Hot steam conveying pipeline leakage risk health monitoring method and system

Abstract

The invention relates to the technical field of pipeline leakage monitoring, and particularly discloses a hot steam conveying pipeline leakage risk health monitoring method and system, and the method comprises the following steps: taking a preset position on a hot steam conveying pipeline as a collection point, and installing pipeline leakage monitoring equipment at the collection point, the pipeline leakage monitoring equipment is used for judging the health state of the hot steam conveying pipeline according to the vibration signal of the hot steam conveying pipeline, and the health state comprises leakage and health; the residual electric quantity of the pipeline leakage monitoring equipment is acquired, a target time point T is determined based on the change condition of the residual electric quantity, the target time point T represents a time point when the residual electric quantity of the pipeline leakage monitoring equipment reaches a preset value C0, and charging of the pipeline leakage monitoring equipment is controlled based on the target time point. The universal degree of pipeline monitoring is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline leakage monitoring, and particularly relates to a method and system for health monitoring of leakage risks in a hot steam transmission pipeline. Background Art

[0002] A hot steam transmission pipeline refers to a pipeline system used to transport high-temperature steam. Such a system usually consists of components such as high-temperature and pressure-resistant pipelines, insulation layers, valves, brackets, etc., with the aim of transporting steam from a boiler or other steam-generating equipment to a process flow, a heat exchange station, or other places of use without reducing the steam temperature and pressure.

[0003] In the field of health monitoring of hot steam pipelines, how to ensure the normal operation of hot steam pipelines has always been an urgent problem to be solved in the pipeline industry. In the prior art, optical fibers are mainly laid along the pipeline, and the jitter of light is detected during the transmission of light in the optical fiber, and then complex algorithms are used to calculate the abnormal conditions of the pipeline. However, the cost of laying optical fibers is expensive, the laying difficulty is high, the equipment cost is expensive, and once damaged, the maintenance cost is also expensive, making it very difficult to popularize and use. Therefore, there are problems of high cost and difficulty in popularization in the prior art. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for health monitoring of leakage risks in a hot steam transmission pipeline, and solve the following technical problems:

[0005] In the prior art, optical fibers are mainly laid along the pipeline, and the jitter of light is detected during the transmission of light in the optical fiber, and then complex algorithms are used to calculate the abnormal conditions of the pipeline. However, the cost of laying optical fibers is expensive, the laying difficulty is high, the equipment cost is expensive, and once damaged, the maintenance cost is also expensive, making it very difficult to popularize and use.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for health monitoring of leakage risks in a hot steam transmission pipeline includes the following steps:

[0008] Taking a preset position on the hot steam transmission pipeline as a collection point, and installing a pipeline leakage monitoring device at the collection point. The pipeline leakage monitoring device is used to judge the health state of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline, and the health state includes leakage and health.

[0009] Obtain the remaining power of the pipeline leakage monitoring device, determine the target time point T based on the change of the remaining power, where the target time point T represents the time point when the remaining power of the pipeline leakage monitoring device reaches the preset value C0, and control the charging of the pipeline leakage monitoring device based on the target time point.

[0010] As a further solution of the present invention: Judging the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline includes:

[0011] When the health status is healthy, collect the vibration signal of the hot steam transmission pipeline, convert the vibration signal into an electrical signal, convert the amplified electrical signal into a digital signal, perform Fourier transform on the digital signal to obtain the spectrum of the vibration signal, extract spectrum features based on the spectrum, the spectrum features include spectrum width, center frequency and amplitude distribution, take the spectrum features as normal features, use the normal state as the label of the normal features, and add the normal features with the normal state as the label as a normal sample, and obtain n normal samples, where n is a preset quantity;

[0012] When the health status is leakage, collect the vibration signal of the hot steam transmission pipeline, and obtain the corresponding spectrum features, denoted as abnormal features, use the abnormal state as the label of the abnormal features, and add the abnormal features with the abnormal state as the label as an abnormal sample, and obtain n abnormal samples;

[0013] Randomly divide the n normal samples and the abnormal samples into a training set and a prediction set according to a preset ratio, establish a leakage warning model based on deep learning, and train and verify the leakage warning model based on the training set and the prediction set;

[0014] Real-time collect the vibration signal of the hot steam transmission pipeline, denoted as the real-time signal, obtain the spectrum features corresponding to the real-time signal, denoted as the real-time features, input the real-time features into the leakage warning model after training and verification, and output the label of the real-time features, denoted as the real-time label;

[0015] When the real-time label is the normal state, determine that the health status of the hot steam transmission pipeline is healthy; when the real-time label is the abnormal state, determine that the health status of the hot steam transmission pipeline is leaking.

