently determining the integrity level of a concrete pile body based on NLP provided in an embodiment of the present invention.

[0044] Figure 2 A schematic diagram of the training process of the integrity level determination model provided in an embodiment of the present invention.

[0045] Figure 3 A schematic diagram of the general format of RSM ultrasound provided in an embodiment of the present invention.

[0046] Figure 4 A schematic diagram of the structure of an intelligent determination system for concrete pile integrity level based on NLP provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] It should be noted that:

[0048] The terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions, for example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to the steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0049] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices. The flowcharts shown in the accompanying drawings are merely exemplary and do not necessarily include all the contents and operations / steps, nor do they necessarily have to be executed in the order described. For example, some operations / steps may be further decomposed, while some operations / steps may be combined or partially combined, so the actual execution order may change according to the actual situation.

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention are arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve it. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that this combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0051] The embodiment of the present invention provides a method for intelligently determining the integrity level of a concrete pile body based on NLP, such as Figure 1 As shown, the method includes: using the sound wave transmission method to detect the concrete pile body to be tested, and obtaining the acoustic parameters of the concrete pile body to be tested and the received waveform signal data; inputting the acoustic parameters of the concrete pile body to be tested and the received waveform signal data into the trained integrity level determination model, and outputting the integrity level of the concrete pile body to be tested.

[0052] Among them, Figure 2 As shown, the training of the integrity level determination model includes:

[0053] Identify the corpus of several testing technical regulations corresponding to the integrity test of concrete piles, and obtain the acoustic parameters and waveform feature description information corresponding to the integrity level;

[0054] Based on the acoustic parameters and waveform feature description information corresponding to the integrity level, the judgment object and its feature keyword set are extracted, and the fuzzy quantity feature keywords are counted;

[0055] Based on the pre-built quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained classification model of receiving waveform state characteristics, the extracted judgment objects and their characteristic keyword sets are quantified to form an integrity level judgment model;

[0056] A training sample set for integrity level determination is constructed, an integrity level determination model is trained with the training sample set, and fuzzy quantitative feature keywords are used as adjustment parameters to obtain a trained integrity level determination model.

[0057] In the embodiment of the present invention, before the integrity determination of the concrete pile body is performed, the following steps are also included: corpus collection in the field of concrete pile body detection, and design of determination objects and attribute keywords thereof.

[0058] Specifically, the corpus in the field of concrete pile integrity detection is collected and summarized, such as "Technical Specifications for Highway Engineering Pile Detection (JTG / T 3512-2020)", "Technical Specifications for Railway Engineering Pile Detection (TB10218-2019)", "Building Foundation Detection Specification (DB32T 3916-2020)", etc. The design judgment object and its attribute keyword dictionary {E|E has A quantity feature attribute, B state feature attribute, C distribution state feature attribute}, as shown in Table 1.

[0059] Table 1. Keyword table of judgment objects and their attributes

[0060]

[0061] In the embodiment of the present invention, before determining the integrity of the concrete pile body, the method further includes: designing a quantitative value standard of the state characteristic of the acoustic parameter. Specifically, when V i When '<1, the sound velocity is judged as an abnormal value, A' i When the sound amplitude is less than 0, it is judged as an abnormal value. i When '<1, the sound is judged as an abnormal value. i When '<1, the main frequency is judged as an abnormal value. i When '<1, PSD is judged as an abnormal value.

[0062] The calculation of the state characteristic quantization value corresponding to the acoustic parameter is as follows:

[0063] Quantitative value of sound speed state characteristics: (Where V i is the sound velocity at the measuring point, V c is the critical value of sound velocity calculated according to the test procedures);

[0064] Amplitude state characteristic quantization value: A′ i =A i -Ac (A i is the sound amplitude at the measuring point, A c is the critical value of sound amplitude calculated according to the test procedures);

[0065] Quantitative value of sound time state characteristics: (where T i When the sound is measured at the measuring point, T c is the critical value of sound time calculated according to the test procedures);

[0066] Quantitative value of main frequency state characteristics: (where F i is the main frequency of the measuring point, F c is the critical value of main frequency calculated according to the test procedure);

[0067] PSD state characteristic quantization value: (where P i is the PSD of the measuring point, P c is the PSD cut-off value calculated according to the detection procedure).

