Quality detection method and system for water conservancy construction engineering

Abstract

The invention relates to the technical field of engineering detection, in particular to a quality detection method and system for water conservancy construction engineering, and the method mainly comprises the steps: judging whether a current dam body has cracks or water stains or not through an image recognition model, obtaining a plurality of basic images through the set next image obtaining time interval and obtaining times, and carrying out the detection of the current dam body. And judging whether the current dam body is deformed or not through the plurality of basic images, obtaining basic data of the current dam body, if not, entering a first detection process, and if so, entering a second detection process. By means of the method, depth judgment can be conducted again on the basis of the result of judging whether the current dam body is qualified or not, the judged result is analyzed, more follow-up content can be obtained through the two selected detection processes no matter whether the current dam body is qualified or not, the follow-up content can be analyzed by workers, and the detection efficiency is improved. And more subsequent results are output in combination with the current condition, so that the final detection result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering detection, and in particular to a quality detection method and system for a water conservancy construction project. Background Art

[0002] Water conservancy construction projects are projects related to the research, planning, design and construction of water resources management, utilization and prevention, including water supply, drainage, irrigation, river management, flood control and water environment protection, mainly involving construction projects related to dams and reservoirs.

[0003] This plan mainly involves the inspection work related to the dam body. The safety performance of the dam body is the top priority of the entire water conservancy construction project. In the existing technology, data is generally directly collected and compared with conventional historical data to determine whether the current dam body is qualified. However, in the existing technology, there is no subsequent output of more results for the current situation, that is, if it is currently qualified, there is no prediction of the future use of the dam body, and if it is unqualified, there is no relevant warning level result judgment. Summary of the invention

[0004] The purpose of the present invention is to provide a quality detection method and system for water conservancy construction projects to solve the above-mentioned problems in the prior art.

[0005] The present invention is achieved through the following technical solutions: In a first aspect, the present invention provides a quality inspection method for a water conservancy construction project, comprising: Obtain the basic image of the current dam body, pre-process the basic image, establish an image recognition model, and use the image recognition model to determine whether the current dam body has cracks or water stains; If there are cracks, the result that the current water conservancy construction project is unqualified is output; if there are only water stains, it is determined whether the current water stains are leakage of the dam body. If so, the result that the current water conservancy construction project is unqualified is output; if not, the time interval and number of acquisitions for the next image acquisition are set; A number of basic images are obtained by setting the next image acquisition time interval and the number of acquisitions. The basic images are used to determine whether the current dam body is deformed, and a number of groups of basic data of the dam body at different periods are obtained. If there is no deformation, the first detection process is entered; if there is deformation, the second detection process is entered; The first detection process includes establishing a dam body use prediction simulation model, and outputting a result of whether the current dam body is qualified through a number of sets of basic data and the dam body use prediction simulation model at different periods; The second detection process includes establishing a dam body quality warning model, outputting an evaluation value through current basic data and the dam body quality warning model, and setting an evaluation value threshold, and outputting a general warning or a serious warning result for the current dam body quality through the evaluation value and the evaluation value threshold.

[0006] Preferably, the preprocessing of the basic image includes: Read the color data of the base image and convert each pixel from the RGB color space to a grayscale value; Assigning the grayscale value to the new image matrix to generate a first image containing only grayscale information; Select the size and standard deviation of the Gaussian filter, and generate a Gaussian kernel based on the size and standard deviation; The first image is convolved with a Gaussian kernel, each pixel in the image is traversed, and the current pixel and its surrounding pixels are weighted averaged by the Gaussian kernel to output a second image.

[0007] Preferably, generating a Gaussian kernel according to the size and the standard deviation comprises:

[0008] In the formula, is the Gaussian kernel, is the horizontal offset of the pixel. is the vertical offset of the pixel. is the standard deviation.

[0009] Preferably, the determining whether the current water stain is leakage of the dam body comprises: Acquire a first image feature currently identified as a water stain and a second image feature currently identified as a dam body; Obtain whether the first image feature and the second image feature overlap in the base image, and if not, output the result that the current water stain is not leakage of the dam body; If they overlap, several first image features are acquired in several time periods to determine whether the first image features are reduced. If they are not reduced, the result that the current water stain is a leakage of the dam body is output; if they are reduced, the result that the current water stain is not a leakage of the dam body is output.

