Detection method and system for fire-fighting pipeline

Abstract

The invention relates to the technical field of fire-fighting pipeline detection, and discloses a detection method and system for a fire-fighting pipeline, and the method comprises the steps: collecting the reflection data of the inner wall of a pipeline through a mobile detection device, extracting a corrosion feature vector and a wall thickness change feature vector, and carrying out the fusion of a corrosion feature vector and a wall thickness change feature vector to generate a comprehensive feature vector; when an abnormal index in the comprehensive feature vector is greater than a preset index threshold value, marking the corresponding pipe section as a defective pipe section; determining the defect type of the defect pipe section according to the comprehensive feature vector and a defect type historical probability distribution model; acquiring coordinate data of the defective pipe section, and calculating a health index of the defective pipe section in combination with the defect type; and obtaining a historical health index change rule of the defective pipe section, and generating a state evaluation report of the defective pipe section in combination with the health index. The problems that pipeline detection depends on manpower, the efficiency is low, and the health state of the pipeline is difficult to quantitatively evaluate are solved, and accurate recognition of defects of the fire-fighting pipeline and quantitative evaluation of the health state are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of fire protection pipeline detection, and in particular to a detection method and system for fire protection pipelines. Background Art

[0002] As core infrastructure for building life safety, the operational status of fire protection piping systems is directly linked to the effectiveness of fire emergency response and the safety of life and property. With the acceleration of urbanization and the increasing complexity of buildings, the scale of fire protection piping networks is growing. Accurately monitoring their health has become a critical component of fire safety management.

[0003] Fire protection pipe inspections are usually performed manually or using fixed sensors. However, the internal space of fire protection pipes is small, making it difficult for manual inspections to penetrate narrow or hidden pipe areas. In addition, when using fixed sensors, their detection coverage is limited, and it is impossible to accurately judge the overall health status of the pipes. Especially when the pipes have multiple potential problems such as corrosion, blockage, and wall thickness changes, the single-dimensional sensor information often leads to misjudgments or missed detections.

[0004] Therefore, how to conduct comprehensive inspections on narrow and complex fire protection pipes, integrate and analyze multi-dimensional inspection data, and generate accurate and reliable pipeline health assessment reports is an urgent problem that needs to be solved in fire protection pipe inspections. Summary of the Invention

[0005] The embodiments of the present application provide a method and system for detecting fire protection pipelines, which solve the problems of pipeline inspection relying on manual labor, low efficiency, and difficulty in quantitatively assessing the health status of pipelines, and achieve accurate identification of fire protection pipeline defects and quantitative assessment of their health status.

[0006] The present invention provides a method for detecting fire protection pipes, the method comprising: collecting pipeline inner wall reflection data by a mobile detection device, and extracting a corrosion feature vector and a wall thickness change feature vector of the pipeline inner wall reflection data; fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector; If it is detected that the comprehensive feature vector has an abnormal indicator, and the abnormal indicator is greater than a preset indicator threshold, the pipe section corresponding to the abnormal indicator is determined to be a defective pipe section; Calculating initial probability distribution data of the defect type of the defective pipe section based on the comprehensive feature vector, and determining the defect type of the defective pipe section in combination with a historical probability distribution model of the defect type; Acquiring coordinate data of the defective pipe section, and calculating a health index of the defective pipe section according to the coordinate data and the defect type; Obtain the historical health index variation pattern of the defective pipe section, and generate a status assessment report of the defective pipe section in combination with the health index.

[0007] Optionally, the step of collecting pipeline inner wall reflection data by a mobile detection device and extracting the corrosion feature vector and the wall thickness change feature vector of the pipeline inner wall reflection data includes: The mobile detection device collects spectral reflection data and wall thickness acoustic wave reflection data of the inner wall of the pipeline; If the infrared band intensity in the spectral reflection data is greater than a preset infrared fluctuation threshold, and the ultrasonic echo time difference of the wall thickness acoustic wave reflection data is greater than a preset standard time difference, the inner wall of the pipeline is marked as a defective pipeline; The corrosion feature vector and the wall thickness variation feature vector of the defective pipe are extracted.

[0008] Optionally, the step of fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector includes: Obtaining an accuracy level value of a sensor in the motion detection device, and adjusting a weight value of the sensor according to the accuracy level value to generate a weight parameter; The corrosion feature vector and the wall thickness change feature vector are fused in combination with the weight parameter to generate a comprehensive feature vector corresponding to the pipeline risk area.

[0009] Optionally, the step of calculating the initial probability distribution data of the defect type of the defective pipe section based on the comprehensive feature vector and determining the defect type of the defective pipe section in combination with a historical probability distribution model of the defect type includes: Generate a probability distribution model of corrosion defects, crack defects and blockage defects of the pipeline based on the comprehensive feature vector, and obtain initial probability distribution data; Obtaining the defect type historical probability distribution model, and calculating the posterior probability values of the corrosion defect, the crack defect, and the blockage defect in combination with the initial probability distribution data; The posterior probability value is compared with a preset threshold value of the defect type, and the defect type is determined according to the comparison result.

[0010] Optionally, the step of obtaining coordinate data of the defective pipe section and calculating the health index of the defective pipe section according to the coordinate data and the defect type includes: Acquire the coordinate data, and fuse the coordinate data with the defect type to generate a defect type data set including a defect type identifier and the coordinate data; The defect weight and defect severity of the defective pipe section are calculated according to the defect type data set to obtain the health index.

