Intelligent early warning indicating device and early warning method for channel siltation

Abstract

The invention discloses a channel siltation intelligent early warning indicating device and an early warning method based on the device, and the device comprises a pile body unit and a multi-mode monitoring unit which is arranged on the pile body unit and is used for calculating the thickness and distribution of siltation by monitoring the underwater pressure change in real time. And the intelligent analysis unit is used for calculating a siltation risk index according to the thickness and distribution of the siltation. The device provided by the invention realizes real-time monitoring, accurate prediction and intelligent early warning of the channel siltation condition. The gradient density PVC pile body structure is adopted, the lightweight and anti-scouring requirements are considered, the water flow impact force is remarkably reduced, the service life of the pile body is prolonged, the manual cleaning frequency can be reduced, and the maintenance cost is reduced. Through the spiral multi-sensor layout scheme, the spatial resolution is remarkably improved, and the accuracy and reliability of early warning are improved. By dynamically adjusting the monitoring frequency and introducing a sliding window mechanism, a sudden deposition event can be quickly responded, and the early warning capability of the system is improved.

Description

Technical Field

[0001] The present invention relates to the field of waterway engineering and intelligent monitoring technology, and in particular to an intelligent early warning indicating device for waterway siltation and an early warning method based on the device. Background Art

[0002] As an important channel for water transportation, the smoothness of waterways is crucial to the development of the shipping industry. However, the problem of waterway siltation has always been one of the key factors affecting waterway safety and efficiency. Traditional waterway siltation monitoring methods mainly rely on manual regular inspections and simple equipment monitoring, which have problems such as low monitoring efficiency, incomplete data, and inability to provide real-time warnings. Especially under complex and changeable natural conditions, waterway siltation may deteriorate rapidly, posing serious safety hazards to ship navigation.

[0003] Therefore, there is an urgent need for a device that can monitor the siltation of waterways in real time and accurately and has an intelligent early warning function to improve the efficiency of waterway maintenance and ensure waterway safety. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide an intelligent early warning indicator device for waterway siltation and an early warning method based on the device, which utilizes an array of sensors to obtain underwater pressure to infer the distribution of silt to achieve early warning of waterway siltation.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The intelligent early warning indicating device for waterway siltation provided by the present invention comprises a pile unit and a multi-modal monitoring unit and an intelligent analysis unit arranged on the pile unit;

[0007] The multimodal monitoring unit is used to calculate the thickness and distribution of silt by real-time monitoring of underwater pressure changes;

[0008] The intelligent analysis unit is used to calculate the sedimentation risk index through the thickness and distribution of sediment, so as to generate a sedimentation monitoring early warning signal.

[0009] Further, the multimodal monitoring unit includes a thin film pressure sensor array, a controllable release tracer particle module, and a micro ADCP flow meter;

[0010] The thin film pressure sensor array is arranged in a spiral form on the outer wall of the pile body to monitor the underwater pressure changes in real time and calculate the thickness and distribution of the sediment;

[0011] The controllable release tracer particle module includes starch-based fluorescent particles and magnetic iron oxide particles, and the release frequency is controlled by a solenoid valve to invert the flow direction and speed of the sediment;

[0012] The described micro ADCP current meter is used to measure the velocity and direction of water flow, providing hydrological data for sedimentation monitoring.

[0013] Furthermore, the intelligent analysis unit calculates the sedimentation risk index through an improved LSTM-GRU fusion model;

[0014] The input data of the improved LSTM-GRU fusion model are: pressure time series data, flow velocity data, and particle loss amount. LSTM extracts long-term dependence features, and GRU captures short-term dynamic features;

[0015] The output data is: sedimentation risk index.

[0016] Furthermore, it also includes a self-maintenance unit arranged on the pile body unit. The self-maintenance unit includes an ultrasonic anti-fouling transducer;

[0017] The ultrasonic anti-fouling transducer uses cavitation to prevent dirt attachment;

[0018] Furthermore, it also includes an energy cabin. The energy cabin includes energy management and a power generation module, a wave energy power generation module, and an energy storage system that are respectively connected to the energy management;

[0019] The power generation module obtains electrical energy through flexible perovskite solar cells;

[0020] The wave energy power generation module captures wave energy through a mechanical structure and converts it into electrical energy;

[0021] The energy storage system realizes hybrid energy storage through supercapacitors and lithium iron batteries;

[0022] The energy management realizes dynamic allocation of the power supply mode by setting an energy management priority strategy.

[0023] Furthermore, the pile body shell of the pile body unit is a hollow cylindrical structure made of gradient density modified PVC. A coating is provided on its outer shell wall. The coating has a composite photocatalytic-hydrophobic dual function. The density of the gradient density modified PVC changes continuously from the pile top to the pile bottom and is formed by a multi-layer co-extrusion process.

[0024] Furthermore, the coating is composed of nano-TiO2 (50 - 70 wt%), fluorosilicone resin (20 - 30 wt%), and graphene quantum dots (5 - 10 wt%), and the thickness is 80 - 120 μm.

