A new modern water conservancy intelligent water supply terminal and matching control system

Abstract

This invention relates to the field of modern water conservancy and intelligent water supply control technology, specifically a novel intelligent water supply terminal and supporting control system for modern water conservancy. It includes data acquisition, state identification, and adaptive control modules. The system filters and reconstructs the phase space of data from fluid network edge nodes, and identifies disturbance characteristics by calculating the deviation between the high-dimensional trajectory and the normal attractor. Its core is to dynamically adjust the inertia, damping, and stiffness parameters of the virtual dynamic model based on these characteristics, enabling the actuator to exhibit matched impedance characteristics. This invention overcomes the limitations of fixed parameter control, achieving adaptive switching between high-damping shock resistance and high-sensitivity pressure replenishment states based on operating conditions, thus achieving coordinated optimization of water supply quality and physical safety.

Description

Technical Field

[0001] This invention relates to the field of modern water conservancy and intelligent water supply control technology, specifically to a new type of intelligent water supply terminal and supporting control system for modern water conservancy. Background Technology

[0002] With the continuous expansion of the scale of modern urban water supply networks, the complexity and volatility of hydraulic conditions have increased significantly. This complexity poses multiple challenges to the stable operation of terminal water supply equipment, especially in terms of balancing pressure control accuracy and physical safety protection. Currently, traditional water supply terminals mostly use PID control logic with preset fixed parameters. When the pipeline network experiences high-frequency pressure surges or abrupt changes in flow demand, the control parameters cannot be dynamically adjusted, making it difficult for the system to balance pressure stabilization accuracy with the rigidity and flexibility of water hammer protection. On the one hand, if the system responds too quickly, it is very easy to induce secondary water hammer effects during valve operation, threatening pipeline safety. On the other hand, if the control is too gentle, it cannot compensate for instantaneous pressure drops in time, leading to a decline in water quality. In addition, traditional monitoring methods are unable to effectively identify deep fluid dynamic characteristics from noisy observation data, and cannot accurately distinguish between normal water use disturbances and abnormal operating conditions such as pipeline leaks, which may lead to increased losses due to blind adjustments when an accident occurs. Therefore, how to accurately identify the characteristics of fluid disturbances in complex pipeline network fluctuation environments and dynamically optimize control strategies to balance system response speed and disturbance resistance stability has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0003] To solve the above-mentioned technical problems, the present invention provides a novel intelligent water supply terminal and supporting control system for modern water conservancy. Specifically, the technical solution of the present invention is as follows: A novel intelligent water supply terminal control system for modern water conservancy includes a data acquisition module, a status identification module, and an adaptive control module. The data acquisition module is configured to acquire the operating status data of the fluid network edge nodes in real time, and to preprocess the operating status data using a digital filter to filter out environmental white noise. The state identification module is configured to perform phase space reconstruction processing on the operating state data to obtain a high-dimensional phase space trajectory, and calculate the deviation between the high-dimensional phase space trajectory and the preset normal operating condition attractor, and identify the current fluid network disturbance characteristics based on the deviation. The adaptive control module is configured to dynamically adjust the model parameters of a preset virtual dynamics model based on the disturbance characteristics of the fluid network; wherein the model parameters include virtual inertial parameters, virtual damping parameters, and virtual stiffness parameters. The adaptive control module is further configured to calculate a target control quantity using the adjusted virtual dynamics model, and generate a drive command based on the target control quantity to control the actuator's action, so that the actuator exhibits impedance characteristics that match the fluid network disturbance characteristics.

[0004] Preferably, the operating status data includes instantaneous pressure data and instantaneous flow rate data of the fluid network edge nodes; The data acquisition module is also used to estimate the equivalent head disturbance of the fluid network based on the instantaneous pressure data and the instantaneous flow rate data using an extended Kalman filter algorithm.

[0005] Preferably, the virtual dynamics model is a second-order differential equation model describing the dynamic relationship between the actuator opening degree and the pressure deviation; The adaptive control module is specifically used to: in response to the fluid network disturbance characteristics indicating a high-frequency pressure impact, increase the virtual inertia parameter and the virtual damping parameter to make the actuator exhibit high damping characteristics to absorb impact energy; and in response to the fluid network disturbance characteristics indicating a step flow demand, decrease the virtual damping parameter and increase the virtual stiffness parameter to make the actuator exhibit fast response characteristics to compensate for pressure loss.

