A hydraulic engineering quality detection method and system

Abstract

The present application relates to a kind of hydraulic engineering quality detection method and system, it is related to hydraulic engineering quality detection field, including: obtaining the pipeline information, environmental parameter and water quality parameter of the first water delivery pipeline of hydraulic engineering with multiple water inlet branch;Obtain the corrosion contribution degree list;According to corrosion contribution degree list and the corrosion material list, pipe section clustering is carried out to first water delivery pipeline, obtains water delivery pipeline segmentation result;Traverse water delivery pipeline segmentation result, process corrosion material concentration list and corrosion contribution degree list, calibrate pipe failure prediction time, send to hydraulic engineering quality detection end and carry out detection task allocation.Solve the technical problem that there is low detection efficiency in prior art due to the need to set as short as possible period detection to ensure detection accuracy when facing multiple water inlet branch pipe corrosion state quality detection, increase and decrease detection length.

Description

Technical Field

[0001] This invention relates to the field of quality testing of water conservancy projects, and in particular to a method and system for quality testing of water conservancy projects. Background Technology

[0002] Traditional water conservancy projects' water pipeline quality inspection lacks guidance and typically involves periodic full inspections.

[0003] The usual inspection method is to inspect the condition of the pipeline at specific intervals and carry out maintenance when quality defects are found. However, since actual water pipelines usually have multiple water inlet branches, the defects inside such pipelines, especially the corrosion state, are multimodal. Longer fixed interval mode inspection may result in missed detections, while shorter fixed interval mode inspection may result in wasted resources and increased inspection costs. Summary of the Invention

[0004] This invention addresses the technical problem in existing technologies where, in order to ensure detection accuracy when inspecting the corrosion status of pipelines with multiple water inlets, it is necessary to set the detection cycle as short as possible, which leads to low detection efficiency due to the adjustment of detection time. The invention provides a water conservancy engineering quality inspection method and system to solve this problem.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides a method for quality testing of water conservancy projects, comprising:

[0007] Obtain pipeline information, environmental parameters, and water quality parameters of the first water transmission pipeline of a water conservancy project with multiple water inlets. Among them, the environmental parameters of the first water transmission pipeline are consistent throughout the entire section, while the water quality parameters of the first water transmission pipeline vary throughout the entire section.

[0008] Based on the environmental parameters and pipeline material information, the contribution of the list of corrosive substances for the water quality parameters is calibrated to obtain a corrosion contribution list.

[0009] Based on the corrosion contribution list and the corrosive substance list, the first water pipeline is clustered to obtain the water pipeline segmentation results.

[0010] The pipeline segmentation results are traversed, and the list of corrosive substance concentrations and the list of corrosion contribution are processed by the pre-trained pipeline thickness loss prediction model to generate pipeline thickness loss time series prediction values. Combined with the pipeline thickness information of the pipeline information, the pipeline failure prediction time is determined.

[0011] The predicted pipeline failure time is sent to the water conservancy project quality inspection terminal for inspection task allocation.

[0012] Secondly, the present invention provides a water conservancy project quality inspection system, comprising:

[0013] The data acquisition module is used to obtain pipeline information, environmental parameters and water quality parameters of the first water conveyance pipeline of a water conservancy project with multiple water inlets. The environmental parameters of the first water conveyance pipeline are consistent throughout the entire pipeline, while the water quality parameters of the first water conveyance pipeline vary throughout the entire pipeline.

[0014] The contribution calibration module is used to calibrate the contribution of the list of corrosive substances in the water quality parameters based on the pipeline material information of the environmental parameters and pipeline information, and obtain the corrosion contribution list.

[0015] The pipe segment clustering module is used to perform pipe segment clustering on the first water transmission pipeline based on the corrosion contribution list and the corrosive substance list to obtain the water transmission pipeline segmentation results.

[0016] The pipeline failure calibration module is used to traverse the segmentation results of the water transmission pipeline, process the list of corrosive substance concentrations and the list of corrosion contribution through a pre-trained pipeline thickness loss prediction model, generate pipeline thickness loss time-series prediction values, and calibrate the pipeline failure prediction time by combining the pipeline thickness information of the pipeline information.

[0017] The detection task allocation module is used to send the predicted pipeline failure time to the water conservancy project quality detection terminal for detection task allocation.