[0025] The present invention provides a method for warning of channel sedimentation using the above-mentioned intelligent warning and indication device for channel sedimentation, including the following steps:

[0026] S1: Deploy indication piles: Calculate a reasonable pile spacing according to the designed flow rate, sediment density, average flow velocity and non-uniformity coefficient of the waterway, and deploy the intelligent early warning indication device for waterway siltation.

[0027] S2: Calibrate the initial flow field: Calibrate the initial flow field through particle image velocimetry technology.

[0028] S3: Dynamically adjust the monitoring frequency: Dynamically adjust the monitoring frequency according to the siltation risk index output by the intelligent analysis unit, set the siltation risk index threshold, and adjust the sampling interval time according to the real-time siltation risk index and the siltation risk index threshold.

[0029] S4: Generate a three-dimensional siltation heat map: Generate a three-dimensional siltation heat map by fitting the spatial distribution with B-spline surfaces.

[0030] Further, a sliding window is adopted in the dynamic adjustment mechanism in step S3, and it is specifically carried out in the following manner:

[0031] Set the window length.

[0032] Calculate the predicted acceleration within the window and determine whether the siltation acceleration exceeds the threshold; when it is detected that the siltation acceleration exceeds the threshold, trigger the emergency mode and start the collaborative monitoring of the surrounding piles.

[0033] Further, the dynamic adjustment of the monitoring frequency in step S3 is specifically carried out in the following manner:

[0034] Calculate the siltation risk index R.

[0035] When R ≤ 30%, the sampling interval is set to 60 - 120 minutes, and routine monitoring is performed.

[0036] When 30% < R ≤ 60%, the interval is shortened to 30 - 60 minutes and the data verification function is activated.

[0037] When 60% < R ≤ 80%, the interval is further compressed to 10 - 30 minutes, and an audible and visual warning is triggered synchronously.

[0038] If R > 80%, ultra-high frequency sampling is performed at 5 - 10 minutes, and collaborative operation of the surrounding piles is linked.

[0039] The beneficial effects of the present invention are as follows:

[0040] The intelligent early warning and indication device for channel siltation provided by the present invention, based on the early warning method of this device, realizes real-time monitoring, accurate prediction and intelligent early warning of the channel siltation situation. In terms of structural design, the first-of-its-kind gradient density PVC pile structure takes into account both the lightweight and anti-scouring requirements, significantly reducing the water flow impact force and extending the service life of the pile. The photocatalytic-hydrophobic dual-functional coating realizes the synergistic effect of "self-cleaning + anti-biological attachment", reducing the manual cleaning frequency and lowering the maintenance cost. In terms of monitoring technology, the spiral multi-sensor layout scheme significantly improves the spatial resolution and provides more comprehensive monitoring data. The hybrid energy system enables uninterrupted power supply throughout the year, and the endurance in extreme weather has been increased by 200%. In terms of system integration, the linkage with the drone inspection system further improves the accuracy and reliability of the early warning. By dynamically adjusting the monitoring frequency and introducing a sliding window mechanism, it can quickly respond to sudden siltation events and improve the early warning ability of the system.

[0041] Other advantages, objectives and features of the present invention will be elaborated to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to make the objectives, technical solutions and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration.

[0043] Figure 1 It is a schematic structural diagram of the present invention.

[0044] Figure 2 It is an internal cross-sectional view.

[0045] Figure 3 It is an enlarged view of the thin film sensor.

[0046] Figure 4 It is an enlarged view of the controllable release tracer particle module of component 5.

[0047] Figure 5 It is a flowchart of the intelligent early warning method for channel siltation.

[0048] In the figure, solar panel 1, super capacitor 2, wave energy generation module 3, transducer 4, controllable release tracer particle module 5, current meter 6, counterweight 7, first chamber 8, solenoid valve 9, second chamber 10, coating 11, pressure sensor 12, sensitive element 13, thin film layer 14, pins and interfaces 15, electrode layer 16, signal conditioning circuit 17, imager 18, pulsed laser 19, tracer particle nozzle 20. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it, but the embodiments cited do not limit the present invention.

[0050] Embodiment 1

[0051] As Figure 1 and Figure 2 shown, the intelligent early warning indication device for channel siltation provided in this embodiment includes a pile body unit and a multi-modal monitoring unit, an intelligent analysis unit, an energy cabin, and a communication unit provided on the pile body unit;

[0052] The pile body unit: The pile body shell is a hollow cylindrical structure made of gradient density modified PVC, and a coating 11 is provided on its outer shell wall. The coating has a composite photocatalytic-hydrophobic dual function, and the interior is divided into a sensor cabin, a counterweight cabin, and an energy cabin. The density of the gradient density modified PVC changes continuously from the pile top (1.2 g / cm 3 ) to the pile bottom (2.3 g / cm 3 ), and is formed by a multi-layer co-extrusion process to enhance the anti-scouring ability and stability of the pile body. The photocatalytic-hydrophobic dual-functional coating is composed of nano-TiO2 (50-70 wt%), fluorosilicone resin (20-30 wt%), and graphene quantum dots (5-10 wt%), with a thickness of 80-120 μm, which can effectively prevent biological attachment and decompose dirt under light, realizing the self-cleaning function.