[0068] In the embodiment of the present invention, the construction of the quantitative discrimination formula of the state characteristics corresponding to the acoustic parameters includes:

[0069] The quantified limit values ​​of the state characteristics of acoustic parameters are calculated, and the data sets of the collected concrete piles that are manually identified are classified and counted according to the integrity level of the identification. The quantified values ​​of the state characteristics of acoustic parameters of Class IV piles are counted as the severely abnormal sample set, the quantified values ​​of the state characteristics of acoustic parameters of Class III piles are counted as the obviously abnormal sample set, the quantified values ​​of the state characteristics of acoustic parameters of Class II piles are counted as the slightly abnormal sample set, and the quantified values ​​of the state characteristics of acoustic parameters of Class I piles are counted as the normal sample set; then, the method of calculating the average value after eliminating the abnormal values ​​by 3 times the standard deviation is used to calculate the quantified limit values ​​of the four state characteristics of sound velocity, sound amplitude, sound time, main frequency and PSD respectively, and the discrimination formula is formulated, as shown in Table 2.

[0070] Table 2 Quantitative discrimination formula of acoustic parameter state characteristics

[0071] Acoustic parameters Severe abnormalities Obvious abnormality Slight abnormality normal Speed ​​of sound <![CDATA[V′1<V3]]> <h2 style=";text-align:left;direction:ltr"><![CDATA[V3≤V′ <h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> <V2]]><h2 style=";text-align:left;direction:ltr"> <![CDATA[V2≤V′ i <V1]]> <![CDATA[V1≤V′ i ]]> Amplitude <![CDATA[A′ i <A3]]> <h2 style=";text-align:left;direction:ltr"><![CDATA[A3≤A′ <h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> <A2]]><h2 style=";text-align:left;direction:ltr"> <![CDATA[A2≤A′ i <A1]]> <![CDATA[A1≤A′ i ]]> Sound Time <![CDATA[T′ i <T3]]> <![CDATA[T3≤T′ i <T2]]> <![CDATA[T2≤T′ i <T1]]> <![CDATA[T1≤T′ i ]]> Frequency <![CDATA[F′ i <F3]]> <h2 style=";text-align:left;direction:ltr"><![CDATA[F3≤F′ <h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> <F2]]><h2 style=";text-align:left;direction:ltr"> <![CDATA[F2≤F′ i <F1]]> <![CDATA[F1≤F′ i ]]> PSD <![CDATA[P′ i <P3]]> <![CDATA[P3≤P′ i <P2]]> <![CDATA[P2≤P′ i <P1]]> <![CDATA[P i ≤P′ i ]]>

[0072] In Table 2, V1, V2, and V3 are empirical values ​​given with reference to the medium sound velocity critical value of the detection technical specifications, and they are all smaller than the sound velocity critical value and decrease in sequence, A1, A2, and A3 are empirical values ​​given with reference to the medium sound amplitude (wave amplitude) critical value of the detection technical specifications, and they are all smaller than the sound amplitude critical value and decrease in sequence, T1, T2, and T3 are empirical values ​​given with reference to the medium first sound time (sound time) critical value of the detection technical specifications, and they are all smaller than the sound time critical value and decrease in sequence, PSD1, PSD, and PSD3 are empirical values ​​given with reference to the medium PSD (the product of the slope on the sound time depth curve and the sound time change) critical value of the detection technical specifications, and they are all smaller than the PSD critical value and decrease in sequence, and F1, F2, and F3 are empirical values ​​given with reference to the medium main frequency critical value of the detection technical specifications, and they are all smaller than the main frequency critical value and decrease in sequence.

[0073] In the embodiment of the present invention, the construction of the quantitative discrimination formula of the acoustic parameter corresponding distribution characteristics includes:

[0074] Calculate the quantitative limit value of the distribution state characteristics, where the quantified value of the vertical distribution abnormal state is the depth value of the continuous abnormal measurement point block (DN e ), the quantified value of the abnormal state of the lateral distribution is the proportion of abnormal profiles (PP e ), the collected data of Class IV piles are used as the sample set of quantitative values ​​of the abnormal distribution state of "very large area", the collected data of Class III piles are used as the sample set of quantitative values ​​of "large area", the collected data of Class II piles are used as the sample set of quantitative values ​​of "small area", and the collected data of Class I piles are used as the sample set of quantitative values ​​of "small spatial distribution range"; then, the method of calculating the average value after eliminating abnormal values ​​by 3 times the standard deviation is used to calculate the quantitative limit values ​​of the four longitudinal and transverse distribution state characteristics of "very large area", "large area", "small area" and "small spatial distribution range" respectively, and the judgment formula is formulated, as shown in Table 3.