[0018] The beneficial effects of this invention are as follows: By collecting pipeline information, environmental parameters, and water quality parameters of the first water conveyance pipeline of a water conservancy project with multiple water inlets, and then using the pipeline material and environmental parameters to calibrate the contribution of a list of corrosive substances, a corrosion contribution list is obtained; further, based on the corrosion contribution list from multiple locations and the list of corrosive substances, pipe segments are clustered to obtain the segmentation results of the first water conveyance pipeline; for each segment, a pipeline thickness loss prediction model is used to process the list of corrosive substance concentrations and the list of corrosion contribution to obtain a time-series predicted value of pipeline thickness loss; further still, by extracting pipeline thickness information, thickness reduction can be performed according to time sequence, thereby calibrating the pipeline failure prediction time as a resource allocation for subsequent quality inspection tasks. Segmentation realizes the spatial positioning of pipe segments, and the prediction of failure time realizes the temporal positioning of pipe segments, achieving targeted quality inspection and improving the technical effect of improving quality inspection efficiency. Attached Figure Description

[0019] Figure 1 A flowchart illustrating a water conservancy project quality testing method provided by the present invention;

[0020] Figure 2 This is a schematic diagram of the structure of a water conservancy project quality testing system provided by the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0024] Example 1, as Figure 1 As shown, this embodiment of the invention provides a method for quality inspection of water conservancy projects, including the following steps:

[0025] S10: Obtain pipeline information, environmental parameters and water quality parameters of the first water transmission pipeline of a water conservancy project with multiple water inlet branches. Among them, the environmental parameters of the first water transmission pipeline are consistent throughout the entire section, while the water quality parameters of the first water transmission pipeline are different throughout the entire section.

[0026] Specifically, the maintenance of water transmission pipelines is an important part of water conservancy projects. In order to ensure the healthy operation of water conservancy projects, it is necessary to conduct quality inspections on water transmission pipelines to facilitate timely maintenance. The scenario of this application embodiment is now defined as a water transmission pipeline with multiple water inlet branches.

[0027] The first water transmission pipeline of the water conservancy project serves as an exemplary description of all pipelines. Pipeline information refers to basic information related to the pipeline, including pipeline material information, pipeline service life information, and pipeline thickness information. Environmental parameters refer to the environmental data inside the pipeline, including temperature, pH, and humidity. Generally, since the external environmental conditions such as temperature, humidity, and soil pH value of the water transmission pipeline remain relatively constant throughout the pipeline length, they are usually achieved through a uniform installation environment or external protection measures (such as insulation layers). Therefore, the environmental parameters in this application embodiment are consistent throughout the entire section. Water quality parameters refer to the water quality parameters at multiple detection locations, including the concentration and proportion of various substances. Due to the presence of multiple inlets, each inlet has different water quality. For example, due to the differences in water sources of different inlet branches (such as industrial wastewater and natural water bodies), the water quality parameters (such as Cl⁻ and SO₄²⁻ concentrations) of each section in the main pipeline change dynamically with location. This leads to differences in water quality parameters throughout the entire section of the water transmission pipeline for multiple inlet branches.

[0028] Preferably, the data acquisition example is described as follows: static data such as pipe material and thickness are obtained through engineering drawings or IoT sensors; temperature and humidity sensors and soil monitoring equipment are deployed to verify the environmental consistency throughout the entire section (e.g., temperature fluctuation ≤ ±2℃); water quality sensors are installed at multiple water inlet branches and key nodes of the main pipeline (e.g., every 100 meters) to detect the concentration of corrosive substances (e.g., Cl⁻, dissolved oxygen) in real time.

[0029] S20: Based on the pipeline material information of the environmental parameters and pipeline information, the contribution of the list of corrosive substances of the water quality parameters is calibrated to obtain a corrosion contribution list;

[0030] Furthermore, based on the environmental parameters and pipeline material information, the contribution of the corrosive substances list of the water quality parameters is calibrated to obtain a corrosion contribution list. Step S20 includes the following steps:

[0031] S21: Using a corrosion table pre-built by the user, retrieve multiple corrosion substance arrays of the pipeline material information, wherein any one of the multiple corrosion substance arrays includes at least one corrosion substance and has a cooperative environmental parameter calibrated by the material corrosion chemical formula, and any environmental attribute of the cooperative environmental parameter has a deviation threshold identifier and an attribute weight identifier.

[0032] S22: Compare the detected substances of the water quality parameters with the plurality of corrosive substance arrays to obtain a plurality of selected corrosive substance arrays, and construct the corrosive substance list by using the plurality of selected corrosive substance arrays as a plurality of list elements;

[0033] S23: Based on the environmental parameters and the cooperative environmental parameters, the contribution of the list of corrosive substances is calibrated to obtain a corrosion contribution list.