[0016] As a further solution of the present invention: Judging the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline further includes:

[0017] When the health state of the hot steam transmission pipeline is leakage, record the time point t when the pipeline leakage monitoring device receives the real-time signal, sort the time points in the order of the time axis to obtain the first sorting, and use the acquisition points corresponding to the first four time points in the first sorting as the selection points;

[0018] Based on the time difference of arrival positioning and the coordinates of the selection points in the preset pipeline position coordinate system, determine the coordinates (distance from the detection point - longitude and latitude) and leakage time T0 of the leakage point, and send the coordinates (distance from the detection point - longitude and latitude) and leakage time T0 of the leakage point to the preset management personnel.

[0019] As a further solution of the present invention: determining the target time point based on the change of the remaining power includes:

[0020] Periodically obtain the remaining power, generate coordinate points (ti, Ci), where ti represents the time point when the remaining power is obtained for the i-th time, Ci represents the remaining power obtained for the i-th time, fit the coordinate points to obtain a fitting curve f(t), and t represents time;

[0021] Substitute the preset value C0 into the fitting curve f(t) to obtain the target time point T.

[0022] As a further solution of the present invention: controlling the charging of the pipeline leakage monitoring device based on the target time point includes:

[0023] Set a monitoring period [T - T1, T + T1], where T1 represents a preset time length, mark the dates within the monitoring period as target dates, and divide the target dates into m time periods with the same length, where m is the preset number of time periods;

[0024] When the current time reaches T - T1, obtain the average temperature within the time period, mark the time periods with the corresponding average temperature greater than the preset temperature threshold as target time periods, and when any two target time periods are adjacent, merge them into a new target time period;

[0025] Obtain the average temperature wd within the target time period, calculate the rechargeable amount W = η * wd * L of the target time period, where η represents a preset empirical coefficient for correcting the influence of temperature on the power generation power of the solar panel, and L represents the length of the target time period;

[0026] Sort the target time periods in descending order according to the rechargeable amount to obtain the second sorting, and control the charging of the pipeline leakage monitoring device by the solar panel based on the second sorting.

[0027] As a further solution of the present invention: The charging of the pipeline leakage monitoring device by the solar panel based on the second sorting includes:

[0028] Obtain the rechargeable amount W1 at the first place in the second sorting, obtain the time point TS1 corresponding to the start of the target time period MB1 corresponding to the rechargeable amount W1, substitute the time point TS1 into the fitting curve f(t), and obtain the remaining power Q1 of the pipeline leakage monitoring device at the time point TS1;

[0029] Obtain the maximum power Qtot of the pipeline leakage monitoring device. If Qtot - Q1 ≤ W1, then start charging the pipeline leakage monitoring device through the solar panel from the time point TS1 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot;

[0030] If Qtot - Q1 > W1, then obtain the rechargeable amount W2 at the second place in the second sorting. If Qtot - Q1 ≤ W1 + W2, then charge the pipeline leakage monitoring device through the solar panel within the target time period MB1 and charge the pipeline leakage monitoring device through the solar panel at the time point TS2 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot. The time point TS2 represents the start of the target time period MB2 corresponding to the rechargeable amount W2.

[0031] As a further solution of the present invention: When the time point TS2 is before the time point TS1, remove the rechargeable amount W2 from the second sorting to obtain a new second sorting P1, obtain the rechargeable amount at the second place in the second sorting P1, and perform subsequent steps.

[0032] A health monitoring system for the leakage risk of a hot steam transmission pipeline, which is applied to the health monitoring method for the leakage risk of a hot steam transmission pipeline as described above, includes a monitoring module and an optimization module. Specifically:

[0033] Monitoring module: Use a preset position on the hot steam transmission pipeline as a collection point, and install a pipeline leakage monitoring device at the collection point. The pipeline leakage monitoring device is used to judge the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline. The health status includes leakage and health;

[0034] Optimization module: Obtain the remaining power of the pipeline leakage monitoring device, determine the target time point based on the change of the remaining power. The target time point represents the time point when the remaining power of the pipeline leakage monitoring device reaches a preset value, and control the charging of the pipeline leakage monitoring device based on the target time point.