[0053] As Figure 3 shown, Figure 3 is an enlarged view of the thin-film pressure sensor. The thin-film pressure sensor array is arranged in a spiral form on the outer wall of the pile body. The axial distance between adjacent sensors is 15 cm, the circumferential interval is 60°, the range is 0-100 kPa, and the accuracy is ±0.5% F. The thin-film pressure sensor 12 is composed of elements such as a thin-film layer 14, a sensitive element 13, an electrode layer 16, a signal conditioning circuit 17, and pins and interfaces 15 in an integrated manner. During operation, the thin-film layer first bears the external pressure and undergoes elastic deformation. The sensitive element is closely attached to the thin-film layer, can sense this deformation, and convert it into an electrical signal. The electrode layer is responsible for conducting the electrical signal generated by the sensitive element to the signal conditioning circuit, and the signal conditioning circuit amplifies, filters, and performs analog-to-digital conversion on the signal to make it a standard signal that can be read by external devices. The shell and encapsulation play the role of protecting the internal components, preventing external interference, and providing mechanical support at the same time. Finally, the pins and interfaces output the conditioned signal to external devices

[0054] The multimodal monitoring unit includes a pressure sensor array arranged in a longitudinally distributed manner, a controllable release tracer particle module, and a micro ADCP flowmeter 6. The pressure sensor is a thin film pressure sensor; the thin film pressure sensor array is arranged in a spiral form on the outer wall of the pile body, the axial spacing between adjacent sensors is 15cm, the circumferential spacing is 60°, the range is 0-100kPa, and the accuracy is ±0.5% FS. It is used to monitor underwater pressure changes in real time and calculate the thickness and distribution of silt. The tracer particle module includes starch-based fluorescent particles and magnetic iron oxide particles, with a particle size distribution of 1-3mm and a density of 1.1-1.8g / cm 3 The release frequency is controlled by the electromagnetic valve 9 to invert the flow direction and speed of the sediment. The micro ADCP flow meter is used to measure the speed and direction of the water flow and provide important hydrological data for sedimentation monitoring.

[0055] The intelligent analysis unit: has a built-in improved LSTM-GRU fusion algorithm for siltation trend modeling. The algorithm uses the input pressure time series data, flow rate data and particle loss, extracts long-term dependent features through the LSTM layer, captures short-term dynamic features through the GRU layer, and fuses these features to output the siltation risk index, thereby achieving accurate prediction and early warning of siltation trends;

[0056] The energy cabin includes a flexible perovskite solar cell and a wave power generation module 3, with a comprehensive conversion efficiency of more than 18%. It is equipped with a supercapacitor and lithium iron battery hybrid energy storage system, which supports 30 days of off-grid operation, ensuring the long-term stable operation of the system, and can work normally even in bad weather or no light conditions.

[0057] The communication unit adopts LoRaWAN communication: it adopts adaptive spreading factor (SF7-SF12) to achieve long-distance transmission (covering radius 5-15km) in low-power mode, which is suitable for near-shore or areas covered by base stations. It supports multi-channel frequency hopping technology, has outstanding anti-interference ability, and a bit error rate of less than 10 -6 .

[0058] Dynamic switching strategy: Based on the joint judgment of Beidou satellite elevation angle (Beidou is activated when >30°) and LoRaWAN signal strength (RSSI>-120dBm is prioritized), the microcontroller is used to achieve seamless switching of transmission modes, with a switching delay of less than 2 seconds. The built-in transmission energy consumption optimization algorithm dynamically adjusts the transmission frequency and power according to the data priority (such as early warning signal > conventional monitoring data).

[0059] Protocol conversion and data encapsulation: Convert the siltation risk index and raw sensor data (pressure, flow rate, particle tracking information) output by the intelligent analysis unit into a standard protocol format (such as MQTT-SN or custom binary frame).

[0060] Interaction with the intelligent analysis unit: Receive the sedimentation risk index, original data packet and early warning instructions processed by the intelligent analysis unit through RS-485 or CAN bus. Feedback the communication status (such as signal strength, transmission success rate) to the intelligent analysis unit to assist in optimizing algorithm parameters (such as adjusting the particle release frequency).

[0061] Coordinated with the power supply of the energy cabin: directly connected to the lithium-titanate battery pack (48V / 20Ah) of the energy cabin, and adapted to the working voltage of the communication unit (5V / 12V) through the DC-DC conversion module.

[0062] Send power consumption logs to the energy cabin to trigger the self-maintenance unit to clean the solar panels or perform battery balancing management.

[0063] Collaboration with multimodal monitoring units: Receive the original hydrological data stream (sampling rate 1Hz) from the micro ADCP flowmeter, package and transmit after preprocessing. Control the solenoid valve of the tracer particle module and adjust the particle release strategy according to the monitoring center instructions fed back by the communication unit.

[0064] Physical integration with the pile unit: The communication module is encapsulated in the sensor cabin of the pile unit. The antenna is extended to the top of the pile through a waterproof feeder and adopts an omnidirectional gain antenna design (gain ≥ 3dBi).

[0065] The external interface uses an IP69K-level waterproof aviation plug, which seamlessly connects to the sensor cable on the outer wall of the pile body and the power supply line of the energy cabin.