[0075] Furthermore, the depth value of the continuous abnormal measurement point block is Among them, i is continuously increasing, that is, 1 == Δi, ΔH is the moving step of the measuring point, and n represents the number of abnormal measuring points. Indicates an abnormal measurement point. When the number of abnormal types of acoustic parameters of the measurement point reaches or exceeds the minimum number of abnormal acoustic parameters in the corresponding integrity category level, the measurement point is recorded as an abnormal measurement point.

[0076] Furthermore, the proportion of abnormal sections is m represents the number of abnormal sections, represents the abnormal section, k represents the number of all detected sections, and P is the detected section. If the total number of abnormal measurement points reaches or exceeds the minimum number of abnormal measurement points in the corresponding integrity category level, the section is recorded as an abnormal section.

[0077] Table 3 Quantitative discrimination formula for longitudinal and transverse distribution states

[0078] Large area Larger area Smaller area Small spatial distribution range Vertical <![CDATA[D3≤DN e ]]> <![CDATA[D2≤DN e <D3]]> <![CDATA[D1≤DN e <D2]]> <![CDATA[DN e <D1]]> Horizontal <![CDATA[P3≤PP e ]]> <![CDATA[P2≤PP e <P3]]> <![CDATA[P1≤PP e <P2]]> <![CDATA[PP e <P1]]>

[0079] In Table 3, D1, D2, and D3 are the depth ranges of continuous abnormal measurement points summarized based on experience, and they are all greater than the minimum calculated value of the depth range and increase in sequence. P1, P2, and P3 are the proportions of the number of abnormal detection profiles summarized based on experience and increase in sequence.

[0080] The embodiment of the present invention also includes: designing a low limit value V of the speed of sound low , all the abnormal sound velocity characteristics of the collected data sets of Class IV and Class III piles are quantified, and the average value calculated after eliminating abnormal values ​​using 3 times the standard deviation is used as the lower limit of sound velocity V low The speed of sound V at a certain measuring point i Greater than or equal to V low When the corresponding keyword is "above the lower limit" or no keyword; the sound velocity V at a certain measuring point i Less than V low , corresponding to the keyword "below the lower limit".

[0081] The embodiment of the present invention further includes: designing the first wave lower limit value T low , at the first sound time T at a certain measuring point i Greater than or equal to T low When the first sound of a certain measuring point is T i Less than T low , the corresponding keyword is "cannot be detected".

[0082] The embodiment of the present invention also includes: establishing a fuzzy quantity feature keyword numerical quantification correction questionnaire (as shown in Table 4). It is difficult to find a reasonable sample set for derivation and estimation of quantifiers such as "continuous", "more", and "a certain depth range". For this reason, the embodiment of the present invention gives a minimum value based on the years of experience of experts, and then adjusts it in the model evaluation stage to obtain the optimal value based on the accuracy rate. Specifically, Table 4 gives the minimum or maximum value of the fuzzy field based on the empirical value, and the detection unit can make appropriate adjustments based on its actual situation.

[0083] Table 4 Numerical quantification correction questionnaire for fuzzy quantitative feature keywords

[0084]

[0085] In an embodiment of the present invention, the construction and training of the received waveform state feature classification model includes:

[0086] Establish a classification model for the state characteristics of the received waveform. The data collected by the concrete pile integrity test of the acoustic wave transmission method is composed of a large number of measurement point data from multiple sections. One measurement point data can be converted into a time series signal, also known as the received waveform RW. i The length of the single-measurement-point timing signal is usually preset to 512, and the sampling period is usually preset to 1us. The received waveform state feature classification model utilizes the feature extraction capability of CNN and the timing modeling capability of LSTM to classify the state features of the collected timing signals of concrete piles.

[0087] Specifically, the training process includes: 1) data collection, collecting the original data files of the concrete pile acquisition data; 2) manual integrity judgment, performing integrity judgment on the collected original data files to generate a verified test sample set; 3) data preprocessing, standardizing or normalizing the original waveform data; 4) data set division, dividing the preprocessed data into a training sample set and a test sample set, and assigning a classification label (normal-B1) to each data according to the result of manual judgment. w , Slight distortion-B2 w , obvious distortion-B3 w , Severe distortion-B4 w );5) Establish a CNN model and use CNN to extract high-level features of received waveform data; 6) Establish an LSTM model and use LSTM to capture the timing information of received waveform data; 6) Train the model and use the training sample set to train the combined model; 7) Evaluate the model and use the test sample set to evaluate the performance of the model with precision as the evaluation standard; 8) Save the model.