[0035] Advantages of the present invention: Compared with the prior art:

[0036] 1) By installing pipeline leakage monitoring devices at multiple preset collection points on the hot steam pipeline and determining the health status of the pipeline based on the vibration signals collected by them, it avoids the high cost and complex construction process of laying optical fibers over a large area along the pipeline in the traditional technology. Compared with the high cost and laying difficulty of the optical fiber system, the monitoring devices in the present invention only need to be installed at specific key positions, which not only reduces the material cost of large-scale layout, but also makes the later maintenance more convenient. Once a monitoring device fails, only the device at the corresponding collection point needs to be replaced or repaired, without large-scale excavation or replacement of optical fibers. At the same time, since there is no need to rely on the high-precision and high-cost laying structure of optical fibers, the harsh requirements for the environment and construction conditions during installation are greatly reduced, making it suitable for large-scale popularization and application;

[0037] 2) By collecting vibration signals and combining with a deep learning model to intelligently identify the real-time health status of the pipeline, it can discover leakage hazards earlier and more accurately. Compared with the traditional detection method mainly based on light intensity jitter, which requires complex algorithm processing and is easily affected by the external environment, the present invention uses time-frequency domain feature extraction and multi-class annotation of sample data to provide richer and more stable feature information for the deep learning model, significantly improving the recognition accuracy of pipeline leakage. And, by performing time difference of arrival positioning on the four collection points that first receive abnormal signals after detecting leakage, the leakage point coordinates and leakage time can be quickly determined in a three-dimensional coordinate system, facilitating the management personnel to receive the accurate position in the first time and make corresponding disposals, reducing the risk of accident expansion;

[0038] 3) Regarding the power supply problem of the monitoring device, a prediction and control strategy based on the change of remaining power is proposed. By periodically collecting the remaining power of the monitoring device and fitting it, the time point when the pipeline leakage monitoring device may drop below the preset power threshold can be obtained, so as to reasonably plan the charging period. On the one hand, it can charge the device in time when the power is about to run out, avoiding the monitoring gap caused by the device being offline for a long time; on the other hand, it can also significantly reduce the situation of blindly and frequently charging the device, scientifically plan the battery, and ensure that the device is charged only when it is really needed, thus effectively reducing the charging times;

[0039] 4) In realizing the reasonable arrangement of the charging period for the pipeline leakage monitoring device, the present invention further comprehensively utilizes the environmental temperature and the power generation law of the solar panel. By sorting the temperature distribution and rechargeable amount during the monitoring period and preferentially replenishing electrical energy to the device during the period with the highest charging efficiency. This refined charging management strategy can not only ensure that the pipeline leakage monitoring device has sufficient power for stable monitoring at critical moments, but also maximize the service life of the battery and reduce the cost and frequency of battery replacement in the later stage. Description of the Drawings

[0040] The present invention will be further described below in conjunction with the drawings.

[0041] Figure 1 It is a schematic flow chart of a method for health monitoring of leakage risks in a hot steam transmission pipeline according to the present invention. Detailed Embodiments

[0042] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] Please refer to Figure 1 As shown, the present invention is a method for health monitoring of leakage risks in a hot steam transmission pipeline, including the following steps:

[0044] Taking a preset position on the hot steam transmission pipeline as a collection point, a pipeline leakage monitoring device is installed at the collection point. The pipeline leakage monitoring device is used to judge the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline, and the health status includes leakage and health;

[0045] It is worth noting that in the actual layout process of the hot steam transmission pipeline, several preset positions can be first determined as collection points according to the pipeline operation environment, pipeline support positions, and parts where stress concentration may occur. Then, a pipeline leakage monitoring device composed of a sensor module and a data processing unit is fixedly installed at these collection points. The sensor module can use a piezoelectric vibration sensor. By adding stable supports around the collection points or using welding seats, etc., the sensor is closely attached to the pipe wall and kept in reliable contact to minimize the influence of additional vibration noise; the transmission of digital signals is realized through short-distance electrical connection cables. At the same time, a front-end amplification and filtering circuit can be integrated, and the vibration signal obtained after amplification is digitized in real time and the vibration spectrum characteristics are extracted. Then, the health status of the currently collected signal is judged according to a pre-trained model;