[0011] Optionally, the step of obtaining a historical health index variation pattern of the defective pipe section and generating a status assessment report of the defective pipe section in combination with the health index includes: Obtaining the historical health index change pattern, and combining it with the coordinate data to generate a health index dataset including a health index time series and the coordinate data; generating a health index prediction value of the defective pipe section according to the health index data set and the health index, to obtain a prediction trend data set; The condition assessment report is generated based on the predicted trend data set, the coordinate data, the health index, and the defect type.

[0012] Optionally, after the step of obtaining the historical health index variation pattern of the defective pipe section and generating a status assessment report of the defective pipe section in combination with the health index, the following steps are included: Obtain the defect type and health index of the pipe section in the condition assessment report, determine the risk level of the pipe section, and obtain risk classification data; If the risk level of the defect point to be maintained in the risk classification data exceeds a preset risk threshold, a maintenance suggestion for the defect point to be maintained is generated in combination with the prediction trend data set to obtain maintenance planning information; The maintenance plan information is updated in the condition assessment report.

[0013] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention further provides a detection system for fire protection pipelines, the system comprising: Data acquisition module, used to collect pipeline inner wall reflection data and pipe section coordinate data; A defect detection module, configured to determine whether a pipe section has defects based on the reflection data from the inner wall of the pipe; a defect type identification module, configured to determine the defect type of the pipe section based on the comprehensive feature vector and the historical defect probability model; A health index calculation and prediction module, configured to output the health index of the pipe section according to the defect type and the coordinate data, and generate a prediction trend data set based on historical health index change patterns; A status assessment module is used to generate the status assessment report based on the predicted trend data set, the coordinate data, the health index and the defect type.

[0014] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention also provides a terminal device, including a memory, a processor, and a detection program for fire protection pipes stored in the memory and executable on the processor. When the processor executes the detection program for fire protection pipes, the method described above is implemented.

[0015] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention further provides a computer-readable storage medium, on which a detection program for fire protection pipes is stored. When the detection program for fire protection pipes is executed by a processor, the method described above is implemented.

[0016] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: (1) The present invention uses a mobile detection device to scan the entire section of the fire protection pipeline, capturing the corrosion and wall thickness change characteristics in the pipeline, and providing a comprehensive data basis for subsequent analysis.

[0017] (2) The present invention fuses the spectral reflection data and wall thickness acoustic reflection data collected by multiple sensors to eliminate the error of a single sensor, generate a feature vector that can comprehensively reflect the health status of the pipeline, and improve the reliability of anomaly detection.

[0018] (3) The present invention realizes automatic determination of defective pipe sections based on quantitative thresholds and matches them with historical data to determine the defect type of the defective pipe section, thereby avoiding the subjectivity of manual judgment and improving detection efficiency.

[0019] (4) The present invention generates a quantifiable health index through coordinate association and defect weighting algorithm, realizing intuitive visualization of pipeline risks.

[0020] (5) The present invention integrates the current pipeline health index with historical trends, predicts the pipeline status change trend and generates a status assessment report, which can help managers maintain risky pipelines in advance and ensure the safety and stability of pipeline operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of Example 1 of the detection method for fire protection pipes of the present application; Figure 2 This is a flow chart of Example 2 of the detection method for fire protection pipes of the present application; Figure 3 This is a schematic diagram of the terminal structure of the hardware operating environment involved in an embodiment of the present application. DETAILED DESCRIPTION

[0022] To address the difficulty in quantifying and assessing the health status of fire protection pipelines, this application uses a mobile detection device to collect spectral and ultrasonic data from the pipeline's inner wall. Feature extraction and fusion generate a comprehensive feature vector, which, combined with a Bayesian probability model, enables intelligent defect identification and classification. Furthermore, through health index calculation and trend analysis, the pipeline's future health status is predicted and a status assessment report is generated. This approach enables intelligent detection, assessment, and early warning of pipeline inner wall defects, providing a scientific basis for pipeline maintenance and effectively improving the safety and reliability of pipeline operations.

[0023] To better understand the above technical solutions, exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0024] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0025] Example 1

[0026] In this embodiment, a detection method for fire protection pipelines is provided.

[0027] Reference Figure 1 The fire protection pipeline detection method of this embodiment includes the following steps: Step S100: collecting pipeline inner wall reflection data by a mobile detection device, and extracting the corrosion feature vector and the wall thickness change feature vector of the pipeline inner wall reflection data; In this embodiment, the mobile detection device, comprised of a multispectral imaging sensor and an ultrasonic sensor, scans the entire pipeline space, acquiring spectral reflectance data and acoustic reflectance data of the inner wall thickness. The corrosion feature vector includes corrosion area percentage, corrosion depth distribution, corrosion morphology, corrosion product spectral characteristics, and surface roughness. The wall thickness variation feature vector includes the remaining wall thickness, wall thinning rate, nonuniformity index, defect area, and axial distribution characteristics.