[0066] Linkage with self-maintenance unit: Receive status reports from self-maintenance unit (such as coating wear, sensor health status) and upload them to the monitoring center simultaneously. Trigger fault repair mechanism of self-maintenance unit (such as restarting abnormal sensor, activating backup power supply).

[0067] Through dual-mode communication, seamless connection between nearshore (LoRaWAN) and offshore (Beidou) is achieved, solving the geographical limitations of traditional single communication methods. The dynamic switching strategy is combined with the data retransmission mechanism to ensure that the success rate of key warning information transmission is ≥ 99.9%. In the intelligent sleep mode, the standby power consumption of the whole machine is less than 0.5W, extending the continuous working time of the equipment under no sunshine conditions to 30 days. The antenna and circuit board adopt triple-proof (anti-salt spray, anti-mildew, anti-moisture) coating, meet the IP68 protection level, and adapt to -40℃~+85℃ operating temperature.

[0068] like Figure 2 As shown, Figure 2The first chamber in it is vertically arranged on the left side of the pile body and is integrated with the solar panel at the top of the pile body, serving as the energy storage of the pile body. It includes a solar power generation panel 1, a super capacitor 2, and a transducer 4 for supplying energy to the pile body. The second chamber 10 is horizontally arranged at the lower left side of the pile body for signal collection, transmission, and processing of the pile body. It includes a multi-modal monitoring unit, an intelligent analysis unit, and a communication unit for data collection, transmission, and processing.

[0069] This embodiment provides a method for warning of channel siltation using the above-mentioned intelligent warning indication device for channel siltation, including the following steps:

[0070] S1: Deploying indicating piles: Obtain the cross-sectional hydrological data (flow rate, flow velocity, bottom slope) of the channel and deploy the intelligent warning indication device for channel siltation. Calculate the theoretical spacing through the critical siltation length formula L c =(Q·ρ s ·η) / (v·ρ w ·S0), where Q represents the flow rate; ρ s represents the sediment density; η represents the sediment settlement efficiency; v represents the flow velocity; ρ w represents the density of water; S0 represents the bottom slope; and introduce a correction factor α (with a value of 0.1 - 0.3) to adjust the actual spacing L = L c / (1 + α·(C u ―1)), where C u represents the non-uniformity coefficient; ensure coverage of the spatial heterogeneity of sediment deposition; during implementation, it is necessary to collect channel hydrological and sediment data, input the algorithm to generate a recommended spacing, verify the monitoring coverage rate through Monte Carlo simulation (≥90% is qualified), and arrange verification piles (at 0.8L, L, 1.2L) in typical siltation sections for engineering verification. Finally, determine a reasonable layout plan in combination with the minimum monitoring spacing constraint (10 - 20 meters). This method can dynamically adapt to different channel scenarios by quantifying and coupling hydrodynamic parameters and sediment characteristics, taking into account both monitoring efficiency and economy, and can accurately guide construction and reduce the trial-and-error cost.

[0071] S2: Calibrate the initial flow field: Calibrate the initial flow field through Particle Image Velocimetry (PIV) technology. By uniformly spreading tracer particles (particle size 10 - 50μm) into the water body, use a double-pulse laser sheet light source (wavelength 532nm) to vertically irradiate the area of the flow field to be measured, synchronously trigger a high-speed CMOS camera (frame rate ≥500Hz, resolution 2048×2048) to capture the particle displacement images, and use the cross-correlation algorithm (window size 32×32 pixels, overlap rate 75%) to calculate the particle displacement between adjacent frames to generate a two-dimensional velocity vector field; before calibration, use a grid calibration board (accuracy ±0.1mm) to correct the spatial coordinates, and optimize the flow velocity mapping relationship by combining the measured data of the micro ADCP. Finally, establish a three-dimensional flow velocity distribution model of the initial flow field with an accuracy of ±2% FS, providing high-precision reference flow field data for sedimentation monitoring. This technology can accurately measure the velocity and direction of water flow, providing basic data for subsequent sedimentation monitoring.

[0072] S3: Dynamically adjust the monitoring frequency: The system dynamically adjusts the monitoring frequency according to the sedimentation risk index R (range 0 - 100%): When R ≤ 30%, the sampling interval is set to 60 - 120 minutes for routine monitoring; when 30% < R ≤ 60%, the interval is shortened to 30 - 60 minutes and the data verification function is activated; when 60% < R ≤ 80%, the interval is further compressed to 10 - 30 minutes, and the acoustic and optical early warnings are synchronously triggered; if R > 80%, then ultra-high-frequency sampling is performed at 5 - 10 minutes (efficiency ≥ 85%), and the surrounding piles are linked for collaborative operation. An emergency response mechanism is added: When the risk index acceleration ΔR / Δt ≥ 5% / min in three consecutive samplings, immediately switch to the intensive monitoring mode of 1 - 3 minutes and push an emergency alarm to the monitoring center. This dynamic adjustment mechanism can optimize the use of monitoring resources according to the actual situation and improve the response speed of the system.

