An intelligent collaborative operation control system for a gate opener of a water network water conservancy hub trash remover

Abstract

This invention discloses an intelligent collaborative operation control system for a water network hydraulic hub's cleaning machine and gate opener. This invention relates to the field of automated control technology for hydraulic hubs, and includes a collaborative sensing layer comprising a cleaning machine status sensing unit, a gate opener status sensing unit, and a hydraulic boundary sensing unit. The cleaning machine status sensing unit is used to acquire the real-time load torque and operating frequency of the cleaning machine. This intelligent collaborative operation control system for a water network hydraulic hub's cleaning machine and gate opener, by constructing a hydraulic-mechanical dynamic impedance mapping model, reconstructs the cleaning machine and gate opener from physically independent devices into a collaborative operating entity with impedance coupling, solving the fundamental defect in existing technologies where the two operate independently and cannot sense each other's status changes. It transforms the control relationship between the cleaning machine and gate opener from a unidirectional response to a bidirectional feedback, achieving dynamic adaptation to the flow rate through the gate and the cleaning load.

Description

Technical Field

[0001] This invention relates to the field of automated control technology for water conservancy projects, specifically to an intelligent collaborative operation control system for cleaning machines and gate openers in water network water conservancy projects. Background Technology

[0002] As key infrastructure for water resource allocation, flood control and drainage, and ecological protection, the safe and efficient operation of water conservancy hubs is of paramount importance. In the daily operation of these hubs, debris-clearing machines are responsible for intercepting and cleaning various debris at the water intake, such as branches, domestic waste, and aquatic plants, to ensure smooth water intake for core equipment such as pump units and turbines. Meanwhile, gate hoists control the opening and closing of various gates, serving as core actuators for water level regulation, flow control, and flood discharge safety. In traditional models, debris-clearing machines and gate hoists are often controlled and operated as independent systems, lacking effective coordination and linkage between them.

[0003] To improve the automation level of water conservancy projects, engineers developed an integrated monitoring system. This system utilizes programmable logic controllers (PLCs), industrial Ethernet, and remote monitoring technology to centrally monitor and control various equipment within the project. Maintenance personnel can remotely query and operate the status of equipment such as bar screeners and gate openers through the monitoring platform in the central control room, thereby achieving automated operation of water conservancy scheduling. For example, patent document CN224005428U describes an "intelligent control system for a bar screener," which monitors the status of moving parts of the bar screener by setting up vibration monitoring and video monitoring mechanisms. Based on the monitoring data, it performs predictive maintenance and real-time fault diagnosis, achieving refined intelligent control of a single bar screener.

[0004] The aforementioned existing technologies are significant in improving the operational reliability of individual devices, but their control logic still primarily focuses on the individual device and fails to address the coordinated control issue between the trash rack cleaner and the gate opener, two highly interconnected subsystems within a water conservancy hub. In actual operation, the operating status of the trash rack cleaner is closely intertwined with the opening degree, opening and closing frequency of the gate opener, and the upstream flow rate. Operating status includes whether there is blockage and the operating load. For example, when changes in gate opening cause abnormal flow velocity through the gate, if the trash rack cleaner cannot adjust its operating frequency in a timely manner, it can easily lead to a large accumulation of debris, causing overload or even damage to the trash rack cleaner. Conversely, when the trash rack cleaner malfunctions or has insufficient cleaning capacity, if the gate opener cannot respond promptly to reduce the flow through the gate or adjust its operating mode, it may exacerbate the accumulation of debris in the intake area, seriously threatening the flood discharge safety of the hub and the stable operation of the pump units.

[0005] Currently, management systems for water conservancy projects often treat the control of cleaning machines and gate openers as independent process steps, lacking an intelligent control strategy that considers them as a coordinated whole. This results in delayed system response when facing complex operating conditions, making it difficult to achieve optimal operating efficiency and proactively avoid failure risks. This poses challenges to the refined operation and maintenance of water conservancy projects, affecting the overall safety and economy of the projects. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent collaborative operation control system for the cleaning machine and gate opener of a water network and water conservancy hub, so as to solve the problems mentioned in the background art.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an intelligent collaborative operation control system for the gate opening and cleaning machine of a water network and water conservancy hub, comprising: The collaborative sensing layer includes a screen cleaner status sensing unit, a gate opener status sensing unit, and a hydraulic boundary sensing unit. The screen cleaner status sensing unit is used to acquire the real-time load torque and operating frequency of the screen cleaner. The gate opener status sensing unit is used to acquire the real-time opening degree and opening and closing speed of the gate. The hydraulic boundary sensing unit is used to acquire the upstream flow velocity distribution and the pressure difference across the gate. The collaborative decision-making layer includes an impedance matching controller, the input of which is communicatively connected to the output of the collaborative sensing layer. The collaborative execution layer includes a cleaning machine execution unit and a door opening machine execution unit. The input terminals of the cleaning machine execution unit and the door opening machine execution unit are respectively communicatively connected to the output terminal of the impedance adapter controller. The impedance matching controller is configured as follows: The real-time impedance of the cleaning machine is calculated based on the real-time load torque and the grid pressure difference. The real-time impedance of the cleaning machine is the load level of the cleaning machine under a unit grid flow rate. Based on the threshold range of the real-time impedance, a control mode is dynamically selected, including impedance matching mode, impedance adaptive mode and impedance decoupling mode. When the impedance matching mode is selected, the impedance matching controller sends a frequency adjustment command to the cleaning machine execution unit, so that the cleaning machine operating frequency follows the target impedance value, and sends an opening adjustment command to the gate opening machine execution unit, so that the gate opening and the cleaning machine operating frequency are in reverse linkage. When the impedance decoupling mode is selected, the impedance adapter controller sends an emergency limiting command to the gate opening actuator to limit the maximum opening of the gate, and sends a load reduction operation command to the cleaning machine actuator. At the same time, it sends a pulse action command to the gate opening actuator to make the gate perform an open-close-open pulse action within a preset time to form a hydraulic impact.

[0008] The real-time impedance of the cleaning machine is obtained by dividing the real-time load torque by the grid flow rate calculated from the grid pressure difference. The sensing devices of each sensing unit directly collect the corresponding physical quantities and transmit them to the impedance adapter controller for data processing.

[0009] Preferably, the collaborative sensing layer further includes a collaborative verification feedback unit, the input of which is communicatively connected to the output of the collaborative execution layer, and the output of which is communicatively connected to the input of the impedance matching controller. The collaborative verification feedback unit is used to obtain the actual load torque change value after the collaborative execution layer executes the instruction, and to feed the actual load torque change value back to the impedance adaptation controller. The impedance matching controller is further configured to dynamically correct the calculation coefficient of the real-time impedance of the cleaning machine based on the difference between the actual load torque change value and the predicted load torque change value.

[0010] The collaborative verification feedback unit collects the actual load torque and cross-gate pressure difference data after the execution command, and calculates the actual load torque change value. The correction process of the calculation coefficient is completed based on the quantitative calculation of the difference between the two.

[0011] Preferably, the impedance matching controller is further configured to: Based on the real-time changes in the gate opening, the changes in the upstream flow velocity distribution, and the changes in the pollutant concentration, a predicted value for the change in the filter cleaning machine impedance is established. The predicted impedance change value of the cleaning machine is used as a feedforward control quantity to send a frequency adjustment command to the cleaning machine execution unit in the impedance matching mode.

[0012] The real-time changes in gate opening and upstream flow velocity distribution are calculated from the time series difference of the corresponding sensing data, while the changes in pollutant concentration are obtained by back-calculation of the real-time impedance of the cleaning machine combined with the preset mapping curve.

[0013] Preferably, in the impedance matching mode, the frequency adjustment command sent by the impedance matching controller to the cleaning machine execution unit has a proportional-derivative control relationship with the difference between the real-time impedance and the target impedance value; the opening adjustment command sent by the impedance matching controller to the door opening machine execution unit has a negative correlation with the change in the operating frequency of the cleaning machine. The proportional-derivative control relationship is realized through the quantitative calculation of the proportional coefficient and the derivative coefficient, and the adjustment direction of the opening adjustment command is adjusted in the opposite direction as the operating frequency of the cleaning machine changes.