[0028] As an optional implementation, a multispectral imaging sensor is used to scan the entire inner wall of the pipeline, obtain spectral reflectance data of the pipeline inner wall, extract infrared band intensity information, and store it in a pre-established database to obtain preliminary information on the distribution of spectral abnormality areas. An ultrasonic sensor is used to scan the entire inner wall of the pipeline, obtain wall thickness data, and record the ultrasonic echo time difference. If the echo time difference is greater than a preset percentage of the echo time standard value, it is marked as a potential defect area, and preliminary information on the wall thickness abnormality area is determined. Based on the spectral abnormality area distribution information and the wall thickness abnormality area information, data fusion processing is performed, and overlapping areas are marked as defective pipelines. The method of determining defective pipelines by fusing data based on spectral abnormality areas and wall thickness abnormality areas can improve the accuracy of defect determination and reduce the risk of misjudgment.

[0029] For example, a multispectral imaging sensor illuminates the inner wall of a pipe with light in different wavelengths, capturing the characteristics of the reflected spectrum, particularly the intensity of the infrared band, to determine material changes or surface anomalies. For example, while inspecting a section of industrial pipe with a diameter of 0.5 meters, the sensor scans and detects an infrared band intensity value in a certain area that reaches 1.2 times the preset threshold, indicating possible surface roughness changes or material deposition caused by corrosion. This area is marked as a spectral anomaly and stored in the database. An ultrasonic sensor infers changes in wall thickness by emitting ultrasonic waves and measuring the time difference of their echoes. For example, in the same section of pipe, if the echo time difference in a certain area reaches 1.3 times the preset echo time standard value, exceeding the preset echo time standard value by a certain percentage, it indicates that the wall thickness may have decreased due to corrosion, and this area is marked as a defective pipe. This non-contact method of rapidly locating potential problem areas directly reflects the structural integrity of the pipeline and provides a physical basis for defect identification.

[0030] As another optional implementation, after the defective pipeline is determined, the corresponding coordinate data can be recorded, and the defect location data can be associated with the spatial coordinates of the inner wall of the pipeline to provide precise positioning support for subsequent maintenance.

[0031] For example, assume a pipeline is 100 meters long and a defect is located at the 25th meter. Through module recording, the defect is precisely mapped to a spatial coordinate system, and a defect distribution map with location identifiers is generated and stored in the database. Assuming that three defect areas are found during the above inspection, located at the 25th, 50th, and 75th meters, the distribution map intuitively displays the defect locations and the corresponding corrosion eigenvectors and wall thickness variation eigenvectors, allowing managers to quickly understand the pipeline's health.

[0032] As another optional implementation, after marking the defective pipeline, the corresponding corrosion feature vector and wall thickness change feature vector are extracted, and feature extraction can be performed on different types of detection data using a convolutional neural network algorithm.

[0033] For example, when extracting corrosion features, a multispectral scan of a natural gas pipeline with a diameter of 0.6 meters was performed, and an abnormal increase in the infrared band intensity in the local area was detected, reaching 1.5 times the preset band threshold. Spectral analysis identified the presence of oxide deposition or surface roughness changes in the area, with its peak intensity 80% higher than that of the normal area. The image processing algorithm calculated that the area of the abnormal area was 15.6 square centimeters, accounting for approximately 8.2% of the total area of the inner wall of the pipeline. At the same time, three-dimensional morphology reconstruction showed that the maximum corrosion depth in the area reached 0.75 mm, and the surface roughness Ra value reached 12.6 microns. To extract the characteristics of wall thickness changes, the ultrasonic sensor measured the wall thickness at intervals of 10 cm along the axial direction of the pipeline. The ultrasonic echo time difference detected in the defective pipe section was 69.6 microseconds, which was 20% higher than the standard value of 58 microseconds. Based on the sound speed conversion formula, the actual wall thickness of the area was calculated to be 5.0 mm, which was 16.7% thinner than the standard wall thickness of 6.0 mm. Continuous measurement values showed that the abnormal area extended about 50 cm along the axial direction of the pipeline, showing a distribution characteristic of being thin in the middle and thick at both ends.

[0034] Step S200: fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector; In this embodiment, after extracting the corrosion feature vector and the wall thickness change feature vector, the initial data such as the spectral reflectance data and the wall thickness data can be quantified. The initial data can be quantified according to dimensions such as data stability, environmental adaptability, and historical consistency. The accuracy level value of each sensor can be determined based on the quantification results. The corrosion feature vector and the wall thickness change feature vector are weightedly fused in combination with the accuracy level value to generate a comprehensive feature vector.

[0035] As an optional implementation, weighted fusion processing is performed based on the corrosion feature vector F1 and the wall thickness change feature vector F2 to obtain the accuracy level value of each sensor, and the weight value of the sensor is adjusted according to the accuracy level value to generate weight parameters W1 and W2. The corrosion feature vector and the wall thickness change feature vector are fused through the calculation formula F=W1*F1+W2*F2 to generate a comprehensive feature vector F.

[0036] For example, for a 0.6-meter section of natural gas pipeline, a multispectral imaging sensor collects visible light and infrared band data. After evaluating these data, it is found that the data fluctuation is less than 5%, so the multispectral imaging sensor is given a higher accuracy level value, such as 0.9. However, the echo data collected by the ultrasonic sensor fluctuates greatly due to environmental noise interference, and the accuracy level value may be 0.7. The accuracy level value of the multispectral imaging sensor is greater than that of the ultrasonic sensor. The weight value of the multispectral data is increased, for example, from 0.5 to 0.6, and the weight value of the ultrasonic data is correspondingly reduced to 0.4. The comprehensive feature vector is 0.6F1+0.4F2。

[0037] Step S300: If it is detected that the comprehensive feature vector has an abnormal indicator, and the abnormal indicator is greater than a preset indicator threshold, the pipe section corresponding to the abnormal indicator is determined to be a defective pipe section; In this embodiment, the abnormality indicator refers to an indicator related to the defect type, such as the comprehensive corrosion index, crack risk value, or flow resistivity. When an abnormality indicator in the comprehensive feature vector is detected to exceed the corresponding preset indicator threshold, the pipe section corresponding to the abnormality indicator is determined to be a defective pipe section.