[0006] Preferably, the state identification module is specifically configured to: map the operating state data to a three-dimensional phase space to construct the phase point trajectory of the current system state; and calculate the Euclidean distance between the phase point trajectory and a preset normal operating condition attractor. If the Euclidean distance is less than a preset safety threshold, the fluid network disturbance characteristic is determined to be a normal water use disturbance; if the Euclidean distance is greater than or equal to the preset safety threshold, the fluid network disturbance characteristic is determined to be an abnormal operating condition disturbance.

[0007] Preferably, the adaptive control module further includes an anomaly processing unit, which is configured to: continuously monitor the evolution trend of the phase point trajectory in response to the fluid network disturbance characteristics being determined as the abnormal operating condition disturbance; If the evolution trend shows a divergent trajectory and fails to return to the neighborhood of the normal operating condition attractor within a preset convergence time window, the system state is determined to be a pipeline leakage state; in response to determining the pipeline leakage state, the operating mode of the control system is switched to pressure holding and current limiting mode, and an alarm signal is generated.

[0008] Preferably, the adaptive control module introduces boundary constraints based on Lyapunov stability theory when calculating the target control quantity; The boundary constraint condition is configured as the upper limit of the virtual stiffness parameter, which is set to be less than the critical stiffness value corresponding to the system resonance frequency, so as to prevent system resonance caused by excessive parameter adjustment.

[0009] Preferably, the actuator includes a regulating valve or a variable frequency water pump; The adaptive control module is also used to establish a nonlinear mapping table between the target control quantity and the physical motion quantity of the actuator, and to convert the target control quantity into a valve opening command or a motor frequency command according to the nonlinear mapping table.

[0010] Preferably, the intelligent water supply terminal control system of the new modern water conservancy further includes a data interaction module, which is used for: The operational status data, the fluid network disturbance characteristics, and the adjustment records of the model parameters are encapsulated into a structured log; the structured log is timestamped and stored locally with encryption; in response to a query command received from the upper-level management system, the structured log is uploaded for global optimization of the pipeline network model.

[0011] A smart water supply terminal device includes a fluid transmission pipeline, a sensor group and an actuator installed on the fluid transmission pipeline, and a supporting control system for the new type of modern water conservancy smart water supply terminal. The intelligent water supply terminal and supporting control system of the new modern water conservancy are electrically connected to the sensor group and the actuator respectively, and are used to control the action of the actuator according to the data collected by the sensor group.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. By introducing a virtual dynamics model and a dynamic impedance adjustment mechanism, the system solves the problem that traditional fixed parameter control cannot balance response speed and system protection. When faced with high-frequency water hammer impact, the system can automatically adjust to a high-damping state to absorb energy and prevent secondary water hammer from damaging the pipeline. When faced with instantaneous water demand, it can quickly switch to a high-sensitivity state to compensate for pressure, thus achieving coordinated optimization of water supply quality and physical safety. 2. The system utilizes phase space reconstruction technology to map one-dimensional pressure and flow data to a high-dimensional space, enabling the extraction of the dynamic essential characteristics of the fluid network from noisy data. Compared with traditional methods that rely solely on threshold judgment, this system can effectively distinguish between normal water usage fluctuations and abnormal operating conditions such as pipeline leaks, significantly reducing the risk of blind adjustments due to misjudgment and improving the system's operating condition perception accuracy. 3. The system is equipped with a dedicated anomaly handling unit, which can determine the leakage status in real time based on the evolution trend of the phase point trajectory. Once a pipe burst or serious leakage is confirmed, the system can automatically switch from constant pressure mode to pressure holding and flow limiting mode and issue an alarm, breaking the vicious cycle of the traditional system where the more it leaks, the more pressure is increased. This effectively reduces water waste and lowers the risk of secondary disasters caused by large-scale leakage. 4. The control algorithm incorporates Lyapunov stability boundary constraints, which mathematically set a safety red line for parameter adjustment to prevent pipeline resonance induced by over-adjustment. At the same time, by establishing a nonlinear mapping table for the actuator, the influence of the inherent physical characteristics of valves or pumps on control accuracy is eliminated, ensuring that the system maintains stable, reliable and consistent regulation performance across the entire range. Attached Figure Description

[0013] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0015] Example 1: Please see Figure 1 A new type of intelligent water supply terminal supporting control system for modern water conservancy, including a data acquisition module, a status identification module and an adaptive control module; The data acquisition module is configured to acquire the operating status data of the fluid network edge nodes in real time, and use a digital filter to preprocess the operating status data to filter out environmental white noise. The state identification module is configured to perform phase space reconstruction processing on the operating state data to obtain a high-dimensional phase space trajectory, and calculate the deviation between the high-dimensional phase space trajectory and the preset normal operating condition attractor, and identify the current fluid network disturbance characteristics based on the deviation. The adaptive control module is configured to dynamically adjust the model parameters of a preset virtual dynamics model based on the characteristics of fluid network disturbances; the model parameters include virtual inertial parameters, virtual damping parameters, and virtual stiffness parameters. The adaptive control module is also configured to calculate the target control quantity using the adjusted virtual dynamics model and generate drive commands based on the target control quantity to control the actuator's action, so that the actuator exhibits impedance characteristics that match the characteristics of the fluid network disturbance.

