CN113076693A - Road surface compaction quality evaluation method based on support vector machine and hidden horse model - Google Patents
Road surface compaction quality evaluation method based on support vector machine and hidden horse model Download PDFInfo
- Publication number
- CN113076693A CN113076693A CN202110361962.XA CN202110361962A CN113076693A CN 113076693 A CN113076693 A CN 113076693A CN 202110361962 A CN202110361962 A CN 202110361962A CN 113076693 A CN113076693 A CN 113076693A
- Authority
- CN
- China
- Prior art keywords
- compaction
- data
- support vector
- vector machine
- algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005056 compaction Methods 0.000 title claims abstract description 172
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 36
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 26
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 41
- 238000012544 monitoring process Methods 0.000 claims abstract description 39
- 238000010276 construction Methods 0.000 claims abstract description 25
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000007476 Maximum Likelihood Methods 0.000 claims abstract description 10
- 230000008859 change Effects 0.000 claims abstract description 9
- 238000011156 evaluation Methods 0.000 claims abstract description 4
- 238000005096 rolling process Methods 0.000 claims description 37
- 238000001514 detection method Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000010801 machine learning Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000007704 transition Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000007635 classification algorithm Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 abstract description 2
- 239000000463 material Substances 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000001066 destructive effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 241000701022 Cytomegalovirus Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000011437 continuous method Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Abstract
The invention relates to a road surface compaction quality evaluation method based on a support vector machine and a hidden horse model, and belongs to the field of traffic infrastructure intelligent construction. The method comprises the steps of acquiring compaction monitoring parameters through a vehicle-mounted sensor and a UWB/GPS combined positioning system; identifying a training sample through an RTK-GPS result; carrying out single-point compaction state classification on the identified training samples by adopting a fuzzy support vector machine algorithm; the hidden horse model parameters are calculated by adopting a maximum likelihood estimation algorithm to describe the compaction quality change rule, and the overall compaction quality evaluation result is obtained by combining the compaction monitoring data. According to the method, membership degree solving is performed on the sample data through the design and application of the fuzzy C-means algorithm, so that the anti-noise capability of the support vector machine algorithm is improved; through the solving and application of the hidden horse model parameters, the change process of the road surface compaction quality is effectively obtained; and the evaluation of the whole process and the whole area pavement compaction quality is realized by combining the application of the combined positioning system.
Description
Technical Field
The invention relates to a road surface compaction quality evaluation method based on a support vector machine and a hidden horse model, and belongs to the field of traffic infrastructure intelligent construction.
Background
Compaction is a key link for the dense forming of pavement materials and the realization of pavement functions, and directly influences the strength, stability and fatigue resistance of pavements. Therefore, the quality control of compaction must be emphasized and enhanced in road construction. At present, pavement compaction quality management is mainly based on post-inspection, and the compaction condition is difficult to know in time and process control is difficult to realize. Thus, intelligent compaction enabling continuous real-time non-destructive monitoring and feedback is of increasing interest, where real-time accurate assessment of road compaction quality is a prerequisite for intelligent compaction feedback control.
In the quality control of pavement compaction construction, the detection of the compaction condition of pavement materials is very important. In practical applications, the compaction quality is usually described by using the degree of compaction, and the detection method comprises a traditional method, an online method, a continuous method and the like.
The traditional compaction quality detection method is simple and easy to implement, but the traditional compaction quality detection method is only limited in result detection, does not have sufficient representativeness, and is difficult to reflect the compaction state in real time. In addition, the cutting ring method, the sand filling method, the core drilling method and the like belong to destructive detection methods, so that the operation is complicated, and simultaneously, a large amount of manpower and material resources are consumed, so that the real-time nondestructive online detection technology becomes an urgent need for pavement construction.
The on-line method avoids the defects of the traditional detection method and can carry out rapid nondestructive detection on the road surface material. However, the on-line method also belongs to a sampling detection and result control method, and the detection result has one-sidedness and hysteresis, and an under-voltage area and an over-voltage area cannot be found in time, so that the requirement of process control is difficult to meet.
