Drainage pipeline inner wall anti-fouling alarm device, method, computer equipment and medium

Abstract

The application provides a sewer inner wall anti-fouling alarm device, method, computer equipment and medium, receives echo signals reflected by the inner surface and the inner wall attachments of the sewer, extracts state indication features from the echo signals, compares the state indication features with a reference model containing a normal working condition fluctuation range established based on long-term historical data of a target monitoring point, obtains a feature deviation, and then maps the feature deviation to a judgment type and thickness estimation result of the current inner wall attachments, determines a reduction coefficient of the current inner wall attachments on the water passage capacity of the sewer based on the judgment type and the thickness estimation result and according to hydraulic design parameters of the sewer, and if the reduction coefficient exceeds a preset reduction threshold of the sewer, a maintenance warning signal is generated. The scheme of the application can stably extract essential features from complex interference to perform health warning on the sewer.

Description

Technical Field

[0001] This application relates to the field of pipeline alarm technology, and in particular to an anti-scaling alarm device, method, computer equipment, and medium for the inner wall of a drainage pipeline. Background Technology

[0002] As the "lifeblood" of urban infrastructure, drainage pipes are prone to scale buildup and siltation, which can lead to reduced water flow capacity, increased energy consumption, and even accidents such as blockages and overflows. Therefore, effective monitoring and early warning of the condition of the pipe walls are crucial for preventive maintenance and ensuring the healthy operation of the pipe network.

[0003] Traditional methods for detecting the internal condition of pipelines mainly rely on closed-circuit television (CCTV) inspection and pipeline robotic inspection. While these methods are intuitive, they are invasive and intermittent operations that require interrupting pipeline operation, resulting in high costs and an inability to achieve long-term, continuous online monitoring. Existing technologies have attempted to detect pipeline conditions using acoustic principles; for example, some schemes infer internal conditions by exciting sound waves outside the pipeline and analyzing their propagation characteristics. However, these existing technologies typically suffer from several drawbacks, such as: a disconnect between the detection principle and engineering practice; insufficient feature recognition capabilities leading to high false alarm rates; and static, rigid models lacking adaptability. Therefore, how to stably extract essential features from complex interference to provide health warnings for drainage pipelines has become a challenge for the industry. Summary of the Invention

[0004] Based on this, this application provides a drainage pipe inner wall anti-scaling alarm device, method, computer equipment, and medium for stably extracting essential features from complex interference to provide health early warning for drainage pipes.

[0005] In a first aspect, this application provides a method for preventing scale buildup on the inner wall of a drainage pipe, comprising the following steps: Receive echo signals reflected by deposits on the inner surface and inner wall of the drainage pipe; State indication features are extracted from the echo signal, including the relative change in the interface reflection coefficient and signal dispersion characteristic parameters; The status indication features are compared with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point to obtain the feature deviation amount. The feature deviation amount is then mapped to the judgment type and thickness estimation result of the current inner wall attachment. Based on the judgment type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, the reduction factor of the current inner wall attachments on the pipe's water flow capacity is determined. If the reduction factor exceeds the preset reduction threshold of the drainage pipe, a maintenance warning signal is generated.

[0006] In some embodiments, state indication features are extracted from the echo signal. These state indication features include the relative change in the interface reflection coefficient and signal dispersion characteristic parameters, specifically including: The echo signal is preprocessed to enhance signal quality and suppress noise; The relative change in the interface reflection coefficient is determined based on the preprocessed echo signal. The signal dispersion characteristic parameters are determined based on the same preprocessed echo signal.

[0007] In some embodiments, the status indication feature is compared with a benchmark model established based on long-term historical data of the target monitoring point, which includes the normal operating condition fluctuation range, to obtain the feature deviation, specifically including: Based on the current operating conditions, obtain the reference values ​​and fluctuation ranges of the corresponding normal state status indication characteristics from the benchmark model; The extracted state indication feature is compared with the reference value and fluctuation range of the state indication feature of the normal state to obtain the feature deviation.

[0008] In some embodiments, mapping the feature deviation to the determination type and thickness estimation result of the current inner wall attachment specifically includes: The feature deviation is input into a pre-trained hybrid recognition model; The feature deviation is processed by the hybrid recognition model to output the preliminary recognition parameters of the current inner wall attachments; The final judgment type and thickness estimation result are determined based on the preliminary identification parameters.

[0009] In some embodiments, determining the reduction factor of the current inner wall deposits on the water flow capacity of the pipe based on the determination type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, specifically includes: The equivalent roughness increment caused by the deposits inside the pipe wall is determined based on the judgment type. The amount of reduction in flow area caused by deposits inside the pipe is determined based on the thickness estimation results; The reduction factor of the current inner wall deposits on the water flow capacity of the pipeline is determined based on the hydraulic design parameters, the equivalent roughness increment, and the reduction in flow area.

[0010] In some embodiments, if the reduction factor exceeds a preset reduction threshold for the drainage pipe, generating a maintenance warning signal specifically includes: The reduction coefficient is compared with a preset reduction threshold. If the reduction coefficient exceeds the reduction threshold, the warning information generation process is triggered. Based on the judgment type, the thickness estimation result, and the degree to which the threshold is exceeded, a maintenance warning signal is generated and output.

[0011] In some embodiments, the method further includes: if the reduction factor does not exceed a preset reduction threshold, it is considered a normal situation and no action is taken.

[0012] Secondly, this application provides a scale prevention alarm device for the inner wall of a drainage pipe, comprising: The receiving module is used to receive echo signals reflected by the deposits on the inner surface and inner wall of the drainage pipe; The processing module is used to extract state indication features from the echo signal, the state indication features including the relative change of the interface reflection coefficient and the signal dispersion characteristic parameter; The processing module is also used to compare the status indication feature with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point, to obtain the feature deviation amount, and then map the feature deviation amount to the judgment type and thickness estimation result of the current inner wall attachment. The processing module is also used to determine the reduction factor of the current inner wall attachment on the water flow capacity of the pipe based on the judgment type and thickness estimation result and according to the hydraulic design parameters of the drainage pipe. The execution module is used to generate a maintenance warning signal if the reduction coefficient exceeds the preset reduction threshold of the drainage pipe.