[0034] Furthermore, based on the environmental parameters and the cooperative environmental parameters, the contribution of the corrosive substance list is calibrated to obtain a corrosion contribution list. Step S23 includes the following steps:

[0035] S231: Obtain the first corrosive substance array of the corrosive substance list, and extract the first attribute deviation threshold and the first attribute weight from the deviation threshold identifier and the attribute weight identifier, up to the Nth attribute deviation threshold and the Nth attribute weight;

[0036] S232: Calculate the first attribute environmental deviation of the environmental parameter and the collaborative environmental parameter up to the Nth attribute environmental deviation;

[0037] S233: Calculate the ratio of the first attribute deviation threshold to the first attribute environmental deviation until the Nth attribute ratio is calculated. Weight the first attribute up to the Nth attribute weight to obtain the contribution of the first corrosive substance array. When the ratio of the first attribute deviation threshold to the first attribute environmental deviation is greater than 1, the first attribute ratio is configured as 1. When it is less than 1, the first attribute ratio is equal to the calculated ratio.

[0038] S234: Add the contribution of the first corrosive substance array to the corrosion contribution list.

[0039] Specifically, the corrosion table is a predefined database that records the chemical reaction relationships between different pipe materials (such as carbon steel and cast iron) and specific corrosive substances (such as Cl⁻ and SO₄²⁻) under different environmental parameters (temperature and humidity). It includes data on corrosion rate and corrosion mechanisms (pitting corrosion, uniform corrosion). Co-environmental parameters refer to the combination of multiple environmental attributes that the reaction between corrosive substances and pipe materials depends on (e.g., when temperature ≥ 30℃ and pH < 6, the corrosivity of Cl⁻ to stainless steel is enhanced). Attribute deviation threshold indicators refer to the allowable deviation range between the actual value of the environmental parameter and the ideal value in the corrosion table (e.g., temperature deviation ± 5℃; exceeding this requires adjustment of the contribution calculation). Attribute weight indicators indicate the influence weight of different environmental parameters on corrosion contribution (e.g., temperature weight 0.6, pH weight 0.4). Contribution calibration quantifies the contribution of corrosive substances to pipe corrosion, dynamically adjusting the weights based on environmental and material characteristics.

[0040] In detail, relevant arrays of corrosive substances are selected from the corrosion table based on the pipe material (e.g., carbon steel). For example, the corrosion substance array for carbon steel might include Cl⁻, dissolved oxygen, and SO₄²⁻, with each array associated with co-environmental parameters (e.g., Cl⁻ corrosion requires a temperature >25℃ and a flow rate <1m / s). The deviation threshold (e.g., allowable temperature deviation ±3℃) and attribute weight (temperature weight 0.7) are extracted from the arrays. Actual water quality parameters (e.g., Cl⁻ detected at 300ppm, SO₄²⁻ at 150ppm) are matched with the arrays in the corrosion table to filter out matching corrosion substances (e.g., Cl⁻, SO₄²⁻), forming a corrosion substance list. The deviation between the actual environmental parameters and the co-environmental parameters is calculated (e.g., measured temperature 28℃ vs. ideal temperature 25℃, deviation +3℃). The deviation threshold (±5℃) is used to determine if the limit is exceeded: if not, the ratio (3 / 5 = 0.6); if exceeded, the ratio is set to 1. A weighted sum is calculated: temperature ratio 0.6 × weight 0.7 + pH ratio 0.8 × weight 0.3 = Overall contribution: 0.66.

[0041] By comparing environmental deviations with thresholds, the contribution value is not distorted due to local environmental fluctuations (such as a sudden rise in temperature in a short period of time). The interaction of multiple factors is comprehensively considered in conjunction with environmental parameters (such as high temperature + low pH aggravating Cl⁻ corrosion), which is closer to the actual corrosion scenario.

[0042] S30: Based on the corrosion contribution list and the corrosive substance list, perform pipe segment clustering on the first water supply pipeline to obtain the water supply pipeline segmentation results;

[0043] Furthermore, based on the corrosion contribution list and the corrosive substance list, the first water pipeline is clustered to obtain the water pipeline segmentation results. Step S0 includes the following steps:

[0044] S31: Extract the first-position corrosion contribution list and the first-position corrosion substance concentration list of the water quality parameters;

[0045] S32: Weight the list of corrosive substances at the first location by the list of corrosion contribution at the first location, and construct the spatial coordinates of the first location;

[0046] S33: Extract the corrosion contribution list and the corrosion substance concentration list of the second position of the water quality parameter adjacent to the first position;

[0047] S34: Weight the corrosion substance concentration list at the second location using the corrosion contribution list at the second location to construct the spatial coordinates of the second location;

[0048] S35: When the Euclidean distance between the spatial coordinates of the first position and the spatial coordinates of the second position is less than or equal to the clustering distance, the first position and the second position are added to the same segment; otherwise, the segment is divided into different segments by the midpoint of the line connecting the first position and the second position.