[0010] Preferably, the setting of the next image acquisition time interval and acquisition times includes:

[0011]

[0012] In the formula, is the time interval for the next image acquisition, is the total number of people currently undergoing dam inspections, The number of people required for a single dam inspection, The area occupied by the dam after completion. is the contact area between the dam and the water surface, is the cost of a single test, is the total cost during the current inspection period, is the current height of the dam, is the current water surface height, is the amount of concrete used in the dam body, The amount of concrete used for the facilities around the dam. To obtain the number of times.

[0013] Preferably, judging whether the current dam body is deformed by using a plurality of basic images includes: The ratio of the dam body features to the entire basic image in several basic images is obtained, and a judgment model is established. The judgment value is output through the judgment model, and a judgment threshold is set. If the judgment value is less than the judgment threshold, the result of the dam body not being deformed is output; if it is not less than the judgment threshold, the result of the dam body being deformed is output.

[0014] Preferably, the judgment model includes:

[0015] In the formula, is the judgment value, is the total number of base images, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of a basic image to the entire basic image.

[0016] Preferably, the establishment of the dam body using a predictive simulation model includes: Normalize several groups of basic data of the dam body at different periods, divide the processed basic data into data sets, and obtain training sets and validation sets; Establish a neural network function model, train the neural network function model through the training set, optimize the neural network function model through the verification set, and update the weights and biases until the number of iterations is reached, and output the optimal neural network function model; The neural network function model includes:

[0017] In the formula, is the confidence value, is the bias from the input layer to the hidden layer, is the proportion of neurons randomly discarded during training, is the learning rate, is the number of samples used to calculate the loss and update the weights in each iteration, is the loss function, For the Layer input layer to the The output of the hidden layer, Input samples for the training set, is the activation function, is the standard deviation of data preprocessing, For the Hidden layers to The calculation weight of the output layer, For the Hidden layers to The calculation weights of the output layer, is the regularization coefficient, To separate the proportion of samples used for validation from the training set, is the number of samples used to calculate the loss and update the weights in each iteration; like ,or, , then the output is the result that the dam tends to deform in the future; like , then the output is the result that the dam tends to be stable in the future.

[0018] Preferably, the dam quality warning model includes:

[0019] In the formula, is the total horizontal and vertical displacement of the dam body, is the height displacement of the dam body over time, is the compressive strength of the dam concrete, is the tensile strength of the dam material, is the strength of the dam material under shear, is the moisture content of the dam material, is the water level change in the reservoir and in front of the dam, is the bearing capacity of the dam foundation soil, is the deformation of the current dam body relative to the first basic image.

[0020] In a second aspect, the present invention provides a quality inspection system for water conservancy construction projects, comprising: An acquisition module is configured to acquire a basic image of the current dam body, pre-process the basic image, establish an image recognition model, and determine whether the current dam body has cracks or water stains through the image recognition model; The judgment module is configured to output the result that the current water conservancy construction project is unqualified if there are cracks, and to judge whether the current water stains are leakage of the dam body if there are only water stains. If so, the result that the current water conservancy construction project is unqualified is output; if not, the next image acquisition time interval and acquisition number are set; a number of basic images are obtained through the set next image acquisition time interval and acquisition number, and whether the current dam body is deformed is judged through the several basic images, and the basic data of the current dam body is obtained; The selection module is configured to enter the first detection process if there is no deformation, and to enter the second detection process if there is deformation. The first detection process includes establishing a dam body use prediction simulation model, and outputting the result of whether the current dam body is qualified through basic data and the dam body use prediction simulation model; the second detection process includes establishing a dam body quality alarm model, outputting an evaluation value through basic data and the dam body quality alarm model, and setting an evaluation value threshold, and outputting the result of a general alarm or a serious alarm for the current dam body quality through the evaluation value and the evaluation value threshold.

[0021] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects: The method provided by the invention mainly includes judging whether the current dam body has cracks or water stains through an image recognition model; obtaining a number of basic images through the set time interval and number of acquisitions for the next image acquisition, judging whether the current dam body is deformed through the several basic images, and obtaining the basic data of the current dam body. If it is not deformed, the first detection process is entered, and if it is deformed, the second detection process is entered. Through the above method of the present invention, a deeper judgment can be made again on the result of judging whether the current dam body is qualified, and the judged result can be analyzed. Whether it is qualified or unqualified, more subsequent content can be obtained through the two selected detection processes for staff to analyze, so that more subsequent results can be output in combination with the current situation to make the final detection result more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0024] 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. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0025] The terms "first, second" and the like in the specification and claims of this application and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0026] The modules or submodules described independently may be physically separated or not: they may be implemented by software or hardware, and some modules or submodules may be implemented by software, and the processor may call the software to implement the functions of these modules or submodules, and other modules or submodules may be implemented by hardware, such as by hardware circuits. In addition, some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0027] Please refer to Figure 1 The present invention provides a quality inspection method for water conservancy construction projects, comprising: S101: obtaining a basic image of the current dam body, preprocessing the basic image, establishing an image recognition model, and determining whether the current dam body has cracks or water stains through the image recognition model; Among them, the presence of cracks or water stains can directly confirm that there is a problem with the current dam body and it is unqualified. The presence of water stains needs to be determined whether it is leakage of the dam body or other interference. Since detection on rainy days will cause greater interference, in this embodiment, the situation of detection on rainy days is ignored.