[0013] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method for preventing scale buildup on the inner wall of a drainage pipe.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for preventing scale buildup on the inner wall of a drainage pipe.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The anti-scaling alarm device, method, computer equipment, and medium for drainage pipes provided in this application first receive echo signals reflected by deposits on the inner surface and inner wall of the drainage pipe. State indication features are extracted from these echo signals, including the relative change in the interface reflection coefficient and signal dispersion characteristics. This step can obtain reflected echoes containing information about the interface between the inner wall and the deposits from a single installation point, and extract essential features directly related to the acoustic impedance and geometric / viscoelastic properties of the deposits. This provides a stable, reliable, and physically meaningful basis for subsequent accurate identification under complex operating conditions. Accurate input information lays the foundation for high-precision detection. Secondly, the state indication features are compared with a benchmark model built upon long-term historical data of the target monitoring point, encompassing the normal operating condition fluctuation range, to obtain the feature deviation. This feature deviation is then mapped to the judgment type and thickness estimation result of the current inner wall deposits. This step effectively filters out interference from normal operating condition fluctuations based on the dynamically adaptive benchmark model. Furthermore, by integrating physical mechanisms and data-driven hybrid recognition models, the acoustic feature deviation is robustly mapped to the specific type and thickness of the deposits, thereby significantly improving the accuracy of dirt identification. The system assesses the pipe's stability, robustness, and adaptability to individual pipe variations and long-term aging, reducing false positives and false negatives. Then, based on the assessment type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, it determines the reduction factor of the current inner wall deposits on the pipe's water flow capacity. This step converts the identified deposit physical parameters into equivalent roughness increments and flow area reductions, further integrating them into a quantified water flow capacity reduction factor—an engineering performance indicator. This transforms the abstract concept of "fouling presence" into an intuitive "performance degradation level," providing an objective, calculable, and unified quantitative basis for risk-based early warning. According to the data, this greatly enhances the engineering guidance value of the assessment results. Finally, if the reduction coefficient exceeds the preset reduction threshold of the drainage pipeline, a maintenance early warning signal is generated. This step compares the calculated performance reduction coefficient with the dynamically set scientific threshold and automatically generates a structured early warning signal containing specific dirt information, risk level, and action suggestions when the limit is exceeded. This directly and efficiently transforms the monitoring results into clear and operable maintenance instructions, improving the timeliness of operation and maintenance response. In summary, the solution of this application can stably extract essential features from complex interference to provide health early warning for drainage pipelines. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a method for preventing scale buildup on the inner wall of a drainage pipe, according to some embodiments of this application. Figure 2 This is a schematic diagram illustrating an application scenario of an anti-scaling alarm data processing system according to some embodiments of this application; Figure 3This is a schematic flowchart illustrating the determination of the reduction factor according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a scale prevention alarm device for the inner wall of a drainage pipe, according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a method for preventing scale buildup on the inner wall of a drainage pipe, according to some embodiments of this application. Detailed Implementation

[0017] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0018] refer to Figure 1 The figure is an exemplary flowchart of a method for preventing scale buildup on the inner wall of a drainage pipe according to some embodiments of this application. The method for preventing scale buildup on the inner wall of a drainage pipe mainly includes the following steps: In practice, the pre-installation of elastic wave sensing devices at the target monitoring points of the drainage pipeline can be achieved in the following way: First, determine the specific location of the target monitoring point. This location should be chosen in a part of the drainage pipeline system that facilitates long-term and stable contact with the outer surface of the pipe wall, preferably a pipe section inside a manhole, an exposed pipe section in a pipe gallery, or a reserved sensor installation interface, to ensure that the device is maintainable and to reduce the shielding and interference of environmental buried objects on the signal; Next, pre-treat the outer surface of the pipe wall at the target monitoring point, cleaning away dirt, rust, old coatings, and other debris, and using an angle grinder or sandpaper to polish a smooth and clean area. The area is slightly larger than the contact bottom surface of the elastic wave sensing device; then, the device is mechanically installed and acoustically coupled. Specifically, the elastic wave sensing device includes a waterproof and corrosion-resistant metal or engineering plastic shell, which encapsulates a piezoelectric ceramic transducer, a pre-conditioning circuit, and connecting cables. During installation, a layer of special acoustic coupling agent, such as silicone grease or epoxy polymer coupling material, is uniformly applied to the pre-treated pipe wall surface. The acoustic impedance value of this coupling agent is designed to be between the acoustic impedance of the piezoelectric ceramic transducer and the acoustic impedance of the pipe body material, such as concrete, cast iron, or PVC, to achieve efficient transmission of sound wave energy and maximize its effectiveness. To minimize sound wave reflection loss due to air gaps, the transmitting / receiving end face of the elastic wave sensing device is then tightly bonded to the pre-treated area of ​​the pipe wall using the coupling agent layer. It is then firmly fixed to the pipe using mechanical fasteners such as stainless steel clamps, strong magnetic bases, or pre-embedded welded bases, ensuring that the device will not shift or detach during pipe vibration and long-term use, thus guaranteeing the stability and repeatability of the sound wave propagation path. Finally, electrical connections and initial testing are completed. Specifically, the device's output cable is connected to a nearby waterproof junction box and ultimately to a data acquisition and processing unit located inside a manhole or on the ground. After power-on, the control unit sends a standard test pulse signal. By observing whether the received echo signal has the expected amplitude and waveform, the installation coupling quality and the normal working status of the device are verified. The signal in this initial state is used as the starting point for the baseline data recording of the monitoring point. The housing protection level of the elastic wave sensing device should not be lower than IP67 to adapt to the humid environment around the drainage pipe, and even the possibility of temporary water immersion. The selection of the mechanical fastening method needs to take into account the pipe material, shape (circular, box culvert) and on-site operating space. Other methods can also be used in other embodiments, which are not limited here.

[0019] In step 101, the echo signal reflected by the inner surface and inner wall of the drainage pipe is received.

[0020] In specific implementation, receiving the echo signal reflected by the attachments on the inner surface and inner wall of the drainage pipe can be achieved in the following way: The installed elastic wave sensing device is controlled, specifically, its internal piezoelectric ceramic transducer is driven to generate a broadband elastic wave excitation signal of the preset waveform according to a preset period and parameters. This broadband elastic wave excitation signal is effectively transmitted to the pipe wall through the acoustic coupling agent layer established during installation, and propagates within the pipe wall in longitudinal and transverse wave modes. When the elastic wave propagates to the inner surface of the pipe, due to the acoustic impedance difference between the pipe wall material and the internal medium (water, air, or a mixture of both), some energy is reflected, forming a primary reflected echo from the inner surface of the pipe. Simultaneously, if there are attachments on the inner wall of the pipe, such as hard scale or soft biofilm, additional acoustic impedance interfaces are formed between the attachment layer and the inner wall of the pipe, and between the attachment layer and the internal medium. The elastic wave will undergo further reflection and scattering at these interfaces, generating subtle signals containing information about the attachments. All this reflected and scattered elastic wave energy returns along the original path, penetrates the pipe wall again, and is detected by the elastic wave sensing device. The same piezoelectric ceramic transducer in the center senses and converts the signal into a weak electrical signal. This signal is then sent to a signal conditioning circuit, where it is initially amplified by a preamplifier, filtered by an anti-aliasing bandpass filter to remove low-frequency vibration noise and high-frequency electromagnetic interference outside the operating frequency band, and then synchronously sampled and digitized by a high-speed analog-to-digital converter under precise clock control. Finally, a discrete time-domain digital echo sequence is obtained. To separate the effective analysis signal, the starting time of the excitation signal is used as the precise zero point. Based on the known or pre-measured pipe wall thickness and pipe sound velocity, the theoretical propagation time of the elastic wave through the pipe wall is calculated, and a fixed-width analysis window is set centered on this theoretical time point. Finally, a signal segment falling within this analysis window is extracted from the complete time-domain digital echo sequence. This signal segment is defined as the echo signal used for subsequent processing, mainly containing the primary reflected wave from the inner surface of the pipe and the fine reflection and scattered wave information superimposed on it by the inner wall attachments. Other methods can be used in other embodiments, which are not limited here.