[0014] Preferably, in the impedance decoupling mode, the emergency limiting command sent by the impedance matching controller to the gate opening actuator includes an upper limit constraint value for the gate opening, which decreases linearly as the real-time impedance increases; the pulse frequency and pulse amplitude of the pulse action command are dynamically determined based on the rate of change of the real-time impedance. The upper limit constraint value for the gate opening is calculated by the amount by which the real-time impedance exceeds the threshold through a linear mapping function, and the pulse frequency and pulse amplitude are obtained by looking up a table in a preset mapping curve using the rate of change of the real-time impedance.

[0015] Preferably, the collaborative execution layer further includes a synchronization latch, which is communicatively connected to both the cleaning machine execution unit and the door opening machine execution unit. When the impedance matching controller selects the impedance decoupling mode, the synchronous latch is activated. The synchronous latch latches the load reduction operation command received by the cleaning machine execution unit and the pulse action command received by the gate opening machine execution unit on the same time base, ensuring that the cleaning machine's operating frequency adjustment and the gate's pulse action remain phase-synchronized. The synchronous latch generates a unified time base signal through a clock synchronization module, calculates the time deviation between the two commands, and performs delay compensation to achieve precise phase synchronization.

[0016] Preferably, the threshold range corresponding to the impedance adaptive mode is between the threshold range corresponding to the impedance matching mode and the threshold range corresponding to the impedance decoupling mode; In the impedance adaptive mode, the impedance adaptation controller is configured as follows: The target impedance reference value is dynamically adjusted based on the changing trend of the real-time impedance, and an opening change rate constraint command is sent to the gate opening actuator to limit the rate of change of the gate opening. The changing trend of the real-time impedance is determined by the rate of change of its time series data, and the opening change rate constraint command sets a specific constraint value based on how close the real-time impedance is to the critical threshold.

[0017] Preferably, the status sensing unit of the cleaning machine includes a torque sensor installed at the drive shaft of the cleaning machine and a speed sensor installed at the reducer of the cleaning machine; the status sensing unit of the gate opener includes a current sensor installed at the drive motor of the gate opener and a displacement sensor installed at the gate linkage; the hydraulic boundary sensing unit includes a multi-point flow meter installed in the forebay of the inlet and a differential pressure transmitter installed on the front and rear sides of the cleaning machine. The physical quantities collected by each sensing device are transmitted to the impedance matching controller after signal conditioning or numerical conversion, and all types of sensors achieve stable communication connection with the collaborative decision-making layer through an industrial fieldbus.

[0018] Preferably, the collaborative decision-making layer further includes an online model learning module, the input of which is communicatively connected to the output of the collaborative verification feedback unit, and the output of which is communicatively connected to the input of the impedance matching controller. The online learning module is used to update the calculation coefficients of the predicted load torque change value based on the difference between the actual load torque change value and the predicted load torque change value. The online learning module uses a recursive least squares method to iteratively update the calculation coefficients, and the updated coefficients are directly written into the hydraulic-mechanical dynamic impedance mapping model to replace the original coefficients.

[0019] Preferably, it further includes: a server, and field control units that are communicatively connected to the server; The field control unit includes any one of the following: the collaborative perception layer, the collaborative decision-making layer, and the collaborative execution layer; The server receives monitoring data uploaded by the collaborative sensing layer and sends control parameter update commands to the collaborative decision-making layer. The server communicates bidirectionally with each field control unit via industrial Ethernet. The field control units, with programmable logic controllers (PLCs) at their core, integrate functions and exchange data across all levels.

[0020] This invention provides an intelligent collaborative operation control system for the cleaning machine and gate opener of a water network and water conservancy hub. It has the following beneficial effects: This system constructs a hydraulic-mechanical dynamic impedance mapping model, reconstructing the trash rack cleaner and gate opener from physically independent devices into a collaborative operating entity with impedance coupling. This solves the fundamental flaw in existing technologies where the two operate independently and cannot perceive each other's state changes. The system calculates the real-time impedance of the trash rack cleaner based on real-time load torque and gate pressure difference, quantifying the load state of the trash rack cleaner into predictable and controllable physical parameters. It dynamically switches control modes according to the threshold range of the impedance: in impedance matching mode, the operating frequency of the trash rack cleaner and the gate opening are linked in reverse, enabling the trash rack cleaner to actively adjust its operating state according to changes in the hydraulic boundary; in impedance decoupling mode, a combination of emergency limiting and pulse backflush control enables the gate opener to actively adjust the hydraulic boundary to assist the trash rack cleaner in reducing its load. This transforms the control relationship between the trash rack cleaner and gate opener from a one-way response to a two-way feedback, achieving dynamic adaptation to the gate flow rate and trash rack load.

[0021] This system simultaneously establishes a collaborative verification feedback and online model learning mechanism, comparing the actual load changes after execution with the predicted load changes to dynamically correct the coupling coefficient of the hydraulic-mechanical dynamic impedance mapping model. This allows the system to adapt to changes in the characteristics of pollutants under different seasons and water quality conditions, enabling continuous optimization of the control strategy. The synchronous latch ensures phase synchronization between the deload operation of the cleaning machine and the pulse action of the gate opener in impedance decoupling mode, avoiding mechanical interference caused by misaligned equipment actions. Through a distributed architecture of server and field control units, the system can uniformly adjust the control parameters of each gate station according to the overall operating conditions of the water network and water conservancy hub, ensuring coordinated operation of the cleaning machine and gate opener across the entire hub, thus improving the operational stability and equipment safety of the water conservancy hub under complex conditions. Attached Figure Description

[0022] Figure 1 This is a system layered architecture diagram of an intelligent collaborative operation control system for a water network and water conservancy hub's cleaning machine and gate opener, as described in this invention. Figure 2 This invention provides a flowchart of the online learning and parameter optimization process for an intelligent collaborative operation control system for a water network and water conservancy hub's cleaning machine and gate opener. Figure 3 This invention relates to a state machine diagram showing the control mode switching of an intelligent collaborative operation control system for a water network and water conservancy hub's cleaning machine and gate opener. Detailed Implementation

[0023] 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.

[0024] Please see Figures 1 to 3 This invention provides a technical solution: an intelligent collaborative operation control system for a cleaning machine and gate opener in a water conservancy hub. The cleaning machine status sensing unit in its collaborative sensing layer includes a torque sensor installed on the cleaning machine's drive shaft and a speed sensor installed on the cleaning machine's reducer. The torque sensor collects the load torque value of the cleaning machine's drive shaft in real time, and the speed sensor collects the output speed value of the cleaning machine's reducer in real time. The load torque value and output speed value are used as the real-time load torque and operating frequency of the cleaning machine.

[0025] The operating frequency of the cleaning machine is obtained by converting the rotational speed collected by the speed sensor of the reducer. Specifically, the number of revolutions per minute collected by the speed sensor is converted into the number of revolutions per second, which is the real-time operating frequency of the cleaning machine. The speed sensor is coaxially connected to the input shaft of the cleaning machine reducer, and the speed data it collects is directly transmitted to the impedance matching controller to complete the conversion.

[0026] The gate operator status sensing unit includes a current sensor installed at the gate operator drive motor and a displacement sensor installed at the gate linkage. The current sensor collects the current value of the gate operator motor in real time to characterize the load status of the gate operator drive motor, and the displacement sensor collects the displacement of the gate linkage in real time to convert it into the real-time opening degree of the gate, and calculates the opening and closing speed of the gate based on the rate of change of the displacement over time.

[0027] The hydraulic boundary sensing unit includes a multi-point velocity meter installed in the inlet forebay and a differential pressure transmitter installed on both sides of the screen cleaner. The multi-point velocity meter acquires velocity distribution data at multiple measuring points in the inlet forebay, and the differential pressure transmitter acquires the differential pressure value across the screen on both sides of the screen cleaner.

[0028] The collaborative sensing layer transmits the collected real-time load torque, operating frequency, real-time gate opening degree, opening and closing speed, flow velocity distribution data, and cross-gate pressure difference value to the impedance matching controller of the collaborative decision layer through the communication network.