[0046] In a preferred embodiment of the present invention, judging the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline includes:

[0047] When the health status is healthy, collect the vibration signal of the hot steam transmission pipeline, convert the vibration signal into an electrical signal, convert the amplified electrical signal into a digital signal, perform Fourier transform on the digital signal to obtain the spectrum of the vibration signal, extract spectral features based on the spectrum, the spectral features include spectral width, center frequency and amplitude distribution, use the spectral features as normal features, use the normal state as the label of the normal features, and add the normal features with the normal state as the label as a normal sample, and obtain n normal samples, where n is a preset quantity;

[0048] When the health status is leakage, collect the vibration signal of the hot steam transmission pipeline and obtain the corresponding spectral features, denoted as abnormal features, use the abnormal state as the label of the abnormal features, and add the abnormal features with the abnormal state as the label as an abnormal sample, and obtain n abnormal samples;

[0049] Randomly divide the n normal samples and the abnormal samples into a training set and a prediction set according to a preset ratio, establish a leakage warning model based on deep learning, and train and validate the leakage warning model based on the training set and the prediction set;

[0050] Collect the vibration signal of the hot steam transmission pipeline in real time, denoted as the real-time signal, obtain the spectral features corresponding to the real-time signal, denoted as the real-time features, input the real-time features into the leakage warning model after training and validation, and output the label of the real-time features, denoted as the real-time label;

[0051] When the real-time label is the normal state, determine that the health status of the hot steam transmission pipeline is healthy; when the real-time label is the abnormal state, determine that the health status of the hot steam transmission pipeline is leaking.

[0052] It should be noted that when it is confirmed that the hot steam transmission pipeline is in a normal working condition, several acquisition points are selected to record their vibration signals. The vibration signals are converted into electrical signals by sensors and amplified through a front-end amplifier circuit. Subsequently, the amplified electrical signals are subjected to analog-to-digital conversion in a data processing unit, and the corresponding spectral features are extracted from the obtained digital signals through Fourier transform. Data such as the spectral width, center frequency, and amplitude distribution in these spectral features are used as normal features and the healthy state is used as a label to generate training data containing n normal samples; when the pipeline leaks or a small leakage point is artificially created under test conditions, the vibration signals recorded at the same position acquisition points are processed using the same amplification and Fourier transform methods to obtain corresponding abnormal spectral features. This feature is marked with a label of abnormal state and n abnormal samples are also collected; subsequently, all samples are randomly divided into a training set and a prediction set according to a certain ratio, and a leakage warning model is established using deep learning algorithms such as convolutional neural networks or long short-term memory networks. The model is repeatedly trained on the training set and verified on the prediction set to ensure that the model has good generalization ability in distinguishing healthy and leakage features; during real-time monitoring, the current vibration signal of the hot steam transmission pipeline is obtained through the same acquisition and signal processing process and converted into real-time features, and then the real-time features are input into the trained and verified model to obtain an output label. If the model determines that the output result belongs to the healthy state, it indicates that there is no obvious leakage risk in the current pipeline. If the output result belongs to the abnormal state, it means that there is a leakage hidden danger in the pipeline;

[0053] In a preferred case of this embodiment, determining the health state of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline further includes:

[0054] When the health state of the hot steam transmission pipeline is leakage, record the time point t when the pipeline leakage monitoring device receives the real-time signal, sort the time points in the order of the time axis to obtain a first sorting, and use the acquisition points corresponding to the first four time points in the first sorting as the selection points;

[0055] Based on the time difference of arrival positioning and the coordinates of the selection points in the preset pipeline position coordinate system, determine the coordinates (distance from the detection point - longitude and latitude) and leakage time T0 of the leakage point, and send the coordinates (distance from the detection point - longitude and latitude) and leakage time T0 of the leakage point to the preset management personnel.