[0021] It should be noted that the waveform of the broadband elastic wave excitation signal in this application can be a single pulse, a frequency-modulated pulse, or a coded pulse. Its frequency band range needs to be selected according to the acoustic characteristics of the pipe and the target attachment to optimize the detection sensitivity. The width of the analysis window needs to be reasonably set according to the width of the transmitted pulse and the time resolution to ensure that the main wave of the first reflection can be completely captured and the interference of multiple reflection waves and other irrelevant noise in the pipe wall can be effectively eliminated. Other methods can also be used in other embodiments, which are not limited here.

[0022] In some embodiments, reference Figure 2As shown in the figure, this figure is a schematic diagram of the application scenario of the anti-scaling alarm data processing system shown in some embodiments of this application. The figure includes three main components: a data acquisition device, a server, and a data storage device. The data acquisition device is responsible for collecting echo signals reflected by the deposits on the inner surface and inner wall of the drainage pipe, and sending the collected echo signals to the server through a communication network. The anti-scaling alarm data processing system runs on the server, and the server stores the processing results in the data storage device and visualizes them.

[0023] In step 102, a state indication feature is extracted from the echo signal. The state indication feature includes the relative change of the interface reflection coefficient and the signal dispersion characteristic parameter.

[0024] In some embodiments, extracting state indication features from the echo signal, the state indication features including the relative change in the interface reflection coefficient and signal dispersion characteristic parameters, can be achieved by the following steps: The echo signal is preprocessed to enhance signal quality and suppress noise; The relative change in the interface reflection coefficient is determined based on the preprocessed echo signal. The signal dispersion characteristic parameters are determined based on the same preprocessed echo signal.

[0025] It should be noted that the relative change in the interface reflection coefficient in this application is an acoustic characteristic parameter used to characterize the change in the acoustic impedance state of the inner wall of the pipe. Its core function is to indicate the relative difference in acoustic impedance between the deposit and the pipe wall substrate, thereby providing a key basis for distinguishing the material type of the deposit, such as high-impedance hard scale and low-impedance soft biofilm. The dispersion characteristic parameter is an acoustic characteristic parameter used to quantify the degree of waveform distortion caused by the dispersion effect of the medium during the propagation of elastic waves. Its core function is to reflect the thickness and viscoelastic structure information of the deposit layer, providing another dimension of characteristics independent of the material type for estimating the thickness of the deposit.

[0026] In specific implementation, the echo signal is preprocessed to enhance signal quality and suppress noise. This can be achieved as follows: First, the DC component of the echo signal, i.e., the signal segment within the analysis window, is eliminated by calculating the arithmetic mean of the signal segment and subtracting it from each data point to eliminate baseline offset introduced during signal acquisition. Next, a digital bandpass filter is applied to the echo signal after DC component elimination. The passband frequency range of this digital bandpass filter is set according to the effective frequency band of the broadband elastic wave excitation signal emitted by the elastic wave sensing device, thereby filtering out low-frequency environmental vibration noise and high-frequency electronic device switching noise outside the passband. Noise is then filtered. Next, the filtered echo signal undergoes time-domain windowing, multiplying the beginning and end of the signal segment by a window function, such as a Hamming window, to reduce spectral leakage caused by signal truncation and improve the accuracy of subsequent frequency domain analysis. Finally, based on the signal quality assessment results, it can be determined whether to perform multi-cycle signal averaging, i.e., averaging multiple echo signal segments collected under the same operating conditions within consecutive monitoring cycles after time-domain alignment, thereby further suppressing random noise and improving the signal-to-noise ratio. The signal processed by the above steps is the pre-processed echo signal. Other methods can also be used in other embodiments, which are not limited here.

[0027] In specific implementation, determining the relative change in the interface reflection coefficient based on the preprocessed echo signal can be achieved in the following way: First, calculate the signal energy value of the preprocessed echo signal within the analysis time window. Specifically, this can be characterized by calculating the sum of the squares of the amplitudes of all data points in the signal segment or by calculating its root mean square value. Then, obtain the reference signal energy value that matches the current pipeline condition and is in a "clean" or "no significant deposits" state from the benchmark model established for the current target monitoring point. The current pipeline condition can be obtained through auxiliary sensors or indirectly estimated through echo signal characteristics, such as pipeline fill degree and temperature. The benchmark model stores different... The statistical values ​​of the echo signal energy under normal conditions within the operating range are calculated, such as the mean and standard deviation. Next, the ratio of the current signal energy value to the reference signal energy value is calculated. Finally, by taking the logarithm of this ratio to base 10 and multiplying the result by 10, the relative change of the interface reflection coefficient in decibels is calculated. The calculated relative change of the interface reflection coefficient is a scalar value. A positive value indicates enhanced reflection, which usually corresponds to deposits with acoustic impedance higher than the pipe wall, such as hard scale. A negative value indicates weakened reflection, which usually corresponds to deposits with lower acoustic impedance, such as soft biofilm. Other methods can also be used in other embodiments, which are not limited here.