[0088] Furthermore, the receiving waveform feature information is extracted from the collected receiving waveform records, and the receiving waveform features are saved in sequence according to the state level as the state feature model B1 w 、B2 w 、B3 w 、B4 w , forming the receiving waveform state characteristic model {B1 w 、B2 w 、B3 w 、B4 w}.

[0089] In the embodiment of the present invention, the training of the integrity level determination model further includes the identification and training of the detection technical procedure integrity determination table.

[0090] Specifically, it includes: 1) using the table recognition function of OCR technology to identify and process the "Integrity Judgment Table of Detection Technical Regulations" picture, such as the integrity judgment table in Table 5 "Technical Regulations for Detection of Highway Engineering Pile Foundations (JTG / T 3512-2020)", to obtain information such as acoustic parameters and waveform feature descriptions corresponding to the integrity category; 2) using the Jieba word segmentation open source library for word segmentation; 3) using the punctuation mark ";" as the end mark of the conditional statement to divide each type of integrity description information into multiple conditional statements; 4) extracting the judgment object and its feature keyword set {E_r|A_r, B_r, C_r} from the conditional statements one by one, such as Table 6 "Technical Regulations for Detection of Highway Engineering Pile Foundations (JTG / T 3512-2020)" feature keyword set. The judgment object has multiple state attribute keywords, and the most serious state shall prevail; 5) Statistical fuzzy quantity feature keywords, such as "continuous", "a certain depth range", "all", "more", etc.

[0091] Table 5 Integrity determination table of "Technical Specifications for Pile Testing of Highway Engineering (JTG / T 3512-2020)"

[0092]

[0093] Table 6 Characteristic keyword set of "Technical Specification for Pile Testing of Highway Engineering (JTG / T 3512-2020)"

[0094]

[0095] In the embodiment of the present invention, generating a concrete pile body integrity grade determination model specifically includes: 1) using "Table 2 Acoustic Parameter Corresponding State Feature Quantitative Discrimination Formula", "Table 3 Transverse and Longitudinal Distribution State Quantitative Discrimination Formula", and the receiving waveform state feature classification model to quantify the extracted detection specification integrity category assessment object and its characteristic keyword set, and the quantization result of Table 6 is shown in Table 7 of the quantization table of "Technical Code for Highway Engineering Foundation Pile Inspection (JTG / T 3512-2020)". 2) Constructing an integrity grade determination model, such as the integrity grade determination model of "Technical Code for Highway Engineering Foundation Pile Inspection (JTG / T 3512-2020)" shown in Table 8, wherein the concrete pile body integrity grade determination rule is to give priority to the high grade, and if the high grade determination condition is not met, the previous grade is determined descendingly. When judging each specific level, it is also necessary to combine the logical relationship of each acoustic parameter in the corresponding detection technical regulations of the concrete pile body to be tested, such as logical AND, logical OR, etc., and use the quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained receiving waveform state characteristic classification model as criteria to judge the level of the concrete pile body to be tested. 3) Evaluate the model, use the collected data of the manually determined integrity level as the test sample set and the accuracy rate as the evaluation standard to evaluate the performance of the model, where the "Fuzzy Quantitative Feature Keyword Numerical Quantification Correction Questionnaire" is used as an adjustment parameter. 4) Save the model.

[0096] Table 7 Quantitative table of "Technical Specification for Pile Testing of Highway Engineering (JTG / T 3512-2020)"

[0097]

[0098] Table 8 Integrity level determination model of "Technical Specification for Pile Testing of Highway Engineering (JTG / T 3512-2020)"

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] As an embodiment, when the above method of the present invention is used to detect the integrity level of the concrete pile body, the sound wave transmission method is used to detect the integrity of the concrete pile and the collected data are used to detect the integrity of the concrete pile. Figure 3The integrity level of the collected data saved in the "RSM ultrasonic general format" shown in the figure is predicted to determine their corresponding integrity levels, including: 1) calculating the critical values ​​of the acoustic parameters of the test data set according to the detection procedures; 2) calculating the quantized values ​​of the acoustic parameters of the test data set according to the aforementioned "acoustic parameter state quantization value standard"; 3) converting all the measurement point data of the test data set into time series signals and using the "received waveform state feature classification model" to predict their corresponding state characteristics; 4) using the "pile body concrete integrity level determination model" to predict the processed test data set to determine their corresponding integrity levels.