[0038] As an optional implementation, each dimension in the comprehensive feature vector corresponds to a specific physical characteristic, such as corrosion depth, wall thinning rate, and spectral anomaly intensity. These characteristics are converted into standardized risk indices using a pre-calibrated mathematical model. These risk indices are then compared with dynamically adjusted indicator thresholds. When one or more risk indices in a pipe section's comprehensive feature vector exceed the preset indicator threshold, a multi-level verification mechanism is activated. This mechanism checks the spatiotemporal continuity of the anomaly characteristics, such as whether the corrosion area exhibits a gradient change between adjacent inspection points. Cross-sensor evidence verification is then performed, such as the simultaneous presence of infrared spectral anomalies and ultrasonic wall thinning.

[0039] For example, in a natural gas pipeline with a diameter of 0.6 meters, the fused data for a certain area indicated that the infrared band intensity exceeded the threshold and the wall thickness change rate reached 25%. These abnormal indicators exceeded the preset indicator threshold, and the corresponding pipe section in this area was determined to be defective. This cross-validation method of verifying pipeline status based on infrared spectral anomalies and ultrasonic wall thickness changes can effectively eliminate most false alarms caused by transient interference.

[0040] It should be noted that the preset indicator thresholds are not fixed values but benchmarks dynamically calculated through an aging model based on pipeline material properties, service life, and environmental factors. For example, the corrosion risk threshold for a 10-year-old carbon steel pipeline will be 15%-20% higher than that of a new pipeline to account for the natural degradation of material properties.

[0041] Optionally, after identifying a defective pipe section, the coordinates of the defective pipe can be recorded and the defect distribution information can be associated with the spatial coordinates of the pipe. For example, if the pipe is 80 meters long and the defect distribution information shows defective sections at meters 30 and 60, these locations can be recorded and a distribution map with location markers generated and stored in the database. This map allows managers to quickly locate the defect.

[0042] Step S400: calculating the initial probability distribution data of the defect type of the defective pipe section according to the comprehensive feature vector, and determining the defect type of the defective pipe section in combination with the defect type historical probability distribution model; In this embodiment, the Bayesian classification algorithm can be used to identify the defect type of the comprehensive feature vector, establish a probability distribution model for three types of defects: corrosion, cracks, and blockage, and then calculate the posterior probability of each type of defect. If the posterior probability of a certain type of defect exceeds the preset threshold of the defect type, the defect type corresponding to the pipe section is determined.

[0043] As an optional implementation, a probability distribution model of corrosion defects, crack defects and blockage defects is generated based on the comprehensive feature vector, and initial probability distribution data is obtained. The historical probability distribution model of each defect type is obtained, and the posterior probability values of corrosion defects, crack defects and blockage defects are calculated in combination with the initial probability distribution data. The posterior probability values are compared with the preset threshold of the defect type to determine the specific defect type.

[0044] For example, for a natural gas pipeline with a diameter of 0.8 meters, the comprehensive feature data includes surface corrosion features detected by a multispectral imaging sensor and wall thickness anomaly information captured by an ultrasonic sensor. A probability distribution model is established based on historical data, analyzing the distribution patterns of surface corrosion features and wall thickness anomaly information for different defect types to generate initial probability distribution data. For example, the initial probability distribution shows a probability of 0.6 for corrosion defects, 0.3 for crack defects, and 0.1 for blockage defects, indicating that the current data favors corrosion defects. The historical probability distribution model for the three defect types is then called, and the prior probability distribution data for the three defect types are 0.5 for corrosion, 0.3 for cracks, and 0.2 for blockages. Based on the initial and prior probability distribution data, the posterior probability values calculated are 73.2% for corrosion defects, 22.0% for cracks, and 4.8% for blockages. If the current preset thresholds for the defect type are 65% for corrosion, 45% for cracks, and 35% for blockages, comparing the posterior probability values with the preset thresholds can determine that the defect type is corrosion.

[0045] Optionally, after determining the defect type, a corresponding defect type identifier can be generated and associated with the spatial coordinates of the pipeline to create a defect location record. For example, if a pipeline is 100 meters long and analysis reveals that corrosion defects are concentrated between 40 and 45 meters, a record containing the specific location will be generated. This record shows abnormal infrared band intensity and a 20% wall thickness reduction in this area, indicating severe corrosion. This information is combined with the pipeline coordinate system to generate a visual distribution map and store it in a database. This precise positioning allows maintenance personnel to quickly identify problem areas and optimize maintenance plans.

[0046] Step S500: obtaining coordinate data of the defective pipe section, and calculating a health index of the defective pipe section according to the coordinate data and the defect type; In this embodiment, the database stores defect information at different locations along the pipeline, such as the corrosion level or crack distribution of a specific section, covering the specific conditions of each section along the entire length of the pipeline. Adjacent defects can affect each other. For example, two corrosion points within a meter of each other can create a synergistic effect, causing the health index to decrease 30%-50% faster than isolated defects. Defects in special locations, such as welds and elbows, require additional weighting. This information is acquired through coordinate positioning. Therefore, the health index is calculated by acquiring coordinate data and combining it with the defect type.