[0049] S36: Repeated analysis: When the Euclidean distance between any two adjacent positions is greater than the cluster distance, output the segmentation result of the water pipeline.

[0050] Specifically, spatial coordinate construction refers to combining the list of corrosion substance concentrations and the list of contributions for each detection point to generate a multi-dimensional vector coordinate. For example, if the corrosion substances are Cl⁻ and SO₄²⁻, the coordinate dimension is the concentration of each substance × contribution (e.g., Cl⁻ concentration 300ppm × contribution 0.7=210, SO₄²⁻ concentration 150ppm × contribution 0.5=75, the coordinate vector is [210,75]); Euclidean distance: measures the similarity of the spatial coordinates of two detection points in multi-dimensional space, used to determine whether they belong to the same corrosion feature segment; Clustering distance: a preset corrosion feature similarity threshold (e.g., distance ≤50), based on historical data or experimental calibration, reflecting the tolerance for differences in corrosion mechanisms; Segment division: when the corrosion features of two points are significantly different, the midpoint of the connecting line is used as the boundary point between the two points with differences, and then the clustering continues in a loop until the clustering condition is met, that is, the Euclidean distance between any two adjacent positions is greater than the clustering distance, ensuring that the corrosion features within the segment are consistent.

[0051] For example, the corrosion contribution list (e.g., Cl⁻:0.7, SO₄²⁻:0.5) and concentration list (Cl⁻:300ppm, SO₄²⁻:150ppm) for the first position are extracted, and weighted to generate coordinates A: [300×0.7, 150×0.5] = [210, 75]; the contribution list (Cl⁻:0.6, SO₄²⁻:0.4) and concentration list (Cl⁻:280ppm, SO₄²⁻:130ppm) for the second position are extracted, and coordinates B: [280×0.6, 130×0.4] = [168, 52]. The Euclidean distance is calculated to be 48. Assuming a clustering distance of 50, the first and second positions are grouped into the same segment. If the distance between the first position and the second position is 55 (exceeding the threshold), then the midpoint of the line connecting the second position and the first position is used as the dividing line between the first position and the second position.

[0052] By using a weighted coordinate system of corrosion contribution and concentration, pipe sections with similar corrosion mechanisms (such as high Cl⁻ contribution areas) are clustered to avoid redundancy in full inspection.

[0053] S40: Traverse the segmentation results of the water pipeline, process the list of corrosive substance concentrations and the list of corrosion contribution through the pre-trained pipeline thickness loss prediction model, generate the pipeline thickness loss time series prediction value, and combine the pipeline thickness information of the pipeline information to determine the pipeline failure prediction time.

[0054] Furthermore, the pipeline segmentation results are traversed, and the list of corrosive substance concentrations and the list of corrosion contribution are processed by a pre-trained pipeline thickness loss prediction model to generate time-series predicted values ​​of pipeline thickness loss, including:

[0055] Extract the first segment corrosion substance concentration list and the first segment corrosion contribution list from the first segment results of the water pipeline segmentation results;

[0056] The pipeline thickness loss prediction model processes the first segment corrosion substance concentration list and the first segment corrosion contribution list, outputs the first segment pipeline thickness loss time series prediction value, and adds it to the pipeline thickness loss time series prediction value.

[0057] Furthermore, the pre-trained pipe thickness loss prediction model includes:

[0058] Configure a pipeline corrosion substance concentration record list, a pipeline corrosion contribution record list, and a pipeline thickness loss record true value time series information with a preset duration;

[0059] Using the true time-series information of the pipeline thickness loss record as supervision, and the list of pipeline corrosion substance concentration records and the list of pipeline corrosion contribution records as input, multiple pipeline thickness loss prediction base models are trained.

[0060] Configure the output mean calculation rules for the multiple pipeline thickness loss prediction base models to generate the pipeline thickness loss prediction model.

[0061] Specifically, the time-series prediction of pipeline thickness loss is as follows: Based on the concentration and contribution of corrosive substances, the pipeline thickness loss (unit: mm / year) is predicted at multiple future time points (e.g., annually), forming time-series data; Base Model: Refers to independent prediction models (e.g., LSTM, Random Forest), each model learns the corrosion kinetics from different perspectives (e.g., LSTM captures time-series dependencies, Random Forest handles nonlinear relationships); Output Mean Calculation Rules: The arithmetic mean or weighted average of the prediction results of multiple base models is taken to reduce the risk of overfitting and improve generalization ability; Failure Prediction Time: When the predicted cumulative thickness loss exceeds the pipeline safety threshold (e.g., original thickness 10mm → remaining thickness ≤ 2mm), it is determined as the failure time.