[0016] This embodiment details the core architecture and basic algorithm flow of the system, aiming to solve the paradox of existing water supply terminals being unable to simultaneously ensure pressure stabilization accuracy and water hammer protection when facing complex pipeline fluctuations. The data acquisition module collects fluid pressure and flow velocity in the pipeline through a high-frequency pressure sensor and an electromagnetic flowmeter, respectively. To eliminate electromagnetic interference and high-frequency noise caused by fluid turbulence commonly found in industrial sites, this module uses a digital filter to preprocess the operating status data. The digital filter, derived from the algorithm processing unit based on discrete-time systems, is physically a signal cleaning mechanism, preferably using an infinite impulse response Butterworth low-pass filter with a cutoff frequency set to 5-10 times the main dynamic frequency of the system. The state identification module extracts the deep dynamic characteristics of the fluid network from the time-domain signal. This module is configured to perform phase space reconstruction processing on the preprocessed operating status data. Phase space reconstruction: derived from the application of Takens' embedding theorem, its physical meaning is a technique for mapping a one-dimensional time series to a high-dimensional geometric space, by selecting an appropriate time delay. and embedding dimension The module constructs a high-dimensional phase space trajectory. It calculates the deviation between this trajectory and a preset normal operating condition attractor. The normal operating condition attractor is derived from the historical data training set during the system calibration phase; physically, it represents the set of convergent trajectories formed by the system's operating state data in phase space under stable water supply and no abnormal disturbances. The deviation is derived from Euclidean geometry calculations; physically, it represents the mathematical distance between the current phase point and the geometric center or manifold surface of the normal operating condition attractor. Based on this, the adaptive control module identifies the current fluid network disturbance characteristics based on this deviation and dynamically adjusts the parameters of the virtual dynamics model. Specifically, the adaptive control module uses a preset mapping function... To achieve dynamic adjustment of parameters; using virtual inertial parameters. For example, the dynamic adjustment formula for the inertia parameter is as follows: in, The reference inertia coefficient; For mapping gain; The normalized phase space deviation is calculated as follows: ,in, The preset safety threshold; The exponential sensitivity factor, with a recommended value range of 0.5 to 2.0, is used to adjust the sensitivity of the inertial parameters to the normalized deviation. By introducing normalization, the overflow of the exponential function calculation is prevented due to the original Euclidean distance value being too large. Through this exponential mapping relationship, the abstract disturbance characteristics are quantified into specific dynamic parameters, ensuring that the control system can quickly improve its anti-disturbance inertia when the deviation increases. The mathematical expression of this model is as follows: in, The physical meaning is the virtual moment of inertia parameter, which characterizes the system's ability to resist changes in opening degree. Its unit is set as [missing unit]. ; The physical meaning is the virtual damping coefficient, which characterizes the ability to dissipate oscillatory energy, and its unit is 1. ; Physically, virtual torsional stiffness characterizes a system's ability to resist deviations and return to a steady state; its unit is 1. ; The physical meaning is virtual angular displacement, corresponding to the valve opening angle of the actuator, and the unit is 1. ; The physical meaning is virtual angular velocity, and the unit is . ; The physical meaning is virtual angular acceleration, and the unit is . ; The physical meaning is virtual load torque, derived from pressure deviation, and the unit is... ; This embodiment introduces a virtual dynamics model that includes virtual inertia, damping, and stiffness, making the water supply terminal no longer a simple actuator, but an intelligent node with impedance characteristics. When faced with different pipeline disturbances, the system can automatically adjust its own stiffness like a biological muscle, thereby ensuring stable water supply while effectively suppressing pipeline resonance and water hammer effects, achieving dual optimization of energy transmission efficiency and system robustness.

[0017] Example 2: Operational status data includes instantaneous pressure and instantaneous flow data of the fluid network edge nodes; The data acquisition module is also used to estimate the equivalent head disturbance of the fluid network based on instantaneous pressure and instantaneous flow data using an extended Kalman filter algorithm.