[0028] S102: if there are cracks, output the result that the current water conservancy construction project is unqualified; if there are only water stains, determine whether the current water stains are leakage of the dam body; if so, output the result that the current water conservancy construction project is unqualified; if not, set the next image acquisition time interval and acquisition times; In this embodiment, as long as cracks appear, it is judged as unqualified, including the case where cracks and water stains appear at the same time. If only water stains appear, further judgment is required.

[0029] The acquisition of basic images can be for photo and video monitoring: regularly take photos or videos of the dam body to observe the changes in the appearance of the dam body; or use tools such as levels and total stations to perform physical measurements of the dam body.

[0030] Of course, satellite images can also be used to obtain information about the dam and its surrounding environment, and regional analysis and large-scale monitoring can be carried out. Unmanned aerial vehicles (UAVs) can be used to collect high-resolution images, which are suitable for detailed observation of the dam and its surroundings.

[0031] Terrestrial laser scanning: Use a laser scanner to obtain a high-precision 3D model of the dam body, and analyze the dam body's geometry and deformation through point cloud data.

[0032] S103: obtaining a number of basic images by setting the next image acquisition time interval and the number of acquisition times, judging whether the current dam body is deformed by the number of basic images, and obtaining a number of groups of basic data of the dam body at different periods, if there is no deformation, entering the first detection process, if there is deformation, entering the second detection process; In this embodiment, if deformation occurs, it can also be judged as unqualified. This solution gradually determines whether there is deformation based on the results of the above steps, from simple detection methods to complex detection methods. On the one hand, it simplifies the detection steps, and on the other hand, it can obtain more accurate results.

[0033] S104: The first detection process includes establishing a dam body use prediction simulation model, and outputting a result of whether the current dam body is qualified through a number of sets of basic data in different periods and the dam body use prediction simulation model; S105: The second detection process includes establishing a dam body quality alarm model, outputting an evaluation value through current basic data and the dam body quality alarm model, and setting an evaluation value threshold, and outputting a general alarm or a serious alarm result for the current dam body quality through the evaluation value and the evaluation value threshold.

[0034] In this embodiment, a general alarm is an alarm level that can be urgently repaired, and a severe alarm is an alarm level that cannot be salvaged. In this way, when the unqualified situation is known, more effective information can be further provided to the staff for response.

[0035] The method provided by the invention mainly includes judging whether the current dam body has cracks or water stains through an image recognition model; obtaining a number of basic images through the set time interval and number of acquisitions for the next image acquisition, judging whether the current dam body is deformed through the several basic images, and obtaining the basic data of the current dam body. If it is not deformed, the first detection process is entered, and if it is deformed, the second detection process is entered. Through the above method of the present invention, a deeper judgment can be made again on the result of judging whether the current dam body is qualified, and the judged result can be analyzed. Whether it is qualified or unqualified, more subsequent content can be obtained through the two selected detection processes for staff to analyze, so that more subsequent results can be output in combination with the current situation to make the final detection result more accurate.

[0036] In an exemplary embodiment of the present invention, preprocessing the base image includes: Read the color data of the base image and convert each pixel from the RGB color space to a grayscale value; Assigning the grayscale value to the new image matrix to generate a first image containing only grayscale information; Select the size and standard deviation of the Gaussian filter, and generate a Gaussian kernel based on the size and standard deviation; The first image is convolved with a Gaussian kernel, each pixel in the image is traversed, and the current pixel and its surrounding pixels are weighted averaged by the Gaussian kernel to output a second image.

[0037] The second image is a preprocessed basic image.

[0038] Specifically, generating a Gaussian kernel based on size and standard deviation includes:

[0039] In the formula, is the Gaussian kernel, is the horizontal offset of the pixel. is the vertical offset of the pixel. is the standard deviation.

