Roadbed compaction problem diagnosis method, device, electronic equipment and storage medium

Abstract

The present application discloses a method, device, electronic device, and storage medium for diagnosing roadbed compaction problems. The method comprises: obtaining roadbed compaction detection data samples and roadbed problem type data samples; establishing an initial Bayesian classification model; inputting the roadbed compaction detection data samples into the initial Bayesian classification model, outputting corresponding roadbed problem type data, training the initial Bayesian classification model, and obtaining a fully trained Bayesian classification model; and diagnosing roadbed compaction problems based on the fully trained Bayesian classification model. By constructing a Bayesian classification model for diagnosing roadbed compaction problems, not only is the detection time shortened, roadbed compaction problems can be diagnosed in a timely manner, and because the learning process of the Bayesian classification model is a process of continuous improvement, the final diagnostic accuracy of the roadbed compaction problem can be guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of highway quality detection, and in particular to a method, device, electronic equipment and storage medium for diagnosing roadbed compaction problems. Background Art

[0002] High-quality roadbed engineering is an important guarantee for the safe and durable service of highway pavement. The quality of roadbed engineering depends to a large extent on the compaction quality of roadbed filling materials. The quality of roadbed compaction is directly related to the construction quality and operational safety of highway engineering.

[0003] At present, the sand injection method, ring knife method and other testing methods are often used in the domestic highway subgrade compaction quality inspection. These methods generally have disadvantages such as long detection time, low accuracy, complex instrument and equipment operation, and high detection cost. They cannot make a comprehensive, rapid and accurate evaluation of the project quality, and it is difficult to keep up with the speed of highway construction, affecting the construction progress.

[0004] Therefore, in the prior art, when testing the roadbed compaction quality, there is a problem that it is difficult to diagnose the roadbed compaction problem in a timely manner due to the long testing time and low testing accuracy. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, device, electronic equipment and storage medium for diagnosing roadbed compaction problems to solve the technical problems in the existing technology when detecting roadbed compaction quality, such as long detection time and low detection accuracy, resulting in uneven product quality and difficulty in ensuring production speed.

[0006] In order to solve the above problems, the present invention provides a method for diagnosing roadbed compaction problems, comprising:

[0007] Obtain roadbed compaction test data samples and roadbed problem type data samples;

[0008] Establish an initial Bayesian classification model;

[0009] Input the roadbed compaction detection data samples into the initial Bayesian classification model, output the corresponding roadbed problem type data, train the initial Bayesian classification model, and obtain a fully trained Bayesian classification model;

[0010] Diagnose roadbed compaction problems based on a well-trained Bayesian classification model.

[0011] Furthermore, the roadbed compaction test data includes: compaction layer thickness, bearing capacity of the lower structural layer, rolling influence depth, particle grading, filler type, moisture content, rolling speed, compaction section, temperature, humidity and measuring instrument operating status parameters.

[0012] Furthermore, the roadbed problem type data includes: insufficient roadbed compaction data, uneven roadbed section data, loose roadbed surface data, and external cause data.

[0013] Furthermore, the roadbed compaction detection data is input into the initial Bayesian classification model, and the corresponding roadbed problem type data is output, including:

[0014] Obtain the detection prior probability of roadbed compaction detection data and the problem prior probability of roadbed problem type data;

[0015] Inputting the detection prior probability, the problem prior probability, the roadbed compaction detection data, and the roadbed problem type data into the initial Bayesian classification model to determine the problem detection probability, wherein the problem detection probability includes at least one;

[0016] According to the problem inspection probability, the corresponding roadbed problem type data is output.

[0017] Furthermore, obtaining the detection prior probability of the roadbed compaction detection data and the problem prior probability of the roadbed problem type data includes:

[0018] Determine the likelihood probability of the subgrade compaction detection data / subgrade problem type data based on the subgrade compaction detection data / subgrade problem type data;

[0019] According to the likelihood probability of the roadbed compaction detection data / roadbed problem type data, the prior probability of the roadbed compaction detection data / roadbed problem type data is determined.