[0062] In detail, the model pre-training process is as follows: Historical data collection: Cl⁻ and SO₄²⁻ concentration records for a carbon steel pipeline over the past 10 years (sampling monthly), a corrosion contribution list (dynamically calculated), and thickness measurement true values ​​(annual detection data); Time series alignment: The input (concentration + contribution) and output (thickness loss) are divided into training samples according to a time window (e.g., 3 years); Base model training: LSTM model: Input is the time series concentration and contribution matrix, output is the thickness loss sequence (e.g., predicting the loss for the next 5 years); Random forest model: Input is the sliding window mean (e.g., annual average concentration, contribution), output is the annual loss value; Ensemble model generation: The prediction results of LSTM and Random Forest are averaged.

[0063] For example, a time series prediction example: The current thickness of the first segment is 8mm, and the predicted thickness loss over the next 5 years is [0.15, 0.18, 0.20, 0.22, 0.25] mm / year; cumulative loss: total loss at the end of the 5th year = 1.0mm, remaining thickness = 7.0mm (assuming a safety threshold of 5mm); failure determination: based on linear or nonlinear regression extrapolation, the time when the remaining thickness is ≤5mm (e.g., the 15th year) is the predicted failure time.

[0064] The integrated model takes into account both temporal and static characteristics, avoiding the bias of a single model (such as the sensitivity of LSTM to noise); dynamic early warning: the predicted value is updated by combining real-time data, and the detection plan is dynamically adjusted (such as the detection is required in advance if the time of prediction failure is shortened from 10 years to 8 years).

[0065] S50: Send the predicted pipeline failure time to the water conservancy project quality inspection terminal for inspection task allocation.

[0066] Furthermore, the predicted pipeline failure time is sent to the water conservancy project quality inspection terminal for inspection task allocation, including:

[0067] Construct a urgency factor function for allocating detection tasks:

[0068] ,

[0069] Where x represents the pipe segment and t represents time. Characterizes the urgency coefficient of the detection task allocation for pipeline segment x at time. The threshold representing the contribution of the i-th attribute corrosive substance preset by the user. Q represents the corrosion contribution of the corrosive substance with the i-th attribute, and Q represents the quantity of corrosive substance. Characterizing the predicted time of pipeline failure, Characterizing the smoothing constant;

[0070] Based on the urgency coefficient function for the detection task allocation, the urgency coefficient for the detection task allocation of any pipeline segment is obtained. Based on the urgency coefficient for the detection task allocation, the detection resources are prioritized from largest to smallest and the pipeline quality detection is performed.

[0071] Specifically, the urgency coefficient function is a mathematical expression that quantifies the priority of pipeline section inspections. It combines the proportion of corrosive substances exceeding the threshold with the proximity of the failure time, and the formula is as follows: Based on the urgency coefficient function for the detection task allocation, the urgency coefficient for the detection task allocation of any pipeline segment is obtained. Based on the urgency coefficient for the detection task allocation, the detection resources are prioritized from largest to smallest and the pipeline quality detection is performed.

[0072] By using a task allocation urgency coefficient function to reflect corrosion risk and time urgency in real time, the lag of fixed-cycle testing can be avoided, and limited resources (such as high-precision testing equipment) can be concentrated on high-risk pipe sections to reduce the probability of missed detection.

[0073] The water conservancy project quality testing method provided in this embodiment of the invention has at least the following technical effects:

[0074] 1. By collecting pipeline information, environmental parameters, and water quality parameters of the first water conveyance pipeline of a water conservancy project with multiple water inlets, and then using pipeline materials and environmental parameters to calibrate the contribution of a list of corrosive substances, a corrosion contribution list is obtained. Further, based on the corrosion contribution lists from multiple locations and the list of corrosive substances, pipe segments are clustered to obtain the segmentation results of the first water conveyance pipeline. For each segment, a pipeline thickness loss prediction model is used to process the list of corrosive substance concentrations and the corrosion contribution list to obtain a time-series predicted value of pipeline thickness loss. Furthermore, pipeline thickness information is extracted, and thickness reduction can be performed according to time sequence, thereby calibrating the pipeline failure prediction time as a resource allocation for subsequent quality inspection tasks. Segmentation enables spatial positioning of pipe segments, and failure time prediction enables temporal positioning of pipe segments, achieving targeted quality inspection and improving the technical effect of improving quality inspection efficiency.