In order to control the compaction quality, real-time online global monitoring is required in rolling construction, so that a continuous compaction detection method is developed. There are various ways to describe the compaction state, such as vibration wheel acceleration, vibration wheel excitation force and displacement, mechanical driving power, rayleigh wave velocity, etc. In view of the factors that the vibration feedback signal is easy to collect, the vibration signal collecting and processing device is easy to integrate with construction equipment and the like, the compactness indexes such as CMV (cytomegalovirus), CCV (CCV) and the like based on the harmonic analysis method have more attention, but the defects of single type of applicable materials, general correlation degree with the results of the conventional compaction detection method and the like still exist, and further intensive research and optimization are needed.
In conclusion, how to combine the continuous compaction monitoring data with the requirement of pavement compaction quality evaluation is an important task for pavement intelligent compaction construction to overcome the defects of the existing pavement compaction quality detection method.
Disclosure of Invention
Aiming at the defects of the traditional compaction detection method, the online compaction detection method and the continuous compaction detection method, the invention provides the road surface compaction quality evaluation method based on the support vector machine and the hidden horse model, and the advantages of the statistical learning method are combined with the requirements of the road surface compaction quality detection to realize the continuous real-time lossless road surface compaction quality evaluation.
The invention adopts the following technical scheme:
the road surface compaction quality evaluation method based on the support vector machine and the hidden horse model comprises the following steps:
acquiring compaction monitoring parameters through a vehicle-mounted sensor and a UWB/GPS combined positioning system; UWB/GPS is Ultra Wide Band (UWB)/Global Positioning System (GPS);
secondly, sample data preprocessing is carried out on compaction monitoring data acquired through a UWB/GPS combined positioning system in the first step, and a training sample is identified through an RTK-GPS result;
step three, single-point compaction state classification: learning and classifying the RTK-GPS result identification training samples in the second step by adopting a fuzzy support vector machine algorithm;
step four, evaluating the integral compaction quality: and calculating hidden horse model parameters by adopting a maximum likelihood estimation algorithm through the classified training samples to describe the compaction quality change rule, and obtaining the overall compaction quality evaluation by combining with the compaction monitoring data.
The invention relates to a pavement compaction quality evaluation method based on a support vector machine and a hidden horse model, wherein in the first step, compaction monitoring parameters comprise a vibration parameter index, a construction temperature parameter and a rolling positioning parameter; the Vibration parameter index includes a Compaction Meter Value (CMV), a Compaction Control Value (CCV), and a Vibration Compaction Energy Value (VCve).
The CMV index is defined as follows:
in the above formula, A2ΩIs the frequency domain amplitude of the second harmonic of the vibration acceleration signal, AΩThe frequency domain amplitude of the fundamental wave, Cal is a calibration coefficient, which is generally set to 300, and calibration needs to be performed by a conventional detection method in practical application.
The CCV index is defined as follows:
in the above formula, A0.5ΩAmplitude in the frequency domain of subharmonics of the vibration feedback signal, A1.5ΩAnd A2.5ΩThe frequency domain amplitude of the inter-harmonic, other harmonic components and the CMV index, Cal is a calibration coefficient which is generally set as 100, and calibration needs to be carried out by a conventional detection method in practical application.
The VCVe index is defined as follows:
In the above formula, AnΩThe amplitude of the N-th harmonic of the acceleration signal in the frequency domain is determined, the size of N depends on the highest-frequency harmonic in the frequency domain, Cal is a calibration coefficient, which is generally set to 100, and calibration needs to be performed by a detection method in practical application.
In the second step, the process of preprocessing the sample data of compaction monitoring is as follows:
The invention relates to a pavement compaction quality evaluation method based on a support vector machine and a hidden horse model, which adopts a fuzzy support vector machine algorithm to carry out machine learning classification and comprises the following steps:
the kernel function selection is a necessary link for supporting the application of the vector machine algorithm, and is a loop of the structural components and the operation process of the vector machine algorithm.