[0020] Furthermore, the formula for determining the probability of problem detection is:

[0021]

[0022] Among them, p(c|x) is the problem detection probability, c is the set of roadbed problem type data {c1, c2, ..., c n}, x is a sample, containing m attributes {x1, x2, …, xm}, corresponding to the parameters of the base compaction test data, p(c) is the prior probability of class label c, p(x|c) is the class conditional probability of sample x relative to class label c, corresponding to the probability that the relevant parameters show a certain state under the roadbed quality problem; p(x) is the prior probability of sample x.

[0023] Furthermore, according to the problem detection probability, the corresponding roadbed problem type data is output, including:

[0024] According to the problem inspection probability, determine the maximum value of the problem inspection probability;

[0025] The roadbed problem type data corresponding to the maximum problem test probability is output, which is the roadbed compaction problem diagnosis result.

[0026] In order to solve the above problems, the present invention further provides a roadbed compaction problem diagnosis device, comprising:

[0027] A data acquisition module is used to obtain roadbed compaction detection data samples and roadbed problem type data samples;

[0028] Model building module, used to build the initial Bayesian classification model;

[0029] The model training module is used to input the roadbed compaction detection data samples into the initial Bayesian classification model, output the corresponding roadbed problem type data, train the initial Bayesian classification model, and obtain a fully trained Bayesian classification model;

[0030] The problem diagnosis module is used to diagnose roadbed compaction problems based on a well-trained Bayesian classification model.

[0031] In order to solve the above problems, the present invention also provides an electronic device, including a processor and a memory, wherein a computer program is stored in the memory. When the computer program is executed by the processor, the roadbed compaction problem diagnosis method described above is implemented.

[0032] In order to solve the above problems, the present invention also provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a computer, the computer executes the roadbed compaction problem diagnosis method as described above.

[0033] The beneficial effects of adopting the above technical solution are as follows: the present invention provides a method, device, electronic device and storage medium for diagnosing roadbed compaction problems, the method comprising: obtaining roadbed compaction detection data samples and roadbed problem type data samples; establishing an initial Bayesian classification model; inputting the roadbed compaction detection data samples into the initial Bayesian classification model, outputting the corresponding roadbed problem type data, training the initial Bayesian classification model, and obtaining a fully trained Bayesian classification model; and diagnosing the roadbed compaction problem based on the fully trained Bayesian classification model. By constructing a Bayesian classification model for diagnosing roadbed compaction problems, not only is the detection time shortened, the roadbed compaction problem can be diagnosed in a timely manner, and because the learning process of the Bayesian classification model is a process of continuous improvement, the final diagnostic accuracy of the roadbed compaction problem can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic flow chart of an embodiment of a method for diagnosing roadbed compaction problems provided by the present invention;

[0035] Figure 2 A schematic diagram of a process for training an initial Bayesian classification model according to an embodiment of the present invention;

[0036] Figure 3 A schematic structural diagram of an embodiment of a device for diagnosing roadbed compaction problems provided by the present invention;

[0037] Figure 4 This is a structural block diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0038] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, and are not used to limit the scope of the present invention.

[0039] Before describing the embodiment, the Bayesian classifier is first described:

[0040] The Bayesian classifier is the one with the lowest probability of misclassification, or the lowest average risk given a predetermined cost. Its design is a fundamental statistical classification method. Its classification principle uses the Bayesian formula to calculate the posterior probability (i.e., the probability that the object belongs to a certain class) based on the prior probability of an object. The class with the highest posterior probability is then selected as the class to which the object belongs.

[0041] At present, in order to timely detect the quality of roadbed compaction, testing methods such as sand injection method and ring knife method are generally used. However, these methods have problems such as long detection time, high destructiveness, and low accuracy, making it difficult to ensure timely diagnosis of roadbed compaction problems.

[0042] Therefore, when testing the roadbed compaction quality in the prior art, there is a problem that it is difficult to diagnose the roadbed compaction problem in a timely manner due to the long testing time and low testing accuracy.

[0043] In order to solve the above problems, the present invention provides a method, device, electronic device and storage medium for diagnosing roadbed compaction problems, which are described in detail below.

[0044] like Figure 1 As shown, Figure 1 A schematic flow chart of an embodiment of a method for diagnosing roadbed compaction problems provided by the present invention includes:

[0045] Step S101: Acquire roadbed compaction detection data samples and roadbed problem type data samples.