[0075] 2. By using the urgency coefficient function for task allocation, the corrosion risk and time urgency can be reflected in real time, avoiding the lag of fixed-cycle testing. Limited resources (such as high-precision testing equipment) can be concentrated on high-risk pipe sections, reducing the probability of missed detection.

[0076] Example 2, as Figure 2 As shown, based on the same inventive concept as the water conservancy project quality inspection method provided in Embodiment 1, this embodiment of the invention also provides a water conservancy project quality inspection system, including:

[0077] The data acquisition module is used to obtain pipeline information, environmental parameters and water quality parameters of the first water conveyance pipeline of a water conservancy project with multiple water inlets. The environmental parameters of the first water conveyance pipeline are consistent throughout the entire pipeline, while the water quality parameters of the first water conveyance pipeline vary throughout the entire pipeline.

[0078] The contribution calibration module is used to calibrate the contribution of the list of corrosive substances in the water quality parameters based on the pipeline material information of the environmental parameters and pipeline information, and obtain the corrosion contribution list.

[0079] The pipe segment clustering module is used to perform pipe segment clustering on the first water transmission pipeline based on the corrosion contribution list and the corrosive substance list to obtain the water transmission pipeline segmentation results.

[0080] The pipeline failure calibration module is used to traverse the segmentation results of the water transmission pipeline, process the list of corrosive substance concentrations and the list of corrosion contribution through a pre-trained pipeline thickness loss prediction model, generate pipeline thickness loss time-series prediction values, and calibrate the pipeline failure prediction time by combining the pipeline thickness information of the pipeline information.

[0081] The detection task allocation module is used to send the predicted pipeline failure time to the water conservancy project quality detection terminal for detection task allocation.

[0082] Furthermore, the contribution rating module execution steps include:

[0083] By using a pre-built corrosion table, multiple corrosion substance arrays of the pipeline material information are retrieved. Each of the multiple corrosion substance arrays includes at least one corrosion substance and has a cooperative environmental parameter calibrated by the material corrosion chemical formula. Any environmental attribute of the cooperative environmental parameter has a deviation threshold identifier and an attribute weight identifier.

[0084] The detected substances of the water quality parameters are compared with the multiple arrays of corrosive substances to obtain multiple selected arrays of corrosive substances. The multiple selected arrays of corrosive substances are used as multiple list elements to construct the list of corrosive substances.

[0085] Based on the environmental parameters and the cooperative environmental parameters, the contribution of the list of corrosive substances is calibrated to obtain a corrosion contribution list.

[0086] Furthermore, the contribution rating module execution steps include:

[0087] Obtain a first corrosive substance array from the list of corrosive substances, and extract the first attribute deviation threshold and the first attribute weight from the deviation threshold identifier and the attribute weight identifier, up to the Nth attribute deviation threshold and the Nth attribute weight;

[0088] Calculate the environmental deviation of the first attribute of the environmental parameter and the collaborative environmental parameter up to the environmental deviation of the Nth attribute;

[0089] Calculate the ratio of the first attribute deviation threshold to the first attribute environmental deviation, and continue until the Nth attribute ratio is calculated. Weight the first attribute up to the Nth attribute weight to obtain the contribution of the first corrosive substance array. When the ratio of the first attribute deviation threshold to the first attribute environmental deviation is greater than 1, the first attribute ratio is configured as 1. When it is less than 1, the first attribute ratio is equal to the calculated ratio.

[0090] Add the contribution of the first corrosive substance array to the corrosion contribution list.

[0091] Furthermore, the pipe segment clustering module performs the following steps:

[0092] Extract the first-position corrosion contribution list and the first-position corrosion substance concentration list of the water quality parameters;

[0093] The list of corrosive substances at the first location is weighted by the list of corrosion contribution at the first location to construct the spatial coordinates of the first location;

[0094] Extract the corrosion contribution list and the corrosion substance concentration list of the second position of the water quality parameter adjacent to the first position;

[0095] The list of corrosion substances at the second location is weighted by the list of corrosion contribution at the second location to construct the spatial coordinates of the second location;

[0096] When the Euclidean distance between the spatial coordinates of the first position and the spatial coordinates of the second position is less than or equal to the clustering distance, the first position and the second position are added to the same segment; otherwise, the midpoint of the line connecting the first position and the second position is used to divide them into different segments.

[0097] Repeat the analysis. When the Euclidean distance between any two adjacent positions is greater than the cluster distance, output the segmentation result of the water pipeline.