[0018] This embodiment refines the data acquisition and estimation process, aiming to reconstruct the hidden state of the system that cannot be directly measured from noisy observation data; the data acquisition module acquires the instantaneous pressure data of the fluid network edge nodes. and instantaneous flow data This module integrates model-based observer functionality, specifically used to estimate the equivalent head disturbance of the fluid network based on instantaneous pressure and flow data using an extended Kalman filter algorithm. In this embodiment, the equivalent head disturbance is assumed to be... For slowly varying parameters, a random walk model is used to model them, i.e. Equivalent head disturbance: This is derived from the output of the EKF state observer. Physically, it represents the equivalent pressure head change at the current node, calculated by converting all unknown external disturbances in the pipe network into equivalent pressure head changes. The unit is meters of water column. In this process, the state vector of EKF is defined as The equation of state is constructed based on a discretized model of the fluid transport line. To ensure the convergence of the observer and the consistency of physical dimensions, a set of first-order approximate difference equations based on the conservation of momentum and mass is specifically adopted: Among them, matrix elements To linearize the first-order Taylor series of the Darcy-Weisbach friction term; The process noise vector follows a normal distribution. ; The term, in its physical sense, refers to the equivalent head disturbance. This is converted into incremental changes in flow rate within the fluid network, thereby establishing the hidden state and observable flow rate. The coupling prediction relationship between them enables the Kalman gain to correct the disturbance in real time through the pressure residual; For the speed of sound in fluid, The cross-sectional area of ​​the pipe. For fluid density, Darcy's coefficient of friction For pipe diameter, For discrete sampling time, The equivalent characteristic length of the currently monitored pipe section is measured during the equipment installation and commissioning phase. The specific method is as follows: a small water hammer wave is generated by rapidly closing the end valve, and the round-trip time of the pressure wave is measured. Based on the formula The calculations are performed to ensure that the model parameters match the actual physical characteristics of the pipeline. Let gravitational acceleration be the constant; the formula is introduced. and The dimensional matching relationship between the pressure change rate and the flow rate change rate was corrected; in order to construct a complete recursive filtering closed loop, this embodiment also specifically defines the observation equation. , where the observation vector Observation matrix This matrix clarifies that only pressure and flow rate are directly measurable physical quantities, while the equivalent head disturbance is... It is in a hidden state; Meanwhile, the noise covariance matrix during algorithm initialization To characterize the uncertainty of the simplified fluid model, where, Set as a diagonal matrix The values ​​of each element are based on the statistical variance of the model linearization error at the system sampling frequency to ensure the dynamic tracking performance of the filter. To achieve the recursive iteration of the extended Kalman filter, this embodiment also specifically defines the Jacobian matrix of the state transition function. Used for predicting and updating the covariance matrix: The coefficient of the intermediate term becomes Because of the friction term For traffic Find the partial derivative This eliminates the constant 2 in the denominator of the kinetic energy term in the original equation; it also corrects the partial derivative coefficients with respect to the head disturbance term to ensure dimensional consistency; and it sets the measurement noise covariance matrix according to the sensor's datasheet. The system continuously calculates the residual between the predicted and measured values, and corrects the state vector based on the Kalman gain, thereby approximating the actual disturbance in real time. This embodiment constructs a high-precision observation mechanism in a complex pipeline noise environment. Since pressure and flow sensors inevitably have measurement noise, directly using raw data for control will cause actuator jitter. By introducing the EKF algorithm, this system can not only filter out measurement noise, but also observe pipeline disturbances that cannot be directly measured, thereby realizing feedforward compensation control and significantly improving the system's anti-interference ability to pipeline pressure fluctuations.

[0019] Example 3: The virtual dynamics model is a second-order differential equation model that describes the dynamic relationship between the actuator opening degree and the pressure deviation; The adaptive control module is specifically used to: respond to the fluid network disturbance characteristics indicating high-frequency pressure shock, increase the virtual inertia parameter and virtual damping parameter to make the actuator exhibit high damping characteristics to absorb the shock energy; and respond to the fluid network disturbance characteristics indicating step flow demand, decrease the virtual damping parameter and increase the virtual stiffness parameter to make the actuator exhibit fast response characteristics to compensate for pressure loss.