And 3, performing machine learning classification and identification on the result data obtained in the step 1 after the preliminary clustering classification by adopting the fuzzy C mean algorithm by using the support vector machine algorithm obtained in the step 2, and determining the single-point compaction state category of the road surface compaction.
5. The method for evaluating the compaction quality of a road surface based on a support vector machine and a hidden horse model according to claim 1, characterized in that: in the fourth step, the hidden horse model parameters are calculated based on the maximum likelihood estimation algorithm to describe the compaction quality change rule, and the process of evaluating the overall compaction quality by combining the compaction monitoring data is as follows:
and 3, calculating hidden horse model parameters through a maximum likelihood estimation algorithm, and determining a hidden horse model lambda (A, B, pi) of the road compaction quality state, wherein A is a hidden state transition probability matrix, B is an observation state transition probability matrix, and pi is an initial state probability vector.
and 5, according to the result in the step 3, combining the result of the UWB/GPS combined positioning system, realizing the whole process of road surface rolling construction and whole area compaction quality evaluation, and providing a decision basis for intelligent compaction feedback control.
Advantageous effects
The invention adapts to the development trend of intelligent detection of the pavement compaction quality, combines the advantages of statistical machine learning and the requirement of pavement compaction quality evaluation, and provides a pavement compaction quality evaluation method based on a support vector machine and a hidden horse model. By fusing multi-source multi-dimensional compaction monitoring parameters and data preprocessing, a good data base is provided for compaction state classification and quality evaluation, and the dependency of a compaction quality evaluation result on a single compaction monitoring parameter type is avoided; the membership degree of the sample data is solved through the design and application of the fuzzy C-means algorithm, so that the anti-noise capability of the support vector machine algorithm is improved; the variation process of the pavement compaction quality is effectively obtained by solving the hidden horse model parameters and describing the variation rule of the compaction state sequence by application; and the evaluation of the whole process and the whole area pavement compaction quality is realized by combining the application of the combined positioning system.
Drawings
Fig. 1 is an overall work flow chart of a road surface compaction quality evaluation method based on a support vector machine and a hidden horse model, which is implemented by the invention.
FIG. 2 is a distribution diagram of training samples of the fuzzy support vector machine.
Fig. 3 is a hidden horse model diagram of a compaction quality sequence.
Fig. 4 is a flow chart of road surface compaction quality evaluation based on a hidden horse model.
Detailed Description
In order to make the purpose and technical solution of the embodiments of the present invention clearer, the technical solution of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention without any inventive step, are within the scope of protection of the invention.
As shown in fig. 1, the method for evaluating the compaction quality of a road surface based on a support vector machine and a hidden horse model comprises the following steps:
(1) acquiring compaction monitoring parameters, acquiring signals in the road surface rolling construction process, and extracting monitoring parameters;
(2) sample data is preprocessed, multidimensional compaction monitoring data are subjected to fusion analysis, and a training sample is identified through an RTK-GPS result;
(3) classifying single-point compaction states, and performing machine learning classification by adopting a fuzzy support vector machine algorithm on the basis of the step (2);
(4) and (4) evaluating the overall compaction quality, namely calculating hidden horse model parameters based on a maximum likelihood estimation algorithm on the basis of the step (3) to describe a compaction quality change rule, and then evaluating the overall compaction quality by combining compaction monitoring data.
In the step (1), in the road rolling construction process, various compaction monitoring signals are collected through a vehicle-mounted sensor and a positioning system, and compaction monitoring parameters are obtained through digital signal processing and analysis, wherein the compaction monitoring parameters mainly comprise vibration parameter indexes (CMV, CCV and VCV)e) Construction temperature parameters, rolling positioning parameters and the like.