[0029] The real-time impedance of the screen cleaner is redefined as the load torque experienced by the screen cleaner under a unit flow rate. The quantitative calculation method is the ratio of the real-time load torque of the screen cleaner to the flow rate through the screen. The range of this parameter is calibrated based on the rated load torque of the screen cleaner and the actual operating conditions of the water conservancy project. When the value is 0 to 60% of the rated impedance value, the screen cleaner is in a light load operating state; when the value is 60% to 80% of the rated impedance value, the screen cleaner is in a normal load operating state; when the value is 80% to 100% of the rated impedance value, the screen cleaner is in a heavy load operating state; and when the value exceeds 100% of the rated impedance value, the screen cleaner is in an overload operating state. The rated impedance value is determined by the design parameters of the screen cleaner and on-site actual machine testing.

[0030] The pulse amplitude is uniformly defined as the maximum value of the gate's single opening change during the pulse action. This value matches the gate's design opening range and is 5% to 20% of the gate's maximum design opening.

[0031] The first impedance threshold and the second impedance threshold are the core impedance parameters for classifying the operating status and control mode of the cleaning machine. The physical meaning of the first impedance threshold is the critical impedance value for the cleaning machine to transition from normal load to heavy load, and the physical meaning of the second impedance threshold is the critical impedance value for the cleaning machine to transition from heavy load to overload. The first impedance threshold is taken as 80% of the rated impedance value of the cleaning machine, and the second impedance threshold is taken as 100% of the rated impedance value of the cleaning machine. If the characteristics of the sewage and the inflow conditions of the water conservancy project change, the two thresholds can be dynamically adjusted by sending control parameter update commands through the server. The adjustment is based on the actual load response characteristics of the cleaning machine on site and the equipment safety operation requirements.

[0032] The impedance matching controller calculates the real-time impedance of the cleaning machine by dividing the real-time load torque by the grid flow rate value converted from the grid pressure difference value based on the received real-time load torque and grid pressure difference value. This real-time impedance characterizes the load level of the cleaning machine under a unit grid flow rate.

[0033] The initial coupling coefficient of the hydraulic-mechanical dynamic impedance mapping model is calibrated based on historical operation data of water network and water conservancy hubs and on-site actual machine test data. First, the impedance change data of the cleaning machine under different gate opening, upstream flow velocity and debris concentration conditions are collected. Then, the initial coupling coefficients corresponding to each input quantity in the model are determined by data fitting. The random disturbance term in the model is dynamically adjusted according to the intensity of environmental interference on site. The intensity of environmental interference is determined by the degree of water flow turbulence and debris impact intensity on site. The average deviation between the measured impedance value and the fitted impedance value under different interference intensities is taken as the benchmark value of the random disturbance term. Then, it is dynamically corrected according to the real-time monitored interference intensity.

[0034] When converting the flow rate through the grid from the grid pressure difference, the flow area of ​​the screen cleaner is the effective flow area of ​​the grid body, which is directly determined according to the design parameters of the screen body. It is the total flow area of ​​the grid body minus the area occupied by the grid bars. The flow coefficient is calibrated through on-site testing. Actual flow rate data of the grid under different grid pressure differences are collected. The correspondence between the grid pressure difference and the grid flow rate is determined by combining the engineering application method of Bernoulli's equation, and the conversion of the grid flow rate is completed. The conversion process is automatically executed by the calculation module built into the impedance adapter controller. The calculation module completes the real-time conversion from grid pressure difference to grid flow rate based on the preset correspondence.

[0035] The impedance matching controller has a preset first impedance threshold and a second impedance threshold, with the first impedance threshold being lower than the second impedance threshold. When the real-time impedance is less than or equal to the first impedance threshold, the impedance matching controller selects the impedance matching mode; when the real-time impedance is greater than or equal to the second impedance threshold, the impedance matching controller selects the impedance decoupling mode.

[0036] In impedance matching mode, the impedance matching controller generates a frequency adjustment command based on the difference between the real-time impedance and the preset target impedance value using a proportional-derivative control algorithm. This command is then sent to the cleaning machine execution unit in the collaborative execution layer. The cleaning machine execution unit adjusts the cleaning machine's operating frequency according to the frequency adjustment command, ensuring that the operating frequency follows the change in the target impedance value.

[0037] Simultaneously, the impedance matching controller generates an opening adjustment command based on the change in the operating frequency of the screen cleaner, and sends the opening adjustment command to the gate opening actuator unit in the collaborative execution layer. The gate opening actuator unit adjusts the gate opening according to the opening adjustment command, and the adjustment direction of the opening adjustment command is negatively correlated with the change in the operating frequency of the screen cleaner; that is, when the operating frequency of the screen cleaner increases, the gate opening decreases, and when the operating frequency of the screen cleaner decreases, the gate opening increases.

[0038] In impedance decoupling mode, the impedance matching controller generates an emergency limiting command and sends it to the gate hoisting unit. The emergency limiting command includes an upper limit constraint value for the maximum gate opening, which decreases linearly as the real-time impedance increases. The gate hoisting unit limits the gate opening to not exceed this upper limit constraint value.

[0039] The impedance matching controller simultaneously generates a load reduction operation command and sends it to the cleaning machine execution unit. The cleaning machine execution unit then switches the cleaning machine to a high-frequency, short-cycle operation state based on the load reduction operation command.

[0040] The high-frequency, short-cycle operation of the trash rack controlled by the load reduction command is not a continuous high-frequency operation of the trash rack, but rather a short-term, high-frequency, intermittent operation. In this operation mode, the trash rack operates at a high frequency for 0.1 to 0.3 seconds and then pauses for 0.5 to 1 second, repeating this action cyclically. Simultaneously, the hydraulic impact generated by the gate pulse action of the gate opener causes the dirt attached to the trash rack's screen and rake teeth to fall off quickly, reducing the scraping resistance and clogging resistance during the operation of the trash rack, thereby reducing the continuous operating load of the trash rack and achieving the technical objective of load reduction. This operation mode is executed synchronously with the gate pulse action until the real-time impedance of the trash rack drops below the second impedance threshold.

[0041] The impedance matching controller also calculates the pulse frequency and pulse amplitude based on the real-time impedance change rate, generates a pulse action command containing the pulse frequency and pulse amplitude, and sends it to the gate opener actuator unit. The gate opener actuator unit executes an open-close-open gate pulse action sequence within a preset time window according to the pulse action command, forming a water impact on the surface of the cleaning machine.

[0042] The opening-closing-opening pulse action executed by the gate operator has a single opening and closing duration dynamically determined based on the rate of change of the real-time impedance of the screen cleaner. A higher rate of change in real-time impedance results in a shorter duration for each opening and closing action, ranging from 0.5 seconds to 2 seconds. The total execution time of the pulse action is 30 to 60 seconds, adjusted according to the actual load of the screen cleaner. If the real-time impedance of the screen cleaner rapidly drops below the second impedance threshold during the pulse action, the pulse action can be terminated early. If the real-time impedance does not decrease significantly, the total execution time can be appropriately extended using an impedance matching controller, but not exceeding 90 seconds. The gate's pulse action starts and terminates synchronously with the high-frequency, short-cycle operation of the screen cleaner, ensuring a synergistic effect between the hydraulic impact and the screen cleaner's load reduction operation.

[0043] The synchronous latch in the collaborative execution layer is activated when the impedance decoupling mode is activated. The synchronous latch latches the load reduction operation command received by the trash racking machine execution unit and the pulse action command received by the gate opening machine execution unit at the same time base, so that the operation frequency adjustment of the trash racking machine and the pulse action of the gate are kept in phase synchronization.

[0044] The collaborative verification feedback unit in the system includes a torque sensor installed on the drive shaft of the screen cleaner and differential pressure transmitters installed on the front and rear sides of the screen cleaner. The collaborative verification feedback unit is communicatively connected to the outputs of the screen cleaner execution unit and the door opening machine execution unit in the collaborative execution layer, respectively. It is used to obtain the actual load torque value at the drive shaft of the screen cleaner and the actual cross-grid pressure difference value on the front and rear sides of the screen cleaner after the collaborative execution layer executes the command. Based on the actual cross-grid pressure difference value, the actual cross-grid flow rate value is calculated. Finally, the actual load torque value is divided by the actual cross-grid flow rate value to obtain the actual load torque change value.