[0106] The implementation basis of each embodiment of the present invention is to implement programmed processing through a device with a processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention are encapsulated into various modules. Based on this reality, on the basis of the above embodiments, an embodiment of the present invention provides a concrete pile body integrity level intelligent determination system based on NLP, which is used to execute the concrete pile body integrity level intelligent determination method based on NLP in the above method embodiment.

[0107] See also Figure 4 The system comprises: a data acquisition module, which is used to detect the concrete pile body to be tested by using the acoustic wave transmission method, and obtain the acoustic parameters and received waveform signal data of the concrete pile body to be tested; a grade determination module, which is used to input the acoustic parameters and received waveform signal data of the concrete pile body to be tested into the trained integrity grade determination model, and output the integrity grade of the concrete pile body to be tested; wherein the training of the integrity grade determination model comprises: identifying the corpus of several detection technical regulations corresponding to the integrity detection of the concrete pile body, and obtaining the acoustic parameters and waveform feature description information corresponding to the integrity grade; extracting the determination object and its feature keyword set based on the acoustic parameters and waveform feature description information corresponding to the integrity grade, and counting the fuzzy quantity feature keywords at the same time; quantifying the extracted determination object and its feature keyword set based on the pre-constructed state feature quantization discriminant formula corresponding to the acoustic parameter, the distribution feature quantization discriminant formula corresponding to the acoustic parameter and the trained receiving waveform state feature classification model to form the integrity grade determination model; constructing a training sample set for integrity grade determination, training the integrity grade determination model with the training sample set, and using the fuzzy quantity feature keywords as adjustment parameters to obtain the trained integrity grade determination model.

[0108] The NLP-based intelligent determination system for concrete pile integrity level provided by the embodiment of the present invention adopts Figure 4Several modules in it use natural language processing technology and ultrasonic image recognition technology to solve the technical problems of low efficiency in manual judgment of concrete pile integrity level based on acoustic wave transmission method, inability to simultaneously judge the integrity of all piles to be tested in the same project, and inconsistent judgment results.

[0109] It should be noted that the system embodiment provided by the present invention is used to implement the methods in the above method embodiment as well as the methods in other method embodiments provided by the present invention. The only difference is that the corresponding functional modules are set. The principle is basically the same as the principle of the above system embodiment provided by the present invention. As long as the technical personnel in this field refer to the specific technical solutions in other method embodiments on the basis of the above system embodiment, obtain the corresponding technical means and the technical solutions composed of these technical means by combining the technical features, and on the premise of ensuring the practicality of the technical solutions, improve the modules in the above system embodiment to obtain the corresponding system class embodiments, which are used to implement the methods in other method class embodiments. For example:

[0110] Based on the contents of the above system embodiment, as a preferred embodiment, the NLP-based intelligent determination system for concrete pile integrity level provided in the embodiment of the present invention further includes:

[0111] The extraction objects are determined as follows: acoustic parameters, received waveform, sound velocity range and first wave.

[0112] Based on the contents of the above system embodiment, as a preferred embodiment, in the NLP-based intelligent determination system for concrete pile integrity level provided in the embodiment of the present invention, the pre-construction of the quantitative discrimination formula corresponding to the state characteristics of the acoustic parameters also includes:

[0113] Obtain a historical data set of manually identified integrity levels of concrete pile bodies, and classify and count the quantified values ​​of acoustic parameter state characteristics according to the integrity levels identified in the historical data set;

[0114] According to the classification and statistical results, the characteristic quantitative limit values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD in different states are calculated respectively;

[0115] According to the calculated characteristic quantization limit values ​​of different states, a characteristic quantization discrimination formula corresponding to the state of acoustic parameters is constructed.

[0116] Based on the contents of the above system embodiment, as a preferred embodiment, in the NLP-based intelligent determination system for concrete pile integrity level provided in the embodiment of the present invention, the pre-construction of the quantitative discrimination formula corresponding to the distribution characteristics of the acoustic parameters also includes:

[0117] Classifying and counting the characteristic quantitative values ​​of the distribution state according to the integrity level determined in the historical data set;

[0118] According to the classification statistics, the characteristic quantitative limit values ​​of the longitudinal and transverse distribution directions in different distribution states are calculated respectively;

[0119] According to the calculated characteristic quantization limit values ​​of different distribution states, a discriminant formula for the distribution characteristic quantization value corresponding to the acoustic parameter is constructed.