[0046] Step S102: Establish an initial Bayesian classification model.

[0047] Step S103: inputting the roadbed compaction detection data samples into the initial Bayesian classification model, outputting the corresponding roadbed problem type data samples, training the initial Bayesian classification model, and obtaining a fully trained Bayesian classification model.

[0048] Step S104: diagnose the roadbed compaction problem based on the fully trained Bayesian classification model.

[0049] In this embodiment, first, sample data is obtained and classified into roadbed compaction detection data samples and roadbed problem type data samples; then, an initial Bayesian classification model is established, wherein the initial Bayesian classification model includes a Bayesian classifier; next, the roadbed compaction detection data samples are input into the initial Bayesian classification model, and the corresponding roadbed problem type data samples are output, and the iteration is repeated continuously to train the initial Bayesian classification model, thereby obtaining a fully trained Bayesian classification model; finally, the roadbed compaction detection data is obtained in real time and input into the fully trained Bayesian classification model, thereby obtaining the corresponding roadbed problem type data, so as to realize the diagnosis of the roadbed compaction problem.

[0050] In this embodiment, a Bayesian classification model is constructed to identify and learn parameters related to the diagnosis of roadbed compaction problems, and data on the roadbed problem type is output accordingly, so that the roadbed problem type can be quickly identified based on the parameters of the roadbed compaction problem. This not only shortens the detection time and enables timely diagnosis of roadbed compaction problems, but also because the learning process of the Bayesian classification model is a process of continuous improvement, it can ensure the final diagnostic accuracy of the roadbed compaction problem.

[0051] As a preferred embodiment, in step S101, the roadbed compaction detection data includes: compaction layer thickness, lower structural layer bearing capacity, rolling influence depth, particle grading, filler type, moisture content, rolling speed, compaction section, temperature, humidity and measuring instrument operating status parameters.

[0052] Roadbed problem type data include: insufficient roadbed compaction data, uneven roadbed section data, loose roadbed surface data, and external cause data.

[0053] As a preferred embodiment, in step S102, according to the characteristics of the Bayesian classification model itself, in the process of using the Bayesian classifier to perform data analysis, it is necessary to first prepare corresponding prior probabilities.

[0054] As a preferred embodiment, in step S103, in order to train the initial Bayesian classification model, Figure 2 As shown, Figure 2 A schematic diagram of a process for training an initial Bayesian classification model according to an embodiment of the present invention includes:

[0055] Step S131: Acquire the detection priori probability of the roadbed compaction detection data and the problem priori probability of the roadbed problem type data.

[0056] Step S132: Input the detection prior probability, the problem prior probability, the roadbed compaction detection data and the roadbed problem type data into the initial Bayesian classification model to determine the problem detection probability, wherein the problem detection probability includes at least one.

[0057] Step S133: Output corresponding roadbed problem type data according to the problem detection probability.

[0058] In this embodiment, first, the detection prior probability of the roadbed compaction detection data and the problem prior probability of the roadbed problem type data are obtained as the initial operation values ​​of the initial Bayesian classification model; then, the detection prior probability, the problem prior probability, the roadbed compaction detection data and the roadbed problem type data are input into the initial Bayesian classification model, and the problem detection probability is determined through calculation, wherein, since the roadbed compaction detection data includes multiple types, multiple detection results may appear after the calculation, therefore, the problem detection probability includes at least one; finally, the maximum value of all problem detection probabilities is selected, and the corresponding roadbed problem type data is output, which is the final diagnosis result.

[0059] As a preferred embodiment, in step S131, in order to obtain the detection prior probability of the roadbed compaction detection data and the problem prior probability of the roadbed problem type data, first, the likelihood probability of the roadbed compaction detection data / roadbed problem type data is determined based on the roadbed compaction detection data / roadbed problem type data; then, the likelihood probability of the roadbed compaction detection data / roadbed problem type data is directly determined as the prior probability of the roadbed compaction detection data / roadbed problem type data.