[0098] Furthermore, the pipeline failure calibration module performs the following steps:

[0099] Extract the first segment corrosion substance concentration list and the first segment corrosion contribution list from the first segment results of the water pipeline segmentation results;

[0100] The pipeline thickness loss prediction model processes the first segment corrosion substance concentration list and the first segment corrosion contribution list, outputs the first segment pipeline thickness loss time series prediction value, and adds it to the pipeline thickness loss time series prediction value.

[0101] Furthermore, the pipeline failure calibration module performs the following steps:

[0102] Configure a pipeline corrosion substance concentration record list, a pipeline corrosion contribution record list, and a pipeline thickness loss record true value time series information with a preset duration;

[0103] Using the true time-series information of the pipeline thickness loss record as supervision, and the list of pipeline corrosion substance concentration records and the list of pipeline corrosion contribution records as input, multiple pipeline thickness loss prediction base models are trained.

[0104] Configure the output mean calculation rules for the multiple pipeline thickness loss prediction base models to generate the pipeline thickness loss prediction model.

[0105] Furthermore, the steps executed by the detection task allocation module include:

[0106] Construct a urgency factor function for allocating detection tasks:

[0107] ,

[0108] Where x represents the pipe segment and t represents time. Characterizes the urgency coefficient of the detection task allocation for pipeline segment x at time. The threshold representing the contribution of the i-th attribute corrosive substance preset by the user. Q represents the corrosion contribution of the corrosive substance with the i-th attribute, and Q represents the quantity of corrosive substance. Characterizing the predicted time of pipeline failure, Characterizing the smoothing constant;

[0109] Based on the urgency coefficient function for the detection task allocation, the urgency coefficient for the detection task allocation of any pipeline segment is obtained. Based on the urgency coefficient for the detection task allocation, the detection resources are prioritized from largest to smallest and the pipeline quality detection is performed.

[0110] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0111] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0113] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0114] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0115] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0116] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for detecting quality of hydraulic engineering, characterized in that, The method comprises the following steps: obtaining pipeline information, environmental parameters and water quality parameters of a first water conveying pipeline of a water conservancy project having a plurality of water inlet branches, wherein the environmental parameters of the first water conveying pipeline are consistent throughout the pipeline, and the water quality parameters of the first water conveying pipeline differ throughout the pipeline; contributing degree calibration is performed on a corrosion substance list of the water quality parameters according to the environmental parameters and pipeline material information of the pipeline information, and a corrosion contributing degree list is obtained; pipeline segment clustering is performed on the first water conveying pipeline according to the corrosion contributing degree list and the corrosion substance list, and a water conveying pipeline segmentation result is obtained; the water conveying pipeline segmentation result is traversed, the corrosion substance concentration list and the corrosion contributing degree list are processed by a pre-trained pipeline thickness loss prediction model to generate pipeline thickness loss time series prediction values, and pipeline thickness information of the pipeline information is combined to calibrate a pipeline failure prediction time; the pipeline failure prediction time is sent to a water conservancy project quality detection end for detection task allocation; wherein, according to the environmental parameters and the pipeline material information of the pipeline information, the contributing degree of the corrosion substance list of the water quality parameters is calibrated to obtain a corrosion contributing degree list, which comprises: retrieving a plurality of corrosion substance arrays of the pipeline material information through a user-prebuilt corrosion number table, wherein any one of the plurality of corrosion substance arrays includes at least one corrosion substance, and has a cooperative environmental parameter calibrated by a material corrosion chemical formula, and any environmental attribute of the cooperative environmental parameter has a bias threshold value identifier and an attribute weight identifier; comparing the detected substances of the water quality parameters with the plurality of corrosion substance arrays to obtain a plurality of selected corrosion substance arrays, taking the plurality of selected corrosion substance arrays as a plurality of list elements, and constructing the corrosion substance list; based on the environmental parameters and the cooperative environmental parameters, the contributing degree of the corrosion substance list is calibrated to obtain a corrosion contributing degree list; wherein, based on the environmental parameters and the cooperative environmental parameters, the contributing degree of the corrosion substance list is calibrated to obtain a corrosion contributing degree list, which comprises: obtaining a first corrosion substance array of the corrosion substance list, extracting a first attribute bias threshold value and a first attribute weight from the bias threshold value identifier and the attribute weight identifier, until the Nth attribute bias threshold value and the Nth attribute weight are extracted; calculating the first attribute environmental bias of the environmental parameters and the cooperative environmental parameters until the Nth attribute environmental bias is calculated; calculating the first attribute ratio of the first attribute bias threshold value and the first attribute environmental bias until the Nth attribute ratio is calculated, and weighting according to the first attribute weight until the Nth attribute weight to obtain the first corrosion substance array contributing degree, wherein when the ratio of the first attribute bias threshold value and the first attribute environmental bias is greater than 1, the first attribute ratio is configured as 1, and when it is less than 1, the first attribute ratio is equal to the calculated ratio; adding the first corrosion substance array contributing degree to the corrosion contributing degree list; wherein, according to the corrosion contributing degree list and the corrosion substance list, pipeline segment clustering is performed on the first water conveying pipeline to obtain a water conveying pipeline segmentation result, which comprises: extracting a first position corrosion contribution degree list and a first position corrosion material concentration list of the water quality parameters; weighting the first position corrosion material concentration list by the first position corrosion contribution degree list to construct a first position spatial coordinate; extracting a second position corrosion contribution degree list and a second position corrosion material concentration list of the water quality parameters adjacent to the first position; weighting the second position corrosion material concentration list by the second position corrosion contribution degree list to construct a second position spatial coordinate; when the Euclidean distance between the first position spatial coordinate and the second position spatial coordinate is less than or equal to a clustering distance, adding the first position and the second position into the same section, otherwise, dividing the midpoint of the line connecting the first position and the second position into different sections; repeating the analysis, and when the Euclidean distance between any two adjacent positions is greater than the clustering distance, outputting the water pipeline segmentation result.