[0047] As an optional implementation, coordinate data is obtained, and the coordinate data is integrated with the defect type to generate a defect type data set including a defect type identifier and the coordinate data. Based on the defect type data set, the defect weight and defect severity of the defective pipe section are calculated to obtain a health index.

[0048] For example, corrosion defects can cause long-term cumulative damage and easily lead to perforation and leakage in the pipe wall, while crack defects can cause sudden fractures, but the impact is usually localized. Blockage defects primarily affect fluidity and pose a lower structural risk. Therefore, the defect weight values are adjusted based on the defect type: a basic weight of 0.8 for corrosion defects, a basic weight of 0.7 for crack defects, and a basic weight of 0.4 for blockage defects. For example, a defective pipe section has both corrosion and crack defects. Based on the coordinate data, the corrosion defect is located underground in a commercial area, with a weight factor of 0.7. Therefore, the resulting corrosion defect weight is 0.8*1.7=0.56. The crack defect is also located underground, with a weight factor of 0.9. If there is a corrosion defect within 1 meter of the crack defect, the weight is increased by 15%, resulting in a crack defect weight of 0.7*0.9*1.15=0.72. The severity of the defect is determined based on the inspection data and coordinate data. The crack defect is 15 cm long, corresponding to a score of 12, and 2 mm deep, corresponding to a score of 20. Based on the coordinate data, the crack is located near the weld, resulting in a calculated crack severity of (12 + 20) * 2 = 64. The corrosion defect has an area of 8 square centimeters and a depth of 1.5 mm, located in a straight pipe section. The calculated corrosion severity is (8 + 7.5) * 1 = 15.5. Based on the defect weight and defect severity, the health index is 0.56 * 64 + 0.72 * 15.5 = 47.

[0049] Optionally, a health index determines the risk level of fire protection pipes. A lower health index value indicates a greater risk. Health index thresholds are set differently for fire protection pipes made of different materials. For ordinary fire protection water pipes, the health index threshold can be set to 60, for pipes in high-level chemical zones to 30, and for underground trunk pipe networks to 40. When the health index falls below the preset health index threshold, the pipe section is in a risky area and requires repair.

[0050] Step S600: Obtain the historical health index variation pattern of the defective pipe section, and generate a status assessment report of the defective pipe section in combination with the health index.

[0051] In this embodiment, after calculating the health index of each pipe section, the historical change pattern of the health index can be obtained, and a status assessment report of the defective pipe section can be generated in combination with the health index. The status assessment report includes the defect location, defect type, severity and maintenance recommendations.

[0052] As an optional implementation, the historical health index variation pattern and corresponding coordinate data of each pipe segment are obtained from the database, and the two are fused to generate a health index dataset, which includes a health index time series and corresponding coordinate data.

[0053] For example, a database stores the health index records for each section of a 300-meter pipeline over the past 12 months. For example, the health index at the 100-meter mark has gradually decreased from 85 to 70 over the past six months. This data, collected from regular inspection equipment, is combined with location information to form a preliminary dataset. These scattered historical health indexes are then tied to specific locations to form a structured time series dataset.

[0054] As another optional implementation, after the health index data set is generated, a health index prediction value of the defective pipe section is generated in combination with the current health index to obtain a prediction trend data set.

[0055] For example, the historical decline rate of the health index is obtained based on the time series data set. Then, using the time series analysis method, a data stationarity test is first performed. After eliminating the trend through differential operation, a differential regression moving average model is established to predict the health index. In the future, to improve the prediction accuracy, spatial correlation factors can also be introduced. Assuming that there are two corrosion-defective pipe sections within 3 meters of the pipe section, their health indices are 54 and 61 respectively. By establishing a distance-based weight matrix, the spatial influence coefficient is calculated to be 1.18, and the corrected prediction value is adjusted to 56-60. At the same time, considering the continuous decline in soil resistivity in the area, an environmental corrosion correction factor of 0.9 is added. The final prediction result is updated to the risk threshold range of 50-54 for the health index in 6 months. This prediction method that integrates spatiotemporal features improves the accuracy of the prediction.

[0056] As another optional implementation, the prediction results of each defective pipe section are integrated to obtain a prediction trend data set, and a status assessment report can be generated based on the prediction trend data set, coordinate data, health index and defect type.

[0057] For example, the predicted trend dataset is linked to coordinate data and sorted by health index decline to determine a maintenance priority list. The maintenance priority list can also be categorized and stored by defect type, generating a condition assessment report. For example, a fire protection pipe inspection revealed a health index of 65, with a health index threshold of 50, and bottom corrosion depth of 1.8 mm. Predictive analysis indicated that without intervention, the health index would drop to 58 in three months and potentially fall below the safety threshold of 45-49 in six months, increasing the leakage risk to 32%. The assessment identified two associated corrosion points within a 2-meter radius, potentially accelerating overall degradation. Based on this analysis, the report recommended epoxy resin lining repair within two months and simultaneous inspection of the adjacent 3-meter pipe segment. All predicted data is confidence-calibrated to within a ±3% error range, and the assessment results are automatically updated quarterly to ensure the timeliness of maintenance plans. This data-driven assessment approach has proven effective in reducing sudden failure rates by over 30% while optimizing the efficiency of maintenance resource allocation.