[0028] It should be noted that the benchmark model is a statistical model trained based on long-term historical data of the target monitoring point. Its core is to describe the joint probability distribution of the state indication characteristics and pipeline operating conditions under conditions of no attachments or stable attachments. The state indication characteristics are the relative change in the interface reflection coefficient and signal dispersion parameters, while the pipeline operating conditions include fill degree and temperature. As a preferred embodiment, the benchmark model can be constructed in the following manner: First, after the elastic wave sensing device is installed at the target monitoring point and its normal operation is confirmed, a learning phase for the benchmark data is initiated. Within the section, under various operating conditions of the pipeline, such as different pipe filling degrees, different flow velocities, and different temperature conditions caused by diurnal temperature differences, the elastic wave sensing device is actively controlled to emit broadband elastic wave excitation signals and collect echo signals at fixed time intervals, such as once per hour, or triggered, such as when stable operating conditions are detected. Raw data is continuously collected for a sufficiently long time, such as one month or a complete seasonal cycle. For each acquisition, the steps of "receiving echo signals reflected by the inner surface and inner wall attachments of the drainage pipe" and "extracting status indication features from the echo signals" are executed. This process yields a large number of status indicator feature samples under normal pipeline conditions, i.e., without significant new deposits on the inner wall. Each sample contains a relative change in the interface reflection coefficient and a signal dispersion characteristic parameter. Simultaneously, operating parameters at each acquisition time are recorded or estimated, such as pipeline fill degree and ambient temperature. The pipeline fill degree can be obtained through acoustic time-of-flight or an external water level gauge. Then, the collected status indicator feature samples are correlated with the corresponding operating parameters. One specific implementation method is to divide the entire data space into several operating condition intervals based on the operating parameters. For example, the fill degree is divided into intervals of 1... The system is divided into 0% intervals and temperature intervals of 5 degrees Celsius. For all state indication feature samples within each operating condition interval, the relative change of the interface reflection coefficient and the statistical characteristics of the signal dispersion parameter are calculated, including but not limited to: calculating the average value of each feature in all samples within the interval as the feature reference value under the operating condition; calculating the standard deviation of each feature and the covariance between two features, which together constitute a covariance matrix describing the normal fluctuation range of the feature under the operating condition; finally, the mapping relationship between all operating condition intervals and their corresponding feature reference values ​​and fluctuation ranges is stored, which constitutes the benchmark model.In subsequent online monitoring, the baseline model can quickly retrieve corresponding characteristic reference values ​​and fluctuation ranges based on real-time estimated operating parameters for comparison. Furthermore, the baseline model supports online updates; that is, when a normal state is confirmed, such as after maintenance and cleaning, new monitoring data can be incorporated into the model according to a certain strategy, such as the moving average method, to progressively update the statistical parameters of the corresponding operating condition range. This allows the model to track the slow changes in the pipeline's acoustic characteristics and maintain long-term effectiveness. Other methods can also be used in other embodiments, and are not limited here.

[0029] In specific implementation, determining the signal dispersion characteristic parameters based on the same preprocessed echo signal can be achieved in the following way: First, perform time-frequency analysis on the preprocessed echo signal to obtain its time-frequency joint distribution. A specific implementation method is to use continuous wavelet transform, select a suitable mother wavelet, such as the Morlet wavelet, and perform convolution operations on the echo signal at multiple scales to obtain the wavelet coefficient matrix of the echo signal. This wavelet coefficient matrix characterizes the energy density of the signal at different times and different frequency components. Then, based on the wavelet coefficient matrix, extract the instantaneous frequency curve of the signal. Specifically, at each time point... The algorithm first finds the scale with the largest wavelet coefficient amplitude, and then calculates the dominant instantaneous frequency value corresponding to that time point based on the correspondence between the scale and the center frequency. Finally, it calculates the statistical variance of the entire extracted instantaneous frequency curve within the entire analysis window. This statistical variance is the signal dispersion characteristic parameter. The value of this signal dispersion characteristic parameter reflects the degree of time broadening of different frequency components during the propagation of the elastic wave. When there is an attachment with a certain thickness and viscoelasticity on the inner wall, the wave dispersion effect will be aggravated, leading to an increase in the fluctuation of the instantaneous frequency, thereby increasing the value of the variance parameter. Other methods can also be used in other embodiments, which are not limited here.

[0030] It should be noted that this step can obtain reflected echoes containing information about the interface between the inner wall and the attached material from a single installation point, and extract essential features that are directly related to the acoustic impedance characteristics and geometric / viscoelastic properties of the attached material. This provides stable, reliable, and physically meaningful input information for subsequent accurate identification under complex working conditions, laying the foundation for high-precision detection.

[0031] In step 103, the status indication feature is compared with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point to obtain the feature deviation amount, and then the feature deviation amount is mapped to the judgment type and thickness estimation result of the current inner wall attachment.

[0032] In some embodiments, comparing the status indication features with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point to obtain the feature deviation can be achieved by the following steps: Based on the current operating conditions, obtain the reference values ​​and fluctuation ranges of the corresponding normal state status indication characteristics from the benchmark model; The extracted state indication feature is compared with the reference value and fluctuation range of the state indication feature of the normal state to obtain the feature deviation.

[0033] It should be noted that the benchmark model in this application is a statistical reference model established for the target monitoring point to describe the normal fluctuation range of acoustic characteristics of its pipeline system under the condition of no significant deposits. Its core function is to serve as a dynamic baseline for comparison, used to filter out acoustic characteristic fluctuations caused by normal changes in pipeline operating conditions, thereby highlighting abnormal changes caused by deposits. The feature deviation is a comprehensive metric that quantitatively characterizes the deviation of the currently extracted acoustic feature state from the normal range defined by the benchmark model. Its core function is to serve as a bridge from the acoustic monitoring domain to the physical state identification domain, condensing the original, disturbed feature data into an indicator that better reflects the degree of abnormality.

[0034] In specific implementation, the reference values ​​and fluctuation ranges of the corresponding normal state indication features obtained from the benchmark model according to the current operating conditions can be achieved in the following way: First, the operating condition parameters of the pipeline are estimated through sensor or signal analysis, for example, the pipeline fullness is estimated through the time delay characteristics of the echo signal; then, this operating condition parameter is used as input to query the benchmark model. When the benchmark model is constructed, historical data has been divided into multiple intervals according to operating conditions through clustering or regression methods, and corresponding feature statistical descriptions have been established for each operating condition interval. For example, the reference values ​​and fluctuation ranges of two state indication features under this operating condition are stored in the form of a multivariate Gaussian distribution. The reference values ​​are the mean vectors, and the fluctuation ranges are the covariance matrix or the standard deviations of each dimension; finally, the model outputs the reference values ​​and fluctuation ranges of the state indication features of the interval that best matches the current operating condition as a comparison benchmark. Other methods can also be used in other embodiments, which are not limited here.