[0040] In an exemplary embodiment of the present invention, determining whether the current water stain is leakage of the dam body includes: S201: Acquire a first image feature currently identified as a water stain and a second image feature currently identified as a dam body; S202: obtaining whether the first image feature and the second image feature overlap in the basic image, and if not, outputting a result that the current water stain is not leakage of the dam body; Among them, if there is no overlap, it means that the water stains are not on the dam body, and it is judged as an interference item, not water stains seeping out of the dam body.

[0041] S203: If they overlap, then a plurality of first image features are acquired in a plurality of time periods to determine whether the first image features are reduced. If not, then outputting that the current water stain is a result of leakage of the dam body; if reduced, then outputting that the current water stain is not a result of leakage of the dam body.

[0042] If the water stains are on the dam body, a normal judgment can be made to see if the water stains have changed. If they have shrunk, it means that they are not water stains caused by leakage from the dam body. Not shrinking includes getting bigger or not changing. If they get bigger, there is no need to explain, and leakage must have occurred. Under normal circumstances, water will evaporate on its own, so if it does not change, it means that the leaked water and the evaporated water have reached a balance, and only then will the unchanged situation occur. Therefore, it is also judged that there is leakage in the dam body, and the current water stains are leakage of the dam body.

[0043] In an exemplary embodiment of the present invention, setting the next image acquisition time interval and acquisition times includes:

[0044]

[0045] In the formula, is the time interval for the next image acquisition, is the total number of people currently undergoing dam inspections, The number of people required for a single dam inspection, The area occupied by the dam after completion. is the contact area between the dam and the water surface, is the cost of a single test, is the total cost during the current inspection period, is the current height of the dam, is the current water surface height, is the amount of concrete used in the dam body, The amount of concrete used for the facilities around the dam. To obtain the number of times.

[0046] In this embodiment, a variety of factors are comprehensively considered to provide a basis for the interval of the next image acquisition, and combined with some parameters of the dam body, the accuracy and reliability of the results are significantly improved. The integration of various data and variables helps to identify potential risks and optimize decisions, reducing uncertainty. At the same time, it can also achieve more comprehensive analysis, enhance the adaptability and flexibility of the model, make it closer to the actual situation, and support more scientific prediction and planning.

[0047] In an exemplary embodiment of the present invention, judging whether the current dam body is deformed by using a plurality of basic images includes: The ratio of the dam body features to the entire basic image in several basic images is obtained, and a judgment model is established. The judgment value is output through the judgment model, and a judgment threshold is set. If the judgment value is less than the judgment threshold, the result of the dam body not being deformed is output; if it is not less than the judgment threshold, the result of the dam body being deformed is output.

[0048] Specifically, the judgment model includes:

[0049] In the formula, is the judgment value, is the total number of base images, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of a basic image to the entire basic image.

[0050] In this embodiment, the judgment threshold can be set to 60%. The judgment threshold can be changed according to specific circumstances. Different judgment thresholds can be set according to the different proportions of the dam body features and the entire basic image.

[0051] In an exemplary embodiment of the present invention, establishing a dam use prediction simulation model includes: Normalize several groups of basic data of the dam body at different periods, divide the processed basic data into data sets, and obtain training sets and validation sets; The normalized data in this embodiment has a value range of -1 to 1.

[0052] Establish a neural network function model, which includes basic input layer, hidden layer and output layer, etc. Train the neural network function model through the training set, optimize the neural network function model through the verification set, and update the weights and biases until the number of iterations is reached, and output the optimal neural network function model; The neural network function model includes:

[0053] In the formula, is the confidence value, is the bias from the input layer to the hidden layer, is the proportion of neurons randomly discarded during training, is the learning rate, is the number of samples used to calculate the loss and update the weights in each iteration, is the loss function, For the Layer input layer to the The output of the hidden layer, Input samples for the training set, is the activation function, is the standard deviation of data preprocessing, For the Hidden layers to The calculation weight of the output layer, For the Hidden layers to The calculation weights of the output layer, is the regularization coefficient, To separate the proportion of samples used for validation from the training set, is the number of samples used to calculate the loss and update the weights in each iteration; like ,or, , then the output is the result that the dam tends to deform in the future; like , then the output is the result that the dam tends to be stable in the future.

[0054] In an exemplary embodiment of the present invention, the dam quality warning model includes:

[0055] In the formula, is the total horizontal and vertical displacement of the dam body, is the height displacement of the dam body over time, is the compressive strength of the dam concrete, is the tensile strength of the dam material, is the strength of the dam material under shear, is the moisture content of the dam material, is the water level change in the reservoir and in front of the dam, is the bearing capacity of the dam foundation soil, is the deformation of the current dam body relative to the first basic image.