In the step (2), the process of preprocessing the sample data of compaction monitoring comprises the following steps:
(2-1) vibration data cleaning: setting a time sliding window according to the fundamental frequency of the vibration compaction feedback signal, and acquiring multiple groups of CMV, CCV and VCV by taking 1 second as a detection periodeIndexes; then, with 3 sigma as a selection basis, after abnormal data is eliminated, the average value of the compactness index is obtained as the vibration parameter characteristic. In addition, abnormal vibration data are removed through initial stage discrimination, edge joint discrimination and static pressure vibration and pressure separation algorithm respectively, wherein vibration parameter indexes are defined as follows:
in the formula (1), A2ΩIs the frequency domain amplitude of the second harmonic of the vibration acceleration signal, AΩIs the frequency domain amplitude of the fundamental wave. Cal is a calibration coefficient which is usually set as 300, and calibration needs to be carried out by a conventional detection method in practical application.
In the formula (2), A0.5ΩAmplitude in the frequency domain of subharmonics of the vibration feedback signal, A1.5ΩAnd A2.5ΩThe frequency domain amplitude of the inter-harmonic, and other harmonic components are the same as the CMV index. Cal is a calibration coefficientUsually, the initial value is 100, and the calibration needs to be carried out by a conventional detection method in practical application.
In the formula (2), AnΩThe magnitude of the N-th harmonic of the acceleration signal in the frequency domain depends on the highest frequency harmonic in the frequency domain. Cal is a calibration coefficient which is generally set as 100, and calibration needs to be carried out through a detection method in practical application.
(2-2) temperature data processing: and (3) processing abnormal temperature data caused by factors such as construction equipment structure, site environment and wind power based on an algorithm of a 3 sigma criterion.
(2-3) rolling times extraction: on the basis of accurately positioning the rolling position of the road roller, a grid map technology is applied to extract the rolling times.
(2-4) sample data identification: firstly, extracting accurate rolling times through an RTK-GPS calibration system; then, calculating compaction degree calibration data in the rolling process according to a logarithmic relation rule of the pavement compaction degree and the rolling times; and finally, judging the compaction state according to the compaction degree acceptance standard, and respectively defining the grade to further identify the sample data. For the simplest two states of loosening and compacting, the compaction state data samples may be defined as-1 and +1, respectively.
In the step (3), the process of machine learning classification by adopting the fuzzy support vector machine algorithm comprises the following steps:
(3-1) determining the membership degree of the compaction monitoring sample data obtained in the step (2) by using a fuzzy C mean algorithm, and further completing fuzzy clustering division of a training sample set;
(3-2) selecting a proper kernel function, determining a penalty factor C value and a kernel function parameter gamma value, and designing a support vector machine algorithm based on the kernel function;
and (3-3) performing machine learning classification and identification on the sample data obtained in the step (3-1) by using the support vector machine algorithm obtained in the step (3-2), and determining the single-point compaction state category of the road surface compaction.
In the step (4), on the basis of the step (3), calculating hidden horse model parameters based on a maximum likelihood estimation algorithm to describe a compaction quality change rule, and then performing overall compaction quality evaluation by combining compaction monitoring data:
(4-1) directly acquiring the initial compaction state of the road surface in the rolling construction, namely obtaining an initial probability vector pi; acquiring a series of compaction monitoring data through the step (1) in the rolling process, and acquiring classification of single-point compaction states through the step (2) and the step (3), namely an observation sequence;
and (4-2) acquiring accurate rolling times by adopting an RTK-GPS high-precision positioning system in a field test, and then reversely deducing a compaction state as an implicit state sequence through a logarithmic relation between the compaction degree and the compaction times. The RTK-GPS system is only used as a calibration means at the initial construction period and is not used for monitoring the whole construction process;
(4-3) calculating hidden horse model parameters through a maximum likelihood estimation algorithm, and determining a hidden horse model lambda (A, B, pi) of the road surface compaction quality state change;
(4-4) taking the classification result obtained in the step (3) as a compaction state observation sequence O, and obtaining an implicit compaction state sequence of rolling construction by using a Viterbi algorithm by combining the hidden horse model parameters obtained in the step (4-3);
and (4-5) according to the results of the step (4-4), combining the results of the combined positioning system, realizing the whole process of pavement rolling construction and whole-area compaction quality evaluation, and providing decision basis for intelligent compaction feedback control.