[0045] The collaborative verification feedback unit includes a data processing module composed of an embedded microprocessor. The specific processing logic of this module is as follows: First, it collects the actual load torque of the cleaning machine and the actual cross-grid pressure difference between the front and rear sides of the cleaning machine after the collaborative execution layer executes the command. Then, it calculates the actual cross-grid flow rate based on the cross-grid pressure difference and calculates the actual load torque change value. Subsequently, it uses the moving average method to smooth the actual load torque change value and remove abnormal data caused by field interference. Finally, it feeds back the processed actual load torque change value to the impedance adapter controller via industrial Ethernet. The transmission delay of the feedback process is controlled within 100 milliseconds. The sampling frequency of the data processing module is consistent with the sensor sampling frequency of the collaborative sensing layer.

[0046] The collaborative verification feedback unit feeds back the actual load torque change value to the impedance adapter controller via the communication network.

[0047] The impedance matching controller has an internal online learning module, which pre-stores a hydraulic-mechanical dynamic impedance mapping model. This model defines the mapping relationship between the predicted impedance change of the screen cleaner and the real-time changes in gate opening, upstream flow velocity distribution, and pollutant concentration. The mapping relationship includes multiple coupling coefficients.

[0048] The impedance matching controller calculates the predicted load torque change value based on the real-time changes in gate opening, upstream flow velocity distribution, and pollutant concentration using a hydraulic-mechanical dynamic impedance mapping model.

[0049] The online learning module compares the actual load torque change value with the predicted load torque change value, calculates the difference between the two, and uses the recursive least squares method to iteratively update multiple coupling coefficients in the hydraulic-mechanical dynamic impedance mapping model based on the difference. The updated coupling coefficients are then written into the hydraulic-mechanical dynamic impedance mapping model.

[0050] When using the recursive least squares method to update the coupling coefficients of the hydraulic-mechanical dynamic impedance mapping model, iterative updates are performed with the goal of minimizing the sum of squares of the predicted deviations. The initial value of the covariance matrix is ​​based on the identity matrix and adjusted according to the degree of deviation of the initial coupling coefficients. The larger the deviation of the initial coupling coefficients, the larger the values ​​of the elements in the initial covariance matrix. During the iteration process, each time the impedance matching controller collects real-time data from the field, it completes an iterative calculation of the coupling coefficients. First, it calculates the predicted deviation between the actual load torque change value and the predicted load torque change value. Then, it adjusts the coupling coefficients based on this deviation value until the predicted deviation value drops to a preset deviation range. The iteratively updated coupling coefficients are directly written into the hydraulic-mechanical dynamic impedance mapping model, replacing the original coefficients.

[0051] The impedance matching controller uses the updated coupling coefficient for subsequent calculations of the predicted load torque change value.

[0052] The impedance adapter controller has a built-in hydraulic-mechanical dynamic impedance mapping model, which defines the mapping relationship between the predicted value of the filter cleaner impedance change and the real-time gate opening change, the upstream flow velocity distribution change, and the sewage concentration change.

[0053] The impedance matching controller obtains the real-time gate opening degree collected by the gate operator status sensing unit through the collaborative sensing layer, and calculates the difference between the real-time gate opening degree at the current moment and the previous moment as the change in the real-time gate opening degree.

[0054] The impedance matching controller acquires upstream multi-point velocity distribution data collected by the hydraulic boundary sensing unit through the collaborative sensing layer, and calculates the difference between the average velocity of each measuring point at the current moment and the previous moment as the change in upstream velocity distribution.

[0055] The impedance matching controller acquires the real-time load torque collected by the cleaning machine's status sensing unit and the trans-grid pressure difference value collected by the hydraulic boundary sensing unit through the collaborative sensing layer. It then calculates the trans-grid flow rate based on the trans-grid pressure difference and divides the real-time load torque by the trans-grid flow rate to obtain the cleaning machine's real-time impedance. Finally, it uses the time-series data of the real-time impedance to infer the change in dirt concentration. Specifically, it compares the rate of change of real-time impedance with a preset impedance-dirt concentration mapping curve to obtain the change in dirt concentration.

[0056] The change in contaminant concentration is not directly collected by sensors, but is inferred from the real-time impedance of the cleaning machine. The impedance-contaminant concentration mapping curve is constructed as follows: an experimental platform is built on-site to simulate water environments with different contaminant concentrations. Real-time impedance data of the cleaning machine at each concentration is collected. Then, a piecewise fitting method is used to construct the corresponding curves between different impedance ranges and contaminant concentrations. The horizontal axis of the curve represents the real-time impedance of the cleaning machine, and the vertical axis represents the contaminant concentration. When inferring the change in contaminant concentration, the impedance matching controller first calculates the rate and amount of change in the real-time impedance of the cleaning machine. Then, the corresponding change in contaminant concentration is matched in the impedance-contaminant concentration mapping curve based on the change in impedance. After the inference is completed, the change in contaminant concentration is input into the hydraulic-mechanical dynamic impedance mapping model.

[0057] The impedance matching controller inputs the real-time changes in gate opening, upstream flow velocity distribution, and pollutant concentration into the hydraulic-mechanical dynamic impedance mapping model, and the model outputs the predicted value of the filter cleaner impedance change.

[0058] In impedance matching mode, the impedance matching controller uses the predicted impedance change of the screen cleaner as a feedforward control quantity. This is superimposed on the feedback control quantity generated by the difference between the real-time impedance and the preset target impedance value, and together they generate a frequency adjustment command which is sent to the screen cleaner execution unit. This allows the screen cleaner execution unit to pre-adjust the screen cleaner's operating frequency before the actual load of the screen cleaner changes, in order to cope with load fluctuations caused by changes in gate opening, upstream flow velocity, or dirt concentration.

[0059] In impedance matching mode, the impedance matching controller has a built-in proportional-derivative control algorithm module. This module acquires the real-time load torque collected by the cleaning machine status sensing unit and the cross-grid pressure difference value collected by the hydraulic boundary sensing unit. It calculates the cross-grid flow rate based on the real-time load torque and cross-grid pressure difference value, divides the real-time load torque by the cross-grid flow rate value to obtain the real-time impedance, and compares the real-time impedance with the preset target impedance value. The difference between the real-time impedance and the target impedance value is calculated as the impedance deviation value.

[0060] The proportional-derivative (PD) control algorithm module multiplies the impedance deviation value by a preset proportional coefficient to obtain the proportional control component, and simultaneously multiplies the first derivative of the impedance deviation value with respect to time by a preset differential coefficient to obtain the differential control component. The proportional and differential control components are then superimposed to generate a frequency adjustment command. This frequency adjustment command is sent to the inverter of the cleaning machine execution unit in the form of a pulse width modulation signal. The inverter adjusts the power supply frequency of the cleaning machine drive motor according to the frequency adjustment command, so that the operating frequency of the cleaning machine follows the target impedance value. The values ​​of the proportional and differential coefficients are pre-calibrated based on the rated power of the cleaning machine drive motor and the rotational inertia of the cleaning machine rake chain.

[0061] The proportional coefficient of the proportional-derivative control is determined based on the rated power of the drive motor of the cleaning machine and the rotational inertia of the rake chain. First, the initial value of the proportional coefficient is calculated based on the equipment design parameters, and then it is debugged and optimized in combination with the load response characteristics of the actual machine on site. The response speed and stability of the cleaning machine's load torque are used as evaluation indicators until the optimal proportional coefficient is determined. The derivative coefficient is calibrated based on the system's dynamic response speed requirements and load overshoot suppression requirements. A smaller initial value is taken first, and then it is gradually adjusted through field tests until there is no obvious load overshoot during the adjustment of the cleaning machine's operating frequency, and it can quickly follow the changes in the target impedance value. The calibration values ​​of the proportional coefficient and the derivative coefficient are stored in the impedance matching controller and can be dynamically updated by the server according to changes in the field working conditions.

[0062] Meanwhile, the impedance matching controller has a built-in negative correlation control module. This negative correlation control module acquires the operating frequency collected by the status sensing unit of the cleaning machine in real time, calculates the difference between the current operating frequency and the previous operating frequency as the operating frequency change, and multiplies the operating frequency change by a preset negative correlation coupling coefficient to generate an opening adjustment command.