2. The method of claim 1, wherein, traversing the water pipeline segmentation result, processing the corrosion material concentration list and the corrosion contribution degree list by a pre-trained pipeline thickness loss prediction model to generate a pipeline thickness loss time series prediction value, including: extracting a first segment corrosion material concentration list and a first segment corrosion contribution degree list of a first segment result of the water pipeline segmentation result; processing the first segment corrosion material concentration list and the first segment corrosion contribution degree list by the pipeline thickness loss prediction model to output a first segment pipeline thickness loss time series prediction value, and adding the first segment pipeline thickness loss time series prediction value into the pipeline thickness loss time series prediction value.

3. The method of claim 2, wherein, pre-training the pipeline thickness loss prediction model, including: configuring a pipeline corrosion material concentration record list, a pipeline corrosion contribution degree record list, and a pipeline thickness loss record true value time series information with a preset time length; using the pipeline thickness loss record true value time series information as supervision, using the pipeline corrosion material concentration record list and the pipeline corrosion contribution degree record list as input, training a plurality of pipeline thickness loss prediction base models; configuring an output mean calculation rule for the plurality of pipeline thickness loss prediction base models to generate the pipeline thickness loss prediction model.

4. The method of claim 1, wherein, sending the pipeline failure prediction time to a water conservancy project quality detection end for detection task allocation, including: constructing a detection task allocation urgency coefficient function: wherein x represents a pipeline section, t represents time, p(x, t) represents a detection task allocation urgency coefficient of the pipeline section x at time t, c i0 represents a contribution threshold of the i-th attribute corrosive substance preset by the user, c i1 represents a contribution of the i-th attribute corrosive substance, Q represents a quantity of corrosive substances, t(pre) represents a pipeline failure prediction time, and ∈ represents a smoothing constant obtaining a detection task allocation urgency coefficient of any pipeline segment based on the detection task allocation urgency coefficient function, and arranging the detection resource priority based on the detection task allocation urgency coefficient from large to small to perform pipeline quality detection.

5. A hydraulic engineering quality detection system, characterized in that, for implementing the water conservancy project quality detection method according to any one of claims 1-4, including: a data acquisition module for obtaining pipeline information, environmental parameters and water quality parameters of a first water pipeline of a water conservancy project with a plurality of water inlet branches, wherein the environmental parameters of the first water pipeline are consistent throughout the section, and the water quality parameters of the first water pipeline are different throughout the section; a contribution degree calibration module for calibrating the corrosion material list of the water quality parameters according to the environmental parameters and the pipeline material information of the pipeline information to obtain a corrosion contribution degree list; a pipe segment clustering module configured to perform pipe segment clustering on the first water delivery pipeline according to the corrosion contribution degree list and the corrosion substance list, and obtain a water delivery pipeline segmentation result; a pipeline failure calibration module configured to traverse the water delivery pipeline segmentation result, process the corrosion substance concentration list and the corrosion contribution degree list by using a pre-trained pipeline thickness loss prediction model, generate a pipeline thickness loss time series prediction value, and calibrate a pipeline failure prediction time in combination with pipeline thickness information of pipeline information; a detection task allocation module configured to send the pipeline failure prediction time to a water conservancy project quality detection end for detection task allocation.

Images