[0035] In specific implementation, the feature deviation is obtained by comparing the currently extracted state indicator features with the reference values ​​and fluctuation ranges of the state indicator features under normal conditions. This can be achieved by comparing the reference values ​​and fluctuation ranges of the state indicator features under the current operating conditions obtained from the benchmark model with the real-time extracted current state indicator feature vector. One specific way to calculate the feature deviation is to calculate the Mahalanobis distance. Specifically, first, the difference between each value in the current feature vector and the corresponding reference average value is calculated to form a set of difference data. Then, combining the information contained in the fluctuation range, especially the variance of each feature itself and the covariance between features, the Mahalanobis distance is solved by linear algebra operations, specifically by multiplying the difference data, the inverse matrix of the covariance matrix, and the transpose of the difference data and then taking the square root. The Mahalanobis distance is a single, non-negative scalar value, which is the feature deviation. It should be noted that the feature deviation quantitatively represents the degree to which the current state deviates from the "normal" state. Because its calculation process considers the normal fluctuation range of different feature dimensions and the correlation between them, it is more scientific and reasonable than simply calculating the Euclidean distance. It can effectively avoid misjudging an anomaly due to a large normal fluctuation of a certain feature. In addition, the standardized residuals on each dimension can also be calculated at the same time, that is, the difference between the current value of each feature and the reference average value, and then divided by the standard deviation of the feature in the normal state, so as to obtain a deviation vector containing two values. This deviation vector can more precisely reflect the direction and magnitude of the deviation on each feature dimension. Other methods can also be used in other embodiments, which are not limited here.

[0036] In some embodiments, mapping the feature deviation to the determination type and thickness estimation result of the current inner wall attachment can be achieved by the following steps: The feature deviation is input into a pre-trained hybrid recognition model; The feature deviation is processed by the hybrid recognition model to output the preliminary recognition parameters of the current inner wall attachments; The final judgment type and thickness estimation result are determined based on the preliminary identification parameters.

[0037] It should be noted that the hybrid identification model in this application is a computational model that integrates the physical mechanism of sound wave propagation and experimental calibration data. Its core function is to establish an inverse mapping relationship from the acoustic feature space to the physical parameter space of the attachment, so as to infer the type and thickness of the attachment based on the feature deviation. The preliminary identification parameters are a set of preliminary inference results about the material and geometric dimensions of the attachment directly calculated and output by the hybrid identification model based on the current feature deviation. Its core function is to provide a primary solution for the identification of the attachment state.

[0038] In specific implementation, the hybrid recognition model can be pre-trained in the following manner: the construction of the hybrid recognition model integrates physical mechanisms and experimental data. First, based on the physical principle of sound wave propagation in multi-layered media, a forward physical model describing the reflection and dispersion behavior of elastic waves in the "pipe wall-attachment layer-internal fluid" structure is established. This forward physical model takes the key physical parameters of the attachment as input, such as acoustic impedance, thickness, density, and sound velocity, and can theoretically calculate the corresponding acoustic characteristics, namely the relative change of the interface reflection coefficient and the signal dispersion characteristic parameters. Then, a large number of calibration experiments are carried out in a laboratory environment. Using pipe sections of the same material as the target pipe, standard attachment samples of different materials and known thicknesses are artificially prepared on their inner walls to simulate hard scale, soft biofilm, and other types. Using the same elastic wave sensing device and acquisition process as in the field, the actual sound corresponding to each standard sample is measured. The system learns feature data to establish a "attachment parameter-acoustic feature" mapping database covering various types and thickness combinations. Next, this mapping database is used to fit and optimize the undetermined parameters in the aforementioned forward physical model, such as the interface coupling loss coefficient and material attenuation coefficient, so that the predicted values ​​of the theoretical model can best match the measured data. Finally, based on the optimized physical model, its inverse solution mechanism is constructed, that is, an algorithm that can quickly deduce the most probable combination of attachment parameters from the input acoustic features or feature deviations. This algorithm can be implemented based on lookup table matching, optimized search algorithms, or machine learning models. Optimized search algorithms include least squares fitting, and machine learning models include support vector regression. The system obtained through the above process, possessing the ability to solve inversely from features to parameters, is the pre-trained hybrid recognition model. Other methods can also be used in other embodiments, which are not limited here.

[0039] In specific implementation, the feature deviation amount can be input into the pre-trained hybrid recognition model in the following way: the calculated feature deviation amount is used as input data and passed to the hybrid recognition model. The specific input form can be a comprehensive Mahalanobis distance scalar value. More preferably, it is a standardized residual vector that simultaneously contains the deviation direction and magnitude of the deviation in two feature dimensions. This standardized residual vector directly reflects the relative change of the currently measured interface reflection coefficient and the signal dispersion characteristic parameter, relative to their respective deviations from the normal reference values ​​under the current working conditions. The standardized residual vector is encapsulated in a necessary format, for example, converted into the input data dimensions and format expected by the model, and transmitted to the hybrid recognition model that has been loaded in memory through a program call interface, thereby starting the model's recognition calculation process. Other methods can also be used in other embodiments, which are not limited here.

[0040] In specific implementation, the hybrid recognition model processes the feature deviation and outputs the preliminary identification parameters of the current inner wall attachments. This can be achieved in the following way: After receiving the input standardized residual vector or other forms of feature deviation, the hybrid recognition model starts its built-in inverse solution algorithm. This algorithm searches and matches in the feature space defined by the "attachment parameter-acoustic feature" mapping relationship established by pre-training. Its core operation is to find a set of parameters, such as a specific material type code and thickness value, in the preset attachment parameter solution space, including different types of acoustic impedance ranges and reasonable thickness ranges, so that these parameters can be substituted into the forward physical model part of the model. At that time, the feature state represented by the deviation between the theoretically calculated acoustic feature value and the actual feature value of the current input is closest. The standard for measuring "closeness" can be to calculate the Euclidean distance or Mahalanobis distance between the theoretical feature value and the actual feature value and seek to minimize it. Finally, the model outputs one or a set of attachment parameter solutions that minimize this distance. These solutions are the preliminary identification parameters. For example, the model may output a preliminary judgment of "Type A" (corresponding to high impedance material) and "thickness of about 2.5 mm", or output that it is more likely to belong to "Type B" (corresponding to low impedance material) and "thickness of about 1.8 mm". Other methods can also be used in other embodiments, which are not limited here.

[0041] In specific implementation, determining the final judgment type and thickness estimation result based on the preliminary identification parameters can be achieved in the following way: The rationality of one or more preliminary identification parameter solutions output by the hybrid identification model is verified and a decision is made. Specifically, firstly, constraints are set based on physical common sense and engineering experience. For example, certain types of deposits cannot exceed the physical limit thickness, or the environment of a specific pipe section is more prone to producing a certain type of dirt. These constraints are used to filter the preliminary solutions. If, after filtering, there is a clearly optimal solution, for example, the error between its theoretically calculated characteristics and actual characteristics is much smaller than other solutions, then the deposit type and thickness value indicated by this solution are directly used as the final judgment type and thickness estimation result. If multiple competing solutions exist with similar errors, one of the following strategies can be used for decision-making: 1) Call a more refined physical model or more historical data patterns for secondary discrimination; 2) Combine prior knowledge of pipeline operating conditions, such as flow rate and medium composition, for weighted selection; 3) Output the most likely type and a thickness range, such as "likely type A, thickness between 2.0-3.0 mm", and transmit this uncertainty information along with it; Finally, output a definite judgment type or a thickness estimation result with a confidence interval. The judgment type belongs to a predefined set of types, and the thickness estimation result is in millimeters or a range. Other methods can also be used in other embodiments, which are not limited here.

[0042] It should be noted that this step can effectively filter out the interference of normal operating condition fluctuations based on a dynamically adaptive benchmark model, and through a hybrid identification model that integrates physical mechanisms and data-driven approaches, it can robustly map the acoustic feature deviation to the specific type and thickness of the deposits, thereby significantly improving the accuracy, robustness, and adaptability to individual differences and long-term aging of pipelines, and reducing misjudgments and missed judgments.

[0043] In step 104, based on the judgment type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, the reduction factor of the current inner wall attachments on the pipe's water flow capacity is determined.

[0044] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining the reduction factor in some embodiments of this application. In this embodiment, the multidimensional process features are compared with the current standard process parameter range in the pre-built process feature-defect association knowledge base. The cumulative deviation of each process feature within the sliding time window can be determined by the following steps: In step 1031, the equivalent roughness increment caused by the deposits inside the pipe wall is determined according to the judgment type; In step 1032, the amount of reduction in flow area caused by the deposits inside the pipe is determined based on the thickness estimation result; In step 1033, the reduction factor of the current inner wall attachment on the water flow capacity of the pipeline is determined based on the hydraulic design parameters, the equivalent roughness increment, and the reduction in flow area.

[0045] It should be noted that the equivalent roughness increment in this application is an engineering parameter that transforms the material type information of the deposits into the degree of influence on the hydraulic friction resistance of the pipe wall. Its core function is to quantify the amount of increase in pipe wall roughness caused by a specific type of deposit. The reduction in flow area is a geometric parameter that quantifies the reduction in effective flow cross-section caused by the deposits occupying pipe space. Its core function is to characterize the direct impact of deposit thickness on pipe flow capacity. The reduction coefficient is a comprehensive performance index that characterizes the proportion of the decrease in the current water flow capacity of the pipe relative to the clean design state, calculated after considering the dual effects of deposits on pipe wall friction resistance and flow cross-section. Its core function is to transform the physical presence of deposits into an intuitive and measurable degree of engineering performance degradation.

[0046] In specific implementation, determining the equivalent roughness increment caused by the deposits inside the pipe wall based on the judgment type can be achieved in the following way: a deposit type-equivalent roughness lookup table is pre-set. This lookup table is established based on engineering experience, experimental data, and literature. Each deposit judgment type predefined in the hybrid recognition model, such as "Type I - calcium carbonate scale" and "Type II - protein biofilm", is assigned a corresponding equivalent absolute roughness value or an additional value of the Manning roughness coefficient n. When the output judgment type is received, the equivalent roughness increment caused by that type of deposit can be directly obtained by querying this lookup table. For example, for "Type I - calcium carbonate scale", the corresponding Manning coefficient n may increase by 0.015, and for "Type II - protein biofilm", the corresponding n may increase by 0.005. This equivalent roughness increment is a parameter directly related to the material characteristics of the deposit and is used to quantify its influence on the frictional resistance of the pipe wall. Other methods can also be used in other embodiments, which are not limited here.

[0047] It should be noted that the attachment type-equivalent roughness comparison table in this application can be pre-set in the following manner: The construction of this comparison table is a process based on multi-source data fusion and verification. First, laboratory measurements: During the calibration experiment stage of the pre-training of the hybrid recognition model, not only are acoustic features recorded, but hydraulic experiments are also conducted simultaneously on standard attachment samples of different materials and thicknesses. For example, the pressure drop of the test pipe section with the attachment sample under different flow rates is measured in a circulating water tank, and its equivalent Manning roughness coefficient or absolute roughness value is obtained by back-calculation, thereby establishing first-hand core mapping data. Second, literature and standard research: Extensive collection of measured data or relevant information from publicly available engineering manuals, academic papers, and industry standards regarding the influence of various common pipe deposits, such as calcium carbonate scale, sulfate scale, grease, and biofilm, on the hydraulic friction of pipes. First, empirical values ​​are used as a supplement and reference. Second, on-site case studies are accumulated: combining historical maintenance records, such as pipeline flow recovery data before and after cleaning, post-evaluation and data backtracking are conducted on the roughness impact corresponding to specific pipe sections and specific types of deposits, continuously enriching and revising the comparison table. Finally, expert knowledge base integration: the data obtained above are combined with the experience of experts in fluid mechanics and water supply and drainage engineering for comprehensive analysis and standardization. A reasonable equivalent roughness increment value or range is assigned to each defined judgment type, such as "Type I - calcium carbonate scale", and stored in a database table or configuration file to form an initial deposit type-equivalent roughness comparison table. This comparison table supports dynamic updates and optimization based on more on-site verification data. Other methods can also be used in other embodiments, which are not limited here.

[0048] In practice, the reduction in flow area caused by deposits inside the pipe, based on the thickness estimation result, can be achieved as follows: First, the nominal inner diameter of the pipe in the hydraulic design parameters of the current monitored pipe section is read from the pipe basic information database and denoted as D; then, the output thickness estimation result is obtained and denoted as δ, in meters or millimeters. Based on a simplified model of uniform deposit coverage, the effective flow inner diameter of the pipe after the presence of deposits is calculated as d = D - 2δ; then, the original flow area of ​​the clean pipe is calculated as A_clean = π*(D / 2). 2 And the current effective current area A_current = π*(d / 2) 2 Finally, the reduction in flow area is calculated. This reduction can be expressed as the absolute value of the area reduction ΔA = A_clean - A_current. However, it is more commonly calculated as the area reduction rate η_area = 1 - (A_current / A_clean), which is a dimensionless number between 0 and 1. It intuitively reflects the reduction ratio of the flow cross section. Other methods can also be used in other embodiments, which are not limited here.

[0049] In specific implementation, the reduction factor of the current inner wall deposits on the water flow capacity of the pipeline can be determined based on the hydraulic design parameters, the equivalent roughness increment, and the reduction in flow area. This can be achieved in the following way: First, obtain key parameters such as the basic roughness coefficient (e.g., Manning coefficient n_clean) and hydraulic gradient (J, usually equal to the pipeline laying slope) under clean pipe conditions from the hydraulic design parameters. Then, add the basic roughness coefficient under clean conditions to the equivalent roughness increment obtained in the previous steps to obtain the comprehensive roughness coefficient n_current = n_clean + Δn under the current condition. Next, select a hydraulic calculation formula suitable for drainage pipelines, typically for non-full-pipe gravity flow, such as the Manning formula, and convert the clean state parameter (n_clean, D) and the current state parameter (n_current, D) to the hydraulic design parameters. d) Substituting into Manning's formula, under the same hydraulic gradient J, calculate the theoretical flow velocity, and combine it with the corresponding flow area (A_clean, A_current) to calculate the theoretical flow rate Q_clean in the clean state and the theoretical flow rate Q_current in the current state; finally, calculate the reduction factor η_flow of the flow capacity, which is defined as: η_flow = 1 - (Q_current / Q_clean). This reduction factor η_flow of the flow capacity is a quantitative performance index. The closer its value is to 1, the more serious the reduction in flow capacity. Other methods can also be used in other embodiments, which are not limited here.

[0050] It should be noted that this step can convert the physical parameters of the identified deposits into equivalent roughness increments and flow area reductions, and further integrate them into a quantified water flow capacity reduction factor, an engineering performance indicator. This transforms the abstract concept of "fouling presence" into an intuitive concept of "performance degradation," providing an objective, calculable, and unified quantitative basis for risk-based early warning, and greatly enhancing the engineering guidance value of the assessment results.

[0051] In step 105, if the reduction factor exceeds the preset reduction threshold of the drainage pipe, a maintenance warning signal is generated.

[0052] In some embodiments, if the reduction factor exceeds a preset reduction threshold for the drainage pipe, generating a maintenance warning signal can be achieved through the following steps: The reduction coefficient is compared with a preset reduction threshold. If the reduction coefficient exceeds the reduction threshold, the warning information generation process is triggered. Based on the judgment type, the thickness estimation result, and the degree to which the threshold is exceeded, a maintenance warning signal is generated and output.

[0053] It should be noted that the preset reduction threshold in this application is a critical value set in advance based on the pipeline design safety margin, functional importance and operation and maintenance strategy. It is used to determine whether the reduction of water supply capacity has reached the level that requires intervention. Its core function is to serve as a decision-making benchmark for triggering early warning.

[0054] In specific implementation, the reduction coefficient is compared with a preset reduction threshold. If the reduction coefficient exceeds the reduction threshold, the early warning information generation process is triggered. This can be achieved by maintaining a reduction threshold configuration library, where one or more reduction thresholds are preset for each monitored pipe segment or type of pipe. For example, the early warning threshold η_warn = 0.15, and the alarm threshold η_alarm = 0.25. These thresholds are comprehensively set based on the importance of the pipeline, design redundancy, historical operating risks, and operation and maintenance management standards. After calculating the water flow capacity reduction coefficient, the preset threshold for the corresponding pipe segment is retrieved from the threshold configuration library for numerical comparison. The judgment logic is: if η_flow ≥ η_warn (or η_alarm), it is determined to be an abnormal state, and the early warning information generation process is immediately triggered. Other methods can also be used in other embodiments, which are not limited here.

[0055] It should be noted that the reduction threshold configuration library in this application can be constructed in the following manner: the construction of this configuration library follows the principles of hierarchical, object-specific, and dynamically adjustable. First, a basic threshold is set based on the design specifications: the design drawings and calculation sheets of the target drainage network are consulted to obtain the design flow rate (Q_design) and design safety margin of each pipe section, and the water flow capacity is reduced to a certain critical level, for example, the water flow capacity reduction factor corresponding to the flow rate being reduced to 85% of the design flow rate, i.e., 1 - 0.85 = The basic early warning threshold η_warn for this pipe section is set at 0.15. Secondly, the threshold is adjusted based on pipeline function and importance: according to the pipeline network topology and function, such as main pipes, secondary pipes, and branches, the basic threshold is adjusted accordingly. For example, a stricter threshold is used for critical main pipes, such as η_warn=0.10, while a more lenient threshold can be used for secondary branches, such as η_warn=0.20. Thirdly, historical operational data is used for correction: historical monitoring and maintenance data for this pipe section is accessed, and its flow fluctuation patterns and past blockage event records are analyzed. If a pipe section has historically experienced frequent minor siltation when the reduction coefficient reaches 0.18, its threshold can be fine-tuned from 0.15 to 0.16 to provide early warning. Finally, the risk assessment results are integrated: for pipe sections transporting special media, such as grease-containing wastewater or industrial wastewater, the risk of rapid fouling growth is comprehensively assessed, and a more aggressive threshold is set separately. All the thresholds generated by the above rules for each monitored pipe section, including multiple levels such as attention, early warning, and alarm thresholds and their applicable conditions, are structured and stored in a central database or distributed configuration file, which constitutes the reduction threshold configuration library. This reduction threshold configuration library allows administrators to query, modify, and manage versions through the human-machine interface of the monitoring platform, and allows automatic recommendation and adjustment of threshold parameters based on seasonal changes or pipeline renovation information. Other methods can also be used in other embodiments, which are not limited here.

[0056] In specific implementation, generating and outputting a maintenance warning signal based on the judgment type, the thickness estimation result, and the degree of exceeding the threshold can be achieved in the following way: When the warning process is triggered, all relevant diagnostic information is collected, including the judgment type, the thickness estimation result (δ), the calculated reduction coefficient (η_flow), and the preset reduction threshold exceeded. Based on this relevant diagnostic information, the risk level is first determined, such as "attention," "warning," or "serious," usually based on the percentage of η_flow exceeding the threshold. Then, combining the judgment type and risk level, recommended maintenance actions are matched from a preset maintenance measures knowledge base. For example, for "Type I hard scale" at the "warning" level, the recommendation is to "arrange high-pressure water jet cleaning." For "Type II biofilm" at the "Caution" level, the recommendation is "Strengthen inspections and consider handling it during the next scheduled maintenance." Finally, this information is assembled into a structured maintenance warning signal, which includes at least the following fields: monitoring point number, timestamp, identified attachment type, estimated thickness, current reduction factor, threshold reached, risk level, and specific maintenance recommendations. This maintenance warning is uploaded to the remote monitoring center platform in real time via a communication module (such as 4G / 5G, NB-IoT), and can also be sent to the terminal of designated maintenance personnel via SMS or App according to the configuration. At the same time, logs are archived in the local non-volatile memory to complete the entire warning output process. Other methods can also be used in other embodiments, which are not limited here.

[0057] In addition, in some embodiments, if the reduction factor does not exceed the preset reduction threshold, it is considered a normal situation and no action is taken.

[0058] It should be noted that this step can compare the calculated performance reduction factor with a dynamically set scientific threshold, and automatically generate a structured early warning signal containing specific dirt information, risk level and action suggestions when the limit is exceeded. This directly and efficiently transforms the monitoring results into clear and actionable maintenance instructions, improving the timeliness of operation and maintenance response.

[0059] In another aspect, in some embodiments, this application provides an anti-scaling alarm device for the inner wall of a drainage pipe, see reference. Figure 4 The figure is a schematic diagram of the structure of a drainage pipe inner wall anti-scaling alarm device according to some embodiments of this application. The drainage pipe inner wall anti-scaling alarm device includes: a receiving module 401, a processing module 402, and an execution module 403, which are described below: The receiving module 401 in this application is mainly used to receive the echo signal reflected by the inner surface and inner wall of the drainage pipe. Processing module 402, in this application, is mainly used to extract state indication features from the echo signal. The state indication features include the relative change of the interface reflection coefficient and the signal dispersion characteristic parameter. The processing module 402 described in this application is also used to compare the status indication feature with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point, to obtain the feature deviation amount, and then map the feature deviation amount to the judgment type and thickness estimation result of the current inner wall attachment. The processing module 402 described in this application is also used to determine the reduction factor of the current inner wall attachment on the water flow capacity of the pipe based on the judgment type and thickness estimation result and the hydraulic design parameters of the drainage pipe. The execution module 403 in this application is mainly used to generate a maintenance warning signal if the reduction coefficient exceeds the preset reduction threshold of the drainage pipe.

[0060] Each module in the aforementioned anti-scaling alarm device for the inner wall of drainage pipes can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0061] In another embodiment, this application provides a computer device, which may be a server, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores anti-scaling alarm data for the inner wall of drainage pipes. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements an anti-scaling alarm method for the inner wall of drainage pipes.

[0062] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0063] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above embodiment of the anti-scaling alarm method for the inner wall of the drainage pipe.

[0064] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described embodiment of the anti-scaling alarm method for the inner wall of a drainage pipe.

[0065] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in the above-described embodiment of the anti-scaling alarm method for the inner wall of a drainage pipe.

[0066] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0067] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for alarming scale buildup on the inner wall of a drainage pipe, wherein, An elastic wave sensor is pre-installed at a target monitoring point in a drainage pipeline. The elastic wave sensor periodically transmits a broadband excitation signal of a preset waveform to the pipe wall. The method is characterized by the following steps: Receive echo signals reflected by deposits on the inner surface and inner wall of the drainage pipe; State indication features are extracted from the echo signal, including the relative change in the interface reflection coefficient and signal dispersion characteristic parameters; The status indication features are compared with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point to obtain the feature deviation amount. The feature deviation amount is then mapped to the judgment type and thickness estimation result of the current inner wall attachment. Based on the judgment type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, the reduction factor of the current inner wall attachments on the pipe's water flow capacity is determined. If the reduction factor exceeds the preset reduction threshold of the drainage pipe, a maintenance warning signal is generated.

2. The method as described in claim 1, characterized in that, State indication features are extracted from the echo signal. These features include the relative change in the interface reflection coefficient and signal dispersion characteristic parameters, specifically: The echo signal is preprocessed to enhance signal quality and suppress noise; The relative change in the interface reflection coefficient is determined based on the preprocessed echo signal. The signal dispersion characteristic parameters are determined based on the same preprocessed echo signal.

3. The method as described in claim 1, characterized in that, The status indication features are compared with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring points. The feature deviation specifically includes: Based on the current operating conditions, obtain the reference values ​​and fluctuation ranges of the corresponding normal state status indication characteristics from the benchmark model; The extracted state indication feature is compared with the reference value and fluctuation range of the state indication feature of the normal state to obtain the feature deviation.

4. The method as described in claim 1, characterized in that, Mapping the feature deviation to the determination of the current inner wall attachment type and thickness estimation result specifically includes: The feature deviation is input into a pre-trained hybrid recognition model; The feature deviation is processed by the hybrid recognition model to output the preliminary recognition parameters of the current inner wall attachments; The final judgment type and thickness estimation result are determined based on the preliminary identification parameters.

5. The method as described in claim 1, characterized in that, Based on the aforementioned judgment type and thickness estimation results, and according to the hydraulic design parameters of the drainage pipe, the specific factors determining the reduction factor of the current inner wall attachments on the pipe's water flow capacity include: The equivalent roughness increment caused by the deposits inside the pipe wall is determined based on the judgment type. The amount of reduction in flow area caused by deposits inside the pipe is determined based on the thickness estimation results; The reduction factor of the current inner wall deposits on the water flow capacity of the pipeline is determined based on the hydraulic design parameters, the equivalent roughness increment, and the reduction in flow area.

6. The method as described in claim 1, characterized in that, If the reduction factor exceeds the preset reduction threshold of the drainage pipe, a maintenance warning signal will be generated, specifically including: The reduction coefficient is compared with a preset reduction threshold. If the reduction coefficient exceeds the reduction threshold, the warning information generation process is triggered. Based on the judgment type, the thickness estimation result, and the degree to which the threshold is exceeded, a maintenance warning signal is generated and output.

7. The method as described in claim 1, characterized in that, Also includes: If the reduction factor does not exceed the preset reduction threshold, it is considered a normal situation and no action is taken.

8. A scale-prevention alarm device for the inner wall of a drainage pipe, characterized in that, include: The receiving module is used to receive echo signals reflected by the deposits on the inner surface and inner wall of the drainage pipe; The processing module is used to extract state indication features from the echo signal, the state indication features including the relative change of the interface reflection coefficient and the signal dispersion characteristic parameter; The processing module is also used to compare the status indication feature with a benchmark model that includes the normal operating condition fluctuation range and is established based on long-term historical data of the target monitoring point, to obtain the feature deviation amount, and then map the feature deviation amount to the judgment type and thickness estimation result of the current inner wall attachment. The processing module is also used to determine the reduction factor of the current inner wall attachment on the water flow capacity of the pipe based on the judgment type and thickness estimation result and according to the hydraulic design parameters of the drainage pipe. The execution module is used to generate a maintenance warning signal if the reduction coefficient exceeds the preset reduction threshold of the drainage pipe.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the anti-scaling alarm method for the inner wall of the drainage pipe as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the anti-scaling alarm method for the inner wall of the drainage pipe as described in any one of claims 1 to 7.

Images