[0058] Optionally, when associating the predicted trend dataset with the coordinate data, the surrounding environment information of the pipe section can also be associated with the predicted trend dataset based on the coordinate data, such as whether it is close to a residential area. If it is close to a residential area, the maintenance priority can be increased.

[0059] In this implementation, multispectral and ultrasonic composite detection technology, combined with a pipeline feature database, enables automated identification of defects such as corrosion and cracks, with millimeter-level accuracy. Furthermore, by establishing a dynamic pipeline health index calculation model that comprehensively considers defect type, spatial distribution, and environmental factors, the generated risk assessment results are over 90% consistent with actual operating conditions, providing early warning of potential failures 3-6 months in advance. The automatically generated assessment report includes current status analysis, trend predictions, and maintenance recommendations, supporting the precise scheduling of maintenance resources. This has significantly reduced operational costs, shortened fault response times, and significantly improved the safety and reliability of fire protection systems.

[0060] Example 2

[0061] Based on the first embodiment, another embodiment of the present application is proposed, referring to Figure 2 After obtaining the historical health index variation pattern of the defective pipe section and generating a status assessment report of the defective pipe section in combination with the health index, the following steps are included: Step S700: Obtain the defect type and health index of the pipe section in the condition assessment report, determine the risk level of the pipe section, and obtain risk classification data; Step S800: If the risk level of the defect point to be maintained in the risk classification data exceeds a preset risk threshold, a maintenance suggestion for the defect point to be maintained is generated in combination with the prediction trend data set to obtain maintenance planning information; Step S900: updating the maintenance plan information into the status assessment report.

[0062] In this embodiment, after obtaining the current health index data and predicted trend data for each pipeline segment from the database and performing data fusion processing to generate a preliminary dataset containing defect location, defect type, and current health index, a condition assessment report is generated. Based on the condition assessment report, the defect location is correlated with the three-dimensional coordinate data. By analyzing the defect type and current health index, the risk level of each defect point is determined, generating risk classification data. Based on this risk classification data and the predicted trend dataset, maintenance recommendations for the defect points to be maintained are generated, resulting in maintenance planning information. This maintenance planning information is then updated in the condition assessment report, resulting in the final pipeline status profile. This continuously updated pipeline status profile continuously calibrates the assessment model, improving prediction accuracy.

[0063] As an optional implementation method, after generating risk classification data, if the risk level of a defect point in the risk classification data exceeds the preset risk threshold, maintenance recommendations for the defect point are generated in combination with the predicted trend data and processing time requirements to obtain maintenance planning information.

[0064] For example, during the inspection of a fire protection pipeline, the assessment revealed a circumferential crack defect in section 5. The health index had dropped to 45, compared to the preset high-risk threshold of 50, and the risk level was marked as "urgent." Analysis of predictive trend data predicted that the health index of this section would drop to 38 within the next two months, and the crack length could increase by 15%. Based on the on-site construction conditions (a 2.5-meter buried depth, a main road above), and the availability of maintenance resources (epoxy resin repair requires 4 hours per meter, and pipe replacement requires 8 hours per meter), maintenance recommendations were automatically generated, prioritizing carbon fiber composite reinforcement, with completion within 14 days and a work window limited to 12:00 AM to 4:00 AM to minimize traffic impact. Preventive coating maintenance was also recommended for the adjacent 3-meter section with a health index of 58. This planning information was automatically updated to the pipeline status file.

[0065] As another optional implementation, the defect location, three-dimensional coordinates, risk level, severity, maintenance recommendations, and health status data are integrated and processed through maintenance planning information to determine the final pipeline status profile.

[0066] For example, after a comprehensive inspection of a fire protection pipeline at a certain plant, severe corrosion defects were discovered in pipe section 102. Through a series of data fusion operations, a complete pipeline status profile was generated: the pipe section's 3D coordinates (X:287.45, Y:156.32, Z:-2.8) showed a current health index of only 38 (out of a preset risk threshold of 50), resulting in a "critical" risk level. The corrosion area reached 35% of the pipe wall, with a maximum depth of 2.3mm. A predictive model indicated that if untreated, perforation and leakage could occur within two months. Based on the plant's production schedule, the next week's shutdown window, and the corrosion characteristics, a customized maintenance plan was generated: on-site reinforcement using polymer composite wrapping, requiring the closure of working channel 1 for eight hours and the simultaneous replacement of an adjacent two-meter section of deteriorated pipe. This profile was linked to the plant's 3D digital twin model in real time, highlighting high-risk areas with flashing red warnings. The profile was automatically pushed to the maintenance supervisor's mobile device, along with an accompanying emergency response plan, including a temporary water supply plan for the affected fire hydrants. After implementation and verification, the digital archive shortened the actual maintenance time by 20% compared with the estimate. The construction acceptance module integrated in the archive also fully recorded the comparison of inspection data before and after the repair, forming a traceable closed-loop management record, which provided data support for subsequent pipeline network renovation.

[0067] In this embodiment, risky pipeline sections are automatically identified based on risk classification data and maintenance plans are generated. The maintenance plans are updated in the status assessment report so that management personnel can allocate maintenance resources to the pipeline according to the status assessment report, greatly reducing operation and maintenance costs.