[0063] The gate opening adjustment command is sent to the servo driver of the gate hoist execution unit in the form of an analog signal. The servo driver adjusts the output torque of the gate hoist drive motor according to the command, causing the gate opening to move inversely with the change in the operating frequency of the cleaning machine. That is, when the change in operating frequency is positive, the opening adjustment command is negative, causing the gate opening to decrease; when the change in operating frequency is negative, the opening adjustment command is positive, causing the gate opening to increase. The absolute value of the negative correlation coupling coefficient is pre-calibrated based on the rated torque of the gate hoist drive motor and the hydraulic characteristics of the gate.

[0064] In impedance decoupling mode, the impedance matching controller internally presets a first impedance threshold and a second impedance threshold, with the second impedance threshold being greater than the first impedance threshold. When the real-time impedance calculated by the impedance matching controller is greater than or equal to the second impedance threshold, the impedance matching controller triggers the impedance decoupling mode.

[0065] The impedance matching controller has a built-in limiting calculation module. This module obtains the real-time impedance value and compares it with a second impedance threshold to calculate the amount by which the real-time impedance value exceeds the second impedance threshold. The limiting calculation module has a pre-stored linear mapping function that defines a linear relationship between the excess amount and the upper limit constraint value of the gate opening; the larger the excess amount, the smaller the upper limit constraint value of the gate opening.

[0066] The limit calculation module generates an emergency limit command based on the excess amount and a linear mapping function. This emergency limit command includes an upper limit constraint value for the gate opening. This upper limit constraint value is sent as an analog signal to the servo driver of the gate opener execution unit. The servo driver writes this upper limit constraint value into its internal position loop limit register. When the current position feedback value of the gate opener drive motor reaches the upper limit constraint value, the servo driver stops outputting torque, preventing the gate opening from increasing further.

[0067] The impedance matching controller has a built-in pulse parameter calculation module. This module acquires the real-time load torque and operating frequency collected by the cleaning machine's status sensing unit and calculates the real-time impedance change rate. Specifically, it divides the difference between the current real-time impedance and the real-time impedance before a preset time window by the duration of the preset time window to obtain the real-time impedance change rate.

[0068] The pulse parameter calculation module pre-stores pulse frequency mapping curves and pulse amplitude mapping curves. The pulse frequency mapping curve defines the correspondence between the real-time impedance change rate and the pulse frequency, while the pulse amplitude mapping curve defines the correspondence between the real-time impedance change rate and the pulse amplitude.

[0069] The pulse parameter calculation module obtains the pulse frequency from a table in the pulse frequency mapping curve and the pulse amplitude from a table in the pulse amplitude mapping curve based on the real-time impedance change rate, and generates a pulse action command containing the pulse frequency and pulse amplitude. This pulse action command is sent to the servo driver of the gate opener execution unit in the form of a pulse sequence signal. The servo driver controls the gate opener drive motor to execute an open-close-open gate pulse action sequence within a preset time window based on the pulse frequency and pulse amplitude, where the pulse amplitude corresponds to the change in gate opening degree and the pulse frequency corresponds to the repetition rate of the gate action.

[0070] In the impedance decoupling mode, the linear mapping function of the upper limit of the gate opening is constructed based on the actual correspondence between the amount by which the real-time impedance exceeds the second impedance threshold and the upper limit of the gate opening. First, the optimal upper limit of the gate opening is collected under different excess amounts, and then a mapping relationship is formed through linear fitting. The larger the excess amount of the real-time impedance, the smaller the upper limit constraint value of the gate opening. The calibration of this function is based on the hydraulic impact test of the actual machine on site and the equipment safety operation requirements. The mapping curves of pulse frequency and pulse amplitude are drawn through field tests. The horizontal axis of the pulse frequency mapping curve is the rate of change of the real-time impedance of the cleaning machine, and the vertical axis is the frequency of the gate pulse action. The horizontal axis of the pulse amplitude mapping curve is the rate of change of the real-time impedance of the cleaning machine, and the vertical axis is the amplitude of the gate pulse action. During the drawing process, the reduction effect of the cleaning machine load and the cleaning effect of the hydraulic impact are used as evaluation indicators to determine the key feature points of the curve. Then, a complete mapping curve is formed through fitting. The impedance matching controller matches the corresponding pulse frequency and pulse amplitude in the curve according to the real-time impedance change rate and generates pulse action commands.

[0071] The synchronous latch is built using an XC7A35T FPGA chip. Its internal clock synchronization module uses a PLL phase-locked loop circuit to synchronize the global clock signal, converting the global clock signal issued by the collaborative decision-making layer into a clock signal that matches the cleaning machine execution unit and the door opening machine execution unit. The instruction cache module uses a dual-port RAM to temporarily store load reduction operation instructions and pulse action instructions. The read and write speed of the dual-port RAM matches the transmission speed of industrial Ethernet, ensuring that the instructions are stored without delay. The phase alignment module is composed of hardware logic circuits, which can calculate the reception time deviation of two instructions in real time and generate a delay waiting instruction based on the deviation value. The synchronization tolerance threshold of the synchronous latch is 50 milliseconds. That is, when the time deviation between two instructions is less than 50 milliseconds, the instructions are considered to be synchronized and no delay compensation is required. When the time deviation is greater than 50 milliseconds, delay compensation is activated. The response deviation range is ±10%. That is, when the actual action response of the cleaning machine or door opener deviates from the instruction requirement within ±10%, the action is considered to be normal. When it exceeds this range, the phase alignment module recalculates the delay compensation value and performs dynamic phase correction. This threshold and range can be fine-tuned according to the action response characteristics of the field equipment.

[0072] The collaborative execution layer includes a synchronous latch, which is built using a field-programmable gate array chip and is connected to the frequency converter of the cleaning machine execution unit and the servo driver of the door opening machine execution unit via an industrial Ethernet bus.

[0073] The synchronous latch internally includes a clock synchronization module, an instruction buffer module, and a phase alignment module. The clock synchronization module receives the global clock signal from the impedance matching controller of the cooperative decision layer and generates a unified time base signal based on the global clock signal.

[0074] When the impedance matching controller selects the impedance decoupling mode, it sends an activation signal to the synchronous latch. Upon receiving the activation signal, the synchronous latch activates the instruction cache module. The instruction cache module receives the load reduction operation instruction sent by the impedance matching controller to the cleaning machine execution unit and the pulse action instruction sent to the door opening machine execution unit. It temporarily stores the load reduction operation instruction and the pulse action instruction in its internal memory and records the timestamp of the receipt of the two instructions.

[0075] The phase alignment module reads the load reduction operation command and pulse action command stored in the command cache module and calculates the time deviation between the two commands based on a unified time base signal. If the time deviation is greater than a preset synchronization tolerance threshold, the phase alignment module uses the timestamp of the later received command as a reference and sends a delay waiting command to the execution unit corresponding to the earlier received command, so that the earlier received command is delayed until it is time-aligned with the later received command, thereby locking the load reduction operation command and the pulse action command on the same time base.

[0076] After completing the instruction time alignment, the synchronous latch simultaneously releases the latched instructions to the frequency converter of the cleaning machine execution unit and the servo driver of the gate opening machine execution unit. The frequency converter adjusts the power supply frequency of the cleaning machine drive motor according to the load reduction operation instruction, switching the cleaning machine's operating frequency to a high-frequency, short-cycle operating state. The servo driver controls the gate opening machine drive motor to execute a gate pulse action sequence according to the pulse action instruction. Furthermore, because the load reduction operation instruction and the pulse action instruction are released at the same time reference, the adjustment of the cleaning machine's operating frequency and the start time of the gate pulse action remain phase-synchronized.

[0077] After the command is released, the synchronous latch continuously monitors the feedback signals of the cleaning machine and the door opening machine. If the response delay of either machine exceeds the preset response deviation range, the phase alignment module recalculates the new delay waiting command and performs dynamic phase compensation for subsequent commands.