[0056] It is understandable that time difference of arrival positioning is a method of determining the position of a signal source by using the tiny differences in the moments when multiple sensors receive the same signal. Assuming that the signal propagates in the medium at a known speed after being emitted from the leakage point, each acquisition point will receive the signal at different times due to different distances from the leakage point. By accurately recording the reception time of each acquisition point, the time difference required for the signal to reach different acquisition points can be calculated. Each time difference reflects the difference in the signal propagation distance, and these differences correspond to a series of hyperboloids (hyperbolas in two dimensions) with each acquisition point as the foci in mathematics. The leakage point must be located in the intersection area of these hyperboloids;

[0057] Another preferred embodiment of the present invention for determining the target time point based on the change of the remaining power includes:

[0058] Periodically obtain the remaining power to generate coordinate points (ti, Ci), where ti represents the time point of the i-th acquisition of the remaining power, and Ci represents the remaining power acquired at the i-th time. Fit the coordinate points to obtain a fitting curve f(t), where t represents time;

[0059] Substitute the preset value C0 into the fitting curve f(t) to obtain the target time point T;

[0060] A preferred case of this embodiment for controlling the charging of the pipeline leakage monitoring device based on the target time point includes:

[0061] Set a monitoring period [T - T1, T + T1], where T1 represents a preset time length. Mark the dates within the monitoring period as target dates, and divide the target dates into m time periods of the same length, where m is the preset number of time periods;

[0062] When the current time reaches T - T1, obtain the average temperature within the time period, and mark the time periods with the corresponding average temperature greater than the preset temperature threshold as target time periods. When any two of the target time periods are adjacent, merge them into a new target time period;

[0063] Obtain the average temperature wd within the target time period, and calculate the rechargeable amount W = η * wd * L of the target time period, where η represents a preset empirical coefficient for correcting the influence of temperature on the power generation power of the solar panel, and L represents the length of the target time period;

[0064] Sort the target time periods in descending order according to the rechargeable amount to obtain a second sorting, and control the charging of the pipeline leakage monitoring device by the solar panel based on the second sorting;

[0065] Another preferred case of this embodiment, the charging of the pipeline leakage monitoring device by the solar panel based on the second sorting includes:

[0066] Obtain the rechargeable amount W1 at the top of the second sorting, obtain the time point TS1 corresponding to the start of the target time period MB1 corresponding to the rechargeable amount W1, substitute the time point TS1 into the fitting curve f(t), and obtain the remaining power Q1 of the pipeline leakage monitoring device at the time point TS1;

[0067] Obtain the maximum power Qtot of the pipeline leakage monitoring device. If Qtot - Q1 ≤ W1, then start charging the pipeline leakage monitoring device through the solar panel from the time point TS1 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot;

[0068] If Qtot - Q1 > W1, then obtain the rechargeable amount W2 at the second place in the second sorting. If Qtot - Q1 ≤ W1 + W2, then charge the pipeline leakage monitoring device through the solar panel within the target time period MB1 and charge the pipeline leakage monitoring device through the solar panel at the time point TS2 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot. The time point TS2 represents the start of the target time period MB2 corresponding to the rechargeable amount W2;

[0069] It should be noted that first, the remaining power of the monitoring device is collected at fixed time intervals, and the time and power data of each collection are recorded. For example, the corresponding power values are obtained at zero o'clock for ten consecutive days. Using these data points, a curve fitting method is used to obtain a curve describing the power decay trend. This curve can predict that the power will drop to a preset threshold at a certain future moment. For example, when the power drops to 20%, this moment is determined as the target time point T. Then, centered on T, it extends a certain time to the left and right, and each day within this period is divided into several equal-length small segments. Then, the ambient temperature is obtained within each small segment, which can be obtained through weather forecasts, etc., because the power generation efficiency of the solar panel is affected by temperature. It is judged whether the temperature condition required for charging is reached according to the average temperature within each small segment, and the small segments with temperatures exceeding the threshold are marked, and they are merged when adjacent small segments both meet the conditions to form continuous target charging periods. Within each target charging period, the system calculates the charging energy that can be provided during this period, which is obtained by multiplying the average temperature, the length of the period, and an empirical correction coefficient within this period, so as to estimate the power generation potential of the solar panel within this period. Subsequently, all target charging periods are sorted from high to low according to the available charging energy, and the period with the strongest power generation ability is preferentially selected as the main charging window. For example, the system will select the first priority period. After the starting time TS1 is determined, TS1 is substituted into the previously fitted power decay curve to calculate the remaining power Q1 of the monitoring device at this time, and then compared with the maximum power Qtot of the device. If the required charging amount (Qtot - Q1) is less than or equal to the charging energy provided during this period, charging is directly started from TS1 until the device is fully charged. If the required charging amount is greater than the charging capacity of this period, the system will continue to select the second priority period in the sorting, and the target period corresponding to its starting time TS2 will continue to provide charging until the requirement of fully charging the device is met. This method can not only avoid frequent charging operations and extend the battery life by predicting the power decay trend of the device in advance, reasonably allocating charging periods, and accurately calculating the power generation potential within each period, but also ensure that the charging process is started in time before the device runs out of power, realizing the intelligence and efficiency of charging management;