[0073] In the dynamic adjustment mechanism of this embodiment, a sliding window is adopted, the window length is 6 - 24 hours, and the sedimentation acceleration threshold is set to 0.05m / s 2- 0.08m / s 2 ; The data segments are divided according to fixed time slices (default 10 minutes) within the window, and the average acceleration value of each time slice is calculated in real time and stored in the circular buffer; when the acceleration mean value of any three consecutive time slices within the window exceeds the set threshold (0.05 - 0.08m / s 2 ), or the slope of the overall acceleration trend of the window ≥ 0.01m / s 2 ·h-1, it is determined as abnormal acceleration, and the emergency mode is immediately triggered to start the collaborative monitoring of all piles within 3km around; this mechanism can quickly respond to sudden sedimentation events and improve the early warning ability of the system. After each sampling is completed, the window slides forward by one time slice (10 minutes), the earliest data segment is eliminated and the new data is incorporated to achieve dynamic coverage of the latest working conditions.

[0074] S4: Generate a three-dimensional sedimentation heat map: The system first collects data on the sedimentation thickness, flow velocity, and tracer particle distribution of the waterway in real time through multimodal sensors, and discretizes it into high-precision three-dimensional grid cells. An initial control point network is generated based on the thin plate spline interpolation algorithm, and the positions of the control points are adjusted through an optimization algorithm to make the model accurately match the actual sedimentation morphology. During the fitting process, control nodes are automatically densified in areas with drastic sedimentation changes (such as bends and gates) to improve the ability to restore local details. At the same time, sensor noise interference is eliminated through a smoothing constraint algorithm to ensure the continuity and stability of the model. After the fitting is completed, the system maps the sedimentation data to the color-opacity space (high transparency in red indicates high-risk sedimentation, and low transparency in green indicates a safe area), and integrates it into a three-dimensional geographic information platform to achieve real-time rendering and interactive analysis of the dynamic heat map (such as cross-section cutting and contour extraction). A three-dimensional sedimentation heat map is generated by fitting the spatial distribution with B-spline surfaces. This visualization method can intuitively display the distribution of sedimentation, facilitating managers to make decisions quickly.

[0075] Example 2

[0076] In this embodiment, the pile body unit in the intelligent early warning indicator device for waterway sedimentation uses a gradient density modified PVC pile body: a hollow cylindrical structure with a density continuously changing from the pile top (1.2 - 2.0 g / cm 3 ) to the pile bottom (2.3 - 3.0 g / cm 3 ), and is formed by a multi-layer co-extrusion process to enhance the anti-scouring ability and stability. The pile body is provided with a photocatalytic-hydrophobic dual-functional coating: an outer wall composite coating (80 - 120 μm), composed of nano-TiO2 (50 - 70 wt%), fluorosilicone resin (20 - 30 wt%), and graphene quantum dots (5 - 10 wt%), to achieve self-cleaning and anti-biofouling.

[0077] In this embodiment, the gradient density modified PVC pile body provides structural support to ensure the stability of the indicator pile in water, and at the same time enhances the anti-scouring ability through the density gradient design. The pile body is formed by a multi-layer co-extrusion process during manufacturing to ensure that the density continuously changes from the pile top (2 g / cm 3 ) to the pile bottom (3 g / cm 3 ). When the pile body is installed, it is necessary to ensure that it is vertically inserted into the riverbed or seabed, and a certain height of the top is exposed above the water surface for installing the communication unit and solar cells. The photocatalytic-hydrophobic dual coating prevents biofouling, achieves self-cleaning, and reduces maintenance costs. The coating is uniformly compounded on the outer wall of the pile body during manufacturing, with a thickness of 80 - 120 μm. Under light conditions, nano-TiO2 in the coating can decompose dirt, fluorosilicone resin provides hydrophobic properties, and graphene quantum dots enhance the stability of the coating; in the sensor cabin of the compartment, a multimodal monitoring unit is installed for real-time monitoring of underwater pressure, water flow velocity, and the movement of tracer particles.

[0078] The thin-film pressure sensor array is arranged on the bulkhead in a spiral form. The thin-film pressure sensor array calculates the thickness and distribution of sediment by measuring the pressure changes at different positions. The sensors are arranged on the bulkhead in a spiral form to collect pressure data in real time. Using the hydrostatic principle and combining with the pressure formula P = ρgh (where P is the pressure, ρ is the sediment density, g is the acceleration due to gravity, and h is the sediment thickness), the sediment thickness at each position is inversely deduced. The sediment thickness distribution map of the entire area is generated by the inverse distance weighted method, which is used to monitor the underwater pressure changes and calculate the sediment thickness and distribution.

[0079] The tracer particle module and the micro ADCP current meter are installed at the bottom of the cabin. Tracer particles are regularly released into the water body, sound waves are emitted and reflected signals are received to obtain the water flow velocity components in multiple directions, including the horizontal directions (east-west, north-south) and the vertical direction. By calculating the direction angle of the total velocity vector, the direction of the water flow is determined. Since the movement trajectory of the tracer particles is consistent with the water flow direction and the velocity is equivalent to the water flow velocity, the measured water flow direction and velocity can be regarded as the direction and velocity of sediment flow.