[0056] In this embodiment, the evaluation value threshold can be obtained based on past historical data. For example, historical data of the dam body that can be repaired urgently corresponding to a general alarm can be found, and a value is obtained according to the above-mentioned dam body quality alarm model. Then, based on the historical data of the dam body that cannot be rescued corresponding to a serious alarm, a second value is obtained according to the above-mentioned dam body quality alarm model. The average of the two values ​​is calculated to obtain the evaluation value threshold of this embodiment.

[0057] Of course, you can also set a floating evaluation value threshold or set it according to the specific situation. The floating threshold can be set to;

[0058] In the formula, The floating evaluation threshold strengthens the relationship between various parameters. When various parameters change, the evaluation threshold will increase or decrease with a larger fluctuation, so more accurate results can be obtained.

[0059] The present invention provides a quality detection system for water conservancy construction projects, comprising: An acquisition module is configured to acquire a basic image of the current dam body, pre-process the basic image, establish an image recognition model, and determine whether the current dam body has cracks or water stains through the image recognition model; The judgment module is configured to output the result that the current water conservancy construction project is unqualified if there are cracks, and to judge whether the current water stains are leakage of the dam body if there are only water stains. If so, the result that the current water conservancy construction project is unqualified is output; if not, the next image acquisition time interval and acquisition number are set; a number of basic images are obtained through the set next image acquisition time interval and acquisition number, and whether the current dam body is deformed is judged through the several basic images, and the basic data of the current dam body is obtained; The selection module is configured to enter the first detection process if there is no deformation, and to enter the second detection process if there is deformation. The first detection process includes establishing a dam body use prediction simulation model, and outputting the result of whether the current dam body is qualified through basic data and the dam body use prediction simulation model; the second detection process includes establishing a dam body quality alarm model, outputting an evaluation value through basic data and the dam body quality alarm model, and setting an evaluation value threshold, and outputting the result of a general alarm or a serious alarm for the current dam body quality through the evaluation value and the evaluation value threshold.

[0060] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0061] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0062] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A quality inspection method for water conservancy construction projects, characterized in that: include: Obtain the basic image of the current dam body, pre-process the basic image, establish an image recognition model, and use the image recognition model to determine whether the current dam body has cracks or water stains; If there are cracks, the result that the current water conservancy construction project is unqualified is output; if there are only water stains, it is determined whether the current water stains are leakage of the dam body. If so, the result that the current water conservancy construction project is unqualified is output; if not, the time interval and number of acquisitions for the next image acquisition are set; A number of basic images are obtained by setting the next image acquisition time interval and the number of acquisition times, and a number of basic images are used to determine whether the current dam body is deformed, and a number of groups of basic data of the dam body at different periods are obtained. If there is no deformation, the first detection process is entered; if there is deformation, the second detection process is entered; The first detection process includes establishing a dam body use prediction simulation model, and outputting a result of whether the current dam body is qualified through a number of sets of basic data and the dam body use prediction simulation model at different periods; The second detection process includes establishing a dam body quality warning model, outputting an evaluation value through current basic data and the dam body quality warning model, and setting an evaluation value threshold, and outputting a general warning or a serious warning result for the current dam body quality through the evaluation value and the evaluation value threshold.

2. A quality inspection method for water conservancy construction projects according to claim 1, characterized in that: The preprocessing of the basic image comprises: Read the color data of the base image and convert each pixel from the RGB color space to a grayscale value; Assigning the grayscale value to the new image matrix to generate a first image containing only grayscale information; Select the size and standard deviation of the Gaussian filter, and generate a Gaussian kernel based on the size and standard deviation; The first image is convolved with a Gaussian kernel, each pixel in the image is traversed, and the current pixel and its surrounding pixels are weighted averaged by the Gaussian kernel to output a second image.

3. A quality inspection method for water conservancy construction projects according to claim 2, characterized in that: Generating a Gaussian kernel according to the size and standard deviation includes: In the formula, is the Gaussian kernel, is the horizontal offset of the pixel. is the vertical offset of the pixel. is the standard deviation.