[0070] In another preferred case of this embodiment, when the time point TS2 is before the time point TS1, the rechargeable amount W2 is removed from the second sorting to obtain a new second sorting P1, the rechargeable amount of the second place in the second sorting P1 is obtained, and the subsequent steps are executed.

[0071] A health monitoring system for leakage risk of a hot steam transmission pipeline, which is applied to the health monitoring method for leakage risk of a hot steam transmission pipeline as described above, includes a monitoring module and an optimization module. Specifically:

[0072] Monitoring module: Taking a preset position on the hot steam transmission pipeline as the acquisition point, a pipeline leakage monitoring device is installed at the acquisition point. The pipeline leakage monitoring device is used to judge the health status of the hot steam transmission pipeline according to the vibration signal of the hot steam transmission pipeline, and the health status includes leakage and health;

[0073] Optimization module: Obtaining the remaining power of the pipeline leakage monitoring device, determining a target time point based on the change of the remaining power, where the target time point represents the time point when the remaining power of the pipeline leakage monitoring device reaches a preset value, and controlling the pipeline leakage monitoring device to charge based on the target time point.

[0074] The above has described an embodiment of the present invention in detail, but the content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for health monitoring of leakage risk of hot steam transmission pipeline, characterized in that: The following steps are involved: A preset position on the hot steam delivery pipeline is used as a collection point, and a pipeline leakage monitoring device is installed at the collection point, wherein the pipeline leakage monitoring device is used to judge the health status of the hot steam delivery pipeline according to the vibration signal of the hot steam delivery pipeline, and the health status includes leakage and health; Obtain the remaining power of the pipeline leakage monitoring device, determine a target time point T based on the change of the remaining power, the target time point T represents the time point when the remaining power of the pipeline leakage monitoring device reaches a preset value C0, and control the charging of the pipeline leakage monitoring device based on the target time point.

2. A hot steam transmission pipeline leakage risk health monitoring method according to claim 1, characterized in that: Judging the health status of the hot steam delivery pipeline according to the vibration signal of the hot steam delivery pipeline includes: When the health state is healthy, a vibration signal of the hot steam delivery pipeline is collected, the vibration signal is converted into an electrical signal, the amplified electrical signal is converted into a digital signal, the digital signal is Fourier transformed to obtain a spectrum of the vibration signal, spectrum features are extracted based on the spectrum, the spectrum features include spectrum width, center of gravity frequency and amplitude distribution, the spectrum features are used as normal features, the normal state is used as a label of the normal features, the normal features after adding the normal state as the label are used as a normal sample, and n normal samples are obtained, where n is a preset number; When the healthy state is leakage, a vibration signal of the hot steam delivery pipeline is collected, and a corresponding frequency spectrum feature is obtained, recorded as an abnormal feature, the abnormal state is used as a label of the abnormal feature, and the abnormal feature after adding the abnormal state as a label is used as an abnormal sample, and n abnormal samples are obtained; Randomly divide the n normal samples and the abnormal samples into a training set and a prediction set according to a preset ratio, establish a leakage warning model based on deep learning, and train and verify the leakage warning model based on the training set and the prediction set; Collecting the vibration signal of the hot steam transmission pipeline in real time, recording it as the real-time signal, obtaining the frequency spectrum feature corresponding to the real-time signal, recording it as the real-time feature, inputting the real-time feature into the leakage warning model after training and verification, and outputting the label of the real-time feature, recording it as the real-time label; When the real-time tag is in the normal state, the health state of the hot steam delivery pipeline is determined to be healthy; when the real-time tag is in the abnormal state, the health state of the hot steam delivery pipeline is determined to be leaking.