[0080] The counterweight cabin is installed with counterweight blocks 7 to ensure that the pile body remains vertically stable in the water. The counterweight cabin is filled with high-density materials (such as lead blocks or concrete), and through calculation, the center of gravity of the pile body is ensured to be at a suitable position. Drainage holes are reserved on the bulkhead of the counterweight cabin to prevent the accumulated water in the cabin from affecting the stability.

[0081] The energy cabin is installed with flexible perovskite solar cells, wave energy power generation modules and energy storage systems to provide energy for the device. The solar cells are installed on the top of the energy cabin, and the surface is covered with a transparent protective layer to ensure the photoelectric conversion efficiency.

[0082] The wave energy power generation module interacts with the external water flow through a mechanical structure to convert wave energy into electrical energy.

[0083] Supercapacitors and lithium iron batteries are used to store electrical energy, and the charge and discharge are controlled through an intelligent management system to ensure the stability and reliability of the energy supply.

[0084] The multi-modal monitoring unit: The thin-film pressure sensor array monitors the underwater pressure changes in real time and calculates the thickness and distribution of sediment. The sensors are arranged on the outer wall of the pile body in a spiral form, and the axial spacing between adjacent sensors is 15 cm, and the circumferential interval is 60°.

[0085] The sensors convert the pressure signals into digital signals through the data acquisition module and transmit them to the intelligent analysis unit. The change in sediment thickness is calculated through the pressure change rate, and combined with the algorithm of the intelligent analysis unit, the sediment trend is predicted.

[0086] As Figure 4 shown,Figure 4 It is an enlarged view of the tracer particle module for component 5. The tracer particle module is arranged inside the bin, and the tracer particle nozzle 20 extends outwards. After opening the switch of the solenoid valve 9, tracer particles are evenly distributed in the flow field, and they move with the fluid. A pulsed laser 19 is used to generate high-energy laser pulses, and the laser is focused into a thin sheet of light through an optical system to illuminate the tracer particles in the flow field. The tracer particles are imaged in the imager 18 and transmitted to the intelligent analysis unit in the first chamber 8, and the velocity vector of each point in the flow field is calculated based on the displacement and time interval of the particles.

[0087] In this embodiment, the tracer particle module releases tracer particles to invert the flow direction and velocity of the sediment. The tracer particle module includes starch-based fluorescent particles and magnetic iron oxide particles, with a particle size distribution of 1 - 3 mm and a density of 1 - 8 g / cm 3 . The release frequency is controlled by the solenoid valve, and the release amount is dynamically adjusted according to the instructions of the intelligent analysis unit. The released tracer particles move in the water flow, and their movement trajectories are traced through fluorescence detection and magnetic induction technologies to invert the flow conditions of the sediment.

[0088] The micro ADCP flow velocity meter measures the velocity and direction of the water flow, providing hydrological data for sediment deposition monitoring. It is installed at the bottom of the sensor bin, near the water flow inlet. The flow velocity meter measures the water flow velocity and direction through the acoustic Doppler principle, and the data is transmitted to the intelligent analysis unit in real time. By combining the flow velocity and direction data, the impact of the water flow on sediment deposition is analyzed to optimize the sediment deposition prediction model.

[0089] Traditional LSTM-GRU fusion methods usually adopt fixed weight allocation, which is difficult to adapt to the dynamic changes in different scenarios. The present invention introduces an adaptive feature fusion mechanism. By dynamically adjusting the weight allocation of LSTM and GRU, the model can automatically optimize the feature fusion process according to the characteristics of the input data. This adaptive mechanism can better capture the dynamic balance between long-term dependencies and short-term dynamic features, significantly improving the model's adaptability to complex underwater environments. The intelligent analysis unit uses an improved LSTM-GRU fusion algorithm to analyze the monitoring data to achieve accurate prediction and intelligent early warning of sediment deposition trends. The input data includes pressure time series data, flow velocity data, and particle loss amount. The LSTM layer extracts long-term dependency features, the GRU layer captures short-term dynamic features, and the fusion output is the sediment deposition risk index. The monitoring frequency is dynamically adjusted according to the sediment deposition risk index, and the sampling interval is shortened when the risk index is high. A three-dimensional sediment deposition heat map is generated, and the spatial distribution is fitted through B-spline surfaces to visually display the sediment deposition distribution.

[0090] The self-maintenance unit's ultrasonic anti-fouling transducer prevents fouling attachment and reduces the frequency of manual cleaning. It is installed on the outer wall of the pile body, with an operating frequency of 28 kHz ± 5% and a power density of 2 W / cm 2It is automatically activated 3 times a day, each time lasting for 10 minutes, and prevents dirt attachment through the cavitation effect of ultrasonic waves. The anti-fouling effect is evaluated by the reduction rate of the dirt attachment amount, and the reduction rate can reach more than 90%.

[0091] Data transmission includes monitoring data, siltation risk index and early warning information, and supports remote monitoring and management.