[0078] It should be further explained that the impedance matching controller internally presets a first impedance threshold and a second impedance threshold, with the first impedance threshold being lower than the second impedance threshold. The impedance matching controller compares the real-time impedance of the cleaning machine with the first and second impedance thresholds. When the real-time impedance is greater than the first impedance threshold but less than the second impedance threshold, the impedance matching controller determines that the threshold range of the real-time impedance is the threshold range corresponding to the impedance adaptive mode. This range lies between the threshold range corresponding to the impedance matching mode and the threshold range corresponding to the impedance decoupling mode.

[0079] In impedance adaptive mode, the impedance adapter controller acquires the real-time load torque collected by the cleaning machine status sensing unit and the cross-grid pressure difference value collected by the hydraulic boundary sensing unit, calculates the time series data of real-time impedance, and calculates the change trend of real-time impedance based on the time series data of real-time impedance. Specifically, the difference between the real-time impedance at the current moment and the real-time impedance before the preset time window is divided by the duration of the preset time window to obtain the real-time impedance change rate, and then the sign of the real-time impedance change rate is used to determine whether the real-time impedance is in an upward or downward trend.

[0080] The impedance matching controller has a built-in adaptive reference value adjustment module. This module acquires the real-time impedance and the real-time impedance change rate. When the real-time impedance is on an upward trend, the adaptive reference value adjustment module sets the target impedance reference value to the sum of the real-time impedance and a pre-stored safety margin. This safety margin is dynamically adjusted based on the difference between the rated load torque of the cleaning machine and the real-time impedance; the smaller the difference, the smaller the safety margin.

[0081] When the real-time impedance is decreasing, the adaptive reference adjustment module sets the target impedance reference value to the sum of the real-time impedance and the pre-stored recovery coefficient. The recovery coefficient increases linearly according to the absolute value of the rate of change of the real-time impedance, so that the target impedance reference value is dynamically adjusted to follow the trend of the real-time impedance, avoiding a sudden change in control commands due to an excessive difference between the target impedance reference value and the real-time impedance.

[0082] Meanwhile, the impedance matching controller incorporates a rate constraint module. This module acquires the real-time gate opening degree collected by the gate operator's status sensing unit, calculates the difference between the current gate opening degree and the previous gate opening degree, divides it by the sampling time interval to obtain the actual gate opening degree change rate, and compares this actual gate opening degree change rate with a pre-stored upper limit threshold for the opening degree change rate. The upper limit threshold for the opening degree change rate is dynamically determined based on the threshold range position of the real-time impedance of the cleaning machine; the closer the real-time impedance is to the second impedance threshold, the smaller the upper limit threshold for the opening degree change rate.

[0083] When the actual gate opening rate of change exceeds the upper limit threshold, the rate constraint module generates an opening rate of change constraint command. This command, containing the upper limit threshold as a constraint value, is sent as an analog signal to the servo driver of the gate opener execution unit. The servo driver writes this constraint value into its internal speed loop limiting register. When the actual speed of the gate opener drive motor, converted to the gate opening rate of change, reaches the constraint value, the servo driver limits the output torque to ensure that the gate opening rate of change does not exceed the constraint value.

[0084] In the impedance adaptive mode, the safety margin is dynamically adjusted based on the difference between the rated load torque of the screen cleaner and the real-time impedance. The larger the difference, the larger the safety margin, which ranges from 5% to 15% of the screen cleaner's rated impedance. The recovery coefficient is linearly adjusted based on the absolute value of the real-time impedance change rate. The larger the absolute value of the change rate, the larger the recovery coefficient, which ranges from 0.1 to 0.5. The upper limit threshold for the opening change rate is determined based on the closeness between the real-time impedance and the second impedance threshold. The closer the real-time impedance is to the second impedance threshold, the smaller the upper limit threshold for the opening change rate, which ranges from 1% / second to 5% / second of the gate's maximum design opening. All the above parameter adjustments are based on the actual operating conditions of the water network and the calibration of the equipment's rated parameters. Reasonable values ​​for each parameter under different impedance states are determined through multiple field tests on the actual machine. The calibrated parameters are stored in the adaptive reference value adjustment module and the rate constraint module of the impedance adaptation controller.

[0085] The status sensing unit of the cleaning machine includes a torque sensor installed on the drive shaft of the cleaning machine and a speed sensor installed on the reducer of the cleaning machine.

[0086] The torque sensor is a strain gauge type, with its elastic body coaxially mounted to the drive shaft of the cleaning machine. Resistance strain gauges attached to the elastic body form a Wheatstone bridge. When the drive bearing is subjected to torque, the elastic body undergoes torsional deformation, causing a change in the strain gauge resistance. The bridge output is a voltage signal linearly related to the torque value. This voltage signal is amplified and filtered by a signal conditioning circuit before being input to the impedance matching controller in the collaborative decision-making layer, serving as the real-time load torque of the cleaning machine.

[0087] The speed sensor uses an incremental encoder, whose code disk is coaxially connected to the input shaft of the cleaning machine's reducer. When the code disk rotates, the photoelectric detection element outputs a pulse sequence proportional to the speed. This pulse sequence is converted into a speed value by a counter and then input to the impedance matching controller as the operating frequency of the cleaning machine.

[0088] The gate operator status sensing unit includes a current sensor installed at the gate operator drive motor and a displacement sensor installed at the gate linkage.

[0089] The current sensor employs a Hall effect current sensor, whose through-hole structure is connected to the three-phase power supply line of the gate hoist drive motor. It senses the magnetic field generated by the motor current and outputs an analog signal that is linearly related to the current value. This analog signal is input to the impedance matching controller to characterize the load state of the gate hoist drive motor and assist in determining the gate's operating resistance.

[0090] The displacement sensor is a magnetostrictive displacement sensor, with its waveguide parallel to the gate linkage and a magnetic ring fixed to the gate linkage. When the gate linkage moves, the magnetic ring moves along the waveguide, the sensor detects the position of the magnetic ring and outputs an analog signal proportional to the displacement. This analog signal is input to an impedance matching controller, which calculates the real-time opening degree of the gate based on the displacement and calculates the gate's opening and closing speed based on the rate of change of the displacement over time.

[0091] The hydraulic boundary sensing unit includes a multi-point flow meter installed in the forebay at the inlet and a differential pressure transmitter installed on the front and rear sides of the cleaning machine.

[0092] The multi-point flow meter uses an ultrasonic Doppler flow meter, with its transducer array arranged at multiple preset measuring points in the inlet forebay. Each transducer emits ultrasonic waves into the water and receives the reflected echoes. The flow velocity value at the corresponding measuring point is calculated based on the Doppler frequency shift. The flow velocity values ​​from multiple measuring points are then aggregated and input into an impedance matching controller to form upstream flow velocity distribution data.

[0093] The differential pressure transmitter is a capacitive differential pressure transmitter. Its high-pressure side tap is located on the inlet side of the screen cleaner, and its low-pressure side tap is located on the outlet side of the screen cleaner. The transmitter detects the pressure difference between the two taps and outputs an analog signal that is linearly related to the pressure difference. This analog signal is input to an impedance matching controller, which calculates the flow rate through the screen based on the pressure difference and the flow area of ​​the screen cleaner.

[0094] All of the aforementioned sensors and transmitters are connected to the impedance matching controller of the collaborative decision-making layer via an industrial fieldbus. The impedance matching controller collects data from each channel in parallel at a sampling period.

[0095] The impedance matching controller uses a parallel sampling period of 50 milliseconds for each channel's data. The sampling frequency of each sensor is matched with this sampling period. Specifically, the sampling frequency of the torque sensor and speed sensor of the cleaning machine status sensing unit, and the current sensor and displacement sensor of the gate opening machine status sensing unit are all set to 20Hz. The sampling frequency of the multi-point flow meter and differential pressure transmitter of the hydraulic boundary sensing unit is also synchronously set to 20Hz. The sampling data of each sensor is synchronously transmitted to the acquisition module of the impedance matching controller through the industrial fieldbus. The acquisition module synchronously receives and stores the data of each channel to ensure that there is no time difference in the data acquisition of each sensor and to avoid affecting the control accuracy due to data asynchrony.

[0096] The collaborative decision-making layer includes an online model learning module, which uses an embedded microprocessor as its carrier and is integrated with the impedance matching controller in the same control cabinet. It communicates with the impedance matching controller through an internal data bus.