[0120] Based on the contents of the above system embodiment, as a preferred embodiment, in the NLP-based intelligent determination system for concrete pile integrity level provided in the embodiment of the present invention, the training of the received waveform state feature classification model also includes:

[0121] Obtain historical data of concrete pile inspection;

[0122] Constructing a training sample set and a test sample set according to the historical data, and configuring a classification label for each waveform data in the sample set;

[0123] Construct a receiving waveform state feature classification model, including a CNN model for extracting high-level features of the receiving waveform data and an LSTM model for capturing the timing information of the receiving waveform data;

[0124] The receiving waveform state feature classification model is trained using the training sample set, and tested using the test sample set to obtain a trained receiving waveform state feature classification model.

[0125] Based on the contents of the above system embodiment, as a preferred embodiment, in the NLP-based intelligent determination system for the integrity level of concrete pile body provided in the embodiment of the present invention, the concrete pile body to be tested is detected by using the sound wave transmission method, and the acoustic parameters of the concrete pile body to be tested and the received waveform signal data are obtained, and also includes:

[0126] The concrete pile body to be tested is tested by using the acoustic wave transmission method to obtain the test data set saved in a set format;

[0127] Calculate the characteristic quantization values ​​of the acoustic parameter state of the data set to be tested, including the characteristic quantization values ​​of each calculated based on the real-time values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD;

[0128] Calculating the acoustic parameter distribution characteristic quantization value of the data set to be tested, including the characteristic quantization values ​​of the longitudinal distribution and the transverse distribution respectively calculated based on the acoustic parameter state characteristic quantization value of the data set to be tested;

[0129] All the measured point data of the data set to be tested are converted into time series signals and the corresponding state characteristics are predicted using the trained receiving waveform state feature classification model to obtain the quantized value of the receiving waveform state feature.

[0130] Based on the content of the above system embodiment, as a preferred embodiment, in the NLP-based intelligent determination system for the integrity level of concrete pile body provided in the embodiment of the present invention, the integrity level determination model includes multiple determination levels, each determination level corresponds to one or more determination conditions, and when making level determinations, the determination is made in a manner of decreasing from high level to low level step by step. If a certain condition is met, it is determined to be the integrity level corresponding to this condition.

[0131] The system embodiments described above are merely illustrative, wherein the units described as separate components are or are not physically separated, and the components shown as units are or are not physical units, located in one place, or distributed on multiple network units. Some or all of the modules are selected according to actual conditions to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement without creative effort.

[0132] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0134] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0135] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. The NLP-based intelligent determination method for the integrity level of concrete piles is characterized by: include: Using the acoustic wave transmission method to detect the concrete pile body to be tested, and obtaining the acoustic parameters of the concrete pile body to be tested and receiving waveform signal data; Input the acoustic parameters of the concrete pile body to be tested and the received waveform signal data into the trained integrity level determination model, and output the integrity level of the concrete pile body to be tested; wherein the training of the integrity level determination model includes: Identify the corpus of several testing technical regulations corresponding to the integrity test of concrete piles, and obtain the acoustic parameters and waveform feature description information corresponding to the integrity level; Based on the acoustic parameters and waveform feature description information corresponding to the integrity level, the judgment object and its feature keyword set are extracted, and the fuzzy quantity feature keywords are counted; Based on the pre-built quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained classification model of receiving waveform state characteristics, the extracted judgment objects and their characteristic keyword sets are quantified to form an integrity level judgment model; A training sample set for integrity level determination is constructed, an integrity level determination model is trained with the training sample set, and fuzzy quantitative feature keywords are used as adjustment parameters to obtain a trained integrity level determination model.

2. The NLP-based intelligent determination method for concrete pile integrity level according to claim 1 is characterized in that: The extracted judgment objects include: acoustic parameters, received waveform, sound velocity range and first wave.

3. The NLP-based intelligent determination method for concrete pile integrity level according to claim 1 is characterized in that: The pre-construction of the quantitative discrimination formula of the state characteristics corresponding to the acoustic parameters includes: Obtain a historical data set of manually identified integrity levels of concrete pile bodies, and classify and count the quantified values ​​of acoustic parameter state characteristics according to the integrity levels identified in the historical data set; According to the classification and statistical results, the characteristic quantitative limit values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD in different states are calculated respectively; According to the calculated characteristic quantization limit values ​​of different states, a characteristic quantization discrimination formula corresponding to the state of acoustic parameters is constructed.