3. A hot steam transmission pipeline leakage risk health monitoring method according to claim 2, characterized in that: Determining the health status of the hot steam delivery pipeline according to the vibration signal of the hot steam delivery pipeline also includes: When the health status of the hot steam transmission pipeline is leakage, the time point t at which the pipeline leakage monitoring device receives the real-time signal is recorded, the time points are sorted according to the time axis sequence to obtain a first sorting, and the collection points corresponding to the first four time points in the first sorting are used as selection points; Based on the arrival time difference positioning and the coordinates of the selected point in the preset pipeline position coordinate system, the coordinates of the leakage point (the distance from the detection point - longitude and latitude) and the leakage time T0 are determined, and the coordinates of the leakage point (the distance from the detection point - longitude and latitude) and the leakage time T0 are sent to the preset management personnel.

4. A hot steam transmission pipeline leakage risk health monitoring method according to claim 1, characterized in that: Determining the target time point based on the change in the remaining power includes: The remaining power is periodically obtained to generate coordinate points (ti, Ci), where ti represents the time point at which the remaining power is obtained for the i-th time, and Ci represents the remaining power obtained for the i-th time, and the coordinate points are fitted to obtain a fitting curve f(t), where t represents time; Substitute the preset value C0 into the fitting curve f(t) to obtain the target time point T.

5. A hot steam transmission pipeline leakage risk health monitoring method according to claim 4, characterized in that: Controlling the charging of the pipeline leakage monitoring device based on the target time point includes: Set the monitoring period [T-T1, T+T1], where T1 represents the preset time length, mark the date within the monitoring period as the target date, and divide the target date into m time periods of equal length, where m is the number of preset time periods; When the current time reaches T-T1, the average temperature in the time period is obtained, and the time period in which the corresponding average temperature is greater than the preset temperature threshold is marked as the target time period. When any two target time periods are adjacent, they are merged into a new target time period; Obtain the average temperature wd in the target time period, and calculate the chargeable capacity W in the target time period = η*wd*L, where η represents a preset empirical coefficient used to correct the influence of temperature on the power generation of the solar panel, and L represents the length of the target time period; The target time periods are sorted in descending order according to the amount of chargeable energy to obtain a second sorting order, and the solar panel is controlled to charge the pipeline leakage monitoring device based on the second sorting order.

6. A hot steam transmission pipeline leakage risk health monitoring method according to claim 5, characterized in that: Controlling the solar panel to charge the pipeline leakage monitoring device based on the second sequencing includes: Obtain the chargeable capacity W1 at the first position in the second sorting, obtain the time point TS1 corresponding to the starting point of the target time period MB1 corresponding to the chargeable capacity W1, substitute the time point TS1 into the fitting curve f(t), and obtain the remaining capacity Q1 of the pipeline leakage monitoring device at the time point TS1; Obtaining the maximum power Qtot of the pipeline leakage monitoring device, if Qtot-Q1≤W1, charging the pipeline leakage monitoring device through the solar panel starting from the time point TS1 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot; If Qtot-Q1>W1, then obtain the second chargeable amount W2 in the second sorting; if Qtot-Q1≤W1+W2, then charge the pipeline leakage monitoring device through the solar panel within the target time period MB1, and charge the pipeline leakage monitoring device through the solar panel at time point TS2 until the remaining power of the pipeline leakage monitoring device is equal to the maximum power Qtot; the time point TS2 indicates the starting point of the target time period MB2 corresponding to the chargeable amount W2.

7. A hot steam transmission pipeline leakage risk health monitoring method according to claim 6, characterized in that: When the time point TS2 is before the time point TS1, the chargeable capacity W2 is removed from the second ranking to obtain a new second ranking P1, the second chargeable capacity in the second ranking P1 is obtained, and subsequent steps are performed.

8. A hot steam transmission pipeline leakage risk health monitoring system, applied to a hot steam transmission pipeline leakage risk health monitoring method according to any one of claims 1 to 7, characterized in that: It includes monitoring module and optimization module, specifically: Monitoring module: a preset position on the hot steam delivery pipeline is used as a collection point, and a pipeline leakage monitoring device is installed at the collection point. The pipeline leakage monitoring device is used to judge the health status of the hot steam delivery pipeline according to the vibration signal of the hot steam delivery pipeline, and the health status includes leakage and health; Optimization module: obtain the remaining power of the pipeline leakage monitoring device, determine the target time point based on the change of the remaining power, the target time point indicates the time point when the remaining power of the pipeline leakage monitoring device reaches a preset value, and control the pipeline leakage monitoring device to charge based on the target time point.

Images