[0020] This embodiment details the structure and parameter adjustment strategy of the virtual dynamics model, aiming to achieve a control logic that combines rigidity and flexibility based on the characteristics of the working conditions. The virtual dynamics model is concretized as a second-order differential equation model describing the relationship between the actuator opening and pressure deviation. During the instantiation of this model, the virtual displacement in the formula of Embodiment 1... At the physical execution level, it is mapped to the normalized opening of the actuator. The corresponding virtual speed Corresponding opening change rate virtual load torque Mapped as pressure deviation The torque conversion coefficient, i.e., the product of the effective area and the equivalent force arm, establishes the correspondence between physical and virtual quantities. The adaptive control module executes specific adjustment logic. To avoid control discontinuity caused by parameter abrupt changes, a smooth gain scheduling algorithm based on the Sigmoid function is adopted: smooth gain scheduling introduces a first-order lag element within the parameter update cycle, combined with the nonlinear transition of the Sigmoid function, to ensure the virtual parameters... , and The second derivative is continuous during operating condition switching; The pressure change rate is indicated by the characteristics of the fluid network disturbance. Exceeding the preset impact threshold The module will increase the virtual inertial parameters. and virtual damping parameters The specific dynamic mapping formula is as follows: in, The preset inertial boundary value, To adjust the sensitivity coefficient; its dimensions are The typical value range is to This is used to control the steepness of the Sigmoid function near the impact threshold; at this point, the virtual system exhibits heavy and viscous characteristics, even with external pressure deviations. It's very large, the opening degree of the implementing agency Nor will there be any drastic changes; Response to fluid network disturbance characteristics indicates flow demand When a step increase occurs and the rate of pressure change is in the low-frequency range, the module will reduce the virtual damping parameter. And increase the virtual stiffness parameter The specific dynamic mapping formula is as follows: It should be further explained that the target value of the step flow demand in the above formula... It is not set out of thin air, but is estimated in real time by the adaptive control module based on the current pressure drop rate; the specific estimation formula is as follows: in, This represents the current instantaneous flow rate. Set the pressure for the system. This represents the instantaneous pressure extreme value. The equivalent flow resistance coefficient of the pre-calibrated pipe network system; this formula strictly follows the square law relationship of pressure-flow rate in the fluid network, i.e. The theoretical flow rate compensation value required to maintain pressure stability is calculated by back-calculating the square root of the pressure deviation. By introducing the absolute value operator This ensures that regardless of whether traffic demand surges or drops sharply, the system can reduce virtual damping. To improve the responsiveness of the actuators; in, As the reference damping, The damping attenuation coefficient is... As the reference stiffness, This is the gain coefficient. For flow normalization factor, Set as the system's rated flow rate to eliminate the influence of dimensions; The target value of the step flow demand is calculated based on the characteristics of fluid network disturbance; at this time, the virtual system exhibits light and rigid characteristics, and the actuator can quickly respond to pressure deviation and rapidly change the opening to compensate for the flow. This embodiment solves the performance trade-off problem caused by fixed parameters in traditional PID control by using a dynamic parameter scheduling strategy. When facing high-frequency water hammer impact, the system adopts a static braking strategy to absorb the impact energy by utilizing the fluid elasticity of the pipeline network itself, avoiding secondary water hammer caused by rapid valve closure or opening. When users urgently need water, the system can provide an extremely fast dynamic response, limiting the pressure drop to a minimum and ensuring water supply quality.

[0021] Example 4: The state identification module is specifically configured as follows: mapping the operating state data to a three-dimensional phase space to construct the phase point trajectory of the current system state; calculating the Euclidean distance between the phase point trajectory and the preset normal operating condition attractor; If the Euclidean distance is less than the preset safety threshold, the fluid network disturbance characteristic is determined to be normal water use disturbance; if the Euclidean distance is greater than or equal to the preset safety threshold, the fluid network disturbance characteristic is determined to be abnormal operating condition disturbance.

[0022] This embodiment details the specific algorithm flow of the state identification module, aiming to distinguish fluid events with similar physical appearances but different mechanisms using topological geometric features. Before performing phase space mapping, the state identification module performs Z-score normalization on the preprocessed operating state data, transforming instantaneous pressure data... and instantaneous flow data The data is mapped to a dimensionless, unified interval to eliminate the interference of physical dimension differences on the distance calculation weights; the module performs a phase space mapping step to map the operating state data to a three-dimensional phase space, constructing the phase point trajectory vector of the current system state. ,in, For optimal delay time; optimal delay time The first minimum point of the mutual information function of the time series is determined to ensure the independence of the reconstructed coordinates; embedding dimension. Determined by the spurious neighbor method, the typical value range is [value range missing]. to To fully develop the geometric structure of the dynamic system; The module performs the distance calculation step to calculate the phase point trajectory. With the preset normal operating condition attractor Euclidean distance between The calculation formula is as follows: Based on this, the system executes the decision logic: in response to The fluid network disturbance characteristics are determined to be normal water usage disturbance; in response to The system determines that the fluid network disturbance characteristics are abnormal operating condition disturbances; furthermore, when it is determined to be an abnormal operating condition disturbance, the system calculates the absolute value of the rate of change of the pressure data. ;like Greater than the preset impact threshold If so, the disturbance characteristic is determined to be a high-frequency pressure shock; if Less than or equal to Furthermore, if the phase point trajectory exhibits a quasi-static translation, then the disturbance characteristic is determined to be a step-like flow demand. Safety threshold: Physically, it refers to the tolerance value of the maximum deviation distance under normal operating conditions; safety threshold The acquisition method is as follows: during the system calibration phase, obtain the data under normal water supply conditions. Set of phase point trajectories Calculate its geometric center And based on statistical principles, the following settings are established: in, The preset reliability redundancy coefficient ranges from 1.1 to 1.5. This calculation method based on historical envelope gives the threshold a clear statistical and physical meaning, eliminating the blindness of artificial setting. This embodiment overcomes the limitations of traditional threshold determination by utilizing phase space topological features. Traditional threshold determination methods cannot distinguish between low pressure caused by high flow rate water and low pressure caused by pipeline leakage, while this solution can keenly capture the essential changes in fluid dynamic characteristics when leakage occurs, thereby achieving high-precision operating condition identification without adding additional sensors and reducing the false alarm rate.