[0097] The online learning module and the impedance matching controller communicate via the ModbusTCP industrial Ethernet protocol. This protocol's transmission rate matches the on-site industrial Ethernet transmission rate, ensuring data transmission without loss or delay. The online learning module updates its coupling coefficients every 100 milliseconds, calculating the difference between the actual and predicted load torque changes and performing an iterative update of the coupling coefficients. The updated coupling coefficients are transmitted to the impedance matching controller in real time. If the on-site water flow and load data fluctuate significantly, the update cycle can be adjusted to 50 milliseconds via a server to improve the model's real-time adaptation capability.

[0098] The input of the online learning module is connected to the output of the collaborative verification feedback unit via an industrial Ethernet communication connection, and is used to receive the actual load torque change value fed back by the collaborative verification feedback unit.

[0099] The online learning module stores a hydraulic-mechanical dynamic impedance mapping model, which is defined as follows: the predicted value of the filter cleaner impedance change is equal to the real-time gate opening change multiplied by the first coupling coefficient, plus the upstream flow velocity distribution change multiplied by the second coupling coefficient, plus the pollutant concentration change multiplied by the third coupling coefficient, plus the random disturbance term.

[0100] The online learning module of the model simultaneously obtains the predicted load torque change value calculated by the impedance adapter controller based on the hydraulic-mechanical dynamic impedance mapping model from the impedance adapter controller.

[0101] The online learning module of the model subtracts the actual load torque change value fed back by the collaborative verification feedback unit from the predicted load torque change value to obtain the prediction deviation value.

[0102] The online learning module of the model incorporates a recursive least squares algorithm unit. This algorithm unit takes minimizing the sum of squares of the prediction deviations as the objective function, uses the first coupling coefficient, the second coupling coefficient, and the third coupling coefficient as parameters to be identified, and uses the real-time changes in gate opening, upstream flow velocity distribution, and pollutant concentration as input vectors to construct the information matrix and gain matrix of the recursive least squares algorithm.

[0103] The recursive least squares algorithm unit obtains the input vector and prediction deviation value at the current time in each sampling period, updates the covariance matrix and parameter vector according to the recursive formula, and writes the updated first coupling coefficient, second coupling coefficient and third coupling coefficient into the hydraulic-mechanical dynamic impedance mapping model to replace the original coupling coefficient.

[0104] After updating the coupling coefficients, the online model learning module sends a model parameter update confirmation signal to the impedance matching controller. When subsequently calculating and predicting load torque changes, the impedance matching controller reads the updated hydraulic-mechanical dynamic impedance mapping model from the online model learning module and uses the updated first, second, and third coupling coefficients for prediction calculations.

[0105] The server is an industrial-grade server deployed in the central control room of the water conservancy hub. The server communicates with multiple field control units located at various gate stations of the water network hub through industrial Ethernet switches, forming a distributed control system architecture.

[0106] Each field control unit comprises a collaborative sensing layer, a collaborative decision-making layer, and a collaborative execution layer. The field control unit is centered around a programmable logic controller (PLC). Sensors in the collaborative sensing layer are connected to the PLC via analog and digital input modules. The impedance matching controller in the collaborative decision-making layer is implemented as a function block within the PLC. The cleaning machine and door opening machine execution units in the collaborative execution layer are connected to the PLC via analog and digital output modules.

[0107] The server has a built-in data acquisition and monitoring control system platform, which sends data request frames to each field control unit via industrial Ethernet in a polling manner. After receiving the data request frame, each field control unit encapsulates the real-time load torque, operating frequency, real-time gate opening degree, opening and closing speed, upstream flow velocity distribution data, and cross-gate pressure difference value collected by the collaborative sensing layer into a response frame, and transmits it back to the server via industrial Ethernet.

[0108] The server stores the received monitoring data into a historical database and displays it visually on the human-computer interaction interface in the form of trend curves, dashboards, and alarm lists.

[0109] The server also has a built-in model parameter optimization module. This module periodically extracts monitoring data of each field control unit under actual operating conditions from the historical database, as well as the predicted load torque change value calculated by the impedance adapter controller in each field control unit and the actual load torque change value fed back by the collaborative verification feedback unit. It then uses the batch least squares method to perform offline optimization of the coupling coefficient of the hydraulic-mechanical dynamic impedance mapping model to obtain the optimized coupling coefficient value.

[0110] The server encapsulates the optimized coupling coefficient values ​​into control parameter update instructions and sends them to the corresponding field control unit via industrial Ethernet. Upon receiving the control parameter update instructions, the field control unit uses the online model learning module to write the optimized coupling coefficients into the hydraulic-mechanical dynamic impedance mapping model, replacing the original coupling coefficients.

[0111] The online iterative updates of the model's online learning module and the offline optimization updates of the server employ a hierarchical synchronization mechanism. The online iterative updates of the model's online learning module are real-time dynamic updates to adapt to real-time changes in field conditions; the offline optimization updates of the server are periodic batch updates used for overall optimization of coupling coefficients. Once the server distributes the offline-optimized coupling coefficients to the field control unit, the model's online learning module immediately pauses its real-time online iterative updates, uses the offline-optimized coupling coefficients as new initial coupling coefficients, writes them into the hydraulic-mechanical dynamic impedance mapping model, and then resumes online iterative updates, continuing real-time iterations based on these initial coefficients. This avoids parameter conflicts arising from the two update methods and ensures the consistency of model parameters.

[0112] When the server performs offline optimization of the coupling coefficients using the batch least squares method, it operates on a monthly cycle. At the beginning of each month, historical operating data from each field control unit for the previous month is extracted, including data such as changes in gate opening, upstream flow velocity distribution, pollutant concentration, and screen cleaner impedance. Then, with the goal of minimizing overall prediction deviation, batch fitting optimization is performed on the coupling coefficients of the hydraulic-mechanical dynamic impedance mapping model. Abnormal historical data is removed during the optimization process to ensure the accuracy of the optimization results. After optimization, the server encapsulates the new coupling coefficients into control parameter update instructions and sends them to each field control unit, completing the offline update of the model parameters.

[0113] The server is also equipped with an operating condition identification module. This module identifies the overall operating condition category of the current water network water conservancy hub based on the upstream flow velocity distribution data, real-time gate opening, and cross-grid pressure difference value uploaded by each field control unit received by the server. Based on the identified operating condition category, it queries the pre-stored operating condition-parameter mapping table for the target impedance value, first impedance threshold, and second impedance threshold that match the operating condition category.

[0114] The target impedance value, the first impedance threshold, and the second impedance threshold obtained from the query are encapsulated into a control parameter update command and sent to the corresponding field control unit. The impedance adapter controller of the field control unit then updates the internal control parameters.

[0115] The operating condition identification module classifies the operating conditions into four categories based on the combined characteristics of upstream flow velocity distribution, cross-grid pressure difference, and real-time gate opening: low flow and low load, medium flow and medium load, high flow and high load, and sudden high load. The low flow and low load operating condition is the operating state with upstream flow velocity less than 0.5 m / s, cross-grid pressure difference less than 5 kPa, and gate opening less than 30%. The medium flow and medium load operating condition is the operating state with upstream flow velocity of 0.5 m / s to 1.0 m / s, cross-grid pressure difference of 5 kPa to 10 kPa, and gate opening of 30% to 60%. The high flow and high load operating condition is the operating state with upstream flow velocity of 1.0 m / s to 1.5 m / s, cross-grid pressure difference of 10 kPa to 15 kPa, and gate opening of 60% to 90%. The sudden high load operating condition is the operating state with upstream flow velocity greater than 1.5 m / s, cross-grid pressure difference greater than 15 kPa, or gate opening greater than 90%. The operating condition-parameter mapping table is constructed based on the optimal operating data under different operating conditions. The table contains the target impedance value, first impedance threshold, and second impedance threshold corresponding to each operating condition. The construction of this table is completed through field tests under different operating conditions. The optimal coordinated operating parameters of the cleaning machine and the gate opener under each operating condition are collected and filled into the mapping table. The operating condition identification module collects upstream flow velocity, cross-grid pressure difference, and gate opening data in real time. After matching the corresponding operating condition, it extracts the relevant control parameters from the mapping table and sends them to the impedance matching controller.