[0092] The energy management system provides stable energy supply for the device to ensure long-term stable operation. Flexible perovskite solar cells are installed on the top of the energy cabin, and wave energy generation modules are installed in the middle of the energy cabin. A hybrid energy storage system of supercapacitors and lithium iron batteries is used to store electrical energy and support 30 days of off-grid operation. The energy management system monitors the energy status in real time, dynamically adjusts the energy distribution, gives priority to using solar energy and wave energy, and uses supercapacitors and lithium iron batteries when the energy is insufficient. A sliding window mechanism is introduced. When it is detected that the siltation acceleration exceeds the threshold, the emergency mode is immediately triggered, and all pile bodies within 3 km around are started for collaborative monitoring.

[0093] The specific usage process is as follows:

[0094] Deployment phase: Site selection and installation: Calculate reasonable pile spacing according to the designed flow rate of the waterway, sediment density, average flow velocity and non-uniformity coefficient. Use pile driving equipment to vertically insert the indicating piles into the riverbed or seabed to ensure the stability of the pile bodies. Install the communication unit and solar cells to ensure their normal operation. Initial calibration: Use particle image velocimetry (PIV) technology to calibrate the initial flow field and accurately measure the velocity and direction of the water flow.

[0095] Calibrate the sensors to ensure their measurement accuracy.

[0096] Operation phase: Real-time monitoring: The thin-film pressure sensor array monitors the underwater pressure changes in real time, and the data is transmitted to the intelligent analysis unit.

[0097] The tracer particle module releases tracer particles according to instructions, and the micro ADCP current meter measures the water flow velocity and direction, and the data is synchronously transmitted to the intelligent analysis unit. Intelligent analysis and early warning: The intelligent analysis unit receives multi-modal monitoring data.

[0098] The algorithm extracts long-term dependence features and short-term dynamic features, and the fusion output is the siltation risk index. Dynamically adjust the monitoring frequency according to the siltation risk index. When the risk index is low, extend the sampling interval to reduce energy consumption.

[0099] When the risk index is high, shorten the sampling interval. If it is detected that the siltation acceleration exceeds the threshold (such as 0.05 m / s 2 ), immediately trigger the emergency mode, and start the collaborative monitoring of all pile bodies within 3 km around. Generate a three-dimensional siltation thermal map, fit the spatial distribution through B-spline surface, and intuitively display the siltation distribution, which is convenient for managers to make quick decisions.

[0100] The flexible perovskite solar cells for energy management and the wave energy generation module continuously supply power to the device, and the energy management system monitors the energy status in real time. When the solar energy or wave energy is insufficient, the hybrid energy storage system of supercapacitors and lithium iron batteries automatically switches to the power supply mode to ensure that the device can operate normally under bad weather or no-light conditions.

[0101] As Figure 5 shown, the operation process of the intelligent early warning and indication device for channel siltation provided in this embodiment is specifically as follows:

[0102] Deployment and installation stage:

[0103] First step: Calculate a reasonable pile spacing according to the designed flow rate, sediment density, average flow velocity and non-uniformity coefficient of the channel, and determine the installation position. Use pile driving equipment to vertically insert the gradient density modified PVC pile body into the riverbed or seabed to ensure the stability of the pile body, and a certain height of the top is exposed above the water surface for subsequent installation of the communication unit and solar cells.

[0104] Second step: Initial calibration Use particle image velocimetry (PIV) technology to calibrate the initial flow field and accurately measure the velocity and direction of the water flow. Calibrate the sensors such as the thin-film pressure sensor array, tracer particle module, and micro ADCP flowmeter in the sensor cabin.

[0105] Operation stage:

[0106] First step: The thin-film pressure sensor array is arranged in a spiral form on the outer wall of the pile body to monitor the underwater pressure change in real time. The axial spacing between adjacent sensors is 15 cm, and the circumferential interval is 60°. The sensor converts the pressure signal into a digital signal and transmits it to the intelligent analysis unit through the data acquisition module.

[0107] The tracer particle module controls the release frequency through the solenoid valve according to the instructions of the intelligent analysis unit, dynamically adjusts the release amount, and releases tracer particles (starch-based fluorescent particles and magnetic iron oxide particles, with a particle size distribution of 1 - 3 mm and a density of 1 - 8 g / cm 3 ). The micro ADCP flowmeter is close to the water flow inlet, and the data is transmitted to the intelligent analysis unit in real time to determine the risk index.

[0108] When the risk index is low, extend the sampling interval to reduce energy consumption; when the risk index is high, shorten the sampling interval. If it is detected that the siltation acceleration exceeds the threshold value (such as 0.05 m / s 2 ), immediately trigger the emergency mode and start the collaborative monitoring of all pile bodies within 3 km around.

[0109] Self-maintenance operation:

[0110] The ultrasonic anti-fouling transducer is automatically activated according to the set frequency (3 times a day) and duration (10 minutes each time), and prevents dirt from adhering to the outer wall of the pile through the cavitation effect of ultrasonic waves. The flexible perovskite solar cell is installed on the top of the energy module, and the wave energy power generation module is installed in the middle of the energy module, continuously powering the device.

[0111] The energy management system monitors the energy status in real time. When the solar energy or wave energy is insufficient, the hybrid energy storage system of supercapacitor and lithium iron battery automatically switches to the power supply mode to ensure that the device can work normally under bad weather or no light conditions, and supports 30-day off-grid operation.