[0023] Example 5: The adaptive control module also includes an anomaly handling unit, which is configured to continuously monitor the evolution trend of the phase point trajectory in response to the fluid network disturbance characteristics being determined as an abnormal operating condition disturbance. If the evolution trend shows a divergent trajectory and fails to return to the neighborhood of the attractor under normal operating conditions within the preset convergence time window, the system state is determined to be a pipeline leakage state; in response to the determination of the pipeline leakage state, the operating mode of the control system is switched to the pressure holding and flow limiting mode, and an alarm signal is generated.

[0024] This embodiment, based on anomaly identification, adds handling logic for pipeline leaks, aiming to build a self-healing and loss-mitigation mechanism for the system. The anomaly handling unit in the adaptive control module, responding to the fluid network disturbance characteristics being determined as an abnormal operating condition disturbance, initiates a trend monitoring program to continuously monitor the evolution trend of the phase point trajectory. The system executes the leak determination logic, responding to the evolution trend manifested as trajectory divergence, and within a preset convergence time window... If the system fails to return to the neighborhood of the attractor under normal operating conditions, the system state is determined to be a pipeline leak state; trajectory divergence: physically, this means that the system's energy dissipation prevents it from returning to its original dynamic trajectory; the system executes an emergency response, in response to the determination of the pipeline leak state, switching the control system's operating mode from constant pressure water supply mode to pressure holding and flow limiting mode, and generating an alarm signal; pressure holding and flow limiting mode: physically, this means an operating strategy that forcibly reduces the target pressure setpoint and limits the maximum valve opening; This embodiment effectively solves the problem of traditional water supply systems forcibly increasing valve opening to maintain pressure when a pipe bursts, leading to a vicious cycle of leakage and increased pressure. By automatically switching to pressure-maintaining and flow-limiting mode, the system can significantly reduce water waste and the risk of secondary disasters under accident conditions, demonstrating the safety of the control strategy.

[0025] Example 6: When calculating the target control quantity, the adaptive control module introduces boundary constraints based on Lyapunov stability theory. The boundary constraints are configured as the upper limit of the virtual stiffness parameter, which is set to be less than the critical stiffness value corresponding to the system resonance frequency, in order to prevent system resonance caused by excessive parameter adjustment.

[0026] This embodiment introduces stability theory to ensure the safety of parameter adjustment, aiming to set strict mathematical boundaries for the adaptive algorithm; when calculating the target control quantity, the adaptive control module introduces boundary constraints based on Lyapunov stability theory; the system constructs a Lyapunov candidate function containing a virtual energy term. To ensure the asymptotic stability of the closed-loop system, the range of parameter variation must be limited to avoid exciting unmodeled high-frequency dynamics. From an energy perspective, if the virtual stiffness is set too large, causing the system's natural frequency to approach the resonant frequency of the pipe network, the disturbance energy input from the external fluid network will exceed the energy dissipated by damping, leading to an increase in the derivative of the Lyapunov function. A positive value would compromise stability; the system adjusts the virtual damping in real time. Forced This holds true globally, thus satisfying the asymptotic stability requirement of the Lyapunov stability criterion. It is worth noting that the above stability proof is based on the time-scale separation assumption of singular perturbation theory; that is, the update frequency of the adaptive adjustment of the system's set parameters is much lower than the characteristic frequency of the actuator's dynamic response, making... and This allows for the neglect of additional energy terms arising from time-varying parameters in Lyapunov derivative calculations, ensuring... and Effectiveness at the engineering scale; The boundary constraint condition is specifically configured as a virtual torsional stiffness parameter. upper limit of the value The calculation is based on the following formula: To ensure the stability of the system during time-varying parameter processes, the boundary constraints are configured as virtual torsional stiffness parameters. upper limit of the value : in, The physical meaning is the critical torsional stiffness value, and the unit is 1. ; The current virtual moment of inertia, in units of ; This is the lowest-order resonant frequency of the pipeline system, measured in Hz. This frequency must avoid the fundamental frequency of water hammer waves. It is usually set to To ensure a safety margin; this formula dynamically correlates the critical stiffness in the Lyapunov stability constraint with the real-time changing virtual inertia. This embodiment defines the safety red line for parameter adjustment using Lyapunov theory. Although adaptive control is flexible, if the stiffness parameter is increased without constraints, it is easy to excite high-frequency unmodeled dynamics of the fluid network, leading to system instability and oscillation. This scheme mathematically guarantees the asymptotic stability of the closed-loop control system and ensures the reliability of the control strategy under extreme conditions.