4. A quality inspection method for water conservancy construction projects according to claim 3, characterized in that: The determining whether the current water stain is leakage of the dam body comprises: Acquire a first image feature currently identified as a water stain and a second image feature currently identified as a dam body; Obtain whether the first image feature and the second image feature overlap in the base image, and if not, output the result that the current water stain is not leakage of the dam body; If they overlap, several first image features are acquired in several time periods to determine whether the first image features are reduced. If they are not reduced, the result that the current water stain is a leakage of the dam body is output; if they are reduced, the result that the current water stain is not a leakage of the dam body is output.

5. A quality inspection method for water conservancy construction projects according to claim 4, characterized in that: The setting of the next image acquisition time interval and acquisition times includes: In the formula, is the time interval for the next image acquisition, is the initial time interval, is the total number of people currently undergoing dam inspections, The number of people required for a single dam inspection, The area occupied by the dam after completion. is the contact area between the dam and the water surface, is the cost of a single test, is the total cost during the current inspection period, is the current height of the dam, is the current water surface height, is the amount of concrete used in the dam body, The amount of concrete used for the facilities around the dam body, To obtain the number of times.

6. A quality inspection method for water conservancy construction projects according to claim 5, characterized in that: The step of judging whether the current dam body is deformed by using a plurality of basic images includes: The ratio of the dam body features to the entire basic image in several basic images is obtained, and a judgment model is established. The judgment value is output through the judgment model, and a judgment threshold is set. If the judgment value is less than the judgment threshold, the result of the dam body not being deformed is output; if it is not less than the judgment threshold, the result of the dam body being deformed is output.

7. A quality inspection method for water conservancy construction projects according to claim 6, characterized in that: The judgment model includes: In the formula, is the judgment value, is the total number of base images, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of the base image to the entire base image, For the The ratio of the dam features of a basic image to the entire basic image.

8. A quality inspection method for water conservancy construction projects according to claim 7, characterized in that: The establishment of a dam use prediction simulation model includes: Normalize several groups of basic data of the dam body at different periods, divide the processed basic data into data sets, and obtain training sets and validation sets; Establish a neural network function model, train the neural network function model through the training set, optimize the neural network function model through the verification set, and update the weights and biases until the number of iterations is reached, and output the optimal neural network function model; The neural network function model includes: In the formula, is the confidence value, is the bias from the input layer to the hidden layer, is the bias from the hidden layer to the output layer, is the proportion of neurons randomly discarded during training, is the learning rate, is the number of samples used to calculate the loss and update the weights in each iteration, is the loss function, For the Layer input layer to the The output of the hidden layer, Input samples for the training set, is the activation function, is the standard deviation of data preprocessing, For the Hidden layers to The calculation weight of the output layer, For the Hidden layers to The calculation weights of the output layer, is the regularization coefficient, To separate the proportion of samples used for validation from the training set, is the number of samples used to calculate the loss and update the weights in each iteration; like ,or, , then the output is the result that the dam tends to deform in the future; like , then the output is the result that the dam tends to be stable in the future.

9. A quality inspection method for water conservancy construction projects according to claim 8, characterized in that: The dam quality warning model includes: In the formula, is the evaluation value, is the total horizontal and vertical displacement of the dam body, is the height displacement of the dam body over time, is the compressive strength of the dam concrete, is the tensile strength of the dam material, is the strength of the dam material under shear, is the moisture content of the dam material, is the water level change in the reservoir and in front of the dam, is the bearing capacity of the dam foundation soil, is the deformation of the current dam body relative to the first basic image.

10. A quality inspection system for water conservancy construction projects, characterized in that: include: An acquisition module is configured to acquire a basic image of the current dam body, pre-process the basic image, establish an image recognition model, and determine whether the current dam body has cracks or water stains through the image recognition model; The judgment module is configured to output the result that the current water conservancy construction project is unqualified if there are cracks, and to judge whether the current water stains are leakage of the dam body if there are only water stains. If so, the result that the current water conservancy construction project is unqualified is output; if not, the next image acquisition time interval and acquisition number are set; a number of basic images are obtained through the set next image acquisition time interval and acquisition number, and whether the current dam body is deformed is judged through the several basic images, and the basic data of the current dam body is obtained; The selection module is configured to enter the first detection process if there is no deformation, and to enter the second detection process if there is deformation. The first detection process includes establishing a dam body use prediction simulation model, and outputting the result of whether the current dam body is qualified through basic data and the dam body use prediction simulation model; the second detection process includes establishing a dam body quality alarm model, outputting an evaluation value through basic data and the dam body quality alarm model, and setting an evaluation value threshold, and outputting the result of a general alarm or a serious alarm for the current dam body quality through the evaluation value and the evaluation value threshold.

Images