4. The NLP-based intelligent determination method for concrete pile integrity level according to claim 3 is characterized in that: The pre-construction of the quantitative discrimination formula corresponding to the distribution characteristics of the acoustic parameters includes: Classifying and counting the characteristic quantitative values ​​of the distribution state according to the integrity level determined in the historical data set; According to the classification statistics, the characteristic quantitative limit values ​​of the longitudinal and transverse distribution directions in different distribution states are calculated respectively; According to the calculated characteristic quantization limit values ​​of different distribution states, a discriminant formula for the distribution characteristic quantization value corresponding to the acoustic parameter is constructed.

5. The NLP-based intelligent determination method for concrete pile integrity level according to claim 1 is characterized in that: The training of the receiving waveform state feature classification model includes: Obtain historical data of concrete pile inspection; Constructing a training sample set and a test sample set according to the historical data, and configuring a classification label for each waveform data in the sample set; Construct a receiving waveform state feature classification model, including a CNN model for extracting high-level features of the receiving waveform data and an LSTM model for capturing the timing information of the receiving waveform data; The receiving waveform state feature classification model is trained using the training sample set, and tested using the test sample set to obtain a trained receiving waveform state feature classification model.

6. The NLP-based intelligent determination method for concrete pile integrity level according to claim 1 is characterized in that: The acoustic wave transmission method is used to detect the concrete pile body to be tested, and the acoustic parameters of the concrete pile body to be tested and the received waveform signal data are obtained, including: The concrete pile body to be tested is tested by using the acoustic wave transmission method to obtain the test data set saved in a set format; Calculate the characteristic quantization values ​​of the acoustic parameter state of the data set to be tested, including the characteristic quantization values ​​of each calculated based on the real-time values ​​of sound velocity, sound amplitude, sound time, main frequency and PSD; Calculating the acoustic parameter distribution characteristic quantization value of the data set to be tested, including the characteristic quantization values ​​of the longitudinal distribution and the transverse distribution respectively calculated based on the acoustic parameter state characteristic quantization value of the data set to be tested; All the measured point data of the data set to be tested are converted into time series signals and the corresponding state characteristics are predicted using the trained receiving waveform state feature classification model to obtain the quantized value of the receiving waveform state feature.

7. The NLP-based intelligent determination method for concrete pile integrity level according to claim 1 is characterized in that: The integrity level determination model includes multiple determination levels, each determination level corresponds to one or more determination conditions. When making level determinations, the determination is made in a manner of decreasing from high level to low level. If a certain condition is met, the integrity level corresponding to the condition is determined. Among them, one or more of the determination conditions include: a quantitative determination formula for state characteristics corresponding to acoustic parameters, a quantitative determination formula for distribution characteristics corresponding to acoustic parameters, a trained classification model for receiving waveform state characteristics, and a logical relationship between various acoustic parameters in the detection technical regulations corresponding to the concrete pile body to be tested.

8. The intelligent determination system of concrete pile integrity level based on NLP is characterized by: include: The data acquisition module is used to detect the concrete pile body to be tested by using the acoustic wave transmission method, and obtain the acoustic parameters of the concrete pile body to be tested and receive waveform signal data; The grade determination module is used to input the acoustic parameters of the concrete pile body to be tested and the received waveform signal data into the trained integrity grade determination model, and output the integrity grade of the concrete pile body to be tested; wherein the training of the integrity grade determination model includes: Identify the corpus of several testing technical regulations corresponding to the integrity test of concrete piles, and obtain the acoustic parameters and waveform feature description information corresponding to the integrity level; Based on the acoustic parameters and waveform feature description information corresponding to the integrity level, the judgment object and its feature keyword set are extracted, and the fuzzy quantity feature keywords are counted; Based on the pre-built quantitative discrimination formula of state characteristics corresponding to acoustic parameters, the quantitative discrimination formula of distribution characteristics corresponding to acoustic parameters and the trained classification model of receiving waveform state characteristics, the extracted judgment objects and their characteristic keyword sets are quantified to form an integrity level judgment model; A training sample set for integrity level determination is constructed, an integrity level determination model is trained with the training sample set, and fuzzy quantitative feature keywords are used as adjustment parameters to obtain a trained integrity level determination model.

Images