[0027] Example 7: The actuators include regulating valves or variable frequency water pumps; The adaptive control module is also used to establish a nonlinear mapping table between the target control quantity and the physical motion quantity of the actuator, and to convert the target control quantity into a valve opening command or a motor frequency command based on the nonlinear mapping table.

[0028] This embodiment describes the conversion process from algorithm output to physical action, aiming to eliminate the impact of nonlinearity of physical actuators on control accuracy. The actuator is defined as a physical component including a regulating valve or a variable frequency pump. The adaptive control module establishes a nonlinear mapping table between the target control quantity and the physical action quantity of the actuator. The nonlinear mapping table is physically a set of data used to compensate for the inherent nonlinear characteristics of the actuator. Based on this mapping table, the module converts the linear target control quantity into a valve opening command or a motor frequency command. The nonlinear mapping table is generated by recording the actual opening response data of the actuator under different target control quantity inputs during the offline calibration phase of the system and fitting it using the least squares method. This embodiment linearizes the actual hydraulic components with strong nonlinear characteristics through a nonlinear mapping process. Virtual dynamic models typically assume that the controlled object is linear, while actual physical objects often have complex nonlinear relationships. Through this mapping mechanism, the front-end virtual model can more accurately predict and control system behavior, thereby improving the control accuracy and consistency across the entire range.

[0029] Example 8: The new modern water conservancy intelligent water supply terminal supporting control system also includes a data interaction module, which is used for: The system encapsulates operational status data, fluid network disturbance characteristics, and model parameter adjustment records into structured logs; timestamps and encrypts the structured logs for local storage; and uploads the structured logs for global optimization of the pipeline network model in response to query commands received from the upper-level management system.

[0030] This embodiment involves the features of data assetization and collaborative optimization, aiming to explore the derived value of edge-side control data; the data interaction module performs log encapsulation operations, encapsulating the running status data, identified fluid network disturbance features, and model parameter adjustment records into structured logs; the module timestamps the structured logs at the microsecond level and uses the AES-256 algorithm for local encrypted storage; in response to receiving a query command from the upper-level management system, the module uploads the structured logs; This embodiment realizes the evolution from single-point intelligence to swarm intelligence; traditional terminal data is often discrete and unlabeled, while this solution records the complete process of parameter adjustment. These data are extremely valuable samples for the upper-level pipeline network model and can be used for the calibration and optimization of the global hydraulic model, realizing the asset management of data.

[0031] Example 9: A smart water supply terminal device includes a fluid transmission pipeline, a sensor group and an actuator installed on the fluid transmission pipeline, and a new type of modern water conservancy smart water supply terminal supporting control system. The intelligent water supply terminal control system of the new modern water conservancy is electrically connected to the sensor group and the actuator, respectively, and is used to control the action of the actuator based on the data collected by the sensor group.

[0032] This embodiment describes the overall physical equipment including the above-mentioned control system, aiming to provide a directly deployable engineering solution. The intelligent water supply terminal equipment specifically includes the following hardware components: a fluid transmission pipeline as a physical fluid channel; a sensor group, installed on the fluid transmission pipeline, including at least a high-frequency pressure transmitter and an electromagnetic flowmeter, for collecting physical signals; an actuator, installed on the fluid transmission pipeline, for changing flow resistance; and a novel modern water conservancy intelligent water supply terminal supporting control system, as an embedded control unit, electrically connected to the sensor group and the actuator respectively. During operation, the data collected by the sensor group is input to the control system, which internally runs virtual dynamics algorithms and phase space identification algorithms to calculate the optimal control commands and drive the actuator to move. This embodiment, as an integrated intelligent hardware device, has extremely high engineering application value; the terminal can directly replace the existing old mechanical or simple PID control terminals, and can realize the intelligent upgrade of the water supply network without large-scale transformation of the pipeline network, thus reducing the implementation threshold and cost of smart water management construction.

