Intelligent monitoring system for municipal drainage pipe network based on multi-source sensing
By employing multi-source sensing and discrete cross-logic matrix falsification technology, the problem of false signals in drainage network monitoring systems under complex operating conditions was solved, achieving low-power, high-efficiency monitoring and communication, and ensuring the accuracy and reliability of the data.
Patent Information
- Application Number
- CN202611119458.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-27
- Publication Date
- 2026-11-10
AI Technical Summary
Existing drainage network monitoring systems are susceptible to damage from oil films, foam, and floating debris under conditions of no mains power, weak coverage, and complex operation. This can lead to deviations in sensor measurement results, and abnormal data may cause repeated transmissions and increased maintenance frequency, making it difficult to effectively suppress false flooding reports and power consumption.
The municipal drainage network intelligent monitoring system adopts multi-source sensing. It uses primary potential energy sensing components to detect the forced fluid dynamic pressure step awakening system, uses edge fluid logic execution units to perform discrete cross logic matrix falsification, generates semantic masks, cuts off the power supply backbone of the communication module, and suppresses false flooding false alarms and power consumption.
It eliminates power consumption caused by multiple concurrent data acquisitions, ensures that the data acquisition starting point is consistent with the actual disturbance event, eliminates the misalignment of spatial position and response timing of heterogeneous sensing components, avoids the processing of false signals, and reduces interference and power consumption of the communication link.
Smart Images

Figure CN123069181A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of drainage monitoring technology, specifically to an intelligent monitoring system for municipal drainage networks based on multi-source sensing. Background Technology
[0002] Municipal drainage network monitoring and anomaly detection are mainly located in drainage pipes, drainage channels, manholes, drainage outlets, sunken roads, underground parking lots, and receiving water bodies. They periodically transmit information on liquid level, flow velocity, and drainage status, and report this information to the management center. In existing projects, liquid level sensors, flow velocity sensors, and data acquisition units are generally deployed underground or at pipe cross-sections. Data is collected and flow is calculated at predetermined time intervals, and warnings or results are reported. In a few battery-powered scenarios, low-power standby or trigger-activated wake-up designs can be adopted to accommodate long-term monitoring. For urban flood monitoring, overflow warning, drainage scheduling, and facility operation and maintenance, the operating conditions are often long-term underground environments with frequent manual inspections, complex changes in operating conditions, and significant differences in water conditions.
[0003] Chinese patent document CN107727179B discloses a flow monitoring device for urban drainage pipelines. This document considers various operating conditions of urban drainage terminals, including culverts, circular pipes, full pipes, and partially full pipes. It installs an electromagnetic current meter, an ultrasonic Doppler current meter, and a liquid level sensor at the top of the drainage pipeline. A data acquisition and processing unit and a power supply unit are installed below the manhole cover. During partially full pipe drainage, the electromagnetic current meter monitors the surface velocity of the water flow, and the data acquisition and processing unit uses the liquid level to calculate the average cross-sectional velocity, instantaneous flow rate, and cumulative flow rate. During full pipe drainage, the ultrasonic Doppler current meter monitors the flow velocity inside the pipe, and the data acquisition and processing unit calculates the flow parameters. The data acquisition and processing unit uses built-in flow calculation model software to calculate the drainage volume and reports the average velocity, water level, instantaneous flow rate, cumulative flow rate, and drainage time via cellular data transmission.
[0004] While the above solutions can monitor drainage flow, their ultimate technical goal is to acquire and continuously report the most complete flow velocity and liquid level data under different operating conditions. However, during long-term pipeline operation, sensors are affected by oil films, foam, floating debris, sediment, and downhole moisture. This can lead to deviations between local measurement results and the actual flow conditions. Switching between full and partial pipe flow, localized backflow, wave crest propagation, and misalignment of the detection location also mean that the liquid level and flow velocity information obtained at the same moment do not naturally correspond to the same fluid event. Current technologies rely primarily on existing acquisition, computation, and upload links. When local abnormal signals are transmitted as valid monitoring data, issues such as repeated transmissions, prolonged online operation, and increased maintenance frequency may arise under conditions of weak coverage and battery power.
[0005] Furthermore, most existing low-power solutions employ timed power-on or non-triggered standby, which still primarily rely on continuous data acquisition at the front end and complete data upload at the back end. This makes it difficult to avoid the cascading effects of abnormal data entering the subsequent communication processing chain. Therefore, the key technical challenge in this field is how to suppress the continuous interference of local anomaly detection on monitoring, judgment, and communication links in underground drainage networks with no mains power, weak coverage, and complex operating conditions. Summary of the Invention
[0006] (a) Technical problems to be solved
[0007] To address the shortcomings of existing technologies, this invention provides an intelligent monitoring system for municipal drainage networks based on multi-source sensing. This system acquires discrete multimodal environmental parameters; utilizes a discrete cross-logic matrix to physically verify the continuity of the parameters, outputting blinding disturbance flags; redirects the storage path of the original waveform service flow based on the flags and generates a semantic mask; after loading the semantic mask into the communication module, it cuts off the power supply backbone of the communication module and releases the full transmission path before a valid reference waveform is formed, thereby suppressing false flooding reports and power consumption from weak network retransmissions, thus solving the technical problems described in the background section.
[0008] (II) Technical Solution
[0009] To achieve the above objectives, the present invention provides the following technical solution:
[0010] The municipal drainage network intelligent monitoring system based on multi-source sensing includes a primary potential energy sensing component, a first spatial potential energy sensing component, a second flow field kinetic energy sensing component, an edge fluid logic execution unit, a communication module and power switching components on its power supply backbone. The primary potential energy sensing component outputs a forced fluid dynamic pressure step interruption.
[0011] The edge fluid logic execution unit responds to the step interrupt wake-up and sequentially pulls up the first spatial potential energy sensing component and the second flow field kinetic energy sensing component according to the asymmetric compensation time window, acquires discrete multimodal environmental parameters and inputs them into the discrete cross logic matrix, and outputs the blinding disturbance flag bit.
[0012] The edge fluid logic execution unit responds to the blinding disturbance flag, redirects the storage path of the original waveform service flow, and generates a semantic mask; after loading the semantic mask into the communication module, the edge fluid logic execution unit drives the power switching components to cut off the power supply backbone of the communication module.
[0013] Furthermore, after responding to a step interruption, the edge fluid logic execution unit first turns on the controlled power supply branch of the first spatial potential energy sensing component and obtains a stable echo. It then extracts the first spatial potential energy extreme value within the first sampling window, determines the asymmetric compensation time window based on the pipe diameter, Manning roughness coefficient, and installation spacing in the calibration table, and turns on the controlled power supply branch of the second flow field kinetic energy sensing component and releases its interface clock after the asymmetric compensation time window ends.
[0014] Furthermore, the edge fluid logic execution unit quantizes the first spatial potential energy extreme value and the second flow field kinetic energy extreme value into the first spatial potential energy level and the second flow field kinetic energy level, respectively. It then inputs the first spatial potential energy level, the second flow field kinetic energy level, the forced fluid dynamic pressure step characteristics, and whether the asymmetric compensation time window falls within the preset interval into the local preset truth table to output the blinding disturbance flag.
[0015] Furthermore, when the blinding disturbance flag is valid, the edge fluid logic execution unit freezes the current sampling descriptor, rewrites the destination storage address of the direct memory access controller from the original waveform buffer to the invalid empty area, and clears the valid length register of the original waveform buffer; the edge fluid logic execution unit then generates a semantic mask based on the blinding disturbance flag, the hierarchical conflict state, the slope excitation state, and the time window valid state.
[0016] Furthermore, the edge fluid logic execution unit writes the semantic mask, node identifier, event sequence number, and checksum into the short packet register in a fixed field order, and loads the contents of the short packet register into the command first-in-first-out queue of the communication module; the communication module only receives the short packet data corresponding to the fixed field order for the current abnormal event, and does not receive the original waveform service stream corresponding to the current abnormal event.
[0017] Furthermore, after receiving the current abnormal event in the command first-in-first-out queue, the communication module first enters the control plane micro-load transmission state and then outputs the release auxiliary signaling; after the communication module enters the control plane micro-load transmission state, the edge fluid logic execution unit keeps the original waveform service flow from entering the broadband service transmission queue of the communication module and maintains the single-depth shadow register of the short packet register area.
[0018] Furthermore, the communication module encapsulates the control plane micro-load of the current abnormal event using a non-IP data delivery message frame format; the message frame includes at least a node identifier, event sequence number, semantic mask, and checksum, and after the message frame is encapsulated, it no longer receives the original waveform service stream corresponding to the current abnormal event, and keeps the broadband service transmission queue closed until the power switch components operate.
[0019] Furthermore, the power switching device is a high-side switching device connected in series between the positive terminal of the battery and the power supply backbone of the communication module. The control terminal of the high-side switching device is directly connected to the digital control pin of the edge fluid logic execution unit through a current-limiting resistor.
[0020] The power supply input of the communication module is connected in parallel with an energy storage capacitor and a transient suppression device, and the high-side switching device is connected in series with the main power supply backbone of the communication module.
[0021] Furthermore, the edge fluid logic execution unit starts the chopper width limit count after the semantic mask is loaded in the communication module, and flips the digital control pin when the chopper width limit count ends; after the high-side switching device is turned off, the asynchronous serial port, the direct memory access controller sampling channel, and the controlled power supply branches corresponding to the first spatial potential energy sensing component and the second flow field kinetic energy sensing component are turned off in sequence, leaving only the independent real-time clock domain and the external interrupt domain.
[0022] Furthermore, the edge fluid logic execution unit records the sampling records that have not triggered the blinding disturbance flag and have completed a full sampling window written to the original waveform buffer as valid reference waveforms, and accumulates the duration of continuous chopping without forming a valid reference waveform in the independent real-time clock domain; when the accumulated duration reaches 72 hours, it generates a reset release flag and releases the full pass-through path of the complete sampling data in the next wake-up.
[0023] (III) Beneficial Effects
[0024] This invention provides an intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing, which has the following beneficial effects:
[0025] By listening for forced fluid dynamic pressure steps through the primary potential energy sensing component, the system is awakened by the edge fluid logic execution unit when the step occurs, eliminating the power consumption caused by multiple concurrent acquisitions, so that the acquisition starting point of the underground closed pipe network without mains power nodes is consistent with the real disturbance event.
[0026] By activating the first spatial potential energy sensing component and staggering the activation of the second flow field kinetic energy sensing component according to the asymmetric compensation time window, the misalignment of the spatial position and response timing of the heterogeneous sensing components is eliminated, so that subsequent judgments are based on the discrete multimodal environmental parameters corresponding to the same wavefront event. The discrete cross-logic matrix is used to place the extreme values of the first spatial potential energy and the extreme values of the second flow field kinetic energy under the same physical continuity constraint for edge-side falsification, so as to eliminate the false situation caused by local physical blindness in a timely manner and no longer treat the distorted environmental representation as a real flood signal.
[0027] By listening for forced fluid dynamic pressure steps through the primary potential energy sensing component, the system is awakened by the edge fluid logic execution unit when the step occurs, eliminating the power consumption caused by multiple concurrent acquisitions, so that the acquisition starting point of the underground closed pipe network without mains power nodes is consistent with the real disturbance event.
[0028] By activating the first spatial potential energy sensing component and staggering the activation of the second flow field kinetic energy sensing component according to the asymmetric compensation time window, the misalignment of the spatial position and response timing of the heterogeneous sensing components is eliminated, so that subsequent judgments are based on the discrete multimodal environmental parameters corresponding to the same wavefront event. The discrete cross-logic matrix is used to place the extreme values of the first spatial potential energy and the extreme values of the second flow field kinetic energy under the same physical continuity constraint for edge-side falsification, so as to eliminate the false situation caused by local physical blindness in a timely manner and no longer treat the distorted environmental representation as a real flood signal. Attached Figure Description
[0029] Figure 1 This is an overall architecture diagram of the underground closed pipe network edge monitoring node according to an embodiment of the present invention;
[0030] Figure 2 This is a cross-sectional view of the on-site deployment of the inspection well monitoring node according to an embodiment of the present invention;
[0031] Figure 3 This is a timing diagram of forced fluid wavefront triggering and staggered sampling according to an embodiment of the present invention;
[0032] Figure 4 This is a discrete cross-logic matrix decision diagram according to an embodiment of the present invention;
[0033] Figure 5 This is a diagram showing the original waveform service flow redirection and semantic mask generation in an embodiment of the present invention.
[0034] Figure 6 This is a diagram showing the high-side power-off and RF link suppression of the communication module in an embodiment of the present invention;
[0035] Figure 7 This is a flowchart of the cross-cycle baseline recovery and full-volume permeation release process according to an embodiment of the present invention. Detailed Implementation
[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0037] Please see Figures 1-7 This invention provides an intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing, comprising:
[0038] Step 1: Using the forced fluid dynamic pressure step as the sole wake-up starting point, the first spatial potential energy sensing component and the second flow field kinetic energy sensing component are staggered and pulled up using an asymmetric compensation time window, and the discrete multimodal environmental parameters corresponding to the same wavefront event are output.
[0039] Anomalies in municipal drainage networks often initially manifest as short, forced fluid wavefronts rather than sustained high water levels. If the edge fluid logic execution unit simultaneously activates two sensing components at a fixed cycle, the result is often that the power supply busbar experiences an initial surge, and the timeline incorrectly splices the flow field changes that haven't yet reached the second measuring point with the liquid level changes that have already reached the first measuring point. While the resulting values are complete, they do not represent two aspects of the same physical event, rendering the subsequent discrete cross-logic matrix ineffective. Therefore, this step does not employ timed concurrent acquisition but instead binds the acquisition sequence to the wavefront propagation mechanism. All the following actions are performed by the edge fluid logic execution unit installed within the sealed cavity on the side wall of the inspection well.
[0040] The edge fluid logic execution unit refers to the circuit unit responsible for wake-up control, time window scheduling, discrete cross-logic determination, storage path switching, and communication module control after listening for forced fluid dynamic pressure step changes. Exemplarily, it includes a microcontroller or programmable logic device and its associated interrupt, timing, storage, and interface circuits. The first spatial potential energy sensing component in this application refers to a sensor component that outputs an electrical signal characterizing changes in liquid level or liquid surface position; the second flow field kinetic energy sensing component refers to a sensor component that outputs an electrical signal characterizing changes in flow velocity. The radio frequency link suppression operation refers to the control action of cutting off the main power supply backbone of the communication module, or causing the communication module to exit the current service transmission state without introducing broadband service retransmission.
[0041] In one embodiment, the edge fluid logic execution unit employs a microcontroller with an external interrupt pin, a timer / counter, a fixed-point arithmetic unit, and a serial peripheral interface; in a parallel embodiment, the edge fluid logic execution unit employs a programmable logic device with a synchronous counter and a programmable delay chain. The primary potential energy sensing component is fixed to the upstream side of the pipe wall, preferably a piezoelectric thin film sandwiched between a stainless steel substrate and an elastic vibration isolation pad; alternatively, it is replaced by a mechanical contact switch composed of a levitation element, a lever, and noble metal contacts. The first spatial potential energy sensing component is preferably an ultrasonic level probe facing the water surface, and the second flow field kinetic energy sensing component is preferably a Doppler velocity probe arranged against the inner wall of the pipe bottom; both are connected to the edge fluid logic execution unit via independent controlled power supply branches.
[0042] Furthermore, the primary potential energy sensing component continuously listens for changes in the dynamic pressure of the forced fluid; the edge fluid logic execution unit is only interrupted by hardware when it detects a dynamic pressure step that could form a hydraulic jump or a pulsating front; then, the first spatial potential energy sensing component is activated and the first stable echo is read, and then the asymmetric compensation time window is calculated based on the pipe diameter, roughness conditions, and baseline flow velocity. After the time window expires, the second flow field kinetic energy sensing component is activated; finally, the extreme values of the first spatial potential energy and the second flow field kinetic energy are extracted from the two non-overlapping sampling windows, respectively. In this way, what is obtained are not two isolated readings, but parameter pairs bound to the same propagation event.
[0043] To facilitate direct implementation, the edge fluid logic execution unit saves a calibration table grouped by pipe diameter and pipe material during the factory write stage. During operation, it reads the pipe diameter, pipe material, installation spacing and the baseline flow rate written in the most recent maintenance from the local non-volatile memory of the well section, and then uses fixed-point first-order interpolation to obtain the asymmetric compensation time window without calling the floating-point library, thereby maintaining a short duty cycle.
[0044] If the wake-up condition depends on the absolute liquid level of the first spatial potential energy sensing component, oil film adhesion, spider web obstruction, or well wall condensation may all misinterpret the static echo as a flood peak. If it depends on the continuous velocity measurement of the second flow field kinetic energy sensing component, then continuous power supply is required to maintain the velocity measurement link when the fluid is dry. Therefore, this solution moves the wake-up starting point forward to the primary potential energy sensing component, which is more sensitive to changes in the dynamic pressure of the forced fluid and has lower standby power consumption.
[0045] After shaping the output of the primary potential energy sensing component, the edge fluid logic execution unit calculates the dynamic pressure slope using a backward differential process. When the dynamic pressure slope exceeds the trigger threshold and remains there for a continuous holding period, the external interrupt pin generates a falling edge, and the edge fluid logic execution unit exits the stop mode. It first turns on the main clock, then sequentially turns on the interrupt mask, timer / counter, and first controlled power supply branch, without simultaneously turning on the serial peripheral interface clock.
[0046] The calculation relationship is written as follows:
[0047]
[0048] Among them, dynamic pressure slope : No. The forced fluid dynamic pressure change rate at each sampling time is used to determine whether a step event sufficient to trigger hardware wake-up has occurred; the value is a real number. : No. The dynamic pressure sampled value output by the shaping circuit at each sampling moment is within the range of the primary potential energy sensing component; the forced fluid dynamic pressure at the previous sampling moment The sampling value of dynamic pressure immediately preceding the previous moment, and the values are within the same measurement range; sampling interval The equal-interval sampling period of the primary potential energy sensing component is given by a timer / counter and has a value greater than 0; the sampling sequence number... Take a positive integer.
[0049] For example, when surface water suddenly overflows into the upstream branch pipe after the rainwater culvert has dried up, what reaches the inspection well node is a short shock wave. The piezoelectric film is first bent, the shaping circuit is pulled out with a clear edge, the edge fluid logic execution unit is awakened, the liquid level probe and flow velocity probe remain de-energized, and are pulled up in the subsequent sequence after the master clock stabilizes.
[0050] Furthermore, the wake-up action is driven by wavefront events and is not affected by static occlusion. The master clock and the sensing branch are started separately, and the subsequent asymmetric compensation time window obtains a unified start time.
[0051] Furthermore, the first spatial potential energy sensing component faces the water surface position or wave crest rise, while the second flow field kinetic energy sensing component faces the local flow velocity near the bottom of the pipe. The two are neither in the same spatial position nor undergo the same dirt adhesion process. If samples are taken at the same power-on time, the two results often correspond to different stages of the wavefront. The former may have already passed the peak value, while the latter has not yet established a stable scattering field.
[0052] The edge fluid logic execution unit first reads the installation spacing, pipe diameter, and Manning roughness coefficient from the calibration table. Then, combining this with the baseline flow velocity left from the previous round of fault-tolerant inspection, it generates an asymmetric compensation time window using piecewise linear interpolation. When the baseline flow velocity is lower than the lower limit, a preset lower limit is used instead to avoid time window distortion caused by the denominator approaching zero. Subsequently, the edge fluid logic execution unit first turns on the first controlled power supply branch. After the output of the first spatial potential energy sensing component passes the echo stability criterion, it starts a timer counter to accumulate the asymmetric compensation time window. When the timing ends, it turns on the second controlled power supply branch and simultaneously releases its serial peripheral interface clock. The time window relationship is written as:
[0053]
[0054] Among them, the asymmetric compensation time window : The waiting time from the start of effective sampling by the first spatial potential energy sensing component to the allowable power-on of the second flow field kinetic energy sensing component, with a value greater than 0; Installation spacing The equivalent distance between the first spatial potential energy sensing component and the second flow field kinetic energy sensing component along the pipe axis is preferably 0.5D to 1.5D; the wavefront apparent velocity... : The propagation velocity of the forced fluid wavefront in the current well section, with a value greater than 0; Manning roughness coefficient. The degree to which pipe wall roughness impedes wavefront propagation and near-wall flow is measured within the calibration table coverage area; it is used for indexing time windows and is derived from the well section material / configuration.
[0055] Pipe diameter : Well section inner diameter, greater than 0; used for indexing time window, derived from well section installation data or local configuration; baseline flow velocity. The steady-state flow rate of the edge fluid logic execution unit before the last valid inspection is used to correct the time window. It is derived from the most recent valid inspection and its value is not less than the lower limit of the flow rate.
[0056] Asymmetric compensation time window The edge fluid logic execution unit reads the calibration from the local calibration table. This local calibration table is at least based on the pipe diameter of the current well section. Manning roughness coefficient The installation distance between the first spatial potential energy sensing component and the second flow field kinetic energy sensing component is used as an index to obtain the baseline flow velocity. The corresponding entries are corrected periodically. Lower bound of the time window. and the upper bound of the time window The same calibration table is used to define the permissible response range of the second flow field kinetic energy sensing component to the same wavefront event. If the edge fluid logic execution unit does not acquire a new baseline flow velocity in the current cycle, it uses the previously saved corresponding entry.
[0057] For example, inside a circular rainwater pipe, a first spatial potential energy sensing component is positioned above the center of the inspection well, and a second flow field kinetic energy sensing component is positioned on the inner wall of the pipe bottom upstream. The edge fluid logic execution unit first powers the first spatial potential energy sensing component and completes one echo screening. Then, after a delay according to an asymmetric compensation time window, it turns on the power supply branch of the second flow field kinetic energy sensing component. Throughout the process, the clocks of the two serial peripheral interfaces do not overlap, avoiding the formation of superimposed spikes on the flexible cable.
[0058] Furthermore, spatial misalignment is transformed into computable temporal misalignment and compensated for, the startup impact on the power supply bus is de-peaked, and well segment differences are explicitly written into the sampling timing.
[0059] Furthermore, the subsequent discrete cross-logic matrix does not require the complete waveform, but rather the first spatial potential energy extremum and the second flow field kinetic energy extremum that can represent the wavefront event. If the complete waveform is directly transmitted, it will prolong the current wake-up period and also bring in trailing noise unrelated to the current wavefront into the next step.
[0060] The discrete cross-logic matrix is a locally pre-built discrete truth table, which is not updated through online training or online learning. The discrete truth table is based on the first spatial potential level. Second flow field kinetic energy level The main index is based on the dynamic slope. Whether the trigger slope threshold and asymmetric compensation time window are exceeded Whether it falls within the allowable range is the gating condition. When the first spatial potential energy level is higher than the second flow field kinetic energy level and the level difference reaches a preset boundary, simultaneously... Characterized by the existence of a forced fluid dynamic pressure step and When the value falls within the allowable range, output the blinding disturbance flag. Output in other cases .
[0061] Specifically, the first spatial potential energy sensing component outputs an echo amplitude sequence in the first sampling window. The edge fluid logic execution unit first removes the missing echo points between the thresholds and then obtains the first spatial potential energy extreme value that may stably appear in this window. The second flow field kinetic energy sensing component outputs a frequency shift amplitude sequence in the second sampling window. The edge fluid logic execution unit extracts the second flow field kinetic energy extreme value in sequence according to the window rules and simultaneously writes the window start point, window end point and trigger source marker to form a fixed multimodal discrete multimodal environmental parameter.
[0062] Its extraction relationship is written as:
[0063]
[0064] Among them, the first spatial potential energy extreme value First sampling window The maximum effective echo amplitude of the first spatial potential energy sensing component within the sensor is used to represent the spatial potential energy response intensity corresponding to the wavefront event, and the range of the first spatial potential energy sensing component is taken; the first sampling window... The first spatial potential energy sensing component can sample a finite positive time interval; the first spatial potential energy sequence. The first space potential energy sensing component in time Continuous output, taking the range of the first spatial potential energy sensing component; the second flow field kinetic energy extreme value. Second sampling window The maximum effective frequency shift amplitude is used to determine the kinetic energy response intensity of the flow field corresponding to the wavefront event, and the range of the second flow field kinetic energy sensing component is taken; the second sampling window... The velocity sampling interval after the asymmetric compensation time window ends is a finite positive time period; the second flow field kinetic energy sequence The second flow field kinetic energy sensing component in time Continuous output, taking the range of the second flow field kinetic energy sensing component; time Take the continuous sampling times within the corresponding sampling window.
[0065] First sampling window The sampling window begins when the first stable echo is obtained by the first spatial potential energy sensing component and ends when the second controlled power supply branch is turned on; The process begins when the second flow field kinetic energy sensing component outputs the first valid frequency shift sample and ends before triggering step two for judgment or step three for clearing in this cycle.
[0066] For example, when a semi-transparent oil film is present on the well wall, the first spatial potential energy sensing component may exhibit weak reflections in the first few echoes. The edge fluid logic execution unit does not directly package and send these echoes, but instead retains only the maximum effective echo that passes the stability criterion within the first sampling window. Subsequently, the second flow field kinetic energy sensing component obtains the corresponding frequency shift peak value in the second sampling window, and both, along with the window index, are encapsulated into a fixed-length record. As a degraded implementation, if no hardware interruption occurs in step A for several months, the independent watchdog wakes up the edge fluid logic execution unit to perform a low-frequency fault-tolerant inspection, still generating the first spatial potential energy extreme value in the same format. Second flow field kinetic energy extreme value .
[0067] Furthermore, events preceding the same wave are compressed into directly identifiable discrete extreme value pairs. The window index and trigger source remain unchanged. The two parallel implementation methods, mechanical contact switch and watchdog inspection, are consistent on the data interface. Hardware wake-up no longer continuously measures speed, staggered start-up no longer creates concurrent startup impact, and dual-window extreme values no longer push the entire business waveform into the subsequent link in advance.
[0068] Step 2: Input the discrete multimodal environment parameters obtained in Step 1 into the discrete cross-logic matrix, complete the physical repulsion deduction within the edge fluid logic execution unit, and output the blinding perturbation flag. .
[0069] Step one has already aligned the sampling time with the propagation delay, but the first spatial potential energy extremum... With the second flow field kinetic energy extreme value This is still merely an event representation and insufficient to directly distinguish between actual incoming water and false positives caused by partial probe obstruction. In underground closed pipe networks, oil films, foam, floating debris, and silt will first raise the echo surface of the first spatial potential energy sensing component, while no corresponding flow field is formed near the second flow field kinetic energy sensing component. If the wide waveform is retained and waited for remote fitting, not only will the communication load increase, but false events that could have been identified downhole will also be sent into the weak network link.
[0070] Therefore, this step moves the decision-making process forward to the edge fluid logic execution unit. The edge fluid logic execution unit then determines the dynamic pressure slope. As the trigger point, the asymmetric compensation time window As a spacetime boundary, the first spatial potential energy extreme value With the second flow field kinetic energy extreme value As a continuous verification object, the event continues to propagate only when the two can support each other under the same propagation constraint; once a combination occurs, such as a sudden rise in liquid level without flow velocity or a wavefront reaching its peak while the kinetic energy is near the noise floor, the output blinding disturbance flag is activated. .
[0071] The following actions are all performed by the edge fluid logic execution unit. In a preferred embodiment, the discrete cross-logic matrix is pre-written to non-volatile memory and moved to on-chip static random access memory after power-on; in a parallel embodiment, the same matrix is stored in the programmable logic device lookup table array. The dynamic voltage slope output in step one. Asymmetric compensation time window First spatial potential energy extreme value Second flow field kinetic energy extreme value Do not change the name or meaning in this step.
[0072] First, read the pipe diameter from the well section installation parameters. Manning roughness coefficient and baseline flow rate And based on this, retrieve the lower bound of the time window. With the upper bound of the time window Then, the first spatial potential energy extreme value With the second flow field kinetic energy extreme value Compressed into matrix row and column inlets; then superimposed with dynamic pressure slope. With asymmetric compensation time window The system employs dual gating; finally, it accesses the discrete cross-logic matrix to determine whether a false situation of local physical blinding has been entered, and then solidifies this determination as a blinding perturbation flag. .
[0073] If we directly compare the extreme values of the first spatial potential energy Second flow field kinetic energy extreme value The original values, the conditioning gain of different sensing components, the installation distance, and the echo threshold will map the same field action into different amplitude ranges, thus causing the matrix boundary to drift. Therefore, the edge fluid logic execution unit first compresses the two types of extreme values into a unified level before proceeding to the subsequent repulsive deduction.
[0074] Preferably, after receiving the event records packaged in step one, the edge fluid logic execution unit first checks the sampling window marker, and then reads the pipe diameter from the local configuration area. and baseline flow rate And the asymmetric compensation time window Introducing the second flow field kinetic energy extreme value The normalization process. Its relationship can be written as:
[0075]
[0076] Where: First spatial potential energy level : From the first spatial potential energy extreme value Quantized data is used as the row index of the discrete cross-logic matrix, with values ranging from integers, preferably between 0 and 15; second flow field kinetic energy level. : The extreme value of kinetic energy of the second flow field Quantized values are used as column indices of the discrete cross-logic matrix, taking integer values and preferably ranging from 0 to 15; the first spatial potential energy extremum. Step 1: The effective maximum echo amplitude extracted within the first sampling window is used to characterize the spatial potential energy response intensity of the wavefront event.
[0077] Second flow field kinetic energy extreme value Step 1: The maximum effective frequency shift amplitude extracted within the second sampling window is used to characterize the kinetic energy response intensity of the flow field during the wavefront event; pipe diameter. The inner diameter of the pipe in the current well section is used to determine the extreme value of the first spatial potential energy. Provide a geometrically normalized baseline with a value greater than 0; baseline flow velocity. The steady-state flow velocity retained from the most recent effective water flow or low-frequency fault-tolerant inspection is used to provide the extreme value of the kinetic energy of the second flow field. Provide a baseline for operation, with a value no less than the preset lower limit of flow rate;
[0078] Asymmetric compensation time window Step 1 is used to stagger the waiting time for the two types of sensing components to be activated, and is used to write the propagation time delay into the second flow field kinetic energy level. The normalization process takes values greater than 0; the floor operator. : Fixed-point quantization operation that rounds down, used to compress continuous quantities into discrete levels.
[0079] For example, when the wellbore membrane raises the ultrasonic echo surface, the first spatial potential energy extremum... It will map to a higher level of first-space potential energy. However, the second flow field kinetic energy extreme value did not form as the mainstream. Still stuck at the lower second flow field kinetic energy level This does not result in loose values, but rather row and column entries that can be directly accessed through matrix addressing.
[0080] Furthermore, the first spatial potential energy extreme value With the second flow field kinetic energy extreme value The first spatial potential energy level is converged into a uniform scale. Second flow field kinetic energy level .
[0081] Observing the first spatial potential energy level in isolation An increase in elevation is not sufficient to prove a false alarm; observe the second flow field kinetic energy level in isolation. A low value is insufficient to conclude blindness. What truly falsifies the finding is the asymmetric compensation time window given in step one. The two measuring points have been constrained to the same wavefront event, dynamic pressure slope It also shows that there is a sudden triggering, but the two types of levels still diverge for a long time.
[0082] In a preferred embodiment, the discrete cross-logic matrix adopts Truth table, row address is taken from the first spatial potential level The column address is taken from the second flow field kinetic energy level. To avoid misjudging chronic fluid accumulation or slow-flowing swelling as blinding, the edge fluid logic execution unit applies double gating before accessing the matrix and then writes the blinding disturbance flag. .
[0083] Their relationship can be written as:
[0084]
[0085] Where: Blinding perturbation flag : Indicates whether the current event is identified as a local physical blinding false situation, serving as the direct control variable for initiating the underlying redirection and interception operation in step three, and takes the value of a binary integer; First spatial potential energy level : Reflects the degree of rise in liquid level or echo surface, with values ranging from integers, preferably between 0 and 15; Second flow field kinetic energy level : Reflects the degree of kinetic energy response of the flow field, with values ranging from integers, preferably between 0 and 15; level difference threshold The minimum hierarchical difference that separates normal coupling from obvious divergence, with a value that is a positive integer;
[0086] dynamic pressure slope Step 1 triggers the wake-up based on the forced fluid dynamic pressure change rate, which characterizes whether the event has a sudden excitation characteristic; trigger slope threshold. The boundary value separating effective excitation from long-term slow variation is a positive real number; the lower bound of the time window. Based on pipe diameter With Manning roughness coefficient The local boundary table provides the shortest physical duration, with a value greater than 0; the upper bound of the time window. Based on pipe diameter With Manning roughness coefficient The local boundary table gives the longest physical duration, which is greater than the lower bound of the time window. ; Logic and Operators : Indicates that both the preceding and following conditions are true; logical OR operator : indicates that either the preceding or following condition is true;
[0087] Asymmetric compensation time window Used to verify whether the two extreme values still belong to the same wavefront event;
[0088] The level difference threshold, trigger slope threshold, and time window boundary are all locally preset comparison boundaries and are not updated online during the runtime.
[0089] During on-site implementation, if an oil film adheres to the wellbore and condensation droplets above it raise the echo surface, the first spatial potential energy level... It will fall at a higher line address, in the second flow field kinetic energy level. Still stuck at a lower column address; meanwhile, the dynamic pressure slope The data indicates that a brief impact did indeed occur, within the asymmetric compensation time window. This is also within the allowable range for this well section. Therefore, the edge fluid logic execution unit does not interpret the event as a normal rise, but instead directly sets the blinding disturbance flag. Set as Furthermore, the discrete cross-logic matrix divides the first spatial potential energy level... Second flow field kinetic energy level Dynamic pressure slope and asymmetric compensation time window By merging them into the same decision chain, false situations caused by localized physical blinding are identified on the spot at the edge.
[0090] In a preferred embodiment, the microcontroller reads the first spatial potential energy level. Second flow field kinetic energy level Then, press the dynamic pressure slope. With asymmetric compensation time window The corresponding gated boundary access truth table is accessed, and the result is written to the single-byte state register area. Step three then directly reads the blinding perturbation flag bit from this register area. .
[0091] In the parallel implementation scheme, the first spatial potential energy extremum With the second flow field kinetic energy extreme value First, it is converted into a multi-bit high / low level code by a window comparator, and then sent to the programmable logic device lookup table array; dynamic voltage slope Edge shaping results in enable bits, and asymmetric compensation time window. The allowed interval forms the window valid bits. The programmable logic device sends the hierarchical encoding, enable bits, and window valid bits together into the combinational logic network, and the output directly provides the blinding disturbance flag bit. .
[0092] In engineering implementation, the microcontroller path optimization uses unsigned integer truth table addressing, and the time window boundary uses piecewise linear interpolation from the pipe diameter. Manning roughness coefficient The local table is generated; the programmable logic device path optimization selects a combination of window comparator and lookup table array, without multiplier cascading, to shorten the decision chain.
[0093] For example, when main control resources are scarce in old well sections undergoing renovation, on-site wiring personnel only need to connect the two comparator outputs and one time window valid line to the programmable logic device, and then connect the result line back to the main control status pin. Step three still reads the blinding disturbance flag bit with the same meaning. .
[0094] Furthermore, step two can fall into either the fixed-point truth table path of the microcontroller or the pure hardware lookup table path of the programmable logic device; the two paths share the same input and output objects, and the interface for subsequent steps remains unchanged. The normalized entry limit the perturbation of matrix boundaries by dimensional differences, and the dual-gated matrix restricts only mismatch combinations under the same wavefront event to entering the blinding determination. The microcontroller path and the programmable logic device path share the output meaning, thus maintaining consistency in the call of step three.
[0095] Step 3: Response to blinding disturbance flags The edge fluid logic execution unit performs low-level redirection interception and semantic masking of the original waveform service flow before application layer protocol encapsulation. Generate and control the micro-load delivery trigger.
[0096] Step two has already provided the blinding perturbation flag. If at this point the conventional acquisition chain is still followed to find the extreme value of the first spatial potential energy... The second flow field kinetic energy extreme value The corresponding original waveform service flow is then handed over to the service protocol stack. The communication module then treats the spurious events caused by local physical blinding as complete monitoring messages, continuously framing, queuing, and retransmitting them. This processing order renders the physical falsification in step two worthless.
[0097] Therefore, this step does not include the blinding perturbation flag. Instead of serving as a high-level alarm suppression flag, it is moved forward to become a bus control source. All the following actions are performed by the edge fluid logic execution unit. The on-chip static random access memory of the edge fluid logic execution unit is divided into a raw waveform buffer, an invalid empty area, and a short packet register area; the sampling channel of the direct memory access controller is connected to the sensor's first-in-first-out buffer; the ferroelectric memory is connected to the side recording branch via an independent serial peripheral interface; the communication module is connected via an asynchronous serial port. Thus, the blinding disturbance flag bit... After its establishment, before the original waveform service flow enters the higher-level encapsulation, its destination has already been rewritten at the on-chip bus layer and compressed into a single-byte semantic mask. .
[0098] Among them, the edge fluid logic execution unit reads the blinding disturbance flag bit. Then the first spatial potential energy level Second flow field kinetic energy level Dynamic pressure slope and asymmetric compensation time window Moved into the local storage area, when the blinding disturbance marker is located. When the value is zero, the sampling channel still maintains the original waveform buffer write path. When the blinding perturbation flag is zero... For a period of time, the edge fluid logic execution unit freezes the current sampling descriptor, rewrites the destination address of the direct memory access controller into the invalid empty area, and clears the effective length register of the original waveform buffer to zero.
[0099] Next, the edge fluid logic execution unit extracts hierarchical conflict states, slope activation states, and time window valid states from a decision chain and combines them into a semantic mask in a fixed positional order. Then, it is delivered to the short packet storage area in a fixed field format;
[0100] In the preferred embodiment, the ferroelectric memory recording operation occurs simultaneously: the edge fluid logic execution unit extracts the blinding leading edge segment from the tail of the sensor's FIFO buffer, writes it into the ferroelectric memory, and then performs invalid empty area redirection; in the parallel embodiment, without ferroelectric memory, invalid empty area redirection and semantic masking are not performed. For delivery, both paths share a single trigger entry point, namely the blinding perturbation flag. .
[0101] If packets are dropped only at the application layer, bus transfer between the sensor's FIFO buffer and the original waveform buffer will still occur, and the on-chip direct memory access controller, buffer consistency maintenance, and subsequent frame queues will still consume energy. Therefore, the edge fluid logic execution unit does not wait for the service protocol stack to determine the destination address, but directly rewrites the destination address of the direct memory access controller.
[0102] The edge fluid logic execution unit receives a blinding disturbance flag. Within the same processing cycle, the current sampling descriptor is latched first, then the original destination address is replaced with the starting address of the invalid empty area, and the effective length register of the original waveform buffer is rewritten synchronously. The relationship can be written as:
[0103]
[0104] Where: Destination address : Indicates the final write address of the current sampling channel of the direct memory access controller, and its value is an on-chip addressable integer address; raw destination address : Indicates that the blinding perturbation flag has not been triggered. The starting address of the original waveform buffer to which the original waveform service stream should be written is an integer address within the on-chip original waveform buffer address range; invalid empty area address : Indicates the starting address of a specially reserved invalid free area in the on-chip static random access memory, and its value is an integer address within the range of the on-chip invalid free area address; Blinding perturbation flag bit : Indicates whether step two determines the current event as a false situation caused by local physical blinding, with a value of 0 or 1;
[0105] In terms of on-site action, when the oil film adhesion causes the first spatial potential energy sensing component to generate a lifting echo while the second flow field kinetic energy sensing component does not provide a corresponding kinetic energy response, the edge fluid logic execution unit receives a blinding disturbance flag. Then, subsequent sampled bytes are directly written to the invalid empty area. At this time, the effective length of the original waveform buffer is set to zero, and the framing routine reads only an empty length descriptor.
[0106] Furthermore, destination address rewriting removes the bus transport action from the original waveform buffer; clearing the effective length blocks subsequent frame entry; and invalid empty areas absorb residual bytes, causing on-chip sampled data to lose its basis for continued uploading.
[0107] If step three leaves no local information after clearing the field, the subsequent baseline restoration process will only see an interrupted session and will be unable to distinguish between film coverage, physical congestion, or time window mismatch. Conversely, if the entire original waveform traffic flow is retained, the heavy load will be brought back to the system.
[0108] To this end, the edge fluid logic execution unit has two thin chains. The first thin chain extracts a short leading edge segment from the end of the sensor first-in-first-out buffer before the direct memory access controller redirects it and writes it into the fixed page of the ferroelectric memory using an independent serial peripheral interface. Only a few sampling points before and after the trigger are retained. It does not participate in the assembly of uplink messages and serves as a local reference for subsequent downhole reset or long-cycle baseline recovery.
[0109] The second thin chain directly generates the semantic mask from the state determined in step two. Retain large sections of business logic. The relationship is:
[0110]
[0111] In the formula: semantic mask : Represents the compressed single-byte semantic code of the current local physical blinding event, used as the core business content output to the communication module in step three, with a value ranging from 0 to 15; Blinding disturbance flag bit Used to occupy semantic masks The most significant bit, used to indicate whether the current event has been determined by step two to be a false situation of local physical blinding, takes a value of 0 or 1; hierarchical conflict bit. From the first spatial potential energy level With the second flow field kinetic energy level Does the difference exceed the hierarchical difference threshold? The result can be either 0 or 1.
[0112] Slope Excitation Site : By dynamic pressure slope Has the trigger slope threshold been exceeded? The value obtained is either 0 or 1; the effective bits of the time window. : by asymmetric compensation time window Does it fall below the time window? With the upper bound of the time window The value obtained between these parameters is either 0 or 1.
[0113] First spatial potential energy level : indicates the extreme value of the potential energy in the first space in step two. The discrete quantization results are used to participate in hierarchical conflict bits. Generation; second flow field kinetic energy level : indicates the extreme value of the kinetic energy of the second flow field in step two. The discrete quantization results are used to participate in hierarchical conflict bits. Generation; Hierarchical difference threshold : Indicates the boundary that distinguishes between normal coupling and significant divergence, used to limit hierarchical conflict bits. The trigger condition takes a positive integer value;
[0114] dynamic pressure slope : Represents the rate of change of forced fluid dynamic pressure in step one, used to define the slope excitation potential. Triggering conditions; trigger slope threshold : Indicates the slope of dynamic pressure Effective excitation boundaries are used to form slope excitation sites. The discrimination threshold takes the value of a positive real number; the asymmetric compensation time window : Indicates the waiting time for staggered start-up in step one, used to form the effective bit of the time window. The object to be judged has a value greater than 0;
[0115] lower bound of time window : Represents the shortest possible propagation time for the same wavefront event, used to form the effective bit of the time window. The lower bound of the time window takes a value greater than 0; the upper bound of the time window. : Indicates the longest allowed propagation time for the same wavefront event, used to form the effective bit of the time window. The upper boundary of the time window has a value greater than the lower boundary of the time window. ;
[0116] Preferably, the hexadecimal encoding corresponds to the bit definition as follows: when , , , At that time, semantic mask for ;when , , , At that time, semantic mask for .
[0117] During on-site implementation, the edge fluid logic execution unit sets the blinding disturbance flag bit. Hierarchical conflict position , slope excitation site and effective position of time window The same byte is written, and then this byte, along with the node identifier, event sequence number, and checksum, is placed in the short packet register. Ferroelectric memory records only short leading-edge segments, not the complete original waveform traffic flow.
[0118] Furthermore, local references are preserved in the front-end segment recording; semantic masking. Multi-source determination is converted into stable bytes; node identifiers and event sequence numbers enable the short packet register content to be delivered directly, instead of using the original waveform service flow.
[0119] Furthermore, after fixing the fields in the short packet register, the edge fluid logic execution unit no longer runs the business message encapsulation thread. Instead, it uses the communication module to enter the control plane micro-payload path, and the optimal order of the short packet register fields is node identifier, event sequence number, and semantic mask. The checksum and edge fluid logic execution unit writes the above fields into the communication module command first-in-first-out queue and positions the micro-load delivery trigger bit. .
[0120] Micropayload delivery establishment refers to the semantic mask The command has been written into the communication module's first-in-first-out queue and accepted by the communication module's local transmit state machine. The edge fluid logic execution unit does not use the acknowledgment message from the cellular network side as the trigger condition for the RF link suppression operation. After the micro-payload delivery is established, only a local chopping grace period is reserved to avoid immediate power failure; when the local chopping grace period ends, the digital control pin flips, driving the power switching components to cut off the main power supply backbone of the communication module.
[0121] In a preferred embodiment, the communication module and the edge fluid logic execution unit are connected via an asynchronous serial port. The edge fluid logic execution unit first writes the control plane non-service payload instruction header to the communication module, and then writes the node identifier, event sequence number, and semantic mask. The checksum is then immediately checked, and the service layer retransmission threshold is closed, preventing the original waveform service flow from re-entering the queue. In a parallel implementation, if a programmable logic device is used to perform part of the determination in step two, the edge fluid logic execution unit still only receives the blinding disturbance flag bit. The format of the short packet register remains unchanged, along with the corresponding bit status.
[0122] In terms of on-site operation, after the communication module completes the short packet payload, the edge fluid logic execution unit only retains one deep short packet shadow register, and does not generate a second service payload, nor does it convert the front-end segment recording content into an uplink attachment. If the communication module is in a busy state, the edge fluid logic execution unit maintains the micro-payload delivery trigger bit. This is a one-time state and is not extended into a business layer queuing chain; instead, subsequent processing is delegated to step four.
[0123] Furthermore, the fixed field order ensures that the communication module receives narrow payloads; micro-payload delivery trigger bits Located at the beginning of step four; after the service layer retransmission threshold is closed, step three will no longer pull back the broadband service flow due to a single micro-load transmission failure.
[0124] Next, step three will use the blinding perturbation flag output in step two. It is directly converted into bus-level redirection action, ferroelectric memory leading edge segment recording action, and control plane micro-load delivery action, and is determined to be false. The cut-off business flow is also implemented in the same edge fluid logic execution unit.
[0125] Step 4: After the micro-load on the control surface is loaded, the edge fluid logic execution unit drives the power switching components to cut off the power supply backbone of the communication module with an extremely short and wide limit, and releases the full pass-through path when no effective reference waveform is obtained for consecutive cycles.
[0126] Step three has already cut the original waveform traffic flow into a semantic mask. However, if the edge fluid logic execution unit still follows the conventional cellular communication process, waiting for link confirmation, network release, and the end of the upper-layer session, the baseband clock, RF amplification link, and retransmission counter inside the communication module will continue to operate. For underground closed pipe networks with absolutely no mains power access, this tail loss of short packets already sent but hardware still awake is precisely the direct source of passive battery avalanche deadlock.
[0127] Therefore, this step does not consider the control surface micro-load delivery as the process endpoint, but rather as the only grace period entry point before the entity is powered off. All the following actions are performed by the edge fluid logic execution unit.
[0128] In a preferred embodiment, the power switching device is a high-side P-channel field-effect transistor, with its source connected to the positive terminal of the battery, its drain connected to the main power supply trunk of the communication module, and its gate connected to the digital control pin of the edge fluid logic execution unit via a gate current-limiting resistor. A pull-up resistor is connected in parallel between the gate and source, and an energy storage capacitor and a transient suppression device are connected in parallel at the power input of the communication module. A high-side load switch with equivalent conduction function can also be used instead. With this arrangement, the edge fluid logic execution unit is semantically masked. With the communication module command FIFO queue already written, simply flipping one digital control pin can push the RF link suppression operation down to the power supply backbone level, without relying on the service layer cancellation command.
[0129] The edge fluid logic execution unit first reads the micro-load delivery trigger bit. and semantic mask Then, the chopper width limit counter is restarted. During the operation of the chopper width limit counter, the communication module is only allowed to complete the physical layer transmission of the current control plane micro-load, and no new service loads are loaded, the original waveform service flow is not re-expanded, and the network session response is not waited for. After the chopper width limit counter reaches its final value, the edge fluid logic execution unit flips the digital control pin, causing the power switching components to enter the cutoff state, directly cutting off the power supply of the communication module, the RF power amplifier stage and the corresponding high-frequency reference clock.
[0130] After completing the RF link suppression operation, the edge fluid logic execution unit continues to sequentially shut down the asynchronous serial port, the direct memory access controller sampling channel, and the second controlled power supply branch, retaining only the independent real-time clock domain and the lowest-level external interrupt domain. If the independent real-time clock domain repeatedly observes a state where chopping has occurred but no valid reference waveform has appeared within the cumulative threshold duration, the edge fluid logic execution unit suspends the interception path of the discrete cross logic matrix and releases the full pass-through path during the next wake-up round, thereby distinguishing between long-term silence caused by heavy oil pollution, mud, and slow blockage and short-lived pseudo-events caused by local physical blinding.
[0131] A valid reference waveform refers to a waveform that has not triggered the blinding disturbance flag. The original waveform service flow was not cleared and the sampling record that was written to the original waveform buffer after a complete sampling window was not completed is not included. Baseline valid bits. The value is 1 when the valid reference waveform exists in the current cycle, and 0 otherwise. The independent real-time clock domain uses the baseline valid bit... As a reset condition for the cumulative duration across cycles.
[0132] If the digital control foot is in the semantic mask If the command is pulled immediately upon entering the command FIFO queue, the communication module has not yet handed over the control plane micro-load to the underlying transmitting state machine, and the short packet will occur simultaneously with the power cut-off. If the network confirmation is waited for before the cut-off, the communication module will remain resident for a long time under weak coverage. Therefore, the edge fluid logic execution unit is only given a chopping grace period constrained by the local state machine. .
[0133] Preferably, the edge fluid logic execution unit writes the semantic mask. Immediately afterwards, the micro-load delivery trigger bit is set. The chopping moment is determined by the earlier of the communication module's transmitted status feedback and the local grace limit upper limit. The relationship is written as:
[0134]
[0135] In the formula: chopping width duration : Indicates the trigger bit for delivery from the microload The maximum allowed retention period between setting and the power switch component performing RF link suppression operation; a value greater than 0; load confirmation time. : Indicates the semantic mask received by the communication module The local time elapsed until the internal state machine confirms loading completion, with a value greater than 0; the maximum grace period. : Indicates the maximum local waiting time preset by the edge fluid logic execution unit to avoid the communication module from entering a long-term dwell period. It is preferably a radio frequency power-off limit grace period of no more than fifteen milliseconds, and the value is greater than 0.
[0136] In terms of on-site actions, the edge fluid logic execution unit will use semantic masks. After the node identifier and event sequence number are written into the communication module's command first-in-first-out queue, the chopper tolerance counter is immediately started. If the communication module quickly returns loading completion feedback, the chopper tolerance counter ends at that feedback time; if the communication module does not provide feedback for a long time, the chopper tolerance counter remains at the upper tolerance limit for the specified duration. Upon arrival, the process ends. Subsequently, the digital control pin flips, the gate potential of the P-channel MOSFET rises, and the main power supply backbone of the communication module is physically cut off.
[0137] Furthermore, the choke time limit To prevent the control surface micro-load from being lost due to premature power-off; maximum grace period. Prevent the communication module from lingering in weak coverage for extended periods; the digital control pin is directly connected to the gate of the power switching device, so that the cut-off action no longer depends on the upper-layer protocol for completion.
[0138] Furthermore, if only the power supply to the communication module is cut off, while the edge fluid logic execution unit continues to poll the serial port, the direct memory access controller idles, or the sensor-controlled power supply branch remains on, the RF tail power will be reduced, but the on-chip tail power will still be prolonged. Therefore, the edge fluid logic execution unit fixes its actions after power-down to first lock, then stop, and then sleep.
[0139] Specifically, after the digital control pin successfully flips, the edge fluid logic execution unit first applies the semantic mask. Transmission sequence number and blinding perturbation flag Write to the low-power hold register, disable asynchronous serial port transmission interrupt, and freeze the short packet register write entry; stop the direct memory access controller sampling channel, shut down the first controlled power supply branch and the second controlled power supply branch, and the first spatial potential energy sensing component and the second flow field kinetic energy sensing component lose their main power supply respectively; switch the main clock to a low-speed clock, retain the real-time domain and external interrupt domain, and do not supply power to the hold register.
[0140] In the parallel implementation, if step two uses the programmable logic device (PLD) path, the edge fluid logic execution unit (EDU) simultaneously pulls down the clock enable pin of the PLD while performing the aforementioned convergence action, keeping the PLD in a static hold state. If the control plane release auxiliary signaling path is used, the EDU first issues the release auxiliary signaling, and then performs the same chopping width limit. The entity is powered down.
[0141] The observed behavior on-site is as follows: after the communication module status light goes out, the edge fluid logic execution unit no longer retryes serial port loading, no longer reopens the original waveform service flow, and no longer wakes up any sensing components. Instead, it pushes the entire node back to a sleep configuration that only retains the wake-up entry. This, in turn, maintains the minimum state across cycles in the register area; the serial port and direct memory access controller stop sequentially, eliminating on-chip overhead.
[0142] Furthermore, when a node remains in a state of chopping immediately after each short packet for an extended period, only fragmented semantic masks will be seen at the remote end. No new valid reference waveform can be obtained. For gradual flow blockage caused by heavy oil and silt, relying solely on the discrete cross-logic matrix in step two will result in a long-term interception conclusion, hindering the recovery of the baseline at the far end. In this case, independent real-time clock domain observation across cycles is required.
[0143] Preferably, after each RF link suppression, the edge fluid logic execution unit writes whether a valid reference waveform has been obtained this week into the holding register, and the observation time is accumulated by an independent real-time clock domain. If the accumulated observation time exceeds a threshold and a valid reference waveform is never found, the next wake-up will directly release the full pass-through path. The relationship is as follows:
[0144]
[0145] Where: Reset release flag bit : Indicates whether the discrete cross-logic matrix interception path should be suspended and the full pass-through path should be used instead during the next wake-up. This is used to handle long-term silence caused by slow structural deadlocks. The value is 0 or 1; Watchdog cumulative duration. : Represents the cumulative duration across cycles of the independent real-time clock domain since the most recent valid reference waveform appeared, with a value greater than or equal to 0; recovery threshold duration : Indicates the minimum time required to trigger full pass-through release, preferably 72 hours, to avoid bypassing the discrete cross logic matrix in short-term anomalies, and the value is greater than 0;
[0146] Baseline effective bits : Indicates whether a valid reference waveform that can be used to update the remote baseline was obtained in this cycle, used to distinguish between two states: long-term lack of baseline and recovery of existing baseline; value is 0 or 1; Logical AND operator : Indicates that both the preceding and following conditions are met; logical OR operator : indicates that either the preceding or following condition is satisfied;
[0147] In terms of on-site operation, when the independent real-time clock domain continuously observes that a node always completes short packets and chopping but never generates a valid reference waveform, after being woken up by an external interrupt or watchdog timer in the next round, the edge fluid logic execution unit first queries the reset release flag bit. If the reset release flag is bit If the value is 1, then the interception path of the discrete cross-logic matrix will no longer be based on the blinding perturbation flag. Instead of performing the original waveform service flow clearing in step three, the complete data obtained by the first spatial potential energy sensing component and the second flow field kinetic energy sensing component is sent up once along the full transmission path for the remote system to rebuild the baseline.
[0148] Watchdog cumulative time Separate from the long-term silent state; reset the release flag. This provides an entry point for degradation protection; full pass-through is only enabled once after the cross-cycle threshold is triggered, without affecting normal RF link suppression operations. Chopper grace period. To minimize the transmission opportunity for the control plane micro-load, the high-side cutoff of the power switching components pushes the communication tail loss back to the power supply backbone. The global lockout synchronously pushes the remaining on-chip power consumption back to the sleep domain, and the reset release flag is activated. This provides a single-pass pathway for slow-moving structural blockages to recover from baseline over a long period.
[0149] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0150] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0151] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0152] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0153] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A municipal drainage pipe network intelligent monitoring system based on multi-source sensing, comprising a primary potential energy sensing component, a first spatial potential energy sensing component, a second flow field kinetic energy sensing component, an edge fluid logic execution unit, a communication module, and power switching components on its power supply backbone, characterized in that, The output of the primary potential energy sensing component is interrupted by the forced fluid dynamic pressure step jump. The edge fluid logic execution unit responds to the step interrupt wake-up and sequentially pulls up the first spatial potential energy sensing component and the second flow field kinetic energy sensing component according to the asymmetric compensation time window, acquires discrete multimodal environmental parameters and inputs them into the discrete cross logic matrix, and outputs the blinding disturbance flag bit. The edge fluid logic execution unit responds to the blinding disturbance flag, redirects the storage path of the original waveform service flow, and generates a semantic mask; After the edge fluid logic execution unit loads the semantic mask into the communication module, it drives the power switching components to cut off the power supply backbone of the communication module.
2. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 1, characterized in that: After responding to a step interruption, the edge fluid logic execution unit first turns on the controlled power supply branch of the first spatial potential energy sensing component and obtains a stable echo. It then extracts the first spatial potential energy extreme value within the first sampling window, determines the asymmetric compensation time window based on the pipe diameter, Manning roughness coefficient, and installation spacing in the calibration table, and turns on the controlled power supply branch of the second flow field kinetic energy sensing component and releases its interface clock after the asymmetric compensation time window ends.
3. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 2, characterized in that: The edge fluid logic execution unit quantizes the first spatial potential energy extreme value and the second flow field kinetic energy extreme value into the first spatial potential energy level and the second flow field kinetic energy level, respectively. It then inputs the first spatial potential energy level, the second flow field kinetic energy level, the forced fluid dynamic pressure step characteristics, and whether the asymmetric compensation time window falls within the preset interval into the local preset truth table to output the blinding disturbance flag.
4. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 3, characterized in that: When the blinding disturbance flag is valid, the edge fluid logic execution unit freezes the current sampling descriptor, rewrites the destination storage address of the direct memory access controller from the original waveform buffer to the invalid empty area, and clears the effective length register of the original waveform buffer. The edge fluid logic execution unit then generates a semantic mask based on the blinding disturbance flag, hierarchical conflict state, slope excitation state, and time window validity state.
5. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 1, characterized in that: The edge fluid logic execution unit writes the semantic mask, node identifier, event sequence number, and check word into the short packet register in a fixed field order, and loads the contents of the short packet register into the command first-in-first-out queue of the communication module. The communication module only receives the short packet data corresponding to the fixed field order for the current abnormal event, and does not receive the original waveform service flow corresponding to the current abnormal event.
6. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 5, characterized in that: After receiving the current abnormal event in the command first-in-first-out queue, the communication module first enters the control plane micro-payload transmission state, and then outputs the release auxiliary signaling. After the communication module enters the control plane micro-load transmission state, the edge fluid logic execution unit keeps the original waveform service flow from entering the broadband service transmission queue of the communication module and maintains the single-depth shadow register of the short packet register area.
7. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 6, characterized in that: The communication module encapsulates the control plane micro-load of the current abnormal event using a non-IP data delivery message frame format. The message frame includes at least a node identifier, event sequence number, semantic mask, and checksum. After the message frame is encapsulated, the original waveform service stream corresponding to the current abnormal event will no longer be received, and the broadband service transmission queue will remain closed until the power switch device operates.
8. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 1, characterized in that: The power switching device is a high-side switching device connected in series between the positive terminal of the battery and the power supply backbone of the communication module. The control terminal of the high-side switching device is directly connected to the digital control pin of the edge fluid logic execution unit through a current-limiting resistor. The power supply input of the communication module is connected in parallel with an energy storage capacitor and a transient suppression device, and the high-side switching device is connected in series with the main power supply backbone of the communication module.
9. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 8, characterized in that: After the semantic mask is loaded into the communication module, the edge fluid logic execution unit starts the chopper width limit count and flips the digital control pin when the chopper width limit count ends. After the high-side switching device is turned off, the asynchronous serial port, the direct memory access controller sampling channel, and the controlled power supply branches corresponding to the first spatial potential energy sensing component and the second flow field kinetic energy sensing component are turned off in sequence, leaving only the independent real-time clock domain and the external interrupt domain.
10. The intelligent monitoring system for municipal drainage pipe networks based on multi-source sensing according to claim 9, characterized in that: The edge fluid logic execution unit records the sampling records that have not triggered the blinding disturbance flag and have completed a full sampling window and written to the original waveform buffer as valid reference waveforms, and accumulates the duration of continuous chopping without forming a valid reference waveform in the independent real-time clock domain; when the accumulated duration reaches 72 hours, it generates a reset release flag and releases the full pass-through path of the complete sampling data in the next wake-up.
Citation Information
Patent Citations
A flow monitoring device for urban drainage pipeline
CN107727179B