[0060] In a specific embodiment, in order to determine the likelihood probability of the roadbed compaction detection data / roadbed problem type data, first, the roadbed compaction detection data / roadbed problem type data are compared to determine the maximum likelihood probability of the roadbed compaction detection data / roadbed problem type data; then, the likelihood probability is determined based on the maximum likelihood probability.

[0061] Specifically, D c Represents the set of samples of type c in the training set D. Assuming that these samples are independent and identically distributed, the parameter θ c For the dataset D c The likelihood probability is:

[0062]

[0063] Log-likelihood:

[0064]

[0065] Then the parameter θ c The prior probability of is:

[0066]

[0067] In another specific embodiment, the prior probability of the roadbed compaction test data / roadbed problem type data may be determined based on factors such as local construction conditions, hydrogeological conditions, filler sources, and historical data combined with the experience of engineering technicians.

[0068] As a preferred embodiment, in step S132, the formula for determining the question verification probability is:

[0069]

[0070] Among them, p(c|x) is the problem detection probability, c is the set of roadbed problem type data {c1, c2, ..., c n}, x is a sample, containing m attributes {x1, x2, …, xm}, corresponding to the parameters of the base compaction test data, p(c) is the prior probability of class label c, p(x|c) is the class conditional probability of sample x relative to class label c, corresponding to the probability that the relevant parameters show a certain state under the roadbed quality problem; p(x) is the prior probability of sample x.

[0071] As a preferred embodiment, in step S133, first, the maximum problem detection probability is determined based on the problem detection probability; then, the roadbed problem type data corresponding to the maximum problem detection probability is output, which is the roadbed compaction problem diagnosis result.

[0072] Through the above method, the roadbed compaction detection data is analyzed and identified by establishing a Bayesian classification model, and the corresponding roadbed problem type data is determined, so as to realize the diagnosis of roadbed compaction problems. Problem diagnosis is performed according to the Bayesian classification model, which fully utilizes the ultra-high computing power of the network model to ensure the accuracy of the final problem diagnosis results. Problem diagnosis is performed according to the detection data, which greatly improves the diagnosis speed and realizes real-time acquisition of the diagnosis results of the roadbed compaction problem.

[0073] In order to solve the above problems, the present invention also provides a roadbed compaction problem diagnosis device, such as Figure 3 As shown, Figure 3 This is a schematic structural diagram of an embodiment of a roadbed compaction problem diagnosis device provided by the present invention. The roadbed compaction problem diagnosis device 300 includes:

[0074] The data acquisition module 301 is used to acquire roadbed compaction detection data samples and roadbed problem type data samples;

[0075] A model building module 302 is used to build an initial Bayesian classification model;

[0076] The model training module 303 is used to input the roadbed compaction detection data samples into the initial Bayesian classification model, output the corresponding roadbed problem type data, train the initial Bayesian classification model, and obtain a fully trained Bayesian classification model;

[0077] The problem diagnosis module 304 is used to diagnose the roadbed compaction problem based on the well-trained Bayesian classification model.

[0078] The present invention also provides an electronic device, such as Figure 4 As shown, Figure 4 This is a block diagram of an electronic device according to an embodiment of the present invention. The electronic device 400 can be a computing device such as a mobile terminal, desktop computer, notebook computer, PDA, or server. The electronic device 400 includes a processor 401 and a memory 402 , wherein the memory 402 stores a roadbed compaction problem diagnosis program 403 .

[0079] In some embodiments, the memory 402 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory 402 may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Furthermore, the memory 402 may also include both an internal storage unit of the computer device and an external storage device. The memory 402 is used to store application software and various types of data installed on the computer device, such as program codes installed on the computer device. The memory 402 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the roadbed compaction problem diagnosis program 403 may be executed by the processor 401, thereby realizing the roadbed compaction problem diagnosis method of each embodiment of the present invention.

[0080] In some embodiments, the processor 401 may be a central processing unit (CPU), a microprocessor, or other data processing chip, configured to execute program codes or process data stored in the memory 402, such as executing a roadbed compaction problem diagnosis program.

[0081] This embodiment also provides a computer-readable storage medium on which a roadbed compaction problem diagnosis program is stored. When the computer program is executed by a processor, the roadbed compaction problem diagnosis method described in any of the above technical solutions is implemented.