[0033] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A novel intelligent water supply terminal control system for modern water conservancy, characterized in that, It includes a data acquisition module, a status identification module, and an adaptive control module; The data acquisition module is configured to acquire the operating status data of the fluid network edge nodes in real time, and to preprocess the operating status data using a digital filter to filter out environmental white noise. The state identification module is configured to perform phase space reconstruction processing on the operating state data to obtain a high-dimensional phase space trajectory, and calculate the deviation between the high-dimensional phase space trajectory and the preset normal operating condition attractor, and identify the current fluid network disturbance characteristics based on the deviation. The adaptive control module is configured to dynamically adjust the model parameters of a preset virtual dynamics model based on the disturbance characteristics of the fluid network; wherein the model parameters include virtual inertial parameters, virtual damping parameters, and virtual stiffness parameters. The adaptive control module is further configured to calculate a target control quantity using the adjusted virtual dynamics model, and generate a drive command based on the target control quantity to control the actuator's action, so that the actuator exhibits impedance characteristics that match the fluid network disturbance characteristics.

2. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, The operational status data includes instantaneous pressure data and instantaneous flow rate data of the fluid network edge nodes; The data acquisition module is also used to estimate the equivalent head disturbance of the fluid network based on the instantaneous pressure data and the instantaneous flow rate data using an extended Kalman filter algorithm.

3. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, The virtual dynamics model is a second-order differential equation model that describes the dynamic relationship between the actuator opening degree and the pressure deviation. The adaptive control module is specifically used to: in response to the fluid network disturbance characteristics indicating a high-frequency pressure impact, increase the virtual inertia parameter and the virtual damping parameter to make the actuator exhibit high damping characteristics to absorb impact energy; and in response to the fluid network disturbance characteristics indicating a step flow demand, decrease the virtual damping parameter and increase the virtual stiffness parameter to make the actuator exhibit fast response characteristics to compensate for pressure loss.

4. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, The state identification module is specifically configured to: map the operating state data to a three-dimensional phase space to construct the phase point trajectory of the current system state; Calculate the Euclidean distance between the phase point trajectory and the preset normal operating condition attractor; If the Euclidean distance is less than a preset safety threshold, the fluid network disturbance characteristic is determined to be a normal water use disturbance; if the Euclidean distance is greater than or equal to the preset safety threshold, the fluid network disturbance characteristic is determined to be an abnormal operating condition disturbance.

5. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 4, characterized in that, The adaptive control module further includes an anomaly handling unit, which is configured to continuously monitor the evolution trend of the phase point trajectory in response to the fluid network disturbance characteristics being determined as the abnormal operating condition disturbance. If the evolution trend shows a divergent trajectory and fails to return to the neighborhood of the normal operating condition attractor within a preset convergence time window, the system state is determined to be a pipeline leakage state; in response to determining the pipeline leakage state, the operating mode of the control system is switched to pressure holding and current limiting mode, and an alarm signal is generated.

6. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, When calculating the target control quantity, the adaptive control module introduces boundary constraints based on Lyapunov stability theory. The boundary constraint condition is configured as the upper limit of the virtual stiffness parameter, which is set to be less than the critical stiffness value corresponding to the system resonance frequency, so as to prevent system resonance caused by excessive parameter adjustment.

7. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, The actuator includes a regulating valve or a variable frequency water pump; The adaptive control module is also used to establish a nonlinear mapping table between the target control quantity and the physical motion quantity of the actuator, and to convert the target control quantity into a valve opening command or a motor frequency command according to the nonlinear mapping table.

8. The novel intelligent water supply terminal control system for modern water conservancy as described in claim 1, characterized in that, The new type of intelligent water supply terminal control system for modern water conservancy also includes a data interaction module, which is used for: The operational status data, the fluid network disturbance characteristics, and the adjustment records of the model parameters are encapsulated into a structured log; the structured log is timestamped and stored locally with encryption; in response to a query command received from the upper-level management system, the structured log is uploaded for global optimization of the pipeline network model.

9. A smart water supply terminal device, characterized in that, It includes a fluid transmission pipeline, a sensor group and an actuator installed on the fluid transmission pipeline, and a novel intelligent water supply terminal supporting control system for modern water conservancy as described in any one of claims 1-8; The intelligent water supply terminal and supporting control system of the new modern water conservancy are electrically connected to the sensor group and the actuator respectively, and are used to control the action of the actuator according to the data collected by the sensor group.

Images