Fig. 2 is a distribution diagram of input fuzzy support vector machine training samples, fig. 3 is a hidden horse model diagram describing compaction quality changes, and fig. 4 is a flow chart of pavement compaction quality evaluation based on the hidden horse model. The image shows that the invention can effectively inhibit the noise and the interference of the wild points in the sample data, and can realize the continuous real-time nondestructive monitoring and evaluation of the pavement compaction quality by combining the application of the combined positioning system.
Although the technical means are illustrated by the above embodiments, the technical means are only described for clearly showing the details of the invention, and the technical means can be arbitrarily combined or replaced to form the technical scheme. Those skilled in the art may make modifications, combinations, or alterations to the present invention without departing from the principles of the invention, to form other embodiments suitable for use in the art.
Claims (5)
1. The road surface compaction quality evaluation method based on the support vector machine and the hidden horse model is characterized by comprising the following steps: the evaluation procedure was as follows:
acquiring compaction monitoring parameters through a vehicle-mounted sensor and a UWB/GPS combined positioning system;
secondly, sample data preprocessing is carried out on compaction monitoring data acquired through a UWB/GPS combined positioning system in the first step, and a training sample is identified through an RTK-GPS result;
step three, single-point compaction state classification: learning and classifying the training samples identified by the RTK-GPS result in the second step by adopting a fuzzy support vector machine algorithm;
step four, evaluating the integral compaction quality: and for the classified training samples, calculating hidden horse model parameters by adopting a maximum likelihood estimation algorithm to describe compaction quality change rules, and obtaining an overall compaction quality evaluation result by combining compaction monitoring data.
2. The method for evaluating the compaction quality of a road surface based on a support vector machine and a hidden horse model according to claim 1, characterized in that: in the first step, compaction monitoring parameters comprise vibration parameter indexes, construction temperature parameters and rolling positioning parameters, wherein the vibration parameter indexes comprise a compaction value CMV, a compaction control value CCV and a vibration compaction energy value VCve;
the CMV index is defined as follows:
in the above formula, A2ΩIs the frequency domain amplitude of the second harmonic of the vibration acceleration signal, AΩThe frequency domain amplitude of the fundamental wave, Cal is a calibration coefficient which is initially set to 300;
the CCV index is defined as follows:
in the above formula, A0.5ΩAmplitude in the frequency domain of subharmonics of the vibration feedback signal, A1.5ΩAnd A2.5ΩAmplitude in the frequency domain of the inter-harmonics, A3ΩThe frequency domain amplitude is the third harmonic of the vibration acceleration signal; cal is a calibration coefficient, and the initial setting is 100;
the VCVe index is defined as follows:
In the above formula, AnΩThe amplitude of the N-th harmonic of the acceleration signal in the frequency domain is determined, the size of N depends on the highest-frequency harmonic in the frequency domain, Cal is a calibration coefficient which is initially set to 100, and calibration needs to be carried out through a detection method in practical application.
3. The road surface compaction quality evaluation method based on the support vector machine and the hidden horse model according to claim 1 or 2, characterized in that: in the second step, the process of preprocessing the sample data of compaction monitoring comprises the following steps:
step 1, cleaning vibration data: setting a time sliding window according to the fundamental frequency of the vibration compaction feedback signal, and acquiring a plurality of groups of CMV, CCV and VCve data based on the first step; in order to eliminate coarse error data, abnormal data is eliminated according to a 3 sigma criterion, and the average value of the compactness index is obtained as the vibration parameter characteristic;
step 2, temperature data processing: processing the abnormal temperature data by an algorithm based on a 3 sigma criterion aiming at the abnormal temperature data;
step 3, rolling times extraction: accurately positioning a rolling position, and extracting rolling times by using a grid map technology;
step 4, sample data identification: calculating to obtain compaction degree calibration data in the rolling process according to the rolling times obtained in the third step; and judging the compaction state according to the compaction degree acceptance standard, and respectively defining the grade to further identify the sample data.