[0116] This system constructs a hydraulic-mechanical dynamic impedance mapping model, reconstructing the trash rack cleaner and gate opener from physically independent devices into a collaborative operating entity with impedance coupling. This solves the fundamental flaw in existing technologies where the two operate independently and cannot perceive each other's state changes. The system calculates the real-time impedance of the trash rack cleaner based on real-time load torque and gate pressure difference, quantifying the load state of the trash rack cleaner into predictable and controllable physical parameters. It dynamically switches control modes according to the threshold range of the impedance: in impedance matching mode, the operating frequency of the trash rack cleaner and the gate opening are linked in reverse, enabling the trash rack cleaner to actively adjust its operating state according to changes in the hydraulic boundary; in impedance decoupling mode, a combination of emergency limiting and pulse backflush control enables the gate opener to actively adjust the hydraulic boundary to assist the trash rack cleaner in reducing its load. This transforms the control relationship between the trash rack cleaner and gate opener from a one-way response to a two-way feedback, achieving dynamic adaptation to the gate flow rate and trash rack load.

[0117] This system also establishes a collaborative verification feedback and online model learning mechanism, comparing the actual load changes after execution with the predicted load changes to dynamically correct the coupling coefficient of the hydraulic-mechanical dynamic impedance mapping model. This allows the system to adapt to changes in the characteristics of pollutants under different seasons and water quality conditions, achieving continuous optimization of the control strategy. The synchronous latch ensures phase synchronization between the deload operation of the screen cleaner and the pulse action of the gate opener in impedance decoupling mode, avoiding mechanical interference caused by misalignment of equipment actions.

[0118] Through the distributed architecture of servers and field control units, the system can uniformly adjust the control parameters of each gate station according to the overall operating conditions of the water network and water conservancy hub, so that the coordinated operation of the cleaning machine and the gate opening machine can be kept consistent throughout the entire hub, thereby improving the operational stability and equipment safety of the water conservancy hub under complex operating conditions.

[0119] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0120] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart collaborative operation control system for a water network and water conservancy hub's cleaning machine and gate opener, characterized in that, include: The collaborative sensing layer includes a screen cleaner status sensing unit, a gate opener status sensing unit, and a hydraulic boundary sensing unit. The screen cleaner status sensing unit is used to acquire the real-time load torque and operating frequency of the screen cleaner. The gate opener status sensing unit is used to acquire the real-time opening degree and opening and closing speed of the gate. The hydraulic boundary sensing unit is used to acquire the upstream flow velocity distribution and the pressure difference across the gate. The collaborative decision-making layer includes an impedance matching controller, the input of which is communicatively connected to the output of the collaborative sensing layer. The collaborative execution layer includes a cleaning machine execution unit and a door opening machine execution unit. The input terminals of the cleaning machine execution unit and the door opening machine execution unit are respectively communicatively connected to the output terminal of the impedance adapter controller. The impedance matching controller is configured as follows: The real-time impedance of the cleaning machine is calculated based on the real-time load torque and the grid pressure difference. The real-time impedance of the cleaning machine is the load level of the cleaning machine under a unit grid flow rate. Based on the threshold range of the real-time impedance, a control mode is dynamically selected, including impedance matching mode, impedance adaptive mode and impedance decoupling mode. When the impedance matching mode is selected, the impedance matching controller sends a frequency adjustment command to the cleaning machine execution unit, so that the cleaning machine operating frequency follows the target impedance value, and sends an opening adjustment command to the gate opening machine execution unit, so that the gate opening and the cleaning machine operating frequency are in reverse linkage. When the impedance decoupling mode is selected, the impedance adapter controller sends an emergency limiting command to the gate opening actuator to limit the maximum opening of the gate, and sends a load reduction operation command to the cleaning machine actuator. At the same time, it sends a pulse action command to the gate opening actuator to make the gate perform an open-close-open pulse action within a preset time to form a hydraulic impact.

2. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 1, characterized in that: The collaborative sensing layer also includes a collaborative verification feedback unit, the input of which is communicatively connected to the output of the collaborative execution layer, and the output of which is communicatively connected to the input of the impedance matching controller. The collaborative verification feedback unit is used to obtain the actual load torque change value after the collaborative execution layer executes the instruction, and to feed the actual load torque change value back to the impedance adaptation controller. The impedance matching controller is further configured to dynamically correct the calculation coefficient of the real-time impedance of the cleaning machine based on the difference between the actual load torque change value and the predicted load torque change value.

3. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 2, characterized in that: The impedance matching controller is further configured to: Based on the real-time changes in the gate opening, the changes in the upstream flow velocity distribution, and the changes in the pollutant concentration, a predicted value for the change in the filter cleaning machine impedance is established. The predicted impedance change value of the cleaning machine is used as a feedforward control quantity to send a frequency adjustment command to the cleaning machine execution unit in the impedance matching mode.

4. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 3, characterized in that: In the impedance matching mode, the frequency adjustment command sent by the impedance matching controller to the cleaning machine execution unit has a proportional-derivative control relationship with the difference between the real-time impedance and the target impedance value; the opening adjustment command sent by the impedance matching controller to the door opening machine execution unit has a negative correlation with the change in the operating frequency of the cleaning machine.

5. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 4, characterized in that: In the impedance decoupling mode, the emergency limiting command sent by the impedance adapter controller to the gate opening actuator includes an upper limit constraint value for the gate opening, which decreases linearly as the real-time impedance increases; the pulse frequency and pulse amplitude of the pulse action command are dynamically determined according to the rate of change of the real-time impedance.

6. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 5, characterized in that: The collaborative execution layer also includes a synchronization latch, which is communicatively connected to the cleaning machine execution unit and the door opening machine execution unit, respectively. When the impedance matching controller selects the impedance decoupling mode, the synchronous latch is activated. The synchronous latch is used to latch the load reduction operation command received by the cleaning machine execution unit and the pulse action command received by the gate opening machine execution unit at the same time base, so that the operation frequency adjustment of the cleaning machine and the gate pulse action are kept in phase synchronization.

7. The intelligent collaborative operation control system for the gate opener of the trash removal machine of a water network hydraulic complex according to claim 6, characterized in that: The threshold range corresponding to the impedance adaptive mode is between the threshold range corresponding to the impedance matching mode and the threshold range corresponding to the impedance decoupling mode. In the impedance adaptive mode, the impedance adaptation controller is configured as follows: The target impedance reference value is dynamically adjusted according to the real-time impedance change trend, and the gate opening machine execution unit is sent an opening change rate constraint command to limit the rate of change of the gate opening.

8. The intelligent collaborative operation control system for a water network and water conservancy hub's cleaning machine and gate opener according to claim 7, characterized in that: The status sensing unit of the cleaning machine includes a torque sensor installed on the drive shaft of the cleaning machine and a speed sensor installed on the reducer of the cleaning machine; the status sensing unit of the gate opener includes a current sensor installed on the drive motor of the gate opener and a displacement sensor installed on the gate connecting rod; the hydraulic boundary sensing unit includes a multi-point flow meter installed in the forebay of the inlet and a differential pressure transmitter installed on the front and rear sides of the cleaning machine.

9. The intelligent collaborative operation control system for the gate opener of the trash removal machine of a water network hydraulic complex according to claim 8, characterized in that: The collaborative decision-making layer also includes an online model learning module, the input of which is communicatively connected to the output of the collaborative verification feedback unit, and the output of which is communicatively connected to the input of the impedance matching controller. The online learning module of the model is used to update the calculation coefficient of the predicted value of the filter cleaning machine impedance change based on the difference between the actual load torque change value and the predicted load torque change value.

10. The intelligent collaborative operation control system for the cleaning machine and gate opener of a water network water conservancy hub according to claim 9, characterized in that: Also includes: A server, and field control units that are respectively connected in communication with the server; The field control unit includes the collaborative perception layer, collaborative decision-making layer, and collaborative execution layer as described in any one of claims 1 to 9; The server is used to receive monitoring data uploaded by the collaborative sensing layer and to issue control parameter update instructions to the collaborative decision-making layer.

Images