[0068] Example 3

[0069] In an embodiment of the present application, a detection device for fire protection pipes is provided.

[0070] Reference Figure 3 , Figure 3 This is a schematic diagram of the terminal structure of the hardware operating environment involved in an embodiment of the present application.

[0071] like Figure 3 As shown, the control terminal may include: a processor 1001, such as a CPU, a network interface 1003, a memory 1004, and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The network interface 1003 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1004 may be a high-speed RAM memory or a non-volatile memory, such as a disk drive. The memory 1004 may also be a storage device independent of the processor 1001.

[0072] Those skilled in the art will understand that Figure 3The terminal structure shown in the figure does not constitute a limitation to the terminal, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0073] like Figure 3 As shown, the memory 1004 as a computer storage medium may include an operating system, a network communication module, and a detection program for fire protection pipes.

[0074] exist Figure 3 In the hardware structure of the detection device for fire protection pipes shown, the processor 1001 can call the detection program for fire protection pipes stored in the memory 1004 and perform the following operations: collecting pipeline inner wall reflection data by a mobile detection device, and extracting a corrosion feature vector and a wall thickness change feature vector of the pipeline inner wall reflection data; fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector; If it is detected that the comprehensive feature vector has an abnormal indicator, and the abnormal indicator is greater than a preset indicator threshold, the pipe section corresponding to the abnormal indicator is determined to be a defective pipe section; Calculating initial probability distribution data of the defect type of the defective pipe section based on the comprehensive feature vector, and determining the defect type of the defective pipe section in combination with a historical probability distribution model of the defect type; Acquiring coordinate data of the defective pipe section, and calculating a health index of the defective pipe section according to the coordinate data and the defect type; Obtain the historical health index variation pattern of the defective pipe section, and generate a status assessment report of the defective pipe section in combination with the health index.

[0075] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: The mobile detection device collects spectral reflection data and wall thickness acoustic wave reflection data of the inner wall of the pipeline; If the infrared band intensity in the spectral reflection data is greater than a preset infrared fluctuation threshold, and the ultrasonic echo time difference of the wall thickness acoustic wave reflection data is greater than a preset standard time difference, the inner wall of the pipeline is marked as a defective pipeline; The corrosion feature vector and the wall thickness variation feature vector of the defective pipe are extracted.

[0076] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: Obtaining an accuracy level value of a sensor in the motion detection device, and adjusting a weight value of the sensor according to the accuracy level value to generate a weight parameter; The corrosion feature vector and the wall thickness change feature vector are fused in combination with the weight parameter to generate a comprehensive feature vector corresponding to the pipeline risk area.

[0077] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: Generate a probability distribution model of corrosion defects, crack defects and blockage defects of the pipeline based on the comprehensive feature vector, and obtain initial probability distribution data; Obtaining the defect type historical probability distribution model, and calculating the posterior probability values of the corrosion defect, the crack defect, and the blockage defect in combination with the initial probability distribution data; The posterior probability value is compared with a preset threshold value of the defect type, and the defect type is determined according to the comparison result.

[0078] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: Acquire the coordinate data, and fuse the coordinate data with the defect type to generate a defect type data set including a defect type identifier and the coordinate data; The defect weight and defect severity of the defective pipe section are calculated according to the defect type data set to obtain the health index.

[0079] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: Obtaining the historical health index change pattern, and combining it with the coordinate data to generate a health index dataset including a health index time series and the coordinate data; generating a health index prediction value of the defective pipe section according to the health index data set and the health index, to obtain a prediction trend data set; The condition assessment report is generated based on the predicted trend data set, the coordinate data, the health index, and the defect type.

[0080] Optionally, the processor 1001 may call a detection program for fire protection pipes stored in the memory 1004, and further perform the following operations: Obtain the defect type and health index of the pipe section in the condition assessment report, determine the risk level of the pipe section, and obtain risk classification data; If the risk level of the defect point to be maintained in the risk classification data exceeds a preset risk threshold, a maintenance suggestion for the defect point to be maintained is generated in combination with the prediction trend data set to obtain maintenance planning information; The maintenance plan information is updated in the condition assessment report.

[0081] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention further provides a system for detecting fire protection pipelines, comprising: Data acquisition module, used to collect pipeline inner wall reflection data and pipe section coordinate data; A defect detection module, configured to determine whether a pipe section has defects based on the reflection data from the inner wall of the pipe; a defect type identification module, configured to determine the defect type of the pipe section based on the comprehensive feature vector and the historical defect probability model; A health index calculation and prediction module, configured to output the health index of the pipe section according to the defect type and the coordinate data, and generate a prediction trend data set based on historical health index change patterns; A status assessment module is used to generate the status assessment report based on the predicted trend data set, the coordinate data, the health index and the defect type.

[0082] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention further provides a device, comprising a memory, a processor, and a detection program for fire protection pipes stored in the memory and executable on the processor. When the processor executes the detection program for fire protection pipes, the above-mentioned detection method for fire protection pipes is implemented.

[0083] In addition, to achieve the above-mentioned purpose, an embodiment of the present invention further provides a computer-readable storage medium, on which a detection program for fire protection pipes is stored. When the detection program for fire protection pipes is executed by a processor, the detection method for fire protection pipes as described above is implemented.