4. The method for evaluating the compaction quality of a road surface based on a support vector machine and a hidden horse model according to claim 1, characterized in that: the process of machine learning classification by adopting the fuzzy support vector machine algorithm comprises the following steps:
step 1, determining the membership degree of sample data by adopting a fuzzy C-means algorithm for compaction monitoring sample data, and further completing the fuzzy clustering preliminary division of training compaction monitoring sample data to be used as basic data of a classification algorithm for a compaction state of a support vector machine;
step 2, selecting a proper kernel function for classifying the sample data according to the characteristics of compaction monitoring sample data and the actual requirements of compaction classification, determining a penalty factor C value and a kernel function parameter gamma value, and designing a support vector machine algorithm based on the kernel function;
and 3, performing machine learning classification and identification on the result data obtained after the preliminary clustering classification in the step 1 by adopting the fuzzy C mean algorithm by using the support vector machine algorithm obtained in the step 2, and determining the single-point compaction state category of the road surface compaction.
5. The method for evaluating the compaction quality of a road surface based on a support vector machine and a hidden horse model according to claim 1, characterized in that: in the fourth step, the hidden horse model parameters are calculated based on the maximum likelihood estimation algorithm to describe the compaction quality change rule, and the process of evaluating the overall compaction quality by combining the compaction monitoring data is as follows:
step 1, directly obtaining an initial compaction state of a road surface in rolling construction, namely obtaining an initial probability vector pi; in the rolling process, acquiring a series of compaction monitoring data through the first step, and acquiring classification results of single-point compaction states through the second step and the third step, namely an observation sequence;
step 2, obtaining accurate rolling times according to a field test, and further reversely deducing a compaction state as an implicit state sequence through a logarithmic relation between the compaction degree and the compaction times;
and 3, calculating hidden horse model parameters through a maximum likelihood estimation algorithm, and determining a hidden horse model lambda (A, B, pi) of the road compaction quality state, wherein A is a hidden state transition probability matrix, B is an observation state transition probability matrix, and pi is an initial state probability vector.
Step 4, taking the classification result of the step three as a compaction state observation sequence O, obtaining hidden horse model parameters by combining the step 3, and calculating by using a Viterbi algorithm to obtain a hidden compaction state sequence of rolling construction;
and 5, according to the result in the step 4, combining the result of the UWB/GPS combined positioning system, realizing the whole process of road surface rolling construction and whole area compaction quality evaluation, and providing a decision basis for intelligent compaction feedback control.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110361962.XA CN113076693A (en) | 2021-04-02 | 2021-04-02 | Road surface compaction quality evaluation method based on support vector machine and hidden horse model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110361962.XA CN113076693A (en) | 2021-04-02 | 2021-04-02 | Road surface compaction quality evaluation method based on support vector machine and hidden horse model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113076693A true CN113076693A (en) | 2021-07-06 |
Family
ID=76615040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110361962.XA Pending CN113076693A (en) | 2021-04-02 | 2021-04-02 | Road surface compaction quality evaluation method based on support vector machine and hidden horse model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113076693A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114908726A (en) * | 2022-05-16 | 2022-08-16 | 广西邕洲高速公路有限公司 | Real-time global intelligent detection method and system for construction quality of highway broad-width roadbed |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885867A (en) * | 2014-03-28 | 2014-06-25 | 渤海大学 | Online evaluation method of performance of analog circuit |
| CN110598902A (en) * | 2019-08-02 | 2019-12-20 | 浙江工业大学 | Water quality prediction method based on combination of support vector machine and KNN |
| US20200380185A1 (en) * | 2016-06-30 | 2020-12-03 | Bridgestone Corporation | Road surface state determination method |
-
2021
- 2021-04-02 CN CN202110361962.XA patent/CN113076693A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885867A (en) * | 2014-03-28 | 2014-06-25 | 渤海大学 | Online evaluation method of performance of analog circuit |
| US20200380185A1 (en) * | 2016-06-30 | 2020-12-03 | Bridgestone Corporation | Road surface state determination method |
| CN110598902A (en) * | 2019-08-02 | 2019-12-20 | 浙江工业大学 | Water quality prediction method based on combination of support vector machine and KNN |
Non-Patent Citations (1)
| Title |
|---|
| 崔华;李建成;郭璐;宋焕生;: "沥青路面碾压质量的支持向量机评价方法", 公路, no. 09, 25 September 2015 (2015-09-25) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114908726A (en) * | 2022-05-16 | 2022-08-16 | 广西邕洲高速公路有限公司 | Real-time global intelligent detection method and system for construction quality of highway broad-width roadbed |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104899437B (en) | Heavy rainfall type landslide disaster method for early warning | |
| CN108932480A (en) | The study of distributing optical fiber sensing signal characteristic and classification method based on 1D-CNN | |
| CN108595414B (en) | Soil heavy metal enterprise pollution source identification method based on source-sink space variable reasoning | |
| CN108960622A (en) | A kind of Assessment method of the reclaimed land in mining area quality based on remote sensing image | |
| CN105426646A (en) | Bridge intelligent degree comprehensive evaluation index selection and quantification method | |
| CN116363524A (en) | Monitoring and early warning method and device for highway geological disaster high-risk area | |
| CN113076693A (en) | Road surface compaction quality evaluation method based on support vector machine and hidden horse model | |
| Jebali et al. | Performance evaluation of detector algorithms of Dust storms in arid lands (Case study: Yazd province) | |
| Kalpoma et al. | Satellite image database creation for road quality measurement of national highways of bangladesh | |
| CN108573285A (en) | A kind of experimental parameter clustering method towards dredging energy consumption | |
| Liu et al. | Pavement surface defect recognition method based on vehicle system vibration data and feedforward neural network | |
| CN109060858A (en) | A kind of quantitative evaluation method of heavy metal-polluted soil concentration space forecast of distribution accuracy | |
| CN115438547A (en) | Overall evaluation method and system based on pavement service state | |
| Renschler et al. | Implementing a process-based decision support tool for natural resource management-the GeoWEPP example | |
| Ribeiro Junior et al. | On the Use of Machine Learning for Damage Assessment in Composite Structures: A Review | |
| CN112926260A (en) | Novel method for accurately estimating physical and mechanical parameters of rock and soil material | |
| Kodikara | Adaptive Fuzzy Inference System for Automated Pavement Condition Evaluation of Large Pavement Sections from Ground Penetrating Radar (GPR) Thickness Data | |
| CN112349098A (en) | Method for estimating accident severity by environmental elements in exit ramp area of expressway | |
| Chongchong et al. | A bridge structural health data analysis model based on semi-supervised learning | |
| Wen | Construction project risk evaluation based on rough sets and artificial neural networks | |
| CN117272840B (en) | Expressway bad weather early warning method and system | |
| Singh et al. | Adaptive Fuzzy Inference System for Automated Pavement Condition Evaluation of Large Pavement Sections from Ground Penetrating Radar (GPR) Thickness Data | |
| de Almeida Cardoso et al. | An enhanced approach for automatic modal identification of structures | |
| CN113420387B (en) | Migration diagnosis method and system for rolling bearing of compacting machine | |
| Poudel | Study of retrofitted system for Intelligent Compaction Analyzer, a machine learning approach for Quality Control of Asphalt Pavement during Construction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |