A method, device, product and medium for processing data of an irrigation drainage liquid level monitoring

Abstract

A kind of flushes drainage level monitoring data processing method, equipment, product and medium, it is related to medical data processing technical field.The method, the initial response data of each flushing path under initial pulse injection is obtained, time-frequency response baseline is established, impedance feature is calculated to generate impedance difference graph;Impedance difference graph is determined according to the target flushing path, velocity data and pressure data are obtained by perturbation injection, and velocity and pressure coupling distribution graph is constructed;Velocity and pressure coupling distribution graph and impedance difference graph are matched in time and space, and convolute core area is identified by time-space image sequence;Secondary injection instruction is generated for convolute core area, secondary response feature data is obtained and image fusion is carried out with impedance difference graph, and risk heat map is generated;Injection control strategy of staggered peak is generated according to risk heat map, and injection control strategy is dynamically adjusted based on the response lag of each path again.The efficiency of fluid flushing is improved.

Description

Technical Field

[0001] This application relates to the field of medical data processing technology, specifically to a method, device, product, and medium for processing data related to flushing and drainage fluid level monitoring. Background Technology

[0002] In the operation of multi-channel flushing and drainage systems (such as medical devices used for cleaning complex industrial pipelines or for flushing target cavities after surgery), the unobstructed flow of fluid lines and the uniformity of fluid distribution are key to ensuring the efficiency of equipment operation.

[0003] Currently, the control methods for such multi-channel flushing and drainage systems are usually based on macroscopic monitoring of the overall liquid level or the total outlet pressure. When the overall liquid level exceeds a preset safety threshold or the total pressure is abnormal, the control system will activate the preset flushing program and perform a uniform and synchronized liquid injection operation on all flushing paths.

[0004] However, this uniform control method ignores the microscopic flow field dynamics within complex target areas and cannot identify and handle fluid dynamic anomalies such as local liquid stagnation, fluid dead zones, or even asymmetric swirling flow caused by significant differences in physical resistance between multiple paths, resulting in low fluid flushing efficiency. Summary of the Invention

[0005] This application provides a method, device, product, and medium for processing flushing and drainage liquid level monitoring data, which has the effect of improving fluid flushing efficiency.

[0006] The first aspect of this application provides a method for processing flushing and drainage fluid level monitoring data, specifically including: Acquire initial response data of multiple flushing paths connected to the target flushing area under initial pulse injection, and establish time-frequency response baselines for each flushing path based on the initial response data; The impedance characteristics of each flushing path are calculated based on the time-frequency response baseline, and an impedance difference map reflecting the path resistance distribution is generated based on the impedance characteristics of each flushing path. The target flushing path is determined from multiple flushing paths based on the impedance difference diagram. The velocity and pressure data of the target flushing path under disturbed injection are obtained, and a velocity and pressure coupling distribution map of the target flushing area is constructed based on the velocity and pressure data. Spatiotemporal matching of velocity-pressure coupling distribution map and impedance difference map is performed to obtain spatiotemporal image sequence, and the swirling core region in the target flushing area is identified based on the spatiotemporal image sequence; A secondary injection command is generated for the cyclone core region, and secondary response characteristic data of the cyclone core region in response to the secondary injection command is obtained. The secondary response characteristic data is then image-fused with the impedance difference map to generate a risk heat map. Based on the risk heat map, a liquid injection control strategy is generated to stagger the injection process across multiple flushing paths. The response hysteresis of each flushing path is obtained, and the liquid injection control strategy is dynamically adjusted based on the response hysteresis.

[0007] By adopting the above technical solution, firstly, initial response data of multiple flushing paths under initial pulse injection are acquired and a time-frequency response baseline is established. The response characteristics of each path in the initial state are solidified into a quantifiable reference standard. Based on this, the impedance characteristics of each flushing path are calculated and an impedance difference map is generated, making the differences in the unobstructed state at each location within the target flushing area globally visible in image form. Then, based on the impedance difference map, the target flushing path is determined, and velocity and pressure data under disturbed injection are acquired to construct a velocity-pressure coupled distribution map. The path resistance information is correlated with the actual flow state information. Subsequently, the velocity-pressure coupled distribution map and the impedance difference map are linked together. The system performs spatiotemporal matching to identify the core area of ​​the swirling flow, synchronously locating the location of resistance anomalies and swirling accumulation. Based on this, a secondary injection command is generated, and the secondary response characteristic data is fused with the impedance difference map to generate a risk heat map. The spatial coupling relationship between swirling accumulation intensity and local resistance anomalies is visualized, enabling differentiated configuration of intervention priorities and injection parameters for each flushing path. The injection sequence of each path is staggered to avoid pressure superposition. Continuous monitoring of response lag allows the strategy parameters to be corrected in real time according to the path status, ensuring that the injection intervention always acts on the area with the most concentrated residual accumulation and maintains effective removal pressure, thereby improving the efficiency of fluid flushing.

[0008] Optionally, the step of acquiring initial response data of multiple flushing paths connected to the target flushing area under the initial pulse injection, and establishing a time-frequency response baseline for each flushing path based on the initial response data, specifically includes: Generate phase-coded micro-amplitude pulse commands containing unique identifiers to instruct each flushing path to perform an initial pulse injection; The input pressure disturbance waveform and output liquid level response curve of each flushing path under the initial pulse injection are obtained, and the input pressure disturbance waveform and output liquid level response curve are used as the initial response data. Based on the time difference between the input pressure disturbance waveform and the output liquid level response curve in the initial response data, the phase delay time of each flushing path is calculated. Based on the rising slope and peak amplitude of the output liquid level response curve in the initial response data, and the disturbance amplitude of the input pressure disturbance waveform, the pressure amplitude attenuation ratio of each flushing path is calculated. The phase delay time and amplitude attenuation ratio of each flushing path are integrated into a two-dimensional response spectrum, and the two-dimensional response spectrum is used as the time-frequency response baseline.

[0009] By employing the above technical solution, a phase-coded micro-amplitude pulse command containing a unique identifier is generated, enabling each flushing path to have traceable marking information when executing the initial pulse injection, thus avoiding response signal confusion when multiple paths operate in parallel. The input pressure disturbance waveform and the output liquid level response curve are acquired as initial response data, directly recording the complete physical transmission process of the liquid from pressure application to liquid level change, establishing a data foundation for subsequent analysis. The phase delay time is calculated based on the time difference between the input pressure disturbance waveform and the output liquid level response curve, quantifying the difference in liquid transmission speed in each flushing path; a longer delay time indicates a more tortuous path or the presence of blockage. The pressure amplitude attenuation ratio is calculated based on the rising slope and peak amplitude of the output liquid level response curve and the disturbance amplitude of the input pressure disturbance waveform, reflecting the degree of pressure energy loss during transmission; a larger attenuation ratio indicates more significant path resistance. The phase delay time and pressure amplitude attenuation ratio are integrated into a two-dimensional response spectrum as a time-frequency response baseline, simultaneously presenting the transmission delay in the time domain and the attenuation loss in the energy domain, constructing a complete reference standard describing the dynamic response characteristics of each flushing path, allowing for a visual comparison of resistance differences between different paths.

[0010] Optionally, the step of calculating the impedance characteristics of each flushing path based on the time-frequency response baseline, and generating an impedance difference map reflecting the path resistance distribution based on the impedance characteristics of each flushing path, specifically includes: Obtain the path length parameters of each flushing path, and based on the path length parameters, convert the phase delay time in the time-frequency response baseline into a phase delay ratio per unit length. Extract the pulse duration segment of the input pressure disturbance waveform, calculate the input energy integral value of the input pressure disturbance waveform within the pulse duration segment, and the output energy integral value of the output liquid level response curve within the pulse duration segment; Based on the ratio of the output energy integral value to the input energy integral value, and combined with the path length parameter, the pressure attenuation gradient per unit length is calculated. The phase hysteresis ratio of each flushing path and the pressure attenuation gradient are combined into a two-dimensional feature vector, and the two-dimensional feature vector is used as the impedance feature of each flushing path. An impedance difference matrix is ​​established based on the impedance characteristics of each flushing path, and the impedance difference matrix is ​​rendered into a two-dimensional heatmap to generate an impedance difference map.

[0011] By employing the above technical solution, the path length parameters of each flushing path are obtained, and the phase delay time is converted into a phase delay ratio per unit length. This eliminates the interference caused by direct comparison of delay times due to different path lengths, making the transmission delay between flushing paths comparable. The input and output energy integral values ​​within the pulse duration are extracted, and the energy transfer during the complete pulse process is statistically analyzed using integration, avoiding random errors introduced by single-point sampling. Based on the ratio of the output to input energy integral values, combined with the path length parameters, the pressure attenuation gradient per unit length is calculated, normalizing energy loss to a unit distance and accurately quantifying the concentration of local resistance within the path. The phase delay ratio and pressure attenuation gradient are combined into a two-dimensional feature vector as the impedance characteristics of each flushing path, simultaneously capturing resistance information in both the time and energy domains, compensating for the inability of a single indicator to comprehensively reflect the path resistance state. Based on the impedance characteristics of each flushing path, an impedance difference matrix is ​​established and rendered as a two-dimensional heatmap to generate an impedance difference map, transforming the numerical differences in resistance between paths into an intuitive spatial distribution, making the resistance distribution of each flushing path clearly distinguishable.

[0012] Optionally, the step of determining the target flushing path from multiple flushing paths based on the impedance difference diagram, obtaining velocity and pressure data of the target flushing path under disturbed injection, and constructing a velocity-pressure coupling distribution map of the target flushing area based on the velocity and pressure data specifically includes: Select the flushing path whose impedance difference is greater than a preset threshold from the impedance difference diagram as the target flushing path; Output control commands to instruct the target flushing path to perform asynchronous directional disturbance injection, and acquire velocity and pressure data of the target flushing area under disturbance injection; Spatial interpolation fitting is performed on the velocity data to generate a velocity vector layer, and a pressure color level layer is constructed based on the pressure data; Overlay the velocity vector layer onto the pressure level layer to generate a time-series velocity-pressure coupled distribution map.

[0013] By employing the above technical solution, flushing paths with impedance differences exceeding a preset threshold are selected from the impedance difference map as target flushing paths. This limits the subsequent analysis to paths with abnormal resistance distribution, eliminating interference from paths with uniform resistance. Control commands are output to instruct the target flushing paths to perform asynchronous directional disturbance injection, ensuring that the injection sequences of each target flushing path are staggered. This avoids the overlap and confusion of velocity and pressure data in the target flushing area caused by synchronous injection, guaranteeing that the response data of each path are independent and identifiable. Spatial interpolation fitting is performed on the velocity data to generate a velocity vector layer, expanding discrete velocity measurement points into a continuous spatial velocity field, thus fully presenting the direction and magnitude of the flow velocity within the target flushing area. A pressure color-gradient layer is constructed based on the pressure data, visually reflecting the pressure gradient distribution within the target flushing area through color-gradient differences. The velocity vector layer is superimposed on the pressure color-gradient layer to generate a time-series velocity-pressure coupled distribution map, synchronously presenting the spatial correspondence between velocity direction and pressure gradient. The time-series format further records the dynamic evolution of the flow field within the target flushing area during the injection process.

[0014] Optionally, the step of spatiotemporally matching the velocity-pressure coupling distribution map with the impedance difference map to obtain a spatiotemporal image sequence, and identifying the swirling core region within the target flushing area based on the spatiotemporal image sequence, specifically includes: Using the liquid outlet coordinates of the flushing path as the reference anchor point, the velocity and pressure coupling distribution map and impedance difference map are spatially mapped and aligned with the time frame sequence to generate a spatiotemporal image sequence with a unified coordinate system. The movement path of fluid particles is traced in a spatiotemporal image sequence to extract the liquid velocity trajectory, and the trajectory curvature, number of direction changes, path closure and velocity fluctuation amplitude of the liquid velocity trajectory are calculated. The trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude are integrated to generate swirling flow indices, and the residence time distribution map is constructed by statistically analyzing the continuous residence time of fluid particles in local areas. Extract overlapping areas where the swirling flow index is greater than a preset index threshold and the continuous residence time in the residence time distribution map is greater than a preset time threshold, and mark the overlapping areas as the swirling core area within the target flushing area.

[0015] By employing the above technical solution, the velocity-pressure coupling distribution map and impedance difference map are spatially mapped and aligned with time frame sequences using the liquid outlet coordinates of the flushing path as the reference anchor point. This generates a spatiotemporal image sequence with a unified coordinate system, ensuring strict correspondence between data at the same spatial location and time in both images, eliminating regional misjudgments caused by coordinate deviations. The movement path of fluid particles is tracked in the spatiotemporal image sequence to extract the liquid velocity trajectory. The trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude are calculated. These four indicators characterize the particle motion features from the perspectives of curvature, turning frequency, trajectory closure state, and velocity stability. Complex swirling patterns that cannot be reflected by a single indicator are covered by combining multiple indicators. These four indicators are integrated to generate a swirling flow index, uniformly measuring the intensity of swirling motion in each region. Simultaneously, the residence time is statistically analyzed to construct a residence time distribution map, reflecting the degree of fluid accumulation in local areas. The overlapping area where both the swirling flow index and the residence time exceed their respective preset thresholds is identified as the swirling core region. The dual threshold constraint excludes interference areas that only have swirling characteristics but where the fluid has not accumulated, or where the fluid accumulates but there is no obvious swirling, so that the calibration results only retain the locations where there is actually continuous swirling accumulation.

[0016] Optionally, the step of generating a secondary injection command for the cyclone core region and acquiring secondary response characteristic data of the cyclone core region in response to the secondary injection command, and then fusing the secondary response characteristic data with the impedance difference map to generate a risk heatmap, specifically includes: Generate a phase-coded micro-amplitude pulse injection command for the cyclotron core region as a secondary injection command; The delayed response time, response amplitude rise rate, and fluid disturbance radius of the cyclone core region under the secondary injection command are obtained as secondary response characteristic data. The swirling flow index is normalized based on the secondary response characteristic data, and the risk level is divided according to the preset multidimensional threshold model to generate a risk level map. By spatially registering the risk level map and the impedance difference map and mixing them with dual-channel colors, a risk heatmap characterizing the coupling distribution of local drag anomalies and swirling flow is generated.

[0017] By employing the above technical solution, a phase-coded micro-amplitude pulse injection command is generated for the swirling core region as a secondary injection command. The micro-amplitude pulse ensures that the injection intensity is insufficient to disturb the existing flow regime in the swirling core region, while the phase coding makes the response signals from different injection points mutually identifiable, ensuring the accurate attribution of the acquired secondary response characteristic data. The delayed response time, response amplitude rise rate, and fluid disturbance radius are acquired as secondary response characteristic data. These three parameters correspond to the local path conduction delay, energy transfer rate, and disturbance spatial range, respectively, comprehensively quantifying the local resistance state of the swirling core region from three dimensions: time, rate, and space. Based on the secondary response characteristic data, the swirling flow indices are normalized to eliminate the magnitude deviation of indices caused by differences in path size in different swirling core regions. Then, a risk level map is generated by classifying risk levels according to a preset multi-dimensional threshold model. The multi-dimensional threshold limitation avoids misjudgment of the level caused by excessively high single indices. A risk heatmap is generated by spatially registering the risk level map and the impedance difference map and mixing them in dual-channel colors. Local resistance anomalies and swirling flow risk levels are superimposed on a unified spatial coordinate system. The spatial overlap area of ​​the two directly marks the high-risk location where high resistance and swirling accumulation coexist.

[0018] Optionally, the step of generating a staggered injection control strategy across multiple flushing paths based on a risk heatmap, obtaining the response hysteresis of each flushing path, and dynamically adjusting the injection control strategy based on the response hysteresis specifically includes: Based on the risk heat map, a comprehensive risk score for each flushing path is extracted, and an intervention priority ranking table is established based on the comprehensive risk score. Based on the intervention priority ranking table, non-equal interval injection start time, injection frequency and injection pressure are assigned to each flushing path to generate an injection control strategy; During the execution of the injection control strategy, the pressure recovery time and liquid removal retention time of each flushing path are obtained, and the response hysteresis of each flushing path is calculated based on the pressure recovery time and liquid removal retention time. If the increase in response lag of any flushing path in a continuous cycle exceeds a preset proportion, the injection cycle interval of any flushing path is extended based on the increase, and the injection frequency of any flushing path is reduced, so as to dynamically adjust the injection control strategy.

[0019] By adopting the above technical solution, a comprehensive risk score is extracted from each flushing path based on the risk heat map, and an intervention priority ranking table is established. The coupling degree between local resistance anomalies and swirling flow in the risk heat map is quantified into a unified score, making the intervention urgency of different flushing paths comparable. The priority ranking table further transforms score differences into an executable treatment sequence. Based on the intervention priority ranking table, non-equidistant injection start times, injection frequencies, and injection pressures are assigned to each flushing path to generate injection control strategies. The non-equidistant timing sequence ensures that high-risk paths receive priority injection intervention, while avoiding pressure superposition within the target flushing area caused by simultaneous injection of multiple paths. The differentiated allocation of injection frequency and injection pressure further adapts to the resistance characteristics of each path. During the execution of the injection control strategy, pressure recovery time and liquid removal retention time are obtained to calculate response lag. Pressure recovery time reflects the rate at which the path recovers to its baseline state after the injection pressure dissipates, and liquid removal retention time reflects the ability to remove fluid residues. The response lag obtained by combining the two comprehensively characterizes the actual unobstructed state of the path under the current injection parameters. If the increase in response lag of any flushing path exceeds a preset proportion within a continuous cycle, the injection control strategy is dynamically adjusted by extending the injection cycle interval and reducing the injection frequency based on the increase. The continuous cycle judgment eliminates erroneous adjustments caused by single fluctuations. The adjustment actions of extending the interval and reducing the frequency provide sufficient liquid discharge and pressure recovery window for high-lag paths, enabling the injection control strategy to continuously adapt to the dynamic changes of each flushing path. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a flushing and drainage level monitoring data processing method provided in an embodiment of this application. Detailed Implementation

[0021] The flushing and drainage fluid level monitoring data processing method provided in this application embodiment can be applied to a flushing and drainage fluid level monitoring data processing system. The system architecture may include electronic devices, a network, and a flushing and drainage monitoring terminal. The network serves as the medium for providing a communication link between the electronic devices and the flushing and drainage monitoring terminal. The network may include various connection types, such as wired, wireless communication links, or fiber optic cables.

[0022] Electronic devices can interact with the flushing and drainage monitoring terminal via a network to receive or send data. The flushing and drainage monitoring terminal can be configured with various data acquisition and execution components, such as micro-pressure sensor arrays, liquid level transducers, vector Doppler probes, and pulse injection control valves.

[0023] The electronic device is hardware and can be any device with data processing and image computing capabilities, including but not limited to smartphones, tablets, laptops, desktop computers, and backend servers. As the executing entity of the data processing method in this application, the electronic device can process and analyze the received initial response data, velocity data, and pressure data to generate impedance difference diagrams, risk heat maps, and injection control strategies, and then issue the generated injection commands.

[0024] The irrigation and drainage monitoring terminal can be a medical hardware device deployed in the target irrigation area (such as the intersphincteric space of the anus). The irrigation and drainage monitoring terminal can collect response data of each irrigation path and send it to the electronic device. At the same time, it can receive the processing results issued by the electronic device (such as secondary injection commands, staggered injection control strategies), and execute corresponding pulse injection and disturbance operations accordingly.

[0025] It should be specifically stated that the flushing and drainage liquid level monitoring data processing method provided in this embodiment of the invention is essentially an automated data processing and equipment control parameter generation method based on sensor data acquisition and fluid dynamics calculation. The steps in this method, such as "injecting pulses," "acquiring data," and "dynamically adjusting the injection cycle," are all executed by automated electronic devices, microprocessors, or fluid control systems, and belong to the signal interaction and control logic processes within the equipment.

[0026] This embodiment provides a method for processing flushing drainage liquid level monitoring data. Figure 1 This is a flowchart illustrating a method for processing flushing and drainage liquid level monitoring data provided in an embodiment of this application. Figure 1 As shown, the method includes steps S101 to S106: S101: Acquire the initial response data of multiple flushing paths connected to the target flushing area under the initial pulse injection, and establish the time-frequency response baseline of each flushing path based on the initial response data.

[0027] In this embodiment, the target flushing area refers to the gap space where flushing and drainage level monitoring is required; multiple flushing paths refer to channels that establish liquid flow connections with the target flushing area through drainage tubes, injection conduits, or intersphincteric spaces; the initial pulse injection refers to the short-duration pulsed liquid injection operation applied to each flushing path during the system initialization phase; the initial response data refers to the physical response information such as pressure changes, flow rate fluctuations, and liquid level oscillations generated by each flushing path after receiving the initial pulse injection; and the time-frequency response baseline is used to represent the comprehensive characteristic curve of the frequency domain characteristics and time domain evolution law presented by each flushing path under the initial pulse excitation.

[0028] Specifically, the electronic device applies initial pulse injection sequentially or synchronously to multiple flushing paths connected to the target flushing area through the flushing and drainage monitoring terminal. The micro-pressure sensor array and vector Doppler probe configured in the flushing and drainage monitoring terminal collect pressure fluctuation sequences and flow velocity change sequences of each flushing path in real time after receiving the initial pulse injection as initial response data. The electronic device performs Fourier transform and wavelet decomposition on the initial response data corresponding to each flushing path, extracts the dominant frequency component and phase lag characteristics of the pressure fluctuation sequence, as well as the peak arrival time and decay rate of the flow velocity change sequence. The extracted dominant frequency component, phase lag characteristics, peak arrival time and decay rate are arranged according to the time axis to form a time-frequency response baseline. In the time-frequency response baseline, the time axis reflects the propagation delay of pressure waves or flow velocity waves in the flushing path, and the frequency axis reflects the response sensitivity of the flushing path to pulse signals of different frequencies.

[0029] Based on the above embodiments, as an optional embodiment, the step of acquiring initial response data of multiple flushing paths connected to the target flushing area under the initial pulse injection, and establishing the time-frequency response baseline of each flushing path based on the initial response data, includes steps S201 to S205: S201: Generate a phase-coded micro-amplitude pulse command containing a unique identifier to instruct each flushing path to perform an initial pulse injection.

[0030] In this embodiment, the unique identifier is used to represent the identity coding information used to distinguish different flushing paths, the phase coding represents the coding method of path marking by changing the phase parameter of the pulse signal, the micro-amplitude pulse command represents the liquid injection control command containing amplitude control parameters, duration parameters and phase parameters, and the phase-coded micro-amplitude pulse command represents the control command with path recognition capability formed by embedding the unique identifier in the phase parameter of the micro-amplitude pulse command.

[0031] Specifically, the electronic device assigns a unique identifier to each of the multiple flushing paths connected to the target flushing area. This unique identifier is mapped to a corresponding phase offset using phase encoding. The electronic device embeds this phase offset as a phase parameter into the waveform configuration of the micro-amplitude pulse command, generating a phase-encoded micro-amplitude pulse command containing the unique identifier. This generated micro-amplitude pulse command is then transmitted via the network to the flushing and drainage monitoring terminal. The flushing and drainage monitoring terminal parses the phase parameter in the phase-encoded micro-amplitude pulse command and identifies the corresponding unique identifier. The flushing and drainage monitoring terminal then controls the pulse injection control valve to drive the liquid injection channel to perform initial pulse injection into the flushing path corresponding to the identified unique identifier. The amplitude of the initial pulse injection micro-pulse signal is set within the range of 1 to 2 kPa, and the duration is controlled within 0.3 to 0.5 seconds to ensure sufficient disturbance detection capability without damaging the structural integrity of the target flushing area.

[0032] S202: Obtain the input pressure disturbance waveform and output liquid level response curve of each flushing path under the initial pulse injection, and use the input pressure disturbance waveform and output liquid level response curve as the initial response data.

[0033] In this embodiment, the input pressure disturbance waveform represents the fluctuation curve of the pressure generated at the injection port of each flushing path as a function of time when receiving the initial pulse injection, and the output liquid level response curve represents the time-series response curve of the liquid level height in the target flushing area as a function of the initial pulse injection.

[0034] Specifically, while the flushing and drainage monitoring terminal performs initial pulse injection into each flushing path, its micro-pressure sensor array collects the pressure value sequence at the injection port of each flushing path in real time. This micro-pressure sensor array adopts a piezoelectric thin film structure with a sensitivity of not less than 0.05 kPa / µV and a sampling frequency set to 5 kHz or higher. The flushing and drainage monitoring terminal connects the collected pressure value sequence in chronological order to form an input pressure disturbance waveform. The liquid level transducer in the flushing and drainage monitoring terminal monitors the change sequence of liquid level height in the target flushing area in real time. This liquid level transducer is based on the piezoresistive strain principle and can sense liquid level fluctuations of less than 0.1 mm. The flushing and drainage monitoring terminal connects the monitored liquid level height change sequence in chronological order to form an output liquid level response curve. The flushing and drainage monitoring terminal sends the formed input pressure disturbance waveform and output liquid level response curve to the electronic device via the network. The electronic device receives the input pressure disturbance waveform and output liquid level response curve corresponding to each flushing path and stores the received input pressure disturbance waveform and output liquid level response curve as initial response data in the database.

[0035] S203: Calculate the phase delay time of each flushing path based on the time difference between the input pressure disturbance waveform and the output liquid level response curve in the initial response data.

[0036] In this embodiment, the phase delay time is used to represent the propagation delay time experienced from the initial pulse injection applying pressure disturbance from the injection port of the flushing path to the liquid level change response generated in the target flushing area, and the time difference represents the time interval between the time point when the input pressure disturbance waveform reaches the peak pressure and the time point when the output liquid level response curve reaches the peak liquid level.

[0037] Specifically, the electronic device extracts the input pressure disturbance waveform and output liquid level response curve corresponding to each flushing path from the initial response data. It identifies the maximum pressure peak and its corresponding peak time point in the input pressure disturbance waveform and the maximum liquid level peak and its corresponding peak time point in the output liquid level response curve through a peak detection algorithm. It calculates the difference between the peak time point of the output liquid level response curve and the peak time point of the input pressure disturbance waveform as the time difference, and determines the phase delay time of the corresponding flushing path and stores it.

[0038] S204: Based on the rising slope and peak amplitude of the output liquid level response curve in the initial response data, and the disturbance amplitude of the input pressure disturbance waveform, calculate the pressure amplitude attenuation ratio of each flushing path.

[0039] In this embodiment, the pressure amplitude attenuation ratio is used to represent the degree of attenuation of pressure energy converted into liquid level change energy during the propagation of the initial pulse injection from the injection port to the target flushing area; the rise slope represents the rate of rise of the output liquid level response curve from the starting point to the peak liquid level; the peak amplitude represents the maximum liquid level change in the output liquid level response curve; and the disturbance amplitude represents the maximum pressure change in the input pressure disturbance waveform.

[0040] Specifically, the electronic device extracts the output liquid level response curve and input pressure disturbance waveform corresponding to each flushing path from the initial response data. It calculates the rate of liquid level change from the starting point to the peak point of the output liquid level response curve using a differential algorithm as the rising slope, identifies the difference between the maximum liquid level value and the reference liquid level value in the output liquid level response curve as the amplitude peak value, and identifies the difference between the maximum pressure value and the reference pressure value in the input pressure disturbance waveform as the disturbance amplitude. The product of the amplitude peak value and the rising slope is divided by the disturbance amplitude to obtain the quotient value, which is then determined as the pressure amplitude attenuation ratio of the corresponding flushing path and stored.

[0041] S205: The phase delay time and amplitude attenuation ratio of each flushing path are integrated into a two-dimensional response spectrum, and the two-dimensional response spectrum is used as the time-frequency response baseline.

[0042] In this embodiment, the two-dimensional response map is used to represent the distribution of each flushing path in the two-dimensional coordinate space composed of the phase delay time dimension and the pressure amplitude attenuation ratio dimension, and the time-frequency response baseline represents the standard response characteristic reference data of each flushing path of the flushing system under unblocked conditions.

[0043] Specifically, the electronic equipment acquires the phase delay time and pressure attenuation ratio corresponding to each flushing path. A two-dimensional coordinate system is established with the phase delay time as the horizontal axis and the pressure attenuation ratio as the vertical axis. The phase delay time value and pressure attenuation ratio value of each flushing path are plotted in the two-dimensional coordinate system to form a scatter plot. Interpolation fitting is performed on each coordinate point in the scatter plot to generate a continuous surface or contour distribution to form a two-dimensional response spectrum. The generated two-dimensional response spectrum is marked as the response characteristics of the initial unblocked state and stored as a time-frequency response baseline.

[0044] S102: Calculate the impedance characteristics of each flushing path based on the time-frequency response baseline, and generate an impedance difference map reflecting the path resistance distribution based on the impedance characteristics of each flushing path.

[0045] In this embodiment, impedance characteristics are used to represent the comprehensive resistance characteristics of the flushing path to the liquid flow, and impedance difference diagrams are visualizations of the resistance distribution differences between each flushing path. The path distribution states of different impedance levels are distinguished by color depth or contour lines.

[0046] Specifically, the electronic device extracts the phase delay time and amplitude attenuation ratio corresponding to each flushing path from the time-frequency response baseline. The phase delay time is multiplied by the reciprocal of the amplitude attenuation ratio to obtain the product value. After normalization, the product value is used as the impedance feature of the corresponding flushing path and stored. A two-dimensional spatial coordinate system is established with the spatial position of the flushing path as the horizontal and vertical axes. The impedance feature value of each flushing path is mapped to the corresponding spatial coordinate point. The impedance feature value is converted into color depth markers at the coordinate point position through a color level mapping algorithm to form an impedance difference map.

[0047] Based on the above embodiments, as an optional embodiment, the step of calculating the impedance characteristics of each flushing path based on the time-frequency response baseline and generating an impedance difference map reflecting the path resistance distribution based on the impedance characteristics of each flushing path includes steps S301 to S305: S301: Obtain the path length parameters of each flushing path, and based on the path length parameters, convert the phase delay time in the time-frequency response baseline into a phase delay ratio per unit length.

[0048] In this embodiment, the path length parameter is used to represent the actual physical distance from the injection port to the target flushing area for each flushing path, and the phase delay ratio represents the amount of phase delay time generated per unit length path. The influence of path length difference on phase delay time is eliminated by normalization processing, thereby realizing the impedance characteristic comparison between paths of different lengths.

[0049] Specifically, the electronic equipment reads the path length parameters corresponding to each flushing path from the flushing system configuration database, extracts the phase delay time corresponding to each flushing path from the time-frequency response baseline, divides the phase delay time of each flushing path by the corresponding path length parameter to obtain the quotient, determines the quotient as the phase delay ratio per unit length of the corresponding flushing path and stores it in the normalized response dataset.

[0050] S302: Extract the pulse duration segment of the input pressure disturbance waveform, calculate the input energy integral value of the input pressure disturbance waveform within the pulse duration segment, and the output energy integral value of the output liquid level response curve within the pulse duration segment.

[0051] In this embodiment, the pulse duration segment represents the complete time interval from the start to the end of the initial pulse injection acting on the flushing system. The input energy integral value is used to represent the total energy applied to the flushing system by the input pressure disturbance waveform within the pulse duration segment. The output energy integral value is used to represent the total energy of the liquid level change reflected by the output liquid level response curve within the pulse duration segment. By comparing the input energy integral value and the output energy integral value, the energy transfer efficiency of the flushing path can be evaluated.

[0052] Specifically, the electronic device identifies the start and end points of the pulse signal from the input pressure disturbance waveform and determines the time interval between the start and end points as the pulse duration segment. It performs time integration on the pressure value of the input pressure disturbance waveform within the pulse duration segment to obtain the input energy integral value and stores it. It also performs time integration on the liquid level response curve within the pulse duration segment to obtain the output energy integral value and stores it.

[0053] S303: Based on the ratio of the output energy integral value to the input energy integral value, and combined with the path length parameter, calculate the pressure attenuation gradient per unit length.

[0054] In this embodiment, the pressure attenuation gradient per unit length represents the degree of energy transfer attenuation per unit length of pipe in the flushing path. The overall energy transfer efficiency of the flushing path is reflected by the ratio of the output energy integral value to the input energy integral value. By combining the path length parameter for normalization, the influence of different path length differences on energy attenuation assessment can be eliminated, thereby realizing the horizontal comparison of pressure attenuation characteristics between flushing paths of different lengths.

[0055] Specifically, the electronic device divides the output energy integral value by the input energy integral value to obtain the energy transfer efficiency ratio, subtracts a value from the energy transfer efficiency ratio to obtain the energy attenuation ratio, and divides the energy attenuation ratio by the path length parameter to obtain the pressure amplitude attenuation gradient per unit length and stores it in the normalized response dataset.

[0056] S304: Combine the phase hysteresis ratio of each flushing path with the pressure attenuation gradient to form a two-dimensional eigenvector, and use the two-dimensional eigenvector as the impedance characteristic of each flushing path.

[0057] In this embodiment, the two-dimensional feature vector is used to represent the comprehensive impedance performance of the flushing path in both the time domain and energy domain dimensions. By using the phase hysteresis ratio as the time domain component to reflect the time delay characteristics of liquid flow, and the pressure amplitude attenuation gradient as the energy domain component to reflect the attenuation characteristics of energy transfer, the combined two-dimensional feature vector can comprehensively characterize the impedance characteristics of the flushing path and provide a quantitative basis for subsequent blockage determination.

[0058] Specifically, the electronic device reads the phase hysteresis ratio and pressure attenuation gradient corresponding to each flushing path from the normalized response dataset. It combines the phase hysteresis ratio as the first dimension component of the two-dimensional feature vector and the pressure attenuation gradient as the second dimension component of the two-dimensional feature vector to obtain a two-dimensional feature vector. The two-dimensional feature vector is then determined as the impedance feature of the corresponding flushing path and stored in the impedance feature dataset.

[0059] S305: Based on the impedance characteristics of each flushing path, establish an impedance difference matrix and render the impedance difference matrix as a two-dimensional heat map to generate an impedance difference map.

[0060] In this embodiment, the impedance difference matrix is ​​used to represent the quantification result of the difference in impedance characteristics between each flushing path. Each element in the matrix represents the distance value between the impedance characteristics of two flushing paths. By rendering the impedance difference matrix into a two-dimensional heat map, the numerical difference can be converted into color depth changes. The generated impedance difference map can intuitively present the impedance difference distribution pattern between each flushing path and quickly identify abnormal paths.

[0061] Specifically, the electronic device reads the impedance characteristics of all flushing paths from the impedance characteristic dataset, calculates the Euclidean distance between the impedance characteristics of any two flushing paths and fills the Euclidean distance into the corresponding position in the impedance difference matrix, maps the values ​​in the impedance difference matrix to a preset color mapping table to obtain the corresponding color values ​​and fills the color values ​​into the corresponding pixel positions of the two-dimensional heat map, and determines the rendered two-dimensional heat map as the impedance difference map and stores it in the visualization result set.

[0062] S103: Determine the target flushing path from multiple flushing paths based on the impedance difference diagram, obtain the velocity and pressure data of the target flushing path under disturbed injection, and construct the velocity and pressure coupling distribution map of the target flushing area based on the velocity and pressure data.

[0063] In this embodiment, the target flushing path refers to the flushing path with the most significant impedance anomaly selected from multiple flushing paths. Abnormal paths whose impedance characteristics deviate from the normal range are identified by analyzing the distribution pattern of differences between each flushing path and other paths in the impedance difference diagram, and are thus designated as key monitoring targets. Disturbance injection refers to the application of a pulsating liquid injection signal containing multi-frequency components to the target flushing path to excite the dynamic fluid response characteristics inside the pipeline. The velocity-pressure coupling distribution map is used to represent the spatial correlation between the velocity field and pressure field within the target flushing area. By superimposing velocity and pressure data in the same coordinate system, the spatial distribution law of fluid dynamic characteristics is presented.

[0064] Specifically, the electronic device reads the impedance difference map from the visualization result set and calculates the mean value of each row element in the impedance difference matrix to obtain the average difference degree of each flushing path. The flushing path with the largest average difference degree is selected as the target flushing path and the corresponding path identifier is marked. The disturbance injection device is controlled to apply a pulsating injection signal containing a preset frequency range to the target flushing path. The instantaneous flow velocity value of each monitoring point is collected by the velocity sensor array distributed along the target flushing path to form velocity data, and the instantaneous pressure value of each monitoring point is collected by the pressure sensor array to form pressure data. The velocity data and pressure data are registered according to the spatial coordinates of the monitoring points and mapped to the same two-dimensional grid to obtain the velocity field distribution and pressure field distribution. The velocity field distribution is superimposed on the contour map of the pressure field distribution in the form of vector arrows to generate a velocity and pressure coupling distribution map and stored in the flow field analysis dataset.

[0065] Based on the above embodiments, as an optional embodiment, the step of determining the target flushing path from multiple flushing paths according to the impedance difference diagram, obtaining the velocity and pressure data of the target flushing path under disturbed injection, and constructing a velocity and pressure coupling distribution map of the target flushing area based on the velocity and pressure data includes steps S401 to S404: S401: Select the flushing path with an impedance difference greater than a preset threshold from the impedance difference diagram as the target flushing path.

[0066] In this embodiment, the impedance difference is used as a quantitative indicator to represent the degree of difference in impedance characteristics between a certain flushing path and all other flushing paths. The statistical values ​​of the corresponding rows or columns in the impedance difference matrix reflect the degree to which the impedance characteristics of that path deviate from the overall level. The larger the impedance difference, the more severe the deviation of the path's impedance characteristics from the normal path. The preset threshold is a critical value used to determine the impedance abnormality of the flushing path. It is determined based on statistical analysis of historical normal operation data, serving as a boundary point that effectively distinguishes between normal and abnormal paths.

[0067] Specifically, the electronic device reads the impedance difference matrix corresponding to the impedance difference map from the visualization result set, traverses each row of the impedance difference matrix, calculates the arithmetic mean of the elements in each row to obtain the impedance difference value corresponding to each flushing path, reads the preset threshold from the configuration file and compares the impedance difference value of each flushing path with the preset threshold one by one, filters out the flushing paths with impedance differences greater than the preset threshold and extracts the corresponding path identifiers, determines the filtered flushing paths as target flushing paths and stores the path identifiers in the target path list.

[0068] S402: Output control commands to instruct the target flushing path to perform asynchronous directional disturbance injection, and acquire velocity and pressure data of the target flushing area under disturbance injection.

[0069] In this embodiment, asynchronous directional disturbance injection refers to a pulsed liquid injection method with time-difference and direction-specific properties applied to the target flushing path. Disturbance signals are applied at different time points from different injection ports according to preset phase differences and flow direction control parameters to specifically stimulate fluid response characteristics in a particular direction. Velocity data is used to represent the velocity change sequence at each monitoring point within the target flushing area during the disturbance injection process, reflecting the spatiotemporal distribution information of the fluid motion state; pressure data is used to represent the pressure change sequence at each monitoring point, reflecting the spatiotemporal distribution information of the fluid dynamics characteristics.

[0070] Specifically, the electronic device reads the path identifier of the target flushing path from the target path list, generates a disturbance injection control command containing the injection start time and phase difference, and sends it to the disturbance injection device through the control signal interface to trigger asynchronous directional disturbance injection. During this process, injection is first initiated in the target flushing path with lower impedance characteristics, and after a delay of 150 milliseconds, injection is initiated in the target flushing path with higher impedance characteristics. The injection pressure is controlled at 12 to 16 kPa, the instantaneous injection volume is controlled at 4 to 6 ml per injection, and the total injection cycle does not exceed 1 second. During this period, the electronic device continuously receives raw flow field data collected and transmitted in real time by an external detection device through a data interface. The external detection device is preferably a phased array vector Doppler probe fixed to the outer wall of the target flushing area, with 128 array elements, a center frequency of 10 MHz, a frequency bandwidth of 4 MHz, a horizontal scanning area width of 20 mm, a vertical depth of 10 mm, and a frame rate of not less than 30 frames per second. After receiving the raw flow field data, the electronic device can calculate the instantaneous flow velocity value of each spatial point in the target flushing area based on the Doppler frequency shift data to assemble velocity data, and estimate the instantaneous pressure value of each spatial point based on the reflection phase time difference data or in combination with the readings of other pressure sensors to assemble pressure data. Finally, the generated velocity data and pressure data are associated with the path identifier of the target flushing path and stored in the flow field measurement dataset.

[0071] S403: Spatial interpolation fitting is performed on the velocity data to generate a velocity vector layer, and a pressure color level layer is constructed based on the pressure data.

[0072] In this embodiment, spatial interpolation fitting refers to the computational process of using mathematical interpolation algorithms to calculate the velocity values ​​of unmonitored locations based on the velocity data of discrete monitoring points, thereby generating a continuous velocity field distribution. Methods such as Kriging interpolation or radial basis functions are used to extend the velocity information of finite monitoring points to a spatial grid covering the entire target flushing area. A velocity vector layer is used to represent the direction and magnitude distribution of the velocity field within the target flushing area. The velocity values ​​obtained through interpolation fitting are visualized as vector arrows, showcasing the directional and intensity characteristics of fluid motion. A pressure color gradient layer is used to represent the numerical distribution of the pressure field within the target flushing area. By mapping pressure data to a preset color gradient scheme, contour maps or filled maps with varying color depths are formed to reflect the spatial distribution of pressure.

[0073] Specifically, the electronic device reads velocity data from the flow field measurement dataset and extracts the spatial coordinates and corresponding velocity components of each monitoring point. Based on the spatial coordinates, it constructs a two-dimensional grid covering the target flushing area and uses the grid nodes as interpolation target points. The Kriging interpolation algorithm is used to calculate the spatial correlation weight between each grid node and the known monitoring point. Based on the spatial correlation weight, the velocity components of each monitoring point are weighted and averaged to obtain the interpolated velocity value of each grid node. The interpolated velocity value is converted into a vector form containing direction angle and velocity magnitude to generate a velocity vector layer and stored in the layer dataset. Pressure data is read from the flow field measurement dataset and the pressure value of each monitoring point is extracted. The pressure value is converted into the corresponding color code according to the preset color mapping rule. Based on the color code, pressure contour lines or filled color areas are drawn on the two-dimensional grid to generate a pressure color level layer and stored in the layer dataset.

[0074] S404: Overlay the velocity vector layer onto the pressure level layer to generate a time-series velocity-pressure coupled distribution map.

[0075] In this embodiment, overlaying a velocity vector layer onto a pressure color level layer refers to a graphic processing procedure that spatially aligns and combines a velocity vector layer representing the direction and magnitude of fluid motion with a pressure color level layer representing pressure distribution. This layer overlay technique simultaneously presents the spatial distribution characteristics of the velocity and pressure fields in the same visualization view, achieving a collaborative display of multiple physics fields. The time-series velocity-pressure coupling distribution map is used to represent the coupling state of the velocity and pressure fields at different moments during the disturbance injection process in the target flushing area. Multiple frames of coupling distribution maps arranged in chronological order constitute a dynamic evolution sequence reflecting the transient characteristics of the flow field changing over time.

[0076] Specifically, the electronic device reads the velocity vector layer and pressure color level layer from the layer dataset and extracts their spatial coordinate system parameters for coordinate system uniform calibration. Based on the time span of the perturbation injection, the sampling interval of the time series is determined and the entire time span is divided into several time segments. For each time segment, the velocity vector sub-layer corresponding to the time moment is extracted from the velocity vector layer and the pressure color level sub-layer corresponding to the time moment is extracted from the pressure color level layer. The pressure color level sub-layer is used as the bottom background and the velocity vector sub-layer is used as the overlay layer. The layers are synthesized using a transparency fusion algorithm to obtain a single-frame coupled distribution map. All single-frame coupled distribution maps are assembled into a time series velocity and pressure coupled distribution map in chronological order and timestamp is added. The generated time series velocity and pressure coupled distribution map is stored in the coupled visualization result set.

[0077] S104: Spatiotemporal matching of velocity and pressure coupling distribution map and impedance difference map is performed to obtain spatiotemporal image sequence, and the swirling core region in the target flushing area is identified based on the spatiotemporal image sequence.

[0078] In this embodiment, spatiotemporal matching refers to the data fusion process of aligning and associating the velocity and pressure coupling distribution map representing the dynamic characteristics of the flow field with the impedance difference map representing the structural characteristics of the pipeline in both time and space dimensions. This is achieved through timestamp alignment and spatial coordinate registration to unify the representation of multi-source heterogeneous image data. The spatiotemporal image sequence represents a multi-dimensional image set that integrates flow field dynamics and structural impedance information. By arranging the velocity and pressure coupling distribution maps and their corresponding impedance difference maps in chronological order and spatially superimposing them, a comprehensive visualization sequence containing both temporal evolution and spatial distribution characteristics is formed. The swirling core region represents a local spatial region within the target flushing area where the fluid undergoes continuous rotational motion and a significant pressure gradient. This region typically corresponds to high-impedance nodes in the pipeline structure, which are key locations where contaminants easily accumulate and the flushing effect is poor.

[0079] Specifically, the electronic device reads the time-series velocity-pressure coupling distribution map from the coupling visualization result set and extracts the timestamp information of each frame image. It reads the impedance difference map from the impedance feature dataset and extracts the spatial coordinate system parameters of the image. Based on the timestamp information, it aligns each frame image of the velocity-pressure coupling distribution map with the corresponding pipeline state at that time. Based on the spatial coordinate system parameters, it performs spatial registration between the velocity-pressure coupling distribution map and the impedance difference map to achieve pixel-level positional correspondence. The registered velocity-pressure coupling distribution map and impedance difference map are then combined into a single-frame spatiotemporal image using a multi-channel fusion method. All single-frame spatiotemporal images are arranged and assembled into a spatiotemporal image sequence according to time order and stored in the spatiotemporal fusion dataset. The velocity of each frame image is then extracted from the spatiotemporal image sequence. The curl field of the vector field and the velocity vector are used to characterize the fluid rotation intensity. Regions in the curl field whose curl values ​​exceed a preset rotation threshold are identified as candidate cyclotron regions. The pressure gradient field of each frame image is extracted from the spatiotemporal image sequence and the magnitude of the pressure gradient is calculated to characterize the severity of pressure changes. Regions whose pressure gradient magnitude exceeds a preset gradient threshold are identified as candidate high-pressure gradient regions. The spatial intersection of the candidate cyclotron regions and the candidate high-pressure gradient regions is performed to obtain regions that simultaneously satisfy the conditions of high curl and high pressure gradient. The spatial proximity of this intersection region to the high-impedance node position in the impedance difference map is analyzed. The intersection region located in the vicinity of the high-impedance node and continuously appearing in the time series is identified as the cyclotron core region and its spatial coordinates and temporal persistence characteristics are marked.

[0080] Based on the above embodiments, as an optional embodiment, the step of spatiotemporally matching the velocity-pressure coupling distribution map and the impedance difference map to obtain a spatiotemporal image sequence, and identifying the swirling core region within the target flushing area based on the spatiotemporal image sequence, includes steps S501 to S504: S501: Using the liquid outlet coordinates of the flushing path as the reference anchor point, spatial mapping and time frame sequence alignment are performed on the velocity and pressure coupling distribution map and impedance difference map to generate a spatiotemporal image sequence with a unified coordinate system.

[0081] In this embodiment, using the injection outlet coordinates as the reference anchor point means using the spatial coordinates of the pipe interface position where the disturbed injection fluid enters the target flushing area in the flushing path as a fixed reference point for spatial registration of multi-source image data. This unifies the spatial reference by aligning the injection outlet positions in images from different sources to the same coordinate position. Spatial mapping refers to establishing a mathematical transformation relationship between the image coordinate system of the velocity-pressure coupling distribution map and the image coordinate system of the impedance difference map. A coordinate transformation matrix is ​​used to convert the pixel coordinates of the two images to a unified physical spatial coordinate system, thereby achieving precise spatial correspondence. Time frame sequence alignment refers to matching and associating the temporal dimension based on the timestamp information of each frame of the velocity-pressure coupling distribution map and the corresponding pipe state time in the impedance difference map, ensuring that each frame in the fused spatiotemporal image sequence reflects the flow field state and structural state at the same moment. The spatiotemporal image sequence with a unified coordinate system is used to represent the multidimensional image set formed after spatial mapping and time alignment processing. All images in this sequence have the same spatial coordinate origin, the same pixel resolution, the same coordinate axis direction, and precisely corresponding timestamp markers.

[0082] Specifically, the electronic device reads flushing path information from the flushing path planning dataset and extracts the physical spatial coordinates of the injection outlet as the reference anchor point coordinates. It reads the velocity-pressure coupling distribution map from the coupling visualization result set and identifies the pixel coordinates corresponding to the injection outlet position in the image as the first image anchor point coordinates. It reads the impedance difference map from the impedance feature dataset and identifies the pixel coordinates corresponding to the injection outlet position in the image as the second image anchor point coordinates. It calculates the coordinate offset between the first image anchor point coordinates and the reference anchor point coordinates, as well as the coordinate offset between the second image anchor point coordinates and the reference anchor point coordinates. Based on the coordinate offsets, it constructs the spatial transformation matrix for the velocity-pressure coupling distribution map and the spatial transformation matrix for the impedance difference map, respectively. It then applies the spatial transformation matrix to perform an affine transformation on the velocity-pressure coupling distribution map and the impedance difference map, making the injection outlet positions in the two images more aligned. By establishing a unified reference anchor point coordinate system, the transformed image is resampled and interpolated to unify the image resolution and pixel size. Ideally, bilinear interpolation is used to unify the image size to 800×600 pixels, with a spatial resolution controlled at 0.1 mm per pixel. Timestamp information for each frame is extracted from the time series of the velocity-pressure coupling distribution map. Based on the timestamp information, the impedance state image at the corresponding moment is retrieved from the temporal evolution data of the impedance difference map. The transformed velocity-pressure coupling distribution map and the corresponding impedance state image at the same moment are fused using multi-channel technology to obtain a single-frame spatiotemporal image. The metadata of the single-frame spatiotemporal image records the unified coordinate origin coordinates, coordinate axis direction vectors, pixel resolution parameters, and timestamp markers. All single-frame spatiotemporal images are assembled into a spatiotemporal image sequence with a unified coordinate system and stored in the spatiotemporal fusion dataset in chronological order.

[0083] S502: Track the movement path of fluid particles in a spatiotemporal image sequence to extract the liquid velocity trajectory, and calculate the trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude of the liquid velocity trajectory.

[0084] In this embodiment, a fluid particle refers to a virtual tracer particle used to characterize the motion characteristics of a fluid in a flow field. This virtual particle moves with the fluid, and its position change trajectory over time reflects the actual flow path of the fluid. A liquid velocity trajectory is a spatial curve formed by tracking the continuous position changes of fluid particles in a spatiotemporal image sequence. This curve contains the coordinate position information of the fluid particles at different times and the corresponding velocity vector information to characterize the fluid's motion history. Trajectory curvature is used to quantify the degree to which the liquid velocity trajectory deviates from a straight path. It is characterized by the ratio of the actual trajectory length to the straight-line distance between the start and end points; a larger value indicates a more tortuous fluid path. The number of direction changes is used to count the frequency of significant changes in the velocity vector direction in the liquid velocity trajectory. It is characterized by the number of events where the angle between velocity vectors at adjacent times exceeds a preset angle threshold to characterize the directional stability of the fluid motion. Path closure measures the degree to which the liquid velocity trajectory forms a closed loop. It is characterized by the ratio of the distance between the trajectory's end point and its starting point to the total length of the trajectory; a smaller value indicates a higher degree of closure. Velocity fluctuation amplitude is used to characterize the degree of drastic change in the magnitude of velocity of a fluid particle as it moves along a trajectory. The instability of velocity is quantified by calculating the ratio of the standard deviation of the velocity values ​​on the trajectory to the average velocity.

[0085] Specifically, the electronic device reads a spatiotemporal image sequence with a unified coordinate system from the spatiotemporal fusion dataset and extracts the velocity vector field data of each frame. In the first frame velocity vector image of the spatiotemporal image sequence, several pixel coordinate points near the flushing outlet are selected as initial fluid particles for tracking the liquid motion trajectory. An image particle tracking algorithm based on velocity direction and angle recursion is used to map and calculate the pixel position of the fluid particle in the next frame image according to the velocity vector direction and magnitude at the current frame pixel. The movement path of each fluid particle in the spatiotemporal image sequence (e.g., at least 30 frames) is continuously tracked to obtain the position sequence of the fluid particles throughout the entire time series. The position sequences of the fluid particles are connected into a spatial curve, and the velocity vector information corresponding to each position is recorded to form the liquid velocity trajectory. The actual velocity trajectory of the liquid is then calculated. The arc length is obtained by accumulating the Euclidean distance between adjacent points. The straight-line distance between the start and end points of the liquid velocity trajectory is calculated. The trajectory curvature is obtained by dividing the total trajectory length by the straight-line distance between the start and end points. The velocity vectors at adjacent moments on the liquid velocity trajectory are traversed and the angle between adjacent velocity vectors is calculated. The number of times the angle exceeds a preset angle threshold is counted to obtain the number of direction changes. The distance between the end point and the start point of the liquid velocity trajectory is calculated. The path closure is obtained by dividing the end point and the start point distance by the total trajectory length. The velocity magnitude values ​​of each point on the liquid velocity trajectory are extracted. The average and standard deviation of the velocity magnitude values ​​are calculated. The velocity fluctuation amplitude is obtained by dividing the standard deviation by the average value. The calculated trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude are stored as feature parameters of the liquid velocity trajectory in the trajectory feature dataset.

[0086] S503: Integrate trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude to generate swirling flow indices, and statistically analyze the residence time distribution of fluid particles in local areas.

[0087] In this embodiment, the swirling core region refers to the set of spatial locations within the target rinsing area that simultaneously meet the conditions of high swirling flow intensity and long residence time. These locations form cleaning dead zones because the fluid forms a stable vortex motion, making it difficult for the fluid to flow out effectively. By extracting the spatial overlap between the high swirling flow region where the swirling flow index exceeds a preset index threshold and the long residence region where the residence time in the residence time distribution map exceeds a preset time threshold, the key locations where the rinsing effect needs to be optimized are precisely located.

[0088] Specifically, the electronic device reads the trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude of each liquid velocity trajectory from the trajectory feature dataset. After standardizing these parameters, they are weighted and integrated according to preset weights to calculate the swirling flow index corresponding to each liquid velocity trajectory, which is used to quantify the intensity of swirling behavior in the region where the trajectory is located. At the same time, a local evaluation region with a preset radius is defined with the starting point of each liquid velocity trajectory as the center. The total time that fluid particles stay in this local evaluation region in consecutive image frames is counted as the continuous residence time. If the continuous residence time exceeds 1.0 second, it is marked as significant residence. The continuous residence time of all liquid velocity trajectories is traversed, and the target flushing area is divided into multiple spatial grid cells. The average continuous residence time in each grid cell is calculated and mapped to different color levels according to the duration, thereby constructing a residence time distribution map reflecting the degree of local fluid accumulation. The swirling flow index and the residence time distribution map are stored in the swirling feature dataset.

[0089] S504: Extract overlapping areas where the swirling flow index is greater than the preset index threshold and the continuous residence time in the residence time distribution map is greater than the preset time threshold, and mark the overlapping areas as the swirling core area within the target flushing area.

[0090] In this embodiment, the swirling core region refers to the set of spatial locations within the target rinsing area that simultaneously meet the conditions of high swirling flow intensity and long residence time. These locations form cleaning dead zones because the fluid forms a stable vortex motion, making it difficult for the fluid to flow out effectively. By extracting the spatial overlap between the high swirling flow region where the swirling flow index exceeds a preset index threshold and the long residence region where the residence time in the residence time distribution map exceeds a preset time threshold, the key locations where the rinsing effect needs to be optimized are precisely located.

[0091] Specifically, the electronic device reads the swirling flow index values ​​and spatial coordinates corresponding to each liquid velocity trajectory from the swirling feature dataset. It compares the swirling flow index values ​​with preset threshold values ​​one by one, filtering out a set of liquid velocity trajectories with swirling flow indices greater than the preset threshold. Trajectories in this set must simultaneously meet the following requirements: closure less than 0.5 mm, number of directional changes no less than 3, velocity fluctuation amplitude greater than 2.5 mm / s, and normalized swirling flow index exceeding 0.7. The spatial coordinates of each trajectory in the filtered liquid velocity trajectory set are extracted, and the corresponding grid cell is marked on the spatial grid of the target flushing area to form a high swirling flow region. Simultaneously, the average position of each grid cell in the residence time distribution map is read. The average residence time of each grid cell is compared with a preset time threshold one by one, and the grid cell sets with an average residence time greater than the preset time threshold are selected to form long residence regions. Spatial cross-matching is performed on the grid cell sets of high swirling flow regions and the grid cell sets of long residence regions, and grid cells that appear in both sets are extracted as overlapping regions. The extracted overlapping regions are marked as swirling core regions within the target flushing area and assigned a unique region identifier. The number of grid cells contained in the swirling core region is counted and the spatial area of ​​the region is calculated. The spatial coordinate range, region identifier, and spatial area information of the marked swirling core region are stored in the swirling core dataset.

[0092] S105: Generate a secondary injection command for the cyclone core region, and acquire secondary response characteristic data of the cyclone core region in response to the secondary injection command. Then, perform image fusion of the secondary response characteristic data and the impedance difference map to generate a risk heat map.

[0093] In this embodiment, the risk heatmap refers to a visual image formed by superimposing secondary response feature data and impedance difference map through image fusion technology. It is used to represent the distribution of cleaning risk level at different locations within the target flushing area. The color depth or color temperature represents the difference in risk level. By performing image fusion between the secondary response feature data collected after applying a secondary injection command to the swirling core area and the impedance difference map, the coupling relationship between fluid dynamics characteristics and electrochemical impedance characteristics can be comprehensively reflected, thereby accurately identifying high-risk areas with poor cleaning effect.

[0094] Specifically, the electronic device reads the spatial coordinate range and region identifier information of the cyclone core region from the cyclone core dataset, calls the injection control module and passes in the spatial coordinate range parameters of the cyclone core region to generate a secondary injection command for the cyclone core region. The secondary injection command includes the injection start time, injection duration, injection flow rate, and injection direction angle information. The generated secondary injection command is sent to the injection execution unit to start the secondary injection operation on the cyclone core region. During the execution of the secondary injection operation, the sensor array is invoked to monitor the fluid velocity changes, pressure fluctuations, and temperature distribution information in the cyclone core region in real time. The monitored fluid velocity changes, pressure fluctuations, and temperature distribution information are then processed according to... The time series data is sampled and stored as secondary response feature data. The impedance difference map corresponding to the spatial coordinate range of the cyclone core region is read from the impedance difference dataset. The fluid velocity change information in the secondary response feature data is converted into a velocity field distribution map. The velocity field distribution map and the impedance difference map are aligned in spatial coordinate system. The aligned velocity field distribution map and the impedance difference map are weighted and fused according to pixel position to obtain a fused image. The values ​​of each pixel in the fused image are normalized and mapped to a color space to form a risk heat map. The darker the color or the higher the color temperature in the risk heat map, the greater the cleaning risk. The generated risk heat map is stored in the risk assessment dataset in image file format.

[0095] Based on the above embodiments, as an optional embodiment, a secondary injection command is generated for the cyclotron core region, and secondary response characteristic data of the cyclotron core region in response to the secondary injection command is obtained. The secondary response characteristic data is then image-fused with the impedance difference map to generate a risk heat map. This step includes steps S601 to S604: S601: Generate a phase-coded micro-amplitude pulse injection command for the cyclotron core region as a secondary injection command.

[0096] In this embodiment, the phase-coded micro-amplitude pulse injection command is a special injection control command designed for the vortex core region. It achieves fine control of the fluid injection process by decomposing the injection action into multiple micro-amplitude pulse injection sequences and assigning different phase coding information to each pulse sequence. The phase coding is used to represent the time interval and relative timing relationship between each pulse injection action, and the micro-amplitude pulse is used to represent the characteristic that the flow rate amplitude of a single injection action is much smaller than the conventional injection flow rate. The phase-coded micro-amplitude pulse injection method can effectively break the stable vortex structure in the vortex core region while avoiding severe flow field disturbances caused by large flow rate injection.

[0097] Specifically, the electronic device reads the spatial coordinate range, region identifier, and spatial area information of the cyclone core region from the cyclone core dataset. It compares the spatial area of ​​the cyclone core region with a preset area threshold to calculate an area ratio coefficient. The area ratio coefficient is multiplied by a preset reference pulse count to obtain the total number of pulse sequences for the current cyclone core region. This total number of pulse sequences is then evenly distributed across a preset injection cycle to calculate the time interval between each pulse. Each pulse in the total pulse sequence is assigned an incremental phase code value to form a phase code sequence. Finally, the preset reference flow rate is divided by the area ratio coefficient. The micro-amplitude flow rate value of a single pulse is obtained. The total number of pulse sequences, time intervals, phase-coded sequences, and micro-amplitude flow rates are combined to generate a phase-coded micro-amplitude pulse injection command. The phase-coded micro-amplitude pulse injection command includes the total number of pulses, the trigger time of each pulse, the phase-coded value, the injection flow rate, and the injection direction angle information. In this command, the injection frequency is controlled at 1 Hz, the pulse injection pressure is set to 1 kPa, the pulse duration does not exceed 0.4 seconds, and the injection cycle is set to start 5 seconds after the previous injection. The generated phase-coded micro-amplitude pulse injection command is marked as a secondary injection command and stored in the injection command dataset.

[0098] S602: Obtain the delayed response time, response amplitude rise rate, and fluid disturbance radius of the cyclone core region under the secondary injection command, as secondary response characteristic data.

[0099] In this embodiment, the secondary response characteristic data refers to the set of dynamic response characteristic parameters exhibited by the cyclone core region after executing the phase-coded micro-amplitude pulse injection command. These parameters include the delayed response time, which represents the length of time from the issuance of the secondary injection command to the start of a significant change in fluid velocity within the cyclone core region; the response amplitude rise rate, which represents the rate of change of fluid velocity within the cyclone core region from its initial value to its peak value; and the fluid disturbance radius, which represents the spatial diffusion range of the secondary injection action's influence on the flow field surrounding the cyclone core region. These three characteristic parameters comprehensively reflect the sensitivity of the cyclone core region to external injection disturbances and its flow field response characteristics.

[0100] Specifically, the electronic device reads the secondary injection command for the cyclone core region from the injection command dataset and extracts the pulse trigger time sequence information from the command. It then sends the secondary injection command to the injection execution unit to initiate the secondary injection operation on the cyclone core region. At the start of the secondary injection operation, it invokes the sensor array to begin real-time monitoring of fluid velocity changes at each monitoring point within the cyclone core region. The fluid velocity values ​​at each monitoring point are compared with a preset velocity benchmark value to calculate the velocity change. When the velocity change exceeds a preset threshold, the current moment is recorded as the response start moment. The time difference between the response start moment and the first pulse trigger moment of the secondary injection command is calculated to obtain the delayed response time. The system continuously monitors the fluid velocity changes at each monitoring point. The system records the time it takes for the velocity to rise from its initial value to its peak value, as well as the velocity change amplitude. The velocity change amplitude is divided by the rise time to obtain the response amplitude rise rate. The monitoring range is expanded to the area surrounding the swirling core region to continuously monitor the fluid velocity values ​​of each monitoring point in the surrounding area. The velocity change of each monitoring point in the surrounding area is compared with a preset disturbance threshold to select a set of monitoring points whose velocity change exceeds the disturbance threshold. The distance between the monitoring point farthest from the center of the swirling core region in the selected set and the center is calculated as the fluid disturbance radius. The delayed response time, response amplitude rise rate, and fluid disturbance radius are structured and encapsulated to form secondary response feature data and stored in the response feature dataset.

[0101] S603: Normalize the swirling flow index based on the secondary response characteristic data, and divide the risk level according to the preset multidimensional threshold model to generate a risk level map.

[0102] In this embodiment, normalization refers to mapping swirling flow indices with different dimensions and numerical ranges to a unified numerical range through mathematical transformation, thereby eliminating the incomparability between indices caused by differences in dimensions. The multidimensional threshold model refers to a set of judgment rules that integrate multiple dimensions such as delayed response time, response amplitude rise rate, fluid disturbance radius, swirling intensity value, swirling radius and velocity gradient, and set different threshold ranges corresponding to low risk, medium risk and high risk levels. The risk level map refers to a visual risk distribution image formed by dividing the swirling core area according to spatial coordinate grids, assigning risk level labels to each sub-region unit and encoding them with color.

[0103] Specifically, the electronic device reads the delayed response time, response amplitude rise rate, and fluid disturbance radius information of the swirling core region from the response feature dataset; it reads the swirling intensity value, swirling radius, and velocity gradient information of the swirling core region from the swirling flow dataset; it divides the delayed response time by a preset maximum delayed response time to obtain a normalized delayed response value; it divides the response amplitude rise rate by a preset maximum response amplitude rise rate to obtain a normalized response amplitude value; it divides the fluid disturbance radius by a preset maximum fluid disturbance radius to obtain a normalized disturbance radius value; it divides the swirling intensity value by a preset maximum swirling intensity to obtain a normalized swirling intensity value; it divides the swirling radius by a preset maximum swirling radius to obtain a normalized swirling radius value; it divides the velocity gradient by a preset maximum velocity gradient to obtain a normalized velocity gradient value; and it reads the boundary parameters of the multidimensional threshold model, including low-risk, medium-risk, and high-risk threshold intervals, from the threshold model dataset. It then normalizes the delayed response value and response amplitude... The normalized values ​​of the swirling flow index, disturbance radius, swirling intensity, swirling radius, and velocity gradient are weighted and summed according to preset weighting coefficients to obtain a comprehensive risk index. The comprehensive risk index is then compared with low-risk, medium-risk, and high-risk threshold ranges to determine the risk level. In the determination, if the swirling flow index is greater than 0.7, the number of velocity direction reversals is not less than 3, the liquid residence time is not less than 1.5 seconds, and the response delay time is more than 20% higher than the initial injection time, it is classified as a high-risk level. If two or three of the above four indicators are met, it is classified as a medium-risk level; otherwise, it is classified as a low-risk level. The swirling core area is divided into multiple sub-regions according to a spatial coordinate grid. The normalization process and risk level determination process are repeated for each sub-region to obtain a risk level label for each sub-region. Low-risk units are assigned a green code, medium-risk units a yellow code, and high-risk units a red code to generate a risk level map, which is then stored in the risk assessment dataset.

[0104] S604: Spatial registration and dual-channel color mixing are performed between the risk level map and the impedance difference map to generate a risk heat map characterizing the coupling distribution of local drag anomalies and swirling flow.

[0105] In this embodiment, spatial registration refers to accurately matching the risk level map and the impedance difference map in spatial location by transforming the coordinate system and aligning the pixels, thereby ensuring that the risk information and impedance information at the same spatial location can correspond accurately. Dual-channel color mixing refers to assigning the color information of the risk level map to one color channel and the color information of the impedance difference map to another color channel, and generating a composite color expression that contains both types of information through a color overlay algorithm. The risk heatmap is a visual image formed by spatial registration and dual-channel color mixing that can intuitively show the coupling distribution relationship between the local resistance anomaly area and the swirling flow risk area.

[0106] Specifically, the electronic device reads the risk level map from the risk assessment dataset and extracts its spatial coordinate system information and pixel matrix data. It also reads the impedance difference map from the impedance analysis dataset and extracts its spatial coordinate system information and pixel matrix data. The spatial coordinate systems of the risk level map and impedance difference map are aligned at the origin and their axes are corrected to achieve coordinate system unification. The pixel matrices of the risk level map and impedance difference map are compared in size. If the two pixel matrices are inconsistent, the smaller pixel matrix is ​​interpolated and enlarged, or the larger pixel matrix is ​​downsampled and reduced to make the two pixel matrices the same size. For size matching, spatial registration is achieved by matching the pixel positions of two pixel matrices after unifying their size. The color information of the risk level map is extracted to the red and green channels, and the color information of the impedance difference map is extracted to the blue channel. The pixel values ​​of the red, green, and blue channels are weighted and summed according to a preset mixing weight coefficient to obtain the mixed RGB color value. The mixed RGB color value is then filled into the corresponding pixel position to generate a risk heat map. Color legends are added to the risk heat map to indicate the coupling relationship between the local resistance anomaly degree represented by different color areas and the swirling flow risk level. The generated risk heat map is stored in the coupling analysis dataset in image file format.

[0107] S106: Generate a liquid injection control strategy that staggers the flow of liquids across multiple flushing paths based on the risk heat map, and obtain the response hysteresis of each flushing path. Then, dynamically adjust the liquid injection control strategy based on the response hysteresis.

[0108] In this embodiment, the injection control strategy refers to a control scheme that determines the injection timing, injection flow rate, and injection pressure parameters of multiple flushing paths by analyzing the distribution of local resistance anomalies and swirling flow coupling in the risk heat map. The staggered advancement refers to starting the injection operation time of multiple flushing paths sequentially according to a preset time interval, rather than starting them simultaneously, so as to avoid fluid interference and pressure fluctuations caused by simultaneous injection of multiple paths. The response lag refers to the degree of time delay from the start of the injection operation to the time when the physical parameters such as fluid pressure, flow rate, or temperature in the flushing path reach the expected target value.

[0109] Specifically, the electronic device reads the risk heatmap from the coupling analysis dataset and extracts the coupling relationship information between the degree of local resistance anomaly represented by different colored areas in the risk heatmap and the risk level of swirling flow. It identifies the spatial distribution of high-risk high-impedance coupling areas, medium-risk normal-impedance coupling areas, and low-risk normal-impedance coupling areas in the risk heatmap. For high-risk high-impedance coupling areas, it assigns high-priority flushing paths and sets the injection flow rate and injection pressure to preset high values. For medium-risk normal-impedance coupling areas, it assigns medium-priority flushing paths and sets the injection flow rate and injection pressure to preset medium values. For low-risk normal-impedance coupling areas, it assigns low-priority flushing paths and sets the injection flow rate and injection pressure to preset low values. The injection initiation sequence is arranged according to the order of high-priority flushing paths, medium-priority flushing paths, and low-priority flushing paths. Preset time intervals are inserted between adjacent priority flushing paths to achieve staggered injection control strategy. The system collects actual fluid pressure, flow rate, and temperature change data after each flushing path's injection operation using sensors. The pressure response lag time is calculated by subtracting the actual fluid pressure change data from the target fluid pressure value; the flow rate response lag time is calculated by subtracting the actual flow rate from the target flow rate value; and the temperature response lag time is calculated by subtracting the actual temperature from the target temperature value. The pressure response lag time, flow rate response lag time, and temperature response lag time are then weighted and averaged using preset weighting coefficients to obtain the overall response lag. When the overall response lag is greater than a preset lag threshold, the injection start time for the corresponding flushing path in the injection control strategy is advanced by the overall response lag time. When the overall response lag is less than the preset advance threshold, the injection start time for the corresponding flushing path in the injection control strategy is delayed by the difference between the preset advance threshold and the overall response lag time. The dynamically adjusted injection control strategy is then stored in the control strategy dataset.

[0110] Based on the above embodiments, as an optional embodiment, the step of generating a staggered injection control strategy among multiple flushing paths according to the risk heat map, obtaining the response hysteresis of each flushing path, and dynamically adjusting the injection control strategy based on the response hysteresis includes steps S701 to S704: S701: Extract the comprehensive risk score of each flushing path based on the risk heat map, and establish an intervention priority ranking table based on the comprehensive risk score.

[0111] In this embodiment of the application, the intervention priority ranking table refers to a data table that arranges multiple flushing paths from high to low according to the comprehensive risk score and assigns a priority number to each flushing path.

[0112] Specifically, the electronic device reads the risk heatmap from the coupling analysis dataset and extracts the spatial location information of the coverage area of ​​each flushing path in the risk heatmap. For each flushing path coverage area, pixel color statistical analysis is performed to obtain the number of high-risk, high-impedance coupling pixels, medium-risk, normal-impedance coupling pixels, and low-risk, normal-impedance coupling pixels. The number of high-risk, high-impedance coupling pixels is multiplied by a preset high-risk weighting coefficient to obtain a high-risk contribution score; the number of medium-risk, normal-impedance coupling pixels is multiplied by a preset medium-risk weighting coefficient to obtain a medium-risk contribution score; and the number of low-risk, normal-impedance coupling pixels is multiplied by a preset low-risk weighting coefficient to obtain a low-risk contribution score. The high-risk contribution score is then multiplied by the preset low-risk weighting coefficient. The comprehensive risk score of the flushing path is calculated by summing the contribution scores of the high-risk, medium-risk, and low-risk paths. The comprehensive risk scores of all flushing paths are sorted in descending order. The flushing path with the highest comprehensive risk score is assigned intervention priority number one, the flushing path with the second highest comprehensive risk score is assigned intervention priority number two, and so on, to generate an intervention priority ranking table. The intervention priority ranking table is supplemented with fields for flushing path identifier, comprehensive risk score value, intervention priority number, and description of recommended intervention measures. The generated intervention priority ranking table is stored in the intervention strategy dataset in tabular file format.

[0113] S702: Based on the intervention priority ranking table, assign non-equal interval injection start time, injection frequency and injection pressure to each flushing path to generate an injection control strategy.

[0114] In this embodiment, non-equal interval means that the injection start time interval between adjacent flushing paths is dynamically calculated based on the difference in their respective comprehensive risk scores and intervention priority numbers, rather than using a fixed time interval. Injection frequency refers to the number of times an injection operation is performed on a certain flushing path within a preset flushing cycle. Injection pressure refers to the pressure value of the fluid applied to the flushing path during the injection operation.

[0115] Specifically, the electronic device reads the intervention priority ranking table from the intervention strategy dataset and extracts the flushing path identifier, comprehensive risk score, and intervention priority number for each flushing path. It sets the injection start time of the flushing path with intervention priority number one as the current system time as the start time of the injection control strategy. It calculates the difference in comprehensive risk scores between the flushing path with intervention priority number one and the flushing path with intervention priority number two. This difference is multiplied by a preset time interval coefficient to obtain the first time interval. Finally, the injection start time of the flushing path with intervention priority number one is added to the first time interval to obtain the intervention priority number. The injection start time for flushing path number two is calculated sequentially to achieve non-equal interval allocation for all flushing paths. For example, the flushing path with intervention priority number one is set as the earliest starting path, with an initial injection interval of once every 8 seconds and a duration of 0.5 seconds, and the injection pressure is limited to the range of 0.9 to 1.1 kPa; the flushing path with intervention priority number two starts with a delay of 3 seconds, and the injection cycle is adjusted to once every 12 seconds, with a duration controlled to 0.6 seconds; the flushing path with intervention priority number three starts with a delay of 6 seconds based on the starting time of path number one, and the injection cycle is once every 15 seconds. The comprehensive risk score is divided into high-risk, medium-risk, and low-risk segments according to preset segment thresholds. A preset high-frequency injection frequency is assigned to the flushing path in the high-risk segment, a preset medium-frequency injection frequency to the flushing path in the medium-risk segment, and a preset low-frequency injection frequency to the flushing path in the low-risk segment. The comprehensive risk score is mapped to the injection pressure range using a linear mapping function to obtain the injection pressure value for each flushing path. The flushing path identifier, injection start time, injection frequency, and injection pressure of each flushing path are combined to form an injection control strategy. An injection duration field is added to the injection control strategy, and the injection duration for each flushing path is calculated based on the flushing path coverage area and the comprehensive risk score. The generated injection control strategy is stored in a structured data format in the control strategy dataset.

[0116] S703: During the execution of the injection control strategy, the pressure recovery time and liquid removal retention time of each flushing path are obtained, and the response hysteresis of each flushing path is calculated based on the pressure recovery time and liquid removal retention time.

[0117] In this embodiment, pressure recovery time refers to the time elapsed from the start of the injection operation to the rise of the fluid pressure in the flushing path from the initial pressure value to the target pressure value; liquid removal retention time refers to the time elapsed from the start of the injection operation to the complete replacement and discharge of the original retained liquid or contaminants in the flushing path by the newly injected liquid; and response hysteresis refers to a quantitative indicator that comprehensively reflects the response speed of the flushing path to the injection operation.

[0118] Specifically, the electronic device reads the injection control strategy from the control strategy dataset and sequentially initiates the injection operation according to the injection start time of each flushing path in the injection control strategy. Simultaneously, pressure sensors collect real-time data on fluid pressure changes within the flushing path, recording the initial pressure value at the start of the injection operation and continuously monitoring the fluid pressure rise. When the fluid pressure reaches the target pressure value set in the injection control strategy, the current time is recorded as the pressure target achievement time. The time difference between the pressure target achievement time and the injection operation start time is calculated to obtain the pressure recovery time. Simultaneously, turbidity or conductivity sensors collect real-time data on the turbidity or conductivity of the liquid at the flushing path outlet, recording the initial turbidity value at the start of the injection operation. Alternatively, an initial conductivity value can be used to continuously monitor the turbidity or conductivity change trend. When the turbidity value drops below the preset cleaning threshold or the conductivity value stabilizes within the preset stable threshold range, the current time is recorded as the liquid cleaning completion time. The time difference between the liquid cleaning completion time and the liquid injection operation start time is calculated to obtain the liquid cleaning residence time. The pressure recovery time is multiplied by a preset pressure weighting coefficient to obtain the pressure response contribution value. The liquid cleaning residence time is multiplied by a preset cleaning weighting coefficient to obtain the cleaning response contribution value. The pressure response contribution value and the cleaning response contribution value are summed to calculate the response hysteresis of each flushing path. The flushing path identifier, pressure recovery time, liquid cleaning residence time and response hysteresis of each flushing path are stored in the response monitoring dataset in a data recording format.

[0119] S704: If the increase in response lag of any flushing path in a continuous cycle exceeds a preset proportion, the injection cycle interval of any flushing path is extended based on the increase, and the injection frequency of any flushing path is reduced, so as to dynamically adjust the injection control strategy.

[0120] Specifically, the electronic device reads the response lag data of each flushing path in multiple consecutive cycles from the response monitoring dataset and arranges them in chronological order to form a response lag time series. The response lag time series of each flushing path is compared with adjacent cycles. The difference between the response lag of the current cycle and the response lag of the previous cycle is extracted to obtain the absolute increment of the response lag. The absolute increment of the response lag is divided by the response lag of the previous cycle and multiplied by 100% to obtain the percentage increase. It is determined whether the percentage increase exceeds a preset percentage threshold. When the percentage increase exceeds the preset percentage threshold (for example, the increase exceeds 15% in two consecutive cycles), the corresponding flushing path is marked as an abnormal response path and a dynamic adjustment mechanism is triggered, that is, the injection cycle interval is extended and the injection frequency is reduced. Conversely, if the response lag decreases and the swirling flow index in the corresponding risk heat map decreases by more than 10%, the injection cycle can be appropriately shortened and the number of pulses increased. Based on the percentage increase, the flushing paths are categorized into mild, moderate, and severe adjustment levels. For mild adjustment levels, the injection cycle interval is extended by a preset mild duration, and the injection frequency is reduced by a preset mild reduction. For moderate adjustment levels, the injection cycle interval is extended by a preset moderate duration, and the injection frequency is reduced by a preset moderate reduction. For severe adjustment levels, the injection cycle interval is extended by a preset severe duration, and the injection frequency is reduced by a preset severe reduction. Simultaneously, an early warning message is generated. The adjusted injection frequency and cycle interval are updated in the parameter fields of the corresponding flushing path in the injection control strategy. An adjustment record field is added to the injection control strategy to record the adjustment time, adjustment reason, parameter values ​​before adjustment, parameter values ​​after adjustment, and adjustment level. The updated injection control strategy is stored in the control strategy dataset with a new version number, and the original version of the injection control strategy is archived in the historical strategy dataset.

[0121] The following describes an exemplary electronic device for processing flushing and drainage level monitoring data provided in the embodiments of this application.

[0122] In some embodiments, the electronic device for processing flushing and drainage fluid level monitoring data is a computer device or includes a computer device. The computer device includes a processor, memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device 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 of the computer device stores data. The network interface of the computer device is used to communicate with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements the methods in the embodiments of this application.

[0123] Those skilled in the art will understand that the structure described above 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 described above, or combine certain components, or have different component arrangements.

Claims

1. A method for processing flushing and drainage liquid level monitoring data, characterized in that, The method includes: Acquire initial response data of multiple flushing paths connected to the target flushing area under initial pulse injection, and establish time-frequency response baselines for each flushing path based on the initial response data; The impedance characteristics of each flushing path are calculated based on the time-frequency response baseline, and an impedance difference map reflecting the path resistance distribution is generated based on the impedance characteristics of each flushing path. The target flushing path is determined from multiple flushing paths based on the impedance difference diagram. The velocity and pressure data of the target flushing path under disturbed injection are obtained, and a velocity and pressure coupling distribution map of the target flushing area is constructed based on the velocity and pressure data. Spatiotemporal matching of velocity-pressure coupling distribution map and impedance difference map is performed to obtain spatiotemporal image sequence, and the swirling core region in the target flushing area is identified based on the spatiotemporal image sequence; A secondary injection command is generated for the cyclone core region, and secondary response characteristic data of the cyclone core region in response to the secondary injection command is obtained. The secondary response characteristic data is then image-fused with the impedance difference map to generate a risk heat map. Based on the risk heat map, a liquid injection control strategy is generated to stagger the injection process across multiple flushing paths. The response hysteresis of each flushing path is obtained, and the liquid injection control strategy is dynamically adjusted based on the response hysteresis.

2. The method for processing flushing and drainage liquid level monitoring data according to claim 1, characterized in that, The acquisition of initial response data for multiple flushing paths connected to the target flushing area under initial pulse injection, and the establishment of time-frequency response baselines for each flushing path based on the initial response data, specifically includes: Generate phase-coded micro-amplitude pulse commands containing unique identifiers to instruct each flushing path to perform an initial pulse injection; The input pressure disturbance waveform and output liquid level response curve of each flushing path under the initial pulse injection are obtained, and the input pressure disturbance waveform and output liquid level response curve are used as the initial response data. Based on the time difference between the input pressure disturbance waveform and the output liquid level response curve in the initial response data, the phase delay time of each flushing path is calculated. Based on the rising slope and peak amplitude of the output liquid level response curve in the initial response data, and the disturbance amplitude of the input pressure disturbance waveform, the pressure amplitude attenuation ratio of each flushing path is calculated. The phase delay time and amplitude attenuation ratio of each flushing path are integrated into a two-dimensional response spectrum, and the two-dimensional response spectrum is used as the time-frequency response baseline.

3. The method for processing flushing and drainage liquid level monitoring data according to claim 2, characterized in that, The calculation of impedance characteristics for each flushing path based on the time-frequency response baseline, and the generation of an impedance difference map reflecting the path resistance distribution based on the impedance characteristics of each flushing path, specifically includes: Obtain the path length parameters of each flushing path, and based on the path length parameters, convert the phase delay time in the time-frequency response baseline into a phase delay ratio per unit length. Extract the pulse duration segment of the input pressure disturbance waveform, calculate the input energy integral value of the input pressure disturbance waveform within the pulse duration segment, and the output energy integral value of the output liquid level response curve within the pulse duration segment; Based on the ratio of the output energy integral value to the input energy integral value, and combined with the path length parameter, the pressure attenuation gradient per unit length is calculated. The phase hysteresis ratio of each flushing path and the pressure attenuation gradient are combined into a two-dimensional feature vector, and the two-dimensional feature vector is used as the impedance feature of each flushing path. An impedance difference matrix is ​​established based on the impedance characteristics of each flushing path, and the impedance difference matrix is ​​rendered into a two-dimensional heatmap to generate an impedance difference map.

4. The method for processing flushing and drainage liquid level monitoring data according to claim 1, characterized in that, The process involves determining the target flushing path from multiple flushing paths based on the impedance difference diagram, acquiring velocity and pressure data of the target flushing path under disturbed injection conditions, and constructing a velocity-pressure coupling distribution map of the target flushing area based on the velocity and pressure data. Specifically, this includes: Select the flushing path whose impedance difference is greater than a preset threshold from the impedance difference diagram as the target flushing path; Output control commands to instruct the target flushing path to perform asynchronous directional disturbance injection, and acquire velocity and pressure data of the target flushing area under disturbance injection; Spatial interpolation fitting is performed on the velocity data to generate a velocity vector layer, and a pressure color level layer is constructed based on the pressure data; Overlay the velocity vector layer onto the pressure level layer to generate a time-series velocity-pressure coupled distribution map.

5. The method for processing flushing and drainage liquid level monitoring data according to claim 1, characterized in that, The process of spatiotemporally matching the velocity-pressure coupling distribution map with the impedance difference map to obtain a spatiotemporal image sequence, and identifying the swirling core region within the target flushing area based on the spatiotemporal image sequence, specifically includes: Using the liquid outlet coordinates of the flushing path as the reference anchor point, the velocity and pressure coupling distribution map and impedance difference map are spatially mapped and aligned with the time frame sequence to generate a spatiotemporal image sequence with a unified coordinate system. The movement path of fluid particles is traced in a spatiotemporal image sequence to extract the liquid velocity trajectory, and the trajectory curvature, number of direction changes, path closure and velocity fluctuation amplitude of the liquid velocity trajectory are calculated. The trajectory curvature, number of direction changes, path closure, and velocity fluctuation amplitude are integrated to generate swirling flow indices, and the residence time distribution map is constructed by statistically analyzing the continuous residence time of fluid particles in local areas. Extract overlapping areas where the swirling flow index is greater than a preset index threshold and the continuous residence time in the residence time distribution map is greater than a preset time threshold, and mark the overlapping areas as the swirling core area within the target flushing area.

6. The method for processing flushing and drainage liquid level monitoring data according to claim 5, characterized in that, The process of generating a secondary injection command for the cyclone core region and acquiring secondary response characteristic data of the cyclone core region in response to the secondary injection command, and then fusing the secondary response characteristic data with the impedance difference map to generate a risk heatmap, specifically includes: Generate a phase-coded micro-amplitude pulse injection command for the cyclotron core region as a secondary injection command; The delayed response time, response amplitude rise rate, and fluid disturbance radius of the cyclone core region under the secondary injection command are obtained as secondary response characteristic data. The swirling flow index is normalized based on the secondary response characteristic data, and the risk level is divided according to the preset multidimensional threshold model to generate a risk level map. By spatially registering the risk level map and the impedance difference map and mixing them with dual-channel colors, a risk heatmap characterizing the coupling distribution of local drag anomalies and swirling flow is generated.

7. The method for processing flushing and drainage liquid level monitoring data according to claim 1, characterized in that, The process involves generating a staggered injection control strategy across multiple flushing paths based on a risk heatmap, acquiring the response hysteresis of each flushing path, and dynamically adjusting the injection control strategy based on the response hysteresis. Specifically, this includes: Based on the risk heat map, a comprehensive risk score for each flushing path is extracted, and an intervention priority ranking table is established based on the comprehensive risk score. Based on the intervention priority ranking table, non-equal interval injection start time, injection frequency and injection pressure are assigned to each flushing path to generate an injection control strategy; During the execution of the injection control strategy, the pressure recovery time and liquid removal retention time of each flushing path are obtained, and the response hysteresis of each flushing path is calculated based on the pressure recovery time and liquid removal retention time. If the increase in response lag of any flushing path in a continuous cycle exceeds a preset proportion, the injection cycle interval of any flushing path is extended based on the increase, and the injection frequency of any flushing path is reduced, so as to dynamically adjust the injection control strategy.

8. An electronic device for processing flushing and drainage liquid level monitoring data, characterized in that, The electronic device includes: a memory and one or more processors; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the electronic device to perform the method as described in any one of claims 1-7.

9. A computer program product containing instructions, characterized in that, When the computer program product is run on an electronic device for processing flushing and drainage level monitoring data, the electronic device causes the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on an electronic device for processing flushing and drainage level monitoring data, the electronic device causes the electronic device to perform the method as described in any one of claims 1-7.

Images