[0084] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0088] It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present application may be implemented by means of hardware comprising several distinct components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second and third etc. does not indicate any order. These words may be interpreted as names.

[0089] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0090] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present application fall within the scope of the claims and their equivalents, the present application is intended to include such modifications and variations.

Claims

1. A method for detecting fire protection pipes, characterized in that: The method comprises: collecting pipeline inner wall reflection data by a mobile detection device, and extracting a corrosion feature vector and a wall thickness change feature vector of the pipeline inner wall reflection data; fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector; If it is detected that the comprehensive feature vector has an abnormal indicator, and the abnormal indicator is greater than a preset indicator threshold, the pipe section corresponding to the abnormal indicator is determined to be a defective pipe section; Calculating initial probability distribution data of the defect type of the defective pipe section based on the comprehensive feature vector, and determining the defect type of the defective pipe section in combination with a historical probability distribution model of the defect type; Acquiring coordinate data of the defective pipe section, and calculating a health index of the defective pipe section according to the coordinate data and the defect type; Obtain the historical health index variation pattern of the defective pipe section, and generate a status assessment report of the defective pipe section in combination with the health index.

2. The detection method for fire protection pipes according to claim 1, characterized in that: The steps of collecting pipeline inner wall reflection data by a mobile detection device and extracting the corrosion feature vector and the wall thickness change feature vector of the pipeline inner wall reflection data include: The mobile detection device collects spectral reflection data and wall thickness acoustic wave reflection data of the inner wall of the pipeline; If the infrared band intensity in the spectral reflection data is greater than a preset infrared fluctuation threshold, and the ultrasonic echo time difference of the wall thickness acoustic wave reflection data is greater than a preset standard time difference, the inner wall of the pipeline is marked as a defective pipeline; The corrosion feature vector and the wall thickness variation feature vector of the defective pipe are extracted.

3. The detection method for fire protection pipes according to claim 1, characterized in that: The step of fusing the corrosion feature vector and the wall thickness change feature vector to generate a comprehensive feature vector includes: Obtaining an accuracy level value of a sensor in the motion detection device, and adjusting a weight value of the sensor according to the accuracy level value to generate a weight parameter; The corrosion feature vector and the wall thickness change feature vector are fused in combination with the weight parameter to generate a comprehensive feature vector corresponding to the pipeline risk area.

4. The detection method for fire protection pipes according to claim 1, characterized in that: The step of calculating the initial probability distribution data of the defect type of the defective pipe section based on the comprehensive feature vector and determining the defect type of the defective pipe section in combination with the defect type historical probability distribution model includes: Generate a probability distribution model of corrosion defects, crack defects and blockage defects of the pipeline based on the comprehensive feature vector, and obtain initial probability distribution data; Obtaining the defect type historical probability distribution model, and calculating the posterior probability values of the corrosion defect, the crack defect, and the blockage defect in combination with the initial probability distribution data; The posterior probability value is compared with a preset threshold value of the defect type, and the defect type is determined according to the comparison result.

5. The detection method for fire protection pipes according to claim 1, characterized in that: The step of obtaining the coordinate data of the defective pipe section and calculating the health index of the defective pipe section according to the coordinate data and the defect type includes: Acquire the coordinate data, and fuse the coordinate data with the defect type to generate a defect type data set including a defect type identifier and the coordinate data; The defect weight and defect severity of the defective pipe section are calculated according to the defect type data set to obtain the health index.

6. The detection method for fire protection pipes according to claim 1, characterized in that: The step of obtaining a historical health index variation pattern of the defective pipe section and generating a status assessment report of the defective pipe section in combination with the health index includes: Obtaining the historical health index change pattern, and combining it with the coordinate data to generate a health index dataset including a health index time series and the coordinate data; generating a health index prediction value of the defective pipe section according to the health index data set and the health index, to obtain a prediction trend data set; The condition assessment report is generated based on the predicted trend data set, the coordinate data, the health index, and the defect type.

7. The detection method for fire protection pipes according to claim 1, characterized in that: After the step of obtaining the historical health index variation pattern of the defective pipe section and generating a status assessment report of the defective pipe section in combination with the health index, the method includes: Obtain the defect type and health index of the pipe section in the condition assessment report, determine the risk level of the pipe section, and obtain risk classification data; If the risk level of the defect point to be maintained in the risk classification data exceeds a preset risk threshold, a maintenance suggestion for the defect point to be maintained is generated in combination with the prediction trend data set to obtain maintenance planning information; The maintenance plan information is updated in the condition assessment report.

8. A detection system for fire protection pipelines, characterized in that: The system comprises: Data acquisition module, used to collect pipeline inner wall reflection data and pipe section coordinate data; A defect detection module, configured to determine whether a pipe section has defects based on the reflection data from the inner wall of the pipe; a defect type identification module, configured to determine the defect type of the pipe section based on the comprehensive feature vector and the historical defect probability model; A health index calculation and prediction module, configured to output the health index of the pipe section according to the defect type and the coordinate data, and generate a prediction trend data set based on historical health index change patterns; A status assessment module is used to generate a status assessment report based on the predicted trend data set, the coordinate data, the health index and the defect type.

9. A terminal device, characterized in that: The invention comprises a memory, a processor and a detection program for fire protection pipes stored in the memory and executable on the processor. When the processor executes the detection program for fire protection pipes, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a detection program for fire protection pipes. When the detection program for fire protection pipes is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images