[0082] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0083] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for diagnosing roadbed compaction problems, characterized in that: include: Acquire samples of roadbed compaction test data and roadbed problem type data. The roadbed compaction test data includes: compaction layer thickness, lower structural layer bearing capacity, rolling impact depth, particle gradation, filler type, moisture content, rolling speed, compaction section, temperature, humidity, and measuring instrument operating status parameters. The roadbed problem type data includes: insufficient roadbed compaction data, uneven roadbed section data, loose roadbed surface data, and external cause data. Establish an initial Bayesian classification model; Inputting the roadbed compaction detection data sample into the initial Bayesian classification model, outputting the corresponding roadbed problem type data, training the initial Bayesian classification model, and obtaining a fully trained Bayesian classification model, including: Obtaining a priori probability of detection of the roadbed compaction detection data and a priori probability of problem of the roadbed problem type data; Inputting the detection prior probability, the problem prior probability, the roadbed compaction detection data, and the roadbed problem type data into the initial Bayesian classification model to determine a problem detection probability, wherein the problem detection probability includes at least one; Outputting the corresponding roadbed problem type data according to the problem inspection probability; The roadbed compaction problem is diagnosed based on the well-trained Bayesian classification model.

2. The method for diagnosing roadbed compaction problems according to claim 1, characterized in that: The obtaining of the detection prior probability of the roadbed compaction detection data and the problem prior probability of the roadbed problem type data includes: Determining the likelihood probability of the roadbed compaction detection data / the roadbed problem type data based on the roadbed compaction detection data / the roadbed problem type data; The prior probability of the roadbed compaction detection data / the roadbed problem type data is determined according to the likelihood probability of the roadbed compaction detection data / the roadbed problem type data.

3. The method for diagnosing roadbed compaction problems according to claim 1, characterized in that: The formula for determining the probability of testing a question is: in, is the problem detection probability, c is the set of roadbed problem type data {c1, c2, ..., c n }, x is a sample, containing m attributes {x1, x2, ..., xm}, corresponding to the parameters of the base compaction detection data, p(c) is the prior probability of class label c, is the class conditional probability of sample x relative to the class label c, corresponding to the probability that the relevant parameters show a certain state under the roadbed quality problem; p(x) is the prior probability of sample x.

4. The method for diagnosing roadbed compaction problems according to claim 1, characterized in that: Outputting the corresponding roadbed problem type data according to the problem detection probability includes: Determining a maximum problem inspection probability based on the problem inspection probability; The roadbed problem type data corresponding to the maximum problem detection probability is output, which is the roadbed compaction problem diagnosis result.

5. A device for diagnosing roadbed compaction problems, characterized in that: include: A data acquisition module is used to acquire roadbed compaction detection data samples and roadbed problem type data samples. The roadbed compaction detection data includes: compaction layer thickness, lower structural layer bearing capacity, rolling impact depth, particle gradation, filler type, moisture content, rolling speed, compaction section, temperature, humidity and measuring instrument operating status parameters. The roadbed problem type data includes: insufficient roadbed compaction data, uneven roadbed section data, loose roadbed surface data, and external cause data. Model building module, used to build the initial Bayesian classification model; A model training module is used to input the roadbed compaction detection data samples into the initial Bayesian classification model, output the corresponding roadbed problem type data, train the initial Bayesian classification model, and obtain a fully trained Bayesian classification model, including: Obtaining a priori probability of detection of the roadbed compaction detection data and a priori probability of problem of the roadbed problem type data; Inputting the detection prior probability, the problem prior probability, the roadbed compaction detection data, and the roadbed problem type data into the initial Bayesian classification model to determine a problem detection probability, wherein the problem detection probability includes at least one; Outputting the corresponding roadbed problem type data according to the problem inspection probability; The problem diagnosis module is used to diagnose the roadbed compaction problem based on the fully trained Bayesian classification model.

6. An electronic device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method for diagnosing roadbed compaction problems according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, characterized in that The storage medium stores computer program instructions, which, when executed by a computer, enable the computer to execute the method for diagnosing roadbed compaction problems according to any one of claims 1 to 4.

Images