[0112] The above-described embodiments are only preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.

Claims

1. Intelligent warning and indication device for channel siltation, characterized in that: It includes a pile body unit, a multimodal monitoring unit and an intelligent analysis unit arranged on the pile body unit; The multimodal monitoring unit is used to calculate the thickness and distribution of sediment by real-time monitoring of underwater pressure changes; The intelligent analysis unit is used to calculate the sedimentation risk index based on the thickness and distribution of sediment, so as to generate a sedimentation monitoring early warning signal.

2. The intelligent early warning and indication device for channel siltation according to claim 1, wherein: The multimodal monitoring unit includes a thin-film pressure sensor array, a controllable release tracer particle module, and a micro ADCP flow velocity meter; The thin-film pressure sensor array is arranged on the outer wall of the pile body in a spiral form, used to real-time monitor underwater pressure changes and calculate the thickness and distribution of sediment; The controllable release tracer particle module includes starch-based fluorescent particles and magnetic iron oxide particles, and the release frequency is controlled by a solenoid valve, used to invert the flow direction and velocity of sediment; The micro ADCP flow velocity meter is used to measure the velocity and direction of water flow and provide hydrological data for sedimentation monitoring.

3. The intelligent warning and indication device for channel siltation according to claim 1, wherein: The intelligent analysis unit calculates the sedimentation risk index through an improved LSTM-GRU fusion model; The input data of the improved LSTM-GRU fusion model are: pressure time series data, flow velocity data and particle loss amount. LSTM extracts long-term dependence features, and GRU captures short-term dynamic features; The output data is: sedimentation risk index.

4. The intelligent warning and indication device for channel siltation according to claim 1, characterized in that: It also includes a self-maintenance unit arranged on the pile body unit. The self-maintenance unit includes an ultrasonic anti-fouling transducer; the ultrasonic anti-fouling transducer uses cavitation to prevent dirt attachment.

5. The intelligent early warning and indication device for channel siltation according to claim 1, wherein: It also includes an energy cabin, and the energy cabin includes energy management, a power generation module, a wave energy power generation module and an energy storage system respectively connected to the energy management; The power generation module obtains electric energy through flexible perovskite solar cells; The wave energy power generation module captures wave energy through a mechanical structure and converts it into electric energy; The energy storage system realizes hybrid energy storage through supercapacitors and lithium iron batteries; The energy management realizes dynamic allocation of power supply modes by setting an energy management priority strategy.

6. The intelligent warning and indication device for channel siltation according to claim 1, wherein: The pile body shell of the pile body unit is a hollow cylindrical structure made of gradient density modified PVC. A coating is provided on the outer shell wall. The coating has a composite photocatalysis-hydrophobic bifunction. The density of the gradient density modified PVC changes continuously from the pile top to the pile bottom and is formed by a multi-layer coextrusion process.

7. The intelligent early warning and indication device for channel siltation according to claim 6, wherein: The coating is composed of nano-TiO2 (50-70wt%), fluorosilicone resin (20-30wt%) and graphene quantum dots (5-10wt%), and the thickness is 80-120μm.

8. The method for warning of channel siltation using the above intelligent warning and indication device for channel siltation is characterized in that: It includes the following steps: S1: Deploy the indicating pile: Calculate a reasonable pile spacing according to the designed flow rate, sediment density, average flow velocity and non-uniformity coefficient of the waterway, and deploy the intelligent early warning device for waterway sedimentation; S2: Calibrate the initial flow field: Calibrate the initial flow field through particle image velocimetry technology; S3: Dynamically adjust the monitoring frequency: Dynamically adjust the monitoring frequency according to the sedimentation risk index output by the intelligent analysis unit, set a sedimentation risk index threshold, and adjust the sampling interval time according to the real-time sedimentation risk index and the sedimentation risk index threshold; S4: Generate a three-dimensional sedimentation heat map: Generate a three-dimensional sedimentation heat map by B-spline surface fitting of the spatial distribution.

9. The intelligent early warning method for channel siltation according to claim 8, wherein: In the dynamic adjustment mechanism in step S3, a sliding window is adopted, and specifically, it is carried out in the following manner: Set the window length; Calculate the predicted acceleration within the window and determine whether the sedimentation acceleration exceeds the threshold; when it is detected that the sedimentation acceleration exceeds the threshold, trigger the emergency mode and start the collaborative monitoring of the surrounding pile bodies.

10. The intelligent early warning method for channel siltation according to claim 8, characterized in that: In step S3, the dynamic adjustment of the monitoring frequency is specifically carried out in the following manner: Calculate the sedimentation risk index R; When R ≤ 30%, the sampling interval is set to 60 - 120 minutes, and routine monitoring is performed; When 30% < R ≤ 60%, the interval is shortened to 30 - 60 minutes and the data verification function is activated; When 60% < R ≤ 80%, the interval is further compressed to 10 - 30 minutes, and the audible and visual alarms are triggered synchronously; If R > 80%, ultra-high frequency sampling is performed at 5 - 10 minutes, and the surrounding pile bodies are linked for